Institutional Logics in Action 9781781909218, 9781781909201

Citation preview

INSTITUTIONAL LOGICS IN ACTION, PART B

RESEARCH IN THE SOCIOLOGY OF ORGANIZATIONS Series Editor: Michael Lounsbury Recent Volumes: Volume 21:

Postmodernism and Management: Pros, Cons and the Alternative

Volume 22:

Legitimacy Processes in Organizations

Volume 23:

Transformation in Cultural Industries

Volume 24:

Professional Service Firms

Volume 25:

The Sociology of Entrepreneurship

Volume 26:

Studying Difference between Organizations: Comparative Approaches to Organizational Research

Volume 27:

Institutions and Ideology

Volume 28:

Stanford’s Organization Theory Renaissance, 1970–2000

Volume 29:

Technology and Organization: Essays in Honour of Joan Woodward

Volume 30A:

Markets on Trial: The Economic Sociology of the US Financial Crisis: Part A

Volume 30B:

Markets on Trial: The Economic Sociology of the US Financial Crisis: Part B

Volume 31:

Categories in Markets: Origins and Evolution

Volume 32:

Philosophy and Organization Theory

Volume 33:

Communities and Organizations

Volume 34:

Rethinking Power in Organizations, Institutions, and Markets

Volume 35:

Reinventing hierarchy and bureaucracy – from the bureau to network organisations

Volume 36:

The Garbage Can Model of Organizational Choice – Looking Forward at Forty

Volume 37:

Managing ‘Human Resources’ by Exploiting and Exploring People’s Potentials

Volume 38:

Configurational Theory and Methods in Organizational Research

Volume 39A:

Institutional Logics in Action, Part A

RESEARCH IN THE SOCIOLOGY OF ORGANIZATIONS VOLUME 39B

INSTITUTIONAL LOGICS IN ACTION, PART B EDITED BY

MICHAEL LOUNSBURY Alberta School of Business, University of Alberta, Edmonton, Canada

EVA BOXENBAUM Centre for Management Studies, ´ Ecole des Mines ParisTech, Paris, France; Department of Organization, Copenhagen Business School, Frederiksberg, Denmark

United Kingdom – North America – Japan India – Malaysia – China

Emerald Group Publishing Limited Howard House, Wagon Lane, Bingley BD16 1WA, UK First edition 2013 Copyright r 2013 Emerald Group Publishing Limited Reprints and permission service Contact: [email protected] No part of this book may be reproduced, stored in a retrieval system, transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without either the prior written permission of the publisher or a licence permitting restricted copying issued in the UK by The Copyright Licensing Agency and in the USA by The Copyright Clearance Center. Any opinions expressed in the chapters are those of the authors. Whilst Emerald makes every effort to ensure the quality and accuracy of its content, Emerald makes no representation implied or otherwise, as to the chapters’ suitability and application and disclaims any warranties, express or implied, to their use. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 978-1-78190-920-1 ISSN: 0733-558X (Series)

ISOQAR certified Management System, awarded to Emerald for adherence to Environmental standard ISO 14001:2004. Certificate Number 1985 ISO 14001

CONTENTS LIST OF CONTRIBUTORS

vii

ADVISORY BOARD

xi

INSTITUTIONAL COMPLEXITY AND ORGANIZATIONAL RESPONSES EMBEDDED IN HYBRID CONTEXTS: HOW INDIVIDUALS IN ORGANIZATIONS RESPOND TO COMPETING INSTITUTIONAL LOGICS Anne-Claire Pache and Filipe Santos INSTITUTIONAL AMBIDEXTERITY: LEVERAGING INSTITUTIONAL COMPLEXITY IN PRACTICE Paula Jarzabkowski, Michael Smets, Rebecca Bednarek, Gary Burke and Paul Spee

3

37

BEYOND THE FAMILY FIRM: REASSERTING THE INFLUENCE OF THE FAMILY INSTITUTIONAL LOGIC ACROSS ORGANIZATIONS Samantha Fairclough and Evelyn R. Micelotta

63

PUTTING NEW WINE IN OLD BOTTLES: UTILIZING RHETORICAL HISTORY TO OVERCOME STIGMA ASSOCIATED WITH A PREVIOUSLY DOMINANT LOGIC Shilo Hills, Maxim Voronov and C. R. (Bob) Hinings

99

v

vi

CONTENTS

IMAGERIES OF CORPORATE SOCIAL RESPONSIBILITY: VISUAL RECONTEXTUALIZATION AND FIELD-LEVEL MEANING Markus A. Ho¨llerer, Dennis Jancsary, Renate E. Meyer and Oliver Vettori

139

LOGIC PLURALISM, ORGANIZATIONAL DESIGN, AND PRACTICE ADOPTION: THE STRUCTURAL EMBEDDEDNESS OF CSR PROGRAMS Mary Ann Glynn and Ryan Raffaelli

175

STRANGE BREW: BRIDGING LOGICS VIA INSTITUTIONAL BRICOLAGE AND THE RECONSTITUTION OF ORGANIZATIONAL IDENTITY Lærke Højgaard Christiansen and Michael Lounsbury

199

INSTANTIATION OF INSTITUTIONAL LOGICS: THE ‘‘BUSINESS CASE’’ FOR DIVERSITY AND THE PREVALENCE OF DIVERSITY MENTORING PRACTICES Shawna Vican and Kim Pernell-Gallagher

233

THE INTERNAL COMPLEXITY OF MARKET LOGICS: FINANCIAL SOPHISTICATION AND PRICE DETERMINATION Vince Feng

275

TAKING STOCK OF INSTITUTIONAL COMPLEXITY: ANCHORING A POOL OF INSTITUTIONAL LOGICS INTO THE INTERINSTITUTIONAL SYSTEM WITH A DESCENDENT HIERARCHICAL ANALYSIS Thibault Daudigeos, Ame´lie Boutinot and Ste´phane Jaumier FROM CULTURAL REPERTOIRES TO INSTITUTIONAL LOGICS: A CONTENT-ANALYTIC METHOD Klaus Weber, Hetal Patel and Kathryn L. Heinze

319

351

LIST OF CONTRIBUTORS Callen Anthony

Management & Organization Department, Boston College, Chestnut Hill, MA, USA

Rebecca Bednarek

Cass Business School, London, UK

Ame´lie Boutinot

Institut Supe´rieur de Gestion, Paris, France Centre for Management Studies, E´cole des Mines ParisTech, Paris, France; Department of Organization, Copenhagen Business School, Frederiksberg, Denmark

Eva Boxenbaum

Gary Burke

Aston Business School, Birmingham, UK

Charlotte K. Coleman

Leeds University Business School, University of Leeds, Leeds, UK

Lærke Højgaard Christiansen

Department of Organization, Copenhagen Business School, Frederiksberg, Denmark

Thibault Daudigeos

Grenoble Ecole de Management, Grenoble, France

Rodolphe Durand

HEC Paris, Paris, France

Samantha Fairclough

School of Business Administration, University of Mississippi, University, MS, USA; Saı¨ d Business School, University of Oxford, Oxford, UK

Vince Feng

Department of Sociology, Harvard University, Cambridge, MA, USA

vii

viii

LIST OF CONTRIBUTORS

Roger Friedland

Religious Studies and Sociology, University of California Santa Barbara, Santa Barbara, CA, USA; Social Research and Public Policy, NYU Abu Dhabi, UAE

Mary Ann Glynn

Carroll School of Management, Boston College, Chestnut Hill, MA, USA

Elizabeth Goodrick

College of Business, Florida Atlantic University, Davie, FL, USA

Royston Greenwood

Alberta School of Business, University of Alberta, Edmonton, Canada

Kathryn L. Heinze

School of Kinesiology, University of Michigan, Ann Arbor, MI, USA

Shilo Hills

Department of Strategic Management & Organization, University of Alberta, Edmonton, Canada

C. R. (Bob) Hinings

Department of Strategic Management & Organization, University of Alberta, Edmonton, Canada

Andrew J. Hoffman

University of Michigan, Ann Arbor, MI, USA

Markus A. Ho¨llerer

Australian School of Business, University of New South Wales, Sydney, Australia

Dennis Jancsary

WU Vienna University of Economics and Business, Vienna, Austria

Paula Jarzabkowski

Cass Business School, City University, London, UK/Science & Technology Studies, Cornell University, Ithaca, USA

Ste´phane Jaumier

Grenoble Ecole de Management, Grenoble, France

P. Devereaux Jennings

Alberta School of Business, University of Alberta, Edmonton, Canada

ix

List of Contributors

Candace Jones

Boston College, Chestnut Hill, MA, USA

Julien Jourdan

Bocconi University, Milan, Italy

Michael Lounsbury

Alberta School of Business, University of Alberta and National Institute for Nanotechnology, Edmonton, Canada

Renate E. Meyer

WU Vienna University of Economics and Business, Vienna, Austria; Copenhagen Business School, Frederiksberg, Denmark

Patricia J. Misutka

Alberta School of Business, University of Alberta, Edmonton, Canada

Evelyn R. Micelotta

School of Business, University of Alberta, Edmonton, Canada

William Ocasio

Kellogg School of Management, Northwestern University, Evanston, IL, USA

Anne-Claire Pache

ESSEC Business School Cergy Pontoise, France

Hetal Patel

Kellogg School of Management, Northwestern University, Evanston, IL, USA

Kim Pernell-Gallagher

Department of Sociology, Harvard University, Cambridge, MA, USA

Vanessa Pouthier

Kellogg School of Management, Northwestern University, Evanston, IL, USA

Ryan Raffaelli

Harvard Business School, Boston, MA, USA

Mia Raynard

Alberta School of Business, University of Alberta, Edmonton, Canada

Trish Reay

Alberta School of Business, University of Alberta, Edmonton, Canada

x

LIST OF CONTRIBUTORS

Filipe Santos

INSEAD Social Entrepreneurship Initiative, INSEAD, Fontainebleau Cedex, France

Marc Schneiberg

Department of Sociology, Reed College, Portland, OR, USA

Michael Smets

Aston Business School, Birmingham, UK

Paul Spee

University of Queensland Business School, The University of Queensland, Brisbane, Australia

Christopher W. J. Steele

Kellogg School of Management, Northwestern University, Evanston, IL, USA

Berangere Szostak

Universite´ Lyon 2, Lyon Cedex, France

Patricia H. Thornton

Fuqua School of Business, Duke University, Durham, NC, USA; Department of Sociology, Stanford University, Stanford, CA, USA

Oliver Vettori

WU Vienna University of Economics and Business, Vienna, Austria

Shawna Vican

Department of Sociology, Harvard University, Cambridge, MA, USA

Maxim Voronov

Goodman School of Business, Brock University, St. Catharines, Canada

Susanne Boch Waldorff

Department of Organization, Copenhagen Business School, Frederiksberg, Denmark

Klaus Weber

Kellogg School of Management, Northwestern University, Evanston, IL, USA

Tammar B. Zilber

Jerusalem School of Business Administration, Hebrew University, Jerusalem, Israel

ADVISORY BOARD SERIES EDITOR Michael Lounsbury Associate Dean of Research Thornton A. Graham Chair University of Alberta School of Business and National Institute for Nanotechnology, Alberta, Canada

ADVISORY BOARD MEMBERS Howard E. Aldrich University of North Carolina, USA

Frank R. Dobbin Harvard University, USA

Stephen R. Barley Stanford University, USA

Royston Greenwood University of Alberta, Canada

Nicole Biggart University of California at Davis, USA

Mauro Guillen The Wharton School, University of Pennsylvania, USA

Elisabeth S. Clemens University of Chicago, USA

Paul M. Hirsch Northwestern University, USA Brayden King Northwestern University, USA

Jeannette Colyvas Northwestern University, USA Barbara Czarniawska Go¨teborg University, Sweden

Renate Meyer Vienna University of Economics and Business Administration, Austria

Gerald F. Davis University of Michigan, USA

Mark Mizruchi University of Michigan, USA

Marie-Laure Djelic ESSEC Business School, France

Walter W. Powell Stanford University, USA xi

xii

ADVISORY BOARD

Hayagreeva Rao Stanford University, USA

W. Richard Scott Stanford University, USA

Marc Schneiberg Reed College, USA

Haridimos Tsoukas ALBA, Greece

INSTITUTIONAL COMPLEXITY AND ORGANIZATIONAL RESPONSES

EMBEDDED IN HYBRID CONTEXTS: HOW INDIVIDUALS IN ORGANIZATIONS RESPOND TO COMPETING INSTITUTIONAL LOGICS Anne-Claire Pache and Filipe Santos ABSTRACT In order to advance the micro-foundations of institutional theory, we explore how individuals within organizations experience and respond to competing institutional logics. Starting with the premises that these responses are driven by the individuals’ degree of adherence to each competing logic (whether novice, familiar, or identified), and that individuals may resort to five types of responses (ignorance, compliance, resistance, combination or, compartmentalization), we develop a comprehensive model that predicts which response organizational members are likely to activate as they face two competing logics. Our model contributes to an emergent political theory of institutional change by predicting what role organizational members are likely to play in the

Institutional Logics in Action, Part B Research in the Sociology of Organizations, Volume 39B, 3–35 Copyright r 2013 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0733-558X/doi:10.1108/S0733-558X(2013)0039A&B014

3

4

ANNE-CLAIRE PACHE AND FILIPE SANTOS

organizational battles for logics dominance or in organizational attempts at crafting hybrid configurations. Keywords: Institutional logics; institutional complexity; practice; institution

In a world characterized by increasing institutional pluralism (Kraatz & Block, 2008), many organizational contexts are becoming embedded in competing institutional logics that impose conflicting demands on organizational members. Institutional scholars have attempted to understand organizational responses to such conflicts (for a review, see Greenwood, Raynard, Kodeih, Micelotta, & Lounsbury, 2011) but have so far paid scant attention to the way in which organizational members experience these conflicts and enact their individual responses. Yet, the growing body of literature on competing institutional logics suggests that under conditions of institutional complexity, individuals play an important role in shaping organizational outcomes. In the context of US community banks, the degree to which the founders of a bank adhered to competing banking and community logics shaped in important ways the likelihood of the new bank’s establishment (Almandoz, 2012). In an Israeli rape crisis center, the collaboration of members adhering to distinct feminist and therapeutic logics led to the emergence of intense internal conflicts as the two groups attempted to promote their own values and practices (Zilber, 2002). In commercial microfinance organizations in Bolivia, the level of socialization of loan officers to competing finance and development logics shaped the sustainability of these organizations (Battilana & Dorado, 2010). Overall, these studies suggest that the degree to which organizational members adhere to competing institutional logics shapes in important ways how organizations respond to them. However, we lack a clear understanding of how individuals within organizations experience and cope with competing institutional logics. The literature on competing institutional logics predominantly rests on the assumption that organizational members’ attitudes toward a given logic are driven by the degree to which they have been embedded in this logic through prior education or professional experience (Bourdieu, 1980; DiMaggio & Powell, 1983). This assumption is based on the premise that education and professional experience introduce individuals to a coherent set of institutional cues. However, a closer look at contemporary career trajectories suggests that

Hybrid Contexts: Competing Institutional Logics

5

they are rarely one-dimensional and that, more often than not, individuals are exposed to multiple competing logics. Management professionals may be asked to get involved in nonprofit activities (Grant, 2012), healthcare professionals may receive training in management (Reay & Hinings, 2009), or executives may join the nonprofit sector (Hwang & Powell, 2009) or the public sector (Meyer & Hammerschmid, 2006) after an initial experience in business. This assumption that individuals get socialized into a main logic through education and experience further obfuscates the fact that individuals get exposed to multiple institutional logics outside of their education and professional lives. They may be embedded in a given logic through the experience of their parents, of their partners, or of their friends (Thornton, Ocasio, & Lounsbury, 2012). Alternatively, they may become embedded in a given logic through the practice of leisure or volunteer activities, such as artistic, political, or environmental commitments. Overall, there is a need for a deeper exploration of how individuals get exposed to institutional logics and how they relate to them. Another assumption of the literature on institutional logics is that organizations are composed of individuals who either adhere to a logic or resist it. A closer look at organizations, however, suggests a more complex reality. Individuals inside organizations may indeed strongly adhere to a logic or strongly resist it, but they may also be indifferent to it, or comply with it without necessarily holding on to all of its core tenets (Lok, 2010). Alternatively, they may also adhere to multiple logics as opposed to only one (Meyer & Hammerschmid, 2006). In summary, the growing field of research on institutional logics lacks a systematic analysis of the repertoire of individuals’ responses to competing institutional logics, as well as of the determinants of these responses. We intend to fill this gap by addressing the following research question: How do individuals, within organizations, experience and respond to competing institutional logics? To address this question, we begin by outlining the different ways in which individuals may experience logics competition. We further identify the types of responses that individuals may resort to as they face competing institutional logics. Finally, we develop a model that predicts individuals’ responses to competing institutional logics as a function of their degree of adherence to each of the logics, as well as of the degree of hybridity of their organizational context. This model further highlights how these responses may afford individuals specific roles inside organizations experiencing institutional complexity. We conclude by discussing the contributions and limitations of the proposed model and suggesting directions for future research.

6

ANNE-CLAIRE PACHE AND FILIPE SANTOS

As the field of research on institutional logics matures, the model that we propose has both conceptual and practical implications. First, by expanding our understanding of how individuals are shaped by and shape institutional logics in plural contexts, this model enriches Thornton et al.’s (2012) model of the micro-foundations of institutional logics and contributes, more broadly, to the micro-foundations of institutional theory (Powell & Colyvas, 2008). More specifically, it complements Thornton et al.’s (2012) model by identifying the individual-level conditions under which embeddedness in competing logics may trigger reflexivity and agency on the part of organizational actors. It further addresses Powell and Colyvas’ (2008, p. 277) call for a ‘‘focus on how the local affairs of members of a field can both sustain and prompt shifts in practices and conventions.’’ Last but not least, this model provides actionable insights into how leaders in organizations embedded in competing institutional logics may think about hiring and socializing their organizational members.

HOW DO INDIVIDUALS EXPERIENCE COMPETING INSTITUTIONAL LOGICS? Institutional logics are broadly defined as patterns of beliefs, practices, values, assumptions, and rules that structure cognition and guide decision making in a given field (Thornton & Ocasio, 1999). Institutional logics are instantiated in and carried by individuals through their actions, tools, and technologies. Institutional logics influence individuals’ behaviors by providing them with ready-to-wear means-end prescriptions (Thornton & Ocasio, 2008), by focusing their attention on a delimited set of issues and solutions considered as appropriate (Ocasio, 1997), and by providing them with vocabularies of motives and sense of self (Friedland & Alford, 1991). As such, institutional logics may shape altogether individuals’ practices, their interests, as well as their identities (Lok, 2010). Their compliance with logics prescriptions is rewarded by social legitimacy, while their deviance is sanctioned (Ingram & Clay, 2000; Rao, Monin, & Durand, 2003), therefore encouraging the reproduction and stability of logics. While institutional scholars recognize that institutional logics spur isomorphic behaviors at the field level, they also acknowledge that some fields are exposed to multiple and potentially competing logics (Friedland & Alford, 1991), imposing competing demands on organizations and their members, and forcing them to devise responses that may deviate from plain compliance

Hybrid Contexts: Competing Institutional Logics

7

(Greenwood et al., 2011; Pache & Santos, 2010). Current studies on responses to competing institutional demands have so far focused on the organizational level of analysis, pointing to various responses such as acquiescence, compromise, avoidance, defiance, and manipulation (Oliver, 1991; Pache & Santos, 2010). These studies further identified a variety of field level factors (fragmentation, centralization, formal structuring) as well as organizational-level factors (field position, structure, governance, and identity) as determinants of organizational responses. We complement this work by focusing on the individual level of analysis, to understand how individuals respond to the conflicting worldviews provided by competing institutional logics. In doing so, we contribute to building the missing link between institutional logics and their instantiation in human action (Greenwood et al., 2011; Thornton et al., 2012).

How Individuals are Exposed to Institutional Logics To understand how individuals respond to competing institutional logics, it is important to understand first how individuals are exposed to institutional logics. Individuals may be influenced by institutional logics through different channels. First, individuals experience institutional logics through their education and work experience. Formal education is a powerful means through which professional norms are produced, legitimized, and diffused (DiMaggio & Powell, 1983). If greatly influences the behavior of individuals by providing them with ready-to-wear templates about how to behave. Individuals further take cues from institutional logics through their work experience as they receive rewards for logic-congruent behaviors and sanctions for logic-discrepant behaviors. Second, individuals, by way of their life activities, experience institutional logics in the context of organizations embedded in organizational fields, that is, in a collective of organizations that share similar goals, values, and practices. At the field level, institutional influences are reinforced by the strength of professional norms as well as by purposive logics-reinforcing activities (conferences, club membershipsy) of professional organizations which define and promote normative rules about professional behavior (DiMaggio & Powell, 1983). At the organizational level, organizations operate, for their individual members, as filters of institutional logics. Depending on the position of the organization in the field, of its status, and of its governance structure, it may either enhance the influence of field level logics on individuals (for instance through the development of rewards system

8

ANNE-CLAIRE PACHE AND FILIPE SANTOS

that encourage compliant behavior) or buffer individuals from field level logics (for instance through specific socialization processes emphasizing organizational rather than professional norms) (Smets, Morris, & Greenwood, 2012). Finally, individuals are members of a society, and are therefore exposed to the key cornerstone institutions of society such as the family, religion, state, market, profession, corporation (Friedland & Alford, 1991), and community (Thornton et al., 2012). These institutions provide the cultural symbols and practices that govern a commonly recognized arena of life. Individuals are exposed to these influences through direct or mediated social interactions. Direct interactions involve relationships with a father or a mother, who provide the basic structure of family institutions or with a religious leader, as enforcer of a religious logic. More broadly, individuals get embedded in institutional logics through their interactions with their personal social networks (DiMaggio & Powell, 1983; Palmer, Jennings, & Zhou, 1993), taking cues from the compliant behavior of the various members of their webs of relationships. Mediated interactions occur through various media, including books, television, or the internet, which are conduits through which institutional norms are diffused and promoted. As they are embedded in these various institutional ties, individuals are likely to be exposed to multiple and potentially competing institutional logics. However, the ties that individuals develop with each of these logics should not be assumed of equal salience (Greenwood et al., 2011): individuals may be influenced by different logics in a different ways. Building upon the work of Thornton et al. (2012), we propose that the degree of influence of a logic on an individual may vary as a function of the degree of availability, the degree of accessibility and the degree of activation of the logic. Availability refers to the knowledge and information that individuals have about a given logic. Accessibility refers to the degree to which knowledge and information about a given logic may come to mind. Activation refers to whether available and accessible knowledge and information are actually used in social interaction. We propose that combinations of these factors shape how individuals relate to different logics.

How Individuals Relate to Institutional Logics We argue that, depending on the degree of availability, accessibility, and activation of a given logic, individuals may relate to it in three different ways, presented here in increasing order of adherence: they may be novice,

Hybrid Contexts: Competing Institutional Logics

9

they may be familiar, or they may be identified with a given logic. We describe below these three distinct types of ties between individuals and institutional logics. Novice An individual who is novice with respect to a given logic has no (or very little) knowledge or information available about this logic. Such a situation may occur when an individual has not been exposed to the logic and its associated demands nor has interacted with others exposed to them. It may further occur in instances where adherence to another logic is so strong that it may blind the individual – keeping her from seeing alternative ways of viewing the world. In the absence of availability, activation is not even an option: a novice is not likely to adapt her behavior to the demands imposed by the logic. Overall, novice therefore refers to an individual whose degree of adherence to a given logic is null. Lounsbury (2001), in his study of university recycling programs, observed that while some of these programs were staffed with individuals who were knowledgeable about recycling and environmentally sustainable behaviors, others were staffed with novices who knew very little about these issues and had little interest in developing and promoting recycling practices. This led them to a relative detachment toward the goals and means associated with the performance of their recycling functions. Familiar An individual who is familiar with a given logic detains available knowledge about it. Such knowledge was made available to her through direct or mediated social interactions. While available, that knowledge is only moderately accessible to the individual: it does not necessarily come to her mind first because she did not build strong ties to this particular logic. The activation of the logic is therefore possible yet not automatic. In other words, this individual has been acquainted with the logic and its associated prescribed goals and means, but is neither emotionally nor ideologically committed to it: she understands the demands that the logic imposes on individuals and is able to comply with these demands to gain legitimacy yet complying with the logic is neither taken-for-granted nor part of her identity. Her degree of adherence to the logic is therefore intermediary. Interestingly, that intermediary level of logic adherence allows her to take distance from the logic and grants her with some room for strategic action. Such a situation may occur when the individual is exposed to the given logic,

10

ANNE-CLAIRE PACHE AND FILIPE SANTOS

yet her degree of embeddedness is low and/or when she does not rely on that specific institutional order to derive her identity and sense of self. In their study of work integration social enterprises in France, Pache and Santos (2013) pointed to situations where individuals became socialized with a particular logic: some social entrepreneurs with a strong background in business got exposed through partners to the social welfare logic. While their identity was not strongly associated with this logic, they were knowledgeable enough about it to enact selective logic-congruent practices in an attempt to secure legitimacy in a field where the social welfare logic was dominant. Identified An individual who is identified with a given logic is one for whom the logic is available and highly accessible and is therefore likely to be activated. In other words, the individual is acquainted with the logic and its organizing principles and feels emotionally and ideologically committed to it: the logic defines for that individual not only what to do but also who she is, as well as how she relates with the rest of the world. Logic identification reinforces the taken-for-grantedness of the logic, because it reinforces the perception of oneness between the self and the logic. Such a relationship to a given logic is likely to develop when the individual has not only been socialized into the logics’ worldviews but has developed, through training or experience, a connection to the logic which provides her with a positive sense of self. Professions act as powerful conduits of logic identification. Examples of individuals identified with institutional logics are numerous in the institutional literature since they are potent illustrations of the influence and power of institutional logics. In her study of the Atlanta symphony Orchestra, Glynn (2000) depicted the two main constituencies of the orchestra, musicians and administrators, as strongly identified with their respective artistic and managerial logics. Their attachment to these logics went beyond a mere knowledge about how an artist or a manager is expected to behave: it was constitutive of how they viewed the world and their own roles in this world. Importantly, these three categories should not be viewed as discrete, but rather as steps on a continuum. As a novice individual gets exposed to a given logic and becomes socialized into it, she is likely to become familiar with it. In turn, an individual who is familiar with a given logic may, over time, develop strong and positive ties with this logic and become identified with it. This typology of ties that individuals may develop with institutional logics is outlined in Table 1.

11

Hybrid Contexts: Competing Institutional Logics

Table 1.

Individuals’ Relationships to Institutional Logics.

Availability (knowledge and information that people have) Accessibility (degree to which knowledge comes to mind) Activation (degree to which knowledge is used in social interaction)

Novice

Familiar

Identified

0

+

+

0

j

+

0

j

+

k Level of adherence: Null

k Level of adherence: Intermediate

k Level of adherence: High

When individuals operate in environments dominated by a single logic, the tie that they have developed with this logic will likely determine their response to this logic: while novice individuals will not comply with the logics’ prescriptions because they will not be aware of it, members who are familiar or identified will be likely to comply with the logics’ demands in an attempt to secure legitimacy in this monologic environment. Yet, responses become more difficult to predict when individuals operate in environments embedded in multiple and competing logics, since complying with one logic may imply defying the competing one. We turn to these situations next.

INDIVIDUAL-LEVEL RESPONSES TO EMBEDDEDNESS IN COMPETING INSTITUTIONAL LOGICS We now turn to the question of how individuals respond to embeddedness in competing institutional logics. To do so, we build on the stream of research on ‘‘inhabited institutions’’ (Binder, 2007; Hallett & Ventresca, 2006), which emphasizes the role played by individuals in making sense of, interpreting, enacting, or resisting institutional prescriptions within organizations. Although we recognize the constraining influence of institutional logics on individuals’ actions, we also embrace Binder’s (2007, p. 568) view that ‘‘real people, in real contexts, with consequential past experiences of their own, play with [logics], question them, combine them with institutional logics from other domains, take what they can use and make them fit their needs.’’

12

ANNE-CLAIRE PACHE AND FILIPE SANTOS

In particular, we argue that the availability of competing models of action creates latitude for individuals to exercise some level of strategic choice (Dorado, 2005; Pache & Santos, 2010; Seo & Creed, 2002). The antagonistic demands emanating from competing institutional logics challenge the takenfor-granted character of institutional arrangements. They make individuals more aware of alternative courses of action and require them to make decisions about what logic to obey, to alter, to ignore, or to reject, in order to satisfy both their identity as well as organizational legitimacy needs. We build upon these views to develop a model of individual responses to competing institutional logics.

A Repertoire of Individual’s Responses to Competing Institutional Logics As they are embedded in competing institutional logics, individuals may choose from a repertoire of responses which is richer than the rejection/ compliance dichotomy usually assumed. To develop that repertoire, we take into account the work on organizational-level responses (Oliver, 1991; Pache & Santos, 2010), yet contend that individual-level responses differ from organizational-level responses: whereas organizational responses are driven by a concern to satisfy institutional referents to ensure survival (DiMaggio & Powell, 1983), individual responses are mainly driven by concerns related to social acceptance, status, and identity. We therefore propose that, as they face competing institutional logics, individuals may resort to the following micro-level responses: ignorance, compliance, defiance, combination, or compartmentalization. We outline in more detail these various responses below. Ignorance refers to an individual’s lack of reaction vis-a`-vis institutional demands. Ignorance is not a conscious attempt at resisting a given institutional logic. It rather refers to an absence of response due to lack of awareness of the logic’s influence. In their study of the emergence of modern architecture, Jones, Maoret, Massa, and Svejenova (2011) for instance show that at the beginning of the twentieth century, architects adhering to a ‘‘revivalist logic’’ based on historical styles where forms signaled a building function, overlooked the emergent ‘‘modern architecture’’ logic founded upon the introduction of universal, efficient, and economical solutions of industrial production. Revivalists initially neither embraced nor rejected the emergent modern logic: they simply ignored it. Compliance refers to an individual’s full adoption of the values, norms, and practices prescribed by a given logic. While compliance may entail

Hybrid Contexts: Competing Institutional Logics

13

different degrees of consciousness, ranging from taken-for-granted habit, unconscious imitation, and voluntary compliance (Oliver, 1991), we contend that an exposure to competing logics challenges the taken-for-granted character of compliance and forces individuals to exercise some degree of agency when complying with a given logic rather than with an alternative one. Instances where individuals complied with a given logic in the midst of institutional complexity have been highlighted in various contexts, including a rape crisis center (Zilber, 2002), museums (Alexander, 1996), community banks (Almandoz, 2012), architecture (Jones et al., 2011), or mutual funds (Lounsbury, 2007). Defiance refers to an individual’s explicit rejection of the values, norms, and practices prescribed by a given logic. As such, defiance differs from ignorance to the extent that it entails awareness and disagreement with the resisted logic as opposed to lack of awareness. Defiance responses may vary in their degree of resistance, ranging from refusal to comply with the prescriptions of a given logic, to more active attempts at contradicting or attacking them with the goal to make them disappear. Townley (1997), for instance, documented how academics in UK universities resisted the introduction of a managerial logic prescribing the practice of performance appraisal, that came in contradiction with the norms and values of the dominant academic logic promoting professional autonomy. Compartmentalization refers to an individual’s attempt at purposefully segmenting her compliance with competing logics. An individual may display full compliance with a given logic (and reject a competing one) in a given context, and choose to display adherence to the competing logic in other contexts: she enacts all competing logics, yet keeps them separated. Compartmentalization may occur across time and/or across space as individuals may choose to enact competing logics in the same place at different times (for instance, when interacting with different people) or in different places (for instance, in different organizational contexts). In the face of competing logics, compartmentalization allows individuals to secure legitimacy by displaying compliance to multiple logics, even if they are incompatible. This approach is identified by Creed and colleagues (2010) as they analyze the responses of gay, lesbian, bisexual and transgender (GLBT) ministers in Protestant churches in the United States to the contradiction between their role in the church and their marginalized GLBT identities. When facing such institutional contradictions, these ministers resorted to compartmentalization, keeping their sexual identity separated from their ministry: while they intimately embraced and complied with the values of each institutional sphere, contradictions between the two spheres led them

14

ANNE-CLAIRE PACHE AND FILIPE SANTOS

to compartmentalize their personal life from their religious commitments, mainly by hiding their sexual orientation to religious constituents. Combination refers to an individual’s attempt at blending some of the values, norms, and practices prescribed by the competing logics. While combining contradictory logics is not necessarily easy given potential incompatibilities between the prescribed templates, empirical research shows that various strategies allow individuals to deal with these incompatibilities. Such strategies involve the selective coupling of intact elements drawn from each logic (Pache & Santos, 2013) or the development of new values, norms, or practices that synthesize the competing logics (Chen & O’Mahony, 2006). In their study of French gastronomy, Rao et al. (2003) for instance showed that some French chefs responded to competing influences of both the orthodox classical cuisine and libertarian societal trends by combining elements of both, thus creating what became the nouvelle cuisine logic. In the face of competing institutional logics, individuals may thus resort to responses that are available to them when facing a single logic (such as ignorance, compliance, or defiance) or may mobilize more complex responses (such as compartmentalization or combination) that are specifically adapted to instances of institutional complexity. We now turn to the factors that may drive these responses.

A Model of Individuals’ Responses to Competing Institutional Logics A wide range of empirical studies reveal that under similar contexts, different individuals respond to competing institutional demands in different ways (Almandoz, 2012; Binder, 2007; Creed et al., 2010; Meyer & Hammerschmid, 2006). What may lead an individual to comply with a given logic and resist an alternative one? What may lead another individual, in the same institutional context, to combine these two logics in an attempt to balance their competing expectations? The core of our argument is that individuals’ responses to competing logics are driven by their degree of adherence to each of the competing logics, that is, whether they are novice (no adherence), familiar (intermediate adherence), or identified with the logic (high adherence). We argue that the nature of these relationships influences individuals’ degree of awareness about the logic and thus their ability to take distance from it, resist it or alter it. This, in turn, is likely to drive how they may respond to competing templates for action. Individuals’ degree of adherence to a logic is likely to

Hybrid Contexts: Competing Institutional Logics

15

shape the degree to which they are committed to it and willing to take action to see it prevail (Greenwood & Hinings, 1996). We further propose that these responses are moderated by a contextual factor: the degree of hybridity of the context. This factor refers to the degree to which the context in which individuals operate is organized around a dominant logic and challenged by an alternative logic (instance of low hybridity) or exposed to multiple competing logics of relatively similar strength (instance of high hybridity). We contend that the degree of hybridity of the context influences the strength of the logics’ influence, and therefore the degree to which individuals may be able to resist these influences without being too severely sanctioned. Building upon these factors, we propose a model of individuals’ responses to competing institutional logics. For the sake of clarity and parsimony, we outline below a simplified framework, using as baseline a situation where individuals are faced with only two competing logics (logic A and logic B) in organizational context with logic A being the dominant logic in situations of low hybridity. We discuss, when relevant, how a change in the degree of hybridity of the context may induce changes in the response mobilized or in the role enacted. When Individuals are Novice with Logic A First, we focus on situations where individuals are novice with logic A and are thus not knowledgeable about the values, norms, and practices promoted by logic A. Novice with Logic B. If, in addition, these individuals are novice with logic B, they are devoid of the institutional background relevant to the context in which they operate. Such a situation may occur as new recruits, hired outside of the organizational field, join an organization. In such situations, individuals will not be aware of the demands of the two logics and are likely to be agnostic about them. They are therefore likely to ignore the demands from both logics, and behave in a way that is relatively independent from the institutional context. Such a response is likely to apply to contexts of both low and high hybridity, since the individuals’ lack of knowledge about the logics makes them relatively immune to their influence. Battilana and Dorado (2010), in their study of commercial microfinance organizations in Bolivia, refer to such an instance of ‘‘tabula rasa’’ hiring where loan officers were recruited prior to any work experience, before they were steeped in either the banking or the development logics that inform the microfinance field. This lack of a priori knowledge about the competing

16

ANNE-CLAIRE PACHE AND FILIPE SANTOS

logics was a sought-after characteristic for new recruits because, in a context of intense logic competition, it allowed them to avoid getting trapped into competitive commitments (Greenwood & Hinings, 1996). This case also suggests that these members did not remain a blank slate for long, but were soon socialized into combining both logics. We propose that individuals who are novice with both logics A and B are likely to play, inside the organization, the role of ingenuous members. This role may entail two different facets. On the one hand, ingenuous members may be clueless and lack the guile required to successfully operate in complex organizational environments. Yet on the other hand, provided that the organization is able to interact with them and listen to them, these ingenuous members may also be valuable resources for complex organizations: they may see things that people do not see anymore because of the cognitive constraints imposed by institutions, ask questions that people do not ask anymore because of the taken-for-granted character of institutionalized norms and practices, and propose alternative courses of actions, because they are not cognitively constrained by institutional rules. Ingenuous members, if properly attended, may help organizations avoid the cognitive traps induced by institutional processes. Overall, we therefore propose the following. Proposition 1. When individuals are novice with both competing logics A and B, they are more likely to ignore both logics than to resort to other response strategies. Their role, in the organization, is likely to be one of ingenuous member. It is important to note, however, that this role is likely to be short lived. As individuals spend time in the organization, experience the logics through interactions with other organizational members or stakeholders, and understand the negative implications of ignoring some of the institutional demands, they become, over time, familiar, if not identified with one or both logics. They are therefore likely to mobilize different types of responses and shift roles inside the organization. Familiar with Logic B. If individuals are novice with logic A yet familiar with competing logic B, they will bear enough knowledge about competing logic B to be able to operate within its sphere of influence. Such situations may occur as new members get recruited, for instance, as an attempt to coopt representatives of a minority logic (Oliver, 1991; Selznick, 1949) or as an attempt, in a hybrid context, to mobilize individuals acquainted with logic B. Irrespective of the degree of hybridity of the field, knowledge about logic B, combined with a lack of awareness of logic A, will most likely lead these

Hybrid Contexts: Competing Institutional Logics

17

individuals to ignore logic A and comply, although with some distance, with logic B. Pache (2012) illustrates this response in her study of work integration social enterprises in France. In this field where the social welfare and the market logics compete, she shows that the members who joined the sector with a background in business and no prior exposure to the social welfare logic, initially enacted the practices prescribed by the market logic without taking into account social welfare demands, potentially jeopardizing the organization’s legitimacy. This case also illustrates the fact that, as these members became acquainted with the social welfare logic over time, they changed responses by starting to incorporate a combination of social welfare and market practices. The role that individuals novice with logic A yet familiar with logic B are likely to play in the organization depends on the degree of hybridity of the context. In contexts where logic A is dominant, these individuals are likely to play the role of outliers, individuals who are part of a minority coalition and are not particularly able to build ties to the dominant coalition. However, as sought by the co-optation strategy, in the absence of a strong identification with logic B, these individuals may, over time, get easily socialized into logic A and thus change roles to become valuable intermediaries between the majority and minority coalitions. In contexts where both logics compete with equivalent strength, these individuals are likely to play the role of disengaged coalition members: while they may be part of the coalition representing logic B inside the organization, their lack of identification with logic B will keep them from engaging actively in the ideological struggles that are likely to occur in such hybrid organizations (Besharov & Smith, 2012). Overall, we propose the following. Proposition 2. When individuals are novice with logic A and familiar with logic B, they are more likely to ignore logic A and comply with logic B than to resort to other response strategies. Their role, in the organization, is likely to be one of outlier, under conditions of low hybridity and of disengaged coalition member under conditions of high hybridity. Identified with Logic B. If individuals are novice with logic A yet identified with competing logic B, they will not only be knowledgeable about logic B but also very motivated to see it prevail. Again, such a situation may occur through the recruitment of new organizational members, either for co-optation purposes (although a high level of identification with logic B

18

ANNE-CLAIRE PACHE AND FILIPE SANTOS

may make it more difficult for these individuals to become socialized with logic A and thus be turned into logic A supporters), or with the goal of reinforcing the pool of members competent with logic B in hybrid contexts. In fields with low hybridity, where logic A is dominant, identification with logic B combined with a lack of awareness of logic A will most likely lead these individuals to ignore logic A and fully comply with logic B. In his study of university recycling programs in the United States, Lounsbury (1998) shows that the university recycling coordinators who were deeply identified with the environmental logic, actively promoted environmental values and practices at the university level, in a context dominated by facility management professionals embedded in a physical plant logic. Yet their lack of knowledge about facility management led environmental recycling coordinators to ignore facility management demands (at least initially). This marginalized them and kept them from being able to collaborate easily with other university stakeholders. In contexts where logic A is dominant, these individuals are likely to play the role of outsiders, individuals who are actively part of a minority coalition i.e. of their lack of knowledge about logic, however, is likely to keep them from actively opposing it dominant. In fields with high hybridity, given the strength of logic B (equal to logic A), individuals novice with logic A and identified with logic B are likely to be encouraged to take stand against logic A while promoting logic B. They are therefore likely to defy logic A and comply with logic B. This response is illustrated by Glynn’s (2000) study of Atlanta’s Symphony Orchestra which was embedded in strong competing artistic and managerial logics. The musicians, who were deeply identified with the artistic logic, yet novice with the managerial logic promoted by the orchestra’s administrators, were not only committed to comply with the artistic logic but also ready to take action to defy the managerial logic which they perceived as a threat to the Orchestra’s fundamentals: they went as far as going on strike to defy the logic which they were not familiar with. In contexts characterized by a high degree of hybridity, members novice with logic A and identified with logic B are thus likely to play the role of challengers: in the face of logics of similar strength, their identification with logic B will not only lead them to comply with the ideology of logic B, but also encourage them to take action against logic A, in order to see logic B prevail. Overall, we propose the following. Proposition 3a. When individuals are novice with logic A and identified with logic B, under conditions of low hybridity, they are more likely to

Hybrid Contexts: Competing Institutional Logics

19

ignore logic A and comply with logic B than to resort to other response strategies. Their role, in the organization, is likely to be one of outsider. Proposition 3b. When individuals are novice with logic A and identified with logic B, under conditions of high hybridity, they are more likely to defy logic A and comply with logic B than to resort to other response strategies. Their role, in the organization, is likely to be one of challenger. Overall, our model suggests that individuals who are novice with one of the competing logics in which they are embedded may feel estranged from the environment in which they operate and are likely to play dissonant roles inside organizations. While their absence of knowledge about this logic may occasionally turn them into organizational misfits (Kleinbaum, 2012), it may also grant them with opportunities to introduce new worldviews and new practices into organizations, thus enhancing their adaptive capacities. When Individuals are Familiar with Logic A We now turn to situations where individuals are familiar with logic A, being knowledgeable about the values, norms, and practices promoted by this logic. They are, however, not identified with this logic, which means that they do not derive their sense of self from adherence to logic A and from compliance with it. Novice with Logic B. If the individuals familiar with logic A are novice with the competing logic B, they will not be aware of the competing demands from logic B and will focus their attention on the demands from logic A. However, they will not feel strongly committed to actively promote compliance with logic A nor stand against logic B. Such a situation may occur as new organizational members get recruited or as previously novice members get progressively socialized into logic A. Both in cases of low or high hybridity, given their knowledge about logic A and their lack of awareness of logic B’s demands, these individuals are likely to comply with the demands of logic A and ignore the demands of logic B. In the context of a child welfare organization embedded in competing administrative and child welfare logics, Binder (2007) highlights that members of the housing department had a background in accounting and finance and were devoid of any connections to the child welfare logic due to a loose coupling of the housing department from the rest of the organization. This pattern of logic adherence led them to ignore the demands of the child welfare logic, founded upon the principle of ‘‘giving

20

ANNE-CLAIRE PACHE AND FILIPE SANTOS

voice’’ to those who are in need and innocent. As they managed the housing operations to address kids and their families’ housing needs, they did not take into account the difficulties faced by each family as expected by the child welfare logic. Instead, they fully complied with the bureaucratic logic, applying similar renting rules and guidelines to all clients and taking action (including eviction) when rules were not respected. The role that individuals familiar with logic A yet novice with logic B are likely to play in the organization depends on the degree of hybridity of the context. In contexts where logic A is dominant, these individuals are likely to play the role of followers, individuals who are part of the majority coalition yet are not particularly committed to that specific coalition. In contexts where both logics compete with equivalent strength, these members are likely to play the role of disengaged coalition members: while they will probably be part of the coalition representing logic A inside the organization, their lack of identification with logic A will keep them from engaging actively in the ideological struggles likely to occur in hybrid organizations. Overall, we propose the following. Proposition 4. When individuals are familiar with logic A and novice with logic B, they are more likely to comply with logic A and ignore logic B. Their role, in the organization, is likely to be one of follower under conditions of low hybridity and one of disengaged coalition member under conditions of high hybridity. Familiar with Logic B. If individuals are familiar with logic A as well as familiar with logic B, they detain a fair amount of knowledge about the idiosyncrasies of each logic. They know the values, norms, and practices prescribed by each logic and are able to comply with these demands if necessary. Importantly, they master the ‘‘vocabularies of practice’’ (Thornton et al., 2012) of each logic and are therefore likely to interact with champions of both logics while projecting legitimacy. This allows them to engage in productive interactions with each group. However, their lack of commitment to either of the logics is likely to keep them from actively attempting to bring the two logics closely together. Such a situation is likely to occur when individuals have benefited from exposure to both logics, for instance through ‘‘second-hand’’ experience (through networks exposing them to these two logics) or through previous experiences in different institutional spheres. Overall, we predict that these individuals are likely to engage in face-work (Goffman, 1955) and resort to compartmentalization as

Hybrid Contexts: Competing Institutional Logics

21

a response strategy, complying with logic A and with logic B sequentially (for instance, complying with logic A when interacting with logic A champions and with logic B when interacting with logic B champions). In their study of underwriters performing reinsurance trading in the Lloyd’s of London, show that underwriters were caught between competing community and market logics. They further suggest that as underwriters became socialized with the community logic through interactions with peers and with the market logic through clear reward systems, they balanced these competing logics by compartmentalizing their compliance with each logic in different organizational spaces: they enacted the community logic in Lloyd’s communal spaces, while they enacted the market logic when working in their offices. This allowed them to balance the competing demands of the two logics and to bridge Lloyds with their respective companies. Irrespective of the degree of hybridity of the field, their knowledge about both logics, combined with their lack of emotional attachments with any of them, puts individuals familiar with two competing logics in potentially favorable situations: they may able to build bridges between competing coalitions of organizational members identified with one or the other logic. They therefore have the potential to operate as intermediaries (Scott, 2003) between two institutional orders, as they may be able to weigh and arbitrate between competing demands and to educate monocultural coalitions about the worldviews of the other party. As such, they may play an important role in helping to regulate conflict between competing factions, especially in contexts of high hybridity. Overall, we propose the following. Proposition 5. When individuals are familiar with logic A and familiar with logic B, they are more likely to compartmentalize logics A and B than to resort to other strategies. Their role, in the organization, is likely to be one of intermediary. Identified with Logic B. If individuals familiar with logic A are identified with logic B, they will be knowledgeable about both logics but only display a strong commitment to logic B. Such a situation may occur when individuals have developed strong ties to logic B through training or experience, while having been more moderately exposed to logic A. The identity commitment to logic B is likely to lead these individuals to take action to see this logic prevail. In contexts where logic A is dominant, these individuals will find it

22

ANNE-CLAIRE PACHE AND FILIPE SANTOS

difficult to impose logic B over logic A. Their knowledge about logic A, however, will allow these individuals to find ways to comply with logic A while simultaneously complying with B: compliance with logic B will be motivated by ideological and identity commitments while compliance with logic A will be driven by a strategic attempt at securing legitimacy with the dominant coalition. In contexts of low hybridity, they are therefore likely to respond to embeddedness in competing logics by combining the two logics. This response is compellingly illustrated by Zilber (2002) in her analysis of the intra-organizational dynamics of a rape crisis center. As this organization, founded by members deeply committed to a feminist logic, started to recruit members identified with a therapeutic logic through their professional training, the dominant feminist logic became challenged. Interestingly, Zilber observes that as therapy-oriented members became familiar with feminist norms and practices, they started to infuse therapeutic meanings into originally feminist practices, thus finding ways to combine the two logics and subtly infiltrating the therapeutic logic inside the organization. Thus, in contexts where logic A is dominant, combining logics A and B may allow the individuals familiar with A and identified with B to play the role of infiltrators: their ability to display adherence with logic A demands may allow them to gain approval with the dominant coalition, while pushing for logic B’s values and norms. Their knowledge about logic A may further allow them to sell logic B’s values, norms, and practices by connecting them with those promoted by logic A. These responses and roles, however, may differ in contexts where both logics are of equal strength. In such contexts, these individuals are again likely to remain faithful to logic B, because of their identification with it. They will have little interest in altering the logic B way of doing things to satisfy the demands of logic A. Yet because they know enough about how logic A operates, they will be able to demonstrate some compliance with logic A when necessary (e.g., through decoupling). In these contexts of high hybridity, we therefore predict that these individuals are likely to resort to compartmentalization as a response strategy to competing logics. Tilcsik’s (2010) study of the funding practices of a governmental agency in a postcommunist setting provides evidence for this approach. The study shows that the agency’s high ranking officials were caught between two powerful competing logics: the communist logic, promoting the distribution of funds by the agency through discretionary decisions and the capitalist logic, prescribing transparent funding processes. The officials, who were

Hybrid Contexts: Competing Institutional Logics

23

identified with the communist logic, yet who were familiar enough with the capitalist logic to understand its demands, chose to formally adopt the transparent funding policies prescribed by the capitalist logic. However, they kept implementing the discretionary funds distribution as promoted by the communist logic. These individuals thus compartmentalized their compliance with the competing communist and capitalist logics. In these contexts where both logics are of equal strength, we propose that individuals are likely to play, inside the organization, the role of advocates: they will use their knowledge about logic A to promote and advance logic B, in an attempt to avoid identity threats. Overall, we propose the following. Proposition 6a. When individuals are familiar with logic A and identified with logic B, under conditions of low hybridity, they are more likely to combine logic A and B than to resort to other response strategies. Their role, in the organization, is likely to be one of infiltrator. Proposition 6b. When individuals are familiar with logic A and identified with logic B, under conditions of high hybridity, they are more likely to compartmentalize logic A and B than to resort to other response strategies. Their role, in the organization, is likely to be one of advocate. Overall, this part of the model describes the relatively common situation of organizational members being familiar with one of the competing logic in which the organization is embedded. It suggests, in particular, that as individuals become familiar with competing logics, they develop the ability to manipulate institutional templates in order to serve specific needs, be it organizational needs (for instance as they play intermediary roles) or their own identity needs (as they play advocate roles). When Individuals are Identified with Logic A We finally turn to situations where individuals are identified with logic A, being not only knowledgeable about the values, norms, and practices promoted by logic A but also deriving their sense of self from compliance with these values, norms, and practices. They are therefore deeply committed to see the templates of logic A prevail. Novice with Logic B. If these individuals, identified with logic A, are novice with competing logic B, they will not only focus on complying with the demands from logic A but will also feel threatened by the introduction of

24

ANNE-CLAIRE PACHE AND FILIPE SANTOS

alternative norms, values, and practices that they are not familiar with. Such profiles are very common: it is the profile of most professionals who have been trained into a profession and who have exercised that profession in homogeneous fields, thus having had little exposure to other institutional models. Given their strong level of adherence to logic A, these individuals will perceive the competing prescriptions of logic B as problematic and are therefore likely to explicitly reject them to ensure the prevalence of their preferred logic. Irrespective of the degree of hybridity, given their strong commitment to logic A and their reluctance to see it challenged by a logic that they do not know, these individuals are likely to comply with the demands of logic A and defy the demands of logic B. Hallett (2010), for instance, shows that teachers of an urban elementary school, who were deeply identified with the education profession, strongly resisted an attempt by the city mayor to introduce a competing accountability logic inside schools. They perceived this introduction as an attack to their autonomy and experienced a high level of distress as they felt that their very identity was threatened. They therefore did not hesitate to mobilize a political campaign against the person in charge of introducing the accountability logic inside schools. The role that individuals familiar with logic A yet novice with logic B are likely to play in the organization depends on the degree of hybridity of the context. In contexts where logic A is dominant, these individuals are likely to play the role of protectors: individuals willing to take action to make sure that the dominant institutional logic from which they derive their identity is protected from challenging influences. This can be explained by the fact that, in the face of identity threats, individuals are likely to be willing to engage in active identity affirmation tactics (Elsbach & Kramer, 1996). This defender role is responsible for most of the internal conflicts emerging in organizations faced with competing logics. In contexts where both logics compete with equivalent strength, these individuals are likely to play the role of challengers: in the face of logics of similar strength, their identification with logic A will not only lead them to comply with the ideology of logic A, but also encourage them to take action against competing logic B which they do not understand, in order to see logic A prevail. Overall, we propose the following. Proposition 7. When individuals are identified with logic A and novice with logic B, they are more likely to comply with logic A and defy logic B. Their role, in the organization, is likely to be one of protector, under conditions of low hybridity, and of challenger, under conditions of high hybridity.

Hybrid Contexts: Competing Institutional Logics

25

Familiar with Logic B. If, in addition to being identified with logic A, individuals are familiar with logic B, they have the ability to understand the demands of logic B and to interact with individuals who strongly adhere to this logic. Their identification with A, however, will keep them focused on ensuring the prevalence of logic A. Such a situation may, for instance, occur when professionals with training and experience with logic A, have had interactions with colleagues or partners adhering with logic B, or have worked in a different context where logic B was prevalent. In contexts where logic A is dominant, given their degree of adherence to logic A, these individuals are likely to maintain a high level of compliance with logic A and to pay lip service to the minority logic B when necessary (for instance, in presence of organizational champions of logic B, or with external referents promoting B). In contexts where both logics are of equal strength, these individuals are again likely to remain highly faithful to logic A, because of their identification with it. They will have little incentive to alter the logic A way of doing things to satisfy the demands of logic B. Yet because they know enough about how logic B operates, they will be able to demonstrate some compliance with logic B when necessary (e.g., through decoupling). Overall, irrespective of the degree of hybridity of the context, individuals are likely to resort to compartmentalization. Lok (2010) provides a compelling example of this dynamic in his study of the responses of UK investors to the emergence of an ‘‘enlightened shareholder value’’ logic. While the dominant ‘‘shareholder value logic’’ instituted shareholder value maximization as the only legitimate goal for a firm, the competing ‘‘enlightened shareholder value logic’’ promoted the need for investors to consider the interests of stakeholders (employees, suppliers, customers) in addition to shareholders as well as to take a longterm, activist approach to their investments. In this context, the author shows that investors deeply identified with the shareholder value logic, yet socialized by the media with the enlightened shareholder value logic, compartmentalized the enactment of both logics: they either decoupled practices – symbolically endorsing the enlightened investment approach, yet actually not implementing it – or physically separating their ‘‘enlightened’’ practices from their traditional trading activities. This compartmentalization strategy allowed them to display compliance with the new logic, thus securing legitimacy vis-a`-vis important institutional referents, while maintaining the logic with which they were deeply identified. We further propose that these individuals are likely to play, inside the organization, the role of advocates. They may not feel threatened by B to the

26

ANNE-CLAIRE PACHE AND FILIPE SANTOS

same extent as a novice individual, because of their empathy with logic B developed through socialization. However, they will be likely to step in to maintain the integrity of logic A in order to avoid identity threats. Overall, we propose the following. Proposition 8. When individuals are identified with logic A and familiar with logic B, they are more likely to resort to compartmentalization than other response strategies. Their role, in the organization, is likely to be one of advocate.

Identified with Logic B. If individuals are identified with both logic A and logic B, they will be knowledgeable about both logics and, more importantly, committed to seeing both of them prevail. Such profiles may be found among individuals with dual degrees, or among individuals with extensive cross sectoral experience, who may have developed strong ties to various logics as a result of these past experiences. We propose that an important skill that accrues to these individuals by way of their commitment to both logics is social skill, that is, an ability to relate to the situation and beliefs of others and to identify with their state (Fligstein, 1997). Combining a high level of motivation to promote both logics and an ability to understand the needs and interests of the coalitions of members identified with each logic, these individuals are likely to be in a good position to find ways to combine the two logics in a sustainable way, irrespective of the degree of hybridity of the context. This response is illustrated by the activities of ‘‘amphibious entrepreneurs’’ described by Powell and Sandholtz (2012) in the context of biotechnology. They indeed show that the founders of new biotech companies who were both identified with the science logic and the business logic were in a particularly favorable position to carry practices and assumptions across domains, thus creating new organizational forms combining the elements of both logics in a sustainable way. Their ability to combine both competing logics in ways that fully take into account the interests of different organizational coalitions may enable these individuals to play the role of hybridizers, that is, of individuals who are able to change the current institutional order to craft new sustainable hybrid institutional arrangements. Their legitimacy with both coalitions of members as well as their knowledge about the various elements of the logics may enable them to craft a viable synthesis between the two competing logics. Overall, we propose the following.

Hybrid Contexts: Competing Institutional Logics

27

Proposition 9. When individuals are identified with both logics A and B, they are more likely to combine the two logics than to resort to other response strategies. Their role, in the organization, is likely to be one of hybridizer. Overall, strong adherence to a dominant logic through identification makes individuals more willing to expend significant resources protecting the institutional status quo. Yet, individuals’ dual identification with competing logics creates inconsistencies that individuals try to resolve through logics combination, which sometimes requires the creation of new institutional arrangements. Hybridizers are not just strategic thinkers who use institutional logics as tools for their work, but rather passionate and complex individuals who strongly adhere to several logics and create new institutional arrangements as a means to adapt their world to their cultures and identities. The full model summarizing how individuals may respond to competing institutional logics as a function of their degree of adherence to these logics, as well as of the degree of hybridity of the context, is outlined in Table 2.

DISCUSSION In this paper, we cast light on the way in which individuals, within organizations, experience and respond to competing institutional logics. We highlight, in particular, that individuals’ degree of adherence to competing logics as well as the degree of hybridity of their organizational context are important drivers of their responses. Focusing on three levels of adherence (novice, familiar, identified) and five types of responses (ignore, comply, defy, compartmentalize, combine), our model allows us to move beyond the simplistic resistance/compliance response dichotomy. It accounts for a richer repertoire of behaviors, reflecting the complexity of the relationship between individuals and their institutional environment. Our model further allows us to outline the roles individuals may play when they are embedded in competing logics, ranging from rather passive roles (follower, disengaged coalition member) to more proactive roles (intermediary, resistor, infiltrator, hybridizer). We find support for these predictions in prior empirical work. Overall, we contribute to an emergent political theory of institutional change which recognizes the important role played by individuals who can skillfully interpret and exploit the ambiguity and contradictions inherent in institutional logics to further

28

ANNE-CLAIRE PACHE AND FILIPE SANTOS

Table 2.

A Model of Individual-Level Responses to Competing Logics A and B Under Varying Degrees of Hybridity. Logic A

Logic B

Novice

Familiar

Novice

Ignore logics A and B (Battilana & Dorado, 2010) Role: Ingenuous member

Comply with logic A and Ignore logic B (Binder, 2007) Role: Followera Disengaged coalition memberb

Comply with logic A and defy logic B (Hallett, 2010) Role: Protectora Challengerb

Familiar

Ignore logic A and comply with logic B (Pache, 2012) Role: Outliera Disengaged coalition memberb

Compartmentalize logics A and B (Smets et al., 2012)

Compartmentalize logics A and B (Lok, 2010)

Role: Intermediary

Role: Advocate

Combine logics A and Ba (Zilber, 2002)

Combine logics A and B (Powell & Sandholtz, 2012)

Role: Infiltratora Compartmentalize logics A and Bb (Tilcsik, 2010) Role: Advocateb

Role: Hybridizer

Identified

a

Ignore logic A and comply with logic Ba (Lounsbury, 1998) Role: Outsidera Defy logic A and comply with logic Bb (Glynn, 2000) Role: Challengerb

Identified

Only in contexts with low hybridity (i.e., where logic A dominates logic B). Only in contexts with high hybridity (i.e., where logic A and B are of comparable strength).

b

their interests (DiMaggio, 1988; Fligstein, 1997; Suddaby & Greenwood, 2005) and protect their identities. Our model has two important implications. First, it implies that ‘‘monocultural’’ individuals (individuals identified with a single logic) are likely to resist change and resort to defensive behaviors in the face of competing logics. Second, it suggests that ‘‘multicultural’’ individuals (individuals identified with multiple logics) are likely to contribute to institutional adaptation and change. Our model therefore complements recent research which suggests that a key feature of institutional entrepreneurs is their ability to build bridges between various institutional orders and constituencies (Fligstein, 1997; Maguire, Hardy, & Lawrence, 2004; Powell & Sandholtz,

Hybrid Contexts: Competing Institutional Logics

29

2012). It further echoes research on organizational ambidexterity (O’Reilly & Tushman, 2004) by suggesting that some actors, namely those that can embrace two logics at the same time, are in a particularly favorable position to lead organizations through adaptation and innovation. It is important to note that while our model outlines individual responses at a given point in time, it also explains why and when individuals may change responses over time. As individuals spend time in a given institutional environment and in a given organization, and as they interact, over time, with various constituencies and get exposed to various institutional influences, they are likely to evolve in their degree of adherence to the logics surrounding them. As they get socialized with a logic, individuals may evolve from being novice to becoming familiar. Then, depending on the nature of their experience with this logic, they may either become identified, or remain at a distance: while positive experiences may, over time, strengthen their identification to a given logic, negative experiences with the enactment of this logic may lead them to progressively detach themselves from (and sometimes reject) this logic. In turn, this is likely to influence the way they respond to multiple competing logics. If over time, the degree of hybridity of the context evolves, they may also respond differently. Our model therefore allows for a dynamic account of the interplay between individual responses and institutional logics. Another important contribution of our model is to provide practitioners with a framework to think about their hiring and socializing strategies. Depending on the type of context in which an organization operates, it may be more or less beneficial for the organizational leaders to recruit monocultural, multicultural or a-cultural members. While monocultural members are likely to operate well in organizations operating in unitary fields, they may introduce rigidities and conflict in organizations operating in plural fields. For example, in a field with high hybridity, hiring members who are identified with one of the logics and novice about the other logic may lead to the emergence of internal conflict as coalitions will form around espoused logics leading to between-group conflict in the organization. Multicultural members, in contrast, who may get bored and lose motivation in unitary fields, are likely to operate well in plural contexts, where they will be stimulated to enact their multiple identities and to craft new institutional arrangements to enable competing worldviews to coexist. Interestingly, our model also suggests that a-cultural members who join an organization without any previous institutional attachment may not necessarily be a liability for organizations. While they may not be able to comply with institutional demands right away, they may bring a fresh look inside the

30

ANNE-CLAIRE PACHE AND FILIPE SANTOS

organization and encourage organizational reflexivity. As blank slates, they may also be more easily socialized into an organizational identity that combines multiple logics (Battilana & Dorado, 2010). Overall, our model thus encourages the leaders of organizations embedded in competing logics to think carefully about their hiring and socialization processes in order to best adapt these processes to their organizational goals and institutional constraints. The theoretical model that we have developed here is not without limitations. For the sake of clarity and parsimony, our model presents a simplified view of institutional influences. We limit, for instance, our analysis to two competing logics, whereas it is recognized in the literature that individuals, as well as organizations, may be embedded in more than two logics (Friedland & Alford, 1991; Greenwood, Diaz, Li, & Lorente, 2010; Greenwood et al., 2011; Heimer, 1999; Jay, 2013). Although it may be uncommon that more than two logics exert demands of similar salience and strength on organizations and their members, we recognize that our model does not provide direct predictions for these occurrences. We nevertheless contend that the factors and reasoning proposed in our model also hold for such situations. For each situation, a careful analysis of the competing logics at play (including their respective strength) as well as of the degree of adherence of the individual to these logics can help determine the response mobilized. We are further aware that individual-level responses to competing institutional logics may be driven by alternative micro-level factors. For instance, in addition to the degree of adherence to the competing logics, individuals’ responses may be driven – at least partly – by the degree to which the individuals perceive the two logics as incompatible. Whereas they may be willing to combine, under specific circumstances, the two competing logics if they categorize them as compatible, they may be more likely to stick to rejection or compartmentalization if they categorize the two as being fully antagonistic (Turner, 1987). Individuals’ responses may further be driven by micro-level factors, including individuals’ ability to manage identity changes and identity threats (Ethier & Deaux, 1994), their need for group belonging (Tajfel & Turner, 1986), as well as their ability to manage the emotions generated by the competition between alternative worldviews (Huy, 1999). Overall, while we acknowledge the importance of these other factors, we have focused our model on the determinants that we believe are the most important drivers of individual-level responses to competing institutional logics. Future research will need to further explore one of the key mechanisms underlying our model: the process of logic identification. Our efforts to achieve parsimony led us to outline a simplified representation of the

Hybrid Contexts: Competing Institutional Logics

31

identification process. However, given the key role played by identification in driving the behavior of individuals confronted with competing logics, it is important to explore in more depth, not only how individuals become identified with a given logic, but also how that identification evolves over time. This enquiry may also entail exploring whether, and if so, under what conditions, individuals may lose their identification with a given logic or replace it with identification with another one. It may also explore instances where individuals may evolve from identification to rejection following negative experiences with a given logic. Such enquiries would require mobilizing the very rich body of knowledge about self-identity, group identity, and organizational identity formation (Albert, Ashforth, & Dutton, 2000; Ashforth & Mael, 1989; Hogg & Terry, 2000). Additional research may also explore the micro-level work required to mobilize and sustain the responses outlined in our model. The more proactive responses, in particular, are not straightforward to implement in pluralistic contexts and require, on the part of individuals, a combination of identity work (Creed et al., 2010) and institutional work (Jarzabkowski, Matthiesen, & Van De Ven, 2009). While our model suggests when these types of work are likely to happen, it remains silent about the way in which they may be implemented. Identity work is required as individuals experience institutional contradictions, manage dual identities, and adapt to changes in the institutional orders around them. Institutional work is involved in the enactment of combination and compartmentalization because in each field, and potentially each organization, individual actors need to craft specific strategies and practices that may allow competing logics to coexist. Specific attention may be paid, in particular, to the organizational processes put in place to facilitate the productive coexistence of competing institutional logics (Battilana, Pache, Sengul, & Model, 2012). Finally, future research will need to explore how multiple individual-level responses to competing logics ultimately aggregate to organizational-level responses. Since every organization is composed of a mosaic of groups structured by functional tasks and employment status (Greenwood & Hinings, 1996), understanding how each group, depending on its adherence to competing logics, is likely to respond constitutes a first step toward a better understanding of broader organizational dynamics. Such an inquiry would allow institutional scholars to go beyond the description of organizations as arenas in which coalitions with different interests and capacities for influence vie for dominance (Palmer et al., 1993; Selznick, 1949), to explain how and why specific coalitions develop their interests and capacities for influence and how these factors allows them to shape organizational outcomes.

32

ANNE-CLAIRE PACHE AND FILIPE SANTOS

In this paper, we have attempted to address Thornton et al.’s (2012, p. 175) call ‘‘to get inside organizations and understand how social interactions within firms shape understandings of institutional complexity as well as how strategic responses to such complexity are conceptualized and implemented.’’ We have proposed that individuals’ level of adherence to competing logics shapes in important ways individual and organizational responses to institutional complexity, thus giving credence to a conceptualization of logics as networks of learned knowledge structures that are unevenly distributed across a population as a result of differences in social interactions and socialization (Thornton et al., 2012). Given the value and importance of bringing individual-level perspectives to the growing area of competing logics, we hope that our model will not only make a valuable theoretical contribution, but will also convince institutional scholars that additional empirical and conceptual research in this direction is warranted.

ACKNOWLEDGMENTS We are grateful to Kevin Andre´, Julie Battilana, and Arthur Gautier for their valuable feedback at various stages of the development of this paper. We also thank Mike Lounsbury for his advice and editorial suggestions.

REFERENCES Albert, S., Ashforth, B. E., & Dutton, J. E. (2000). Organizational identity and identification: Charting new waters and building new bridges. Academy of Management Review, 25(1), 13. Alexander, V. D. (1996). Pictures at an exhibition: Conflicting pressures in museums and the display of art. American Journal of Sociology, 101, 797–839. Almandoz, J. (2012). Arriving at the starting line: The impact of community and financial logics on new banking ventures. Academy of Management Journal, 55(6), 1381–1406. Ashforth, B. E., & Mael, F. (1989). Social identity theory and the organization. Academy of Management Review, 14(1), 20. Battilana, J., & Dorado, S. (2010). Building sustainable hybrid organizations: The case of commercial microfinance organizations. Academy of Management Journal, 6, 1419–1440. Battilana, J., Pache, A. C., Sengul, M., & Model, J. (2012). Keeping a foot in both camps: Understanding the determinants of social performance in hybrid organizations. Working Paper. Besharov, M., & Smith, W. (2012). Multiple logics within organizations: An integrative framework and model of organizational hybridity. Cornell University Working Paper. Binder, A. (2007). For love and money: Organizations’ creative responses to multiple environmental logics. Theory and society, 36, 547–571.

Hybrid Contexts: Competing Institutional Logics

33

Bourdieu, P. (1980). The logic of practice. Stanford, CA: Stanford University Press. Chen, K. K., & O’Mahony, S. (2006). The selective synthesis of competing logics. Paper presented at the Academy of Management Proceedings. Creed, W. E. D., Dejordy, R., & Lok, J. (2010). Being the change: Resolving institutional contradiction through identity work. Academy of Management Journal, 53(6), 1336– 1364. DiMaggio, P. (1988). Interest and agency in institutional theory. In L. G. Zucker (Ed.), Institutional patterns and organizations: Culture and environment (pp. 3–21). Cambridge, MA: Ballinger Publishing Co/Harper & Row Publishers, Inc. DiMaggio, P., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48, 147–160. Dorado, S. (2005). Institutional entrepreneurship, partaking and convening. Organization Studies, 26(3), 385–414. Elsbach, K. D., & Kramer, R. M. (1996). Members’ responses to organizational identity threats: Encountering and countering the business week rankings. Administrative Science Quarterly, 41(3), 442–476. Ethier, K., & Deaux, K. (1994). Negotiating social identity when contexts change: Maintaining identification and responding to threat. Journal of Personality and Social Psychology, 67(2), 243–251. Fligstein, N. (1997). Social skill and institutional theory. American Behavioral Scientist, 40(4), 397–405. Friedland, R., & Alford, R. R. (1991). Bringing society back in: Symbols, practices, and institutional contradictions. In W. W. Powell & P. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 232–263). Chicago, IL: The University of Chicago Press. Glynn, M. A. (2000). When cymbals become symbols: Conflict over organizational identity within a symphony orchestra. Organization Science, 11(3), 285–298. Goffman, E. (1955). On face-work: An analysis of ritual elements in social interaction. Psychiatry: Journal for the Study of Interpersonal Processes, 18, 213–231. Grant, A. M. (2012). Giving time, time after time: Work design and sustained employee participation in corporate volunteering. Academy of Management Review, 37(4), 589–615. Greenwood, R., Diaz, A. M., Li, S. X., & Lorente, J. C. (2010). The multiplicity of institutional logics and the heterogeneity of organizational responses. Organization Science, 21(2), 521–539. Greenwood, R., & Hinings, C. R. (1996). Understanding radical organizational change: Bringing together the old and the new institutionalism. Academy of Management Review, 21(4), 1022. Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E., & Lounsbury, M. (2011). Institutional complexity and organizational responses. The Academy of Management Annals, 5, 317–371. Hallett, T. (2010). The myth incarnate: Recoupling processes, turmoil, and inhabited institutions in an urban elementary school. American Sociological Review, 75(1), 52–74. Hallett, T., & Ventresca, M. J. (2006). Inhabited institutions: Social interactions and organizational forms in Gouldner’s patterns of industrial bureaucracy. Theory and Society, 35, 213–236. Heimer, C. A. (1999). Competing institutions: Law, medicine and family in neonatal intensive care. Law & Society Review, 33(1), 17–66.

34

ANNE-CLAIRE PACHE AND FILIPE SANTOS

Hogg, M. A., & Terry, D. J. (2000). Social identity and self-categorization processes in organizational contexts. Academy of Management Review, 25, 121. Huy, Q. N. (1999). Emotional capability, emotional intelligence, and radical change. Academy of Management Review, 24(2), 325. Hwang, H., & Powell, W. W. (2009). The rationalization of charity: The influences of professionalism in the nonprofit sector. Administrative Science Quarterly, 54(2), 268–298. Ingram, P., & Clay, K. (2000). The choice-within-constraints new institutionalism and implications for sociology. Annual Review of Sociology, 26(1), 525. Jarzabkowski, P., Matthiesen, J., & Van De Ven, A. H. (2009). Doing which work? A practice approach to institutional pluralism. In T. B. Lawrence, R. Suddaby & B. Leca (Eds.), Institutional work (pp. 284–316). Cambridge: Cambridge University Press. Jay, J. (2013). Navigating paradox as a mechanism of change and innovation in hybrid organizations. Academy of Management Journal, 56(1), 137–159. Jones, C., Maoret, M., Massa, F. G., & Svejenova, S. (2011). Rebels with a cause: Formation, contestation, and expansion of the De Novo category ‘‘Modern Architecture,’’ 1870–1975. Organization Science, 23(6), 1523–1545. Kleinbaum, A. M. (2012). Organizational misfits and the origins of brokerage in intrafirm networks. Administrative Science Quarterly, 57(3), 407–452. Kraatz, M. S., & Block, E. S. (2008). Organizational implications of institutional pluralism. In R. Greenwood, C. Oliver, R. Suddaby & K. Sahlin-Andresson (Eds.), The Sage handbook of organizational institutionalism (p. 840). London: Sage. Lok, J. (2010). Institutional logics as identity projects. Academy of Management Journal, 53(6), 1305–1335. Lounsbury, M. (1998). Collective entrepreneurship: The mobilization of college and university recycling coordinators. Journal of Organizational Change Management, 11(1), 50–69. Lounsbury, M. (2001). Institutional sources of practice variation: Staffing college and university recycling programs. Administrative Science Quarterly, 46, 29–56. Lounsbury, M. (2007). A tale of two cities: Competing logics and practice variation in the professionalizing of mutual funds. Academy of Management Journal, 50(2), 289–307. Maguire, S., Hardy, C., & Lawrence, T. B. (2004). Institutional entrepreneurship in emerging fields: HIV/AIDS treatment advocacy in Canada. Academy of Management Journal, 47(5), 657–679. Meyer, R. E., & Hammerschmid, G. (2006). Changing institutional logics and executive identities. American Behavioral Scientist, 49(7), 1000–1014. Ocasio, W. (1997). Towards and attention-based view of the firm. Strategic Management Journal, 18(Summer Special Issue), 187–206. Oliver, C. (1991). Strategic responses to institutional processes. Academy of Management Review, 16(1), 145–179. O’Reilly, C. A., III., & Tushman, M. L. (2004). The ambidextrous organization. Harvard Business Review, 82(4), 74–81. Pache, A. C. (2012). When competing logics enter organizations: The politics of organizational responses to conflicting institutional demands. ESSEC Business School Working Paper, pp. 1–47. Pache, A. C., & Santos, F. (2010). When worlds collide: The internal dynamics of organizational responses to conflicting institutional demands. Academy of Management Review, 35(3), 455–476. Pache, A. C., & Santos, F. (2013). Inside the hybrid organization: Selective coupling as a response to competing institutional logics. Academy of Management Journal.

Hybrid Contexts: Competing Institutional Logics

35

Palmer, D. A., Jennings, P. D., & Zhou, X. (1993). Late adoption of the multidivisional form by large U.S. corporations: Institutional, political, and economic accounts. Administrative Science Quarterly, 38(1), 100–131. Powell, W. W., & Colyvas, J. A. (2008). Microfoundations of institutional theory. In R. Greenwood, C. Oliver, R. Suddaby & K. Sahlin-Andresson (Eds.), The Sage handbook of organizational institutionalism (p. 840). London: Sage. Powell, W. W., & Sandholtz, K. W. (2012). Amphibious entrepreneurs and the emergence of new organizational forms. Strategic Entrepreneurship Journal, 6(2), 94–115. Rao, H., Monin, P., & Durand, R. (2003). Institutional change in Toque Ville: Nouvelle Cuisine as in identity movement in French gastronomy. American Journal of Sociology, 108(4), 795–843. Reay, T., & Hinings, C. R. (2009). Managing the rivalry of competing institutional logics. Organization Studies, 30(06), 629–652. Scott, W. R. (2003). Institutional carriers: Reviewing modes of transporting ideas over time and space and considering their consequences. Industrial & Corporate Change, 12(4), 879–894. Selznick, P. (1949). TVA and the grass roots: A study of politics and organization. Berkeley, CA: University of California Press. Seo, M., & Creed, W. E. D. (2002). Institutional contradictions, praxis and institutional change: A dialectical perspective. Academy of Management Review, 27(2), 222–247. Smets, M., Burke, G. T., Jarzabkowski, P., & Spee, A. P. (2012). Reinsurance trading in lloyd’s of london: Balancing conflicting-yet-complementary logics in practice. Paper presented at the Academy of Management Proceedings. Smets, M., Morris, T. I. M., & Greenwood, R. (2012). From practice to field: A multilevel model of practice-driven institutional change. Academy of Management Journal, 55(4), 877–904. Suddaby, R., & Greenwood, R. (2005). Rhetorical strategies of legitimacy. Administrative Science Quarterly, 50, 35–67. Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W. Austin (Eds.), Psychology of intergroup relation (pp. 7–24). Chicago, IL: Nelson Hall. Thornton, P. H., & Ocasio, W. (1999). Institutional logics and the historical contingency of power in organizations: Executive succession in the higher education publishing industry, 1958–1990. American Journal of Sociology, 105, 801–844, University of Chicago Press. Thornton, P. H., & Ocasio, W. (2008). Institutional logics. In R. Greenwood, C. Oliver, R. Suddaby & K. Sahlin-Andresson (Eds.), The Sage handbook of organizational institutionalism (p. 840). London: Sage. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure, and process. Oxford: Oxford University Press. Tilcsik, A. (2010). From ritual to reality: Demography, ideology, and decoupling in a post communist government agency. Academy of Management Journal, 6, 1474–1498. Townley, B. (1997). The institutional logic of performance appraisal. Organization Studies, 18(2), 261, Walter de Gruyter GmbH & Co. KG. Turner, J. C. (1987). Rediscovering the social group: Self-categorization theory. Oxford, UK: Basil Blackwell. Zilber, T. B. (2002). Institutionalization as an interplay between actions, meaning and actors: The case of a rape crisis center in Israel. Academy of Management Journal, 45(1), 234–254.

INSTITUTIONAL AMBIDEXTERITY: LEVERAGING INSTITUTIONAL COMPLEXITY IN PRACTICE Paula Jarzabkowski, Michael Smets, Rebecca Bednarek, Gary Burke and Paul Spee ABSTRACT This paper develops a practice approach to institutional ambidexterity. In doing so, it first explores the ‘promise’ of institutional ambidexterity as a concept to address shortcomings with the treatment of complexity in institutional theory. However, we argue that this is an empty promise because ambidexterity remains an organizational level construct that neither connects to the institutional level, or to the practical actions and interactions within which individuals enact institutions. We therefore suggest a practice approach that we develop into a conceptual framework for fulfilling the promise of institutional ambidexterity. The second part of the paper outlines what a practice approach is and the variation in practice-based insights into institutional ambidexterity that we might expect in contexts of novel or routine institutional complexity. Finally, the paper concludes with a research agenda that highlights the potential of practice to extend institutional theory through new research approaches

Institutional Logics in Action, Part B Research in the Sociology of Organizations, Volume 39B, 37–61 Copyright r 2013 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0733-558X/doi:10.1108/S0733-558X(2013)0039A&B015

37

38

PAULA JARZABKOWSKI ET AL.

to well-established institutional theory questions, interests and established-understandings. Keywords: Institutional logics; institutional complexity; practice; ambidexterity

INTRODUCTION Institutional ambidexterity (Greenwood, Raynard, Kodeih, Micellota, & Lounsbury, 2011) is institutionalists’ latest attempt to come to terms with situations of institutional ‘complexity’ or ‘pluralism’, in which incompatible expectations and prescriptions, or ‘logics’, collide (Goodrick & Reay, 2011; Kraatz & Block, 2008; Pache & Santos, 2010; Thornton, Ocasio, & Lounsbury, 2012). Drawing on existing literature in the strategy field (Gibson & Birkinshaw, 2004; Gupta, Smith, & Shalley, 2006; March, 1991; Tushman & O’Reilly, 1996), institutional ambidexterity looks to the potential benefits of being able to operate across coexisting, contradictory logics. This perspective marks a significant departure from previous approaches that saw institutional complexity as problematic and focused on resolving conflict by keeping apart people, practices or audiences that followed contradictory logics (e.g. Greenwood & Suddaby, 2006; Seo & Creed, 2002; Smets, Morris, & Greenwood, 2012). This problem-based focus led institutionalists to overlook the potential benefits for organizations in operating across different logics. Indeed, from healthcare to professional services to academia, there are numerous examples of organizations in which seemingly incompatible logics not only coexist, but also fruitfully complement each other. Therefore, we need to better understand how organizations can capitalize on interdependencies between logics, and not just remedy problems arising from their tensions. The concept of institutional ambidexterity is an important step towards addressing this shortcoming in the institutional literature; albeit, one that still falls short of resolving it. The notion of ambidexterity differs from existing approaches because it acknowledges that working to different prescriptions, while difficult, has the potential to benefit organizations. From this perspective, institutional complexity is not a problem to be resolved, but a naturally occurring condition to be managed, and harnessed (Kraatz & Block, 2008). Indeed, if we see exploitation and exploration (March, 1991) as two distinct and contradictory logics of learning, then

Leveraging Institutional Complexity in Practice

39

ambidexterity has, implicitly, always been looking at logics-in-action and is a useful frame for looking at institutional complexity (Greenwood et al., 2011). Transposing organizational ambidexterity insights into the institutional literature thus appears promising at first sight. Closer scrutiny, however, exposes this as an empty promise because, to date, ambidexterity scholars struggle to suggest practical solutions to the puzzle of integrating contradictory prescriptions. For example, relying on senior management teams to integrate separate efforts lower down the corporate hierarchy may work for exploration and exploitation (e.g. Benner & Tushman, 2003; Jansen, Tempelaar, van den Bosch, & Volberda, 2009; O’Reilly & Tushman, 2004; Siggelkow & Levinthal, 2003), but is a less useful insight for other logics that need to be balanced on a continuous basis by practitioners at work. Similarly, those authors suggesting that ambidexterity hinges on organizational culture, fail to specify what people in those organizations do to create such cultures, or to perform their everyday work under their influence (Gibson & Birkinshaw, 2004). These problems arise because ambidexterity scholars have treated complexity as an organizational phenomenon, without pursuing its practical implications at the individual level or its origins at the institutional level. Hence, in order to fulfil the promise of the institutional ambidexterity concept, we need an approach that leverages institutional complexity in practice. Developing this practice approach to institutional ambidexterity is the focus of this paper, which unfolds as follows: first, we review existing approaches to institutional complexity. Then, we visit the ambidexterity literature to gauge the promise it holds for addressing the puzzle of institutional complexity. Third, we introduce practice theory as a perspective that fruitfully connects individual, organizational and institutional levels of analysis (e.g., Barley, 2008; Bourdieu, 1990; Lounsbury & Crumley, 2007; Smets et al., 2012; Whittington, 2006), building a conceptual model to illustrate our argument. Finally, drawing on the relatively few examples from the institutional theory literature, we outline the types of insights that a practice perspective on institutional ambidexterity can generate and indicate promising avenues for future research.

INSTITUTIONAL COMPLEXITY: THE STORY SO FAR Since Friedland and Alford’s (1991) seminal work, institutional complexity, which is the encounter of ‘incompatible prescriptions from multiple institutional logics’ (Greenwood et al., 2011, p. 317), has captured

40

PAULA JARZABKOWSKI ET AL.

institutionalists’ interest. Such complexity arises from an overlap of institutional orders with incompatible prescriptions (Thornton et al., 2012); for example, where organizations operate across institutional spheres with contradictory prescriptions (e.g., Greenwood & Suddaby, 2006; Smets et al., 2012) or face different constituents with divergent demands (e.g., D’Aunno, Sutton, & Price, 1991; Greenwood, Maga`n Diaz, Li, & Ce´spedes Lorente, 2010; Heimer, 1999). Over time and based on the intellectual priorities in the broader field of institutional theory, this interest has taken three different guises that we outline below.

Institutional Complexity as Occasion for Change Institutional logics are the building blocks of institutional complexity and initially rose to prominence as a means of theorizing about the heterogeneity in institutional environments (Battilana & Dorado, 2010; Seo & Creed, 2002; Thornton & Ocasio, 2008; Thornton et al., 2012). Complexity allowed institutionalists to counter critiques of being overly focused on homogeneity and stasis (Hirsch & Lounsbury, 1997) and to theorize institutional change arising endogenously, rather than from exogenous regulatory, social or technological shocks (Edelman, 1992; Garud, Jain, & Kumaraswamy, 2002; Rao, Monin, & Durand, 2003). Seo and Creed (2002) argue that institutional contradictions can function as an endogenous trigger of change, because they make actors aware of alternatives to their institutionalized, taken-for-granted ways and motivate them to pursue more favourable alternatives (e.g., Djelic & Quack, 2003; Greenwood & Suddaby, 2006; Seo & Creed, 2002; Smets et al., 2012; Thornton, Jones, & Kury, 2005). Given the predominant preoccupation with explaining institutional change, such research depicted complexity as transitory, en route to a state of relative stability in which one logic dominates another (Marquis & Lounsbury, 2007; Reay & Hinings, 2005; Thornton, 2002; Townley, 2002).

Institutional Complexity as Contested Coexistence More recently institutional theorists have come to recognize that institutional complexity may not be transitory, but permanent (Greenwood et al., 2011; Zilber, 2011); especially for those organizations which, by their very nature, are ‘an incarnation or embodiment of multiple logics’ (Kraatz & Block, 2008, p. 244). In these organizations, contradictory prescriptions

Leveraging Institutional Complexity in Practice

41

from different legitimating audiences systematically collide in everyday operations and institutional complexity must be managed continuously. Therefore, studies of such organizations made efforts to leave the field level and ‘get inside’ organizations to understand organizational responses to institutional complexity (e.g., Battilana & Dorado, 2010; Pache & Santos, 2010). However, they retained a focus on the incompatibilities and tensions that characterize the overlap of competing logics (Greenwood et al., 2011), as reflected in the imagery of ‘turmoil’ (Hallett, 2010, p. 52); ‘threat’ (Jarzabkowski, Matthiesen, & Van de Ven, 2009, p. 306) and ‘uneasy truce’ (Reay & Hinings, 2005, p. 364). As a result, suggested organizational responses, building on Oliver’s (1991) seminal work, have been largely defensive. These may involve resistance to institutional pressures (Townley, 1997), use of organizational status to avoid or defy problematic stakeholders (e.g., Greenwood & Suddaby, 2006; Smets et al., 2012) or compartmentalization of compliance with different sets of expectations into different organizational or geographic units (e.g., Binder, 2007; Dunn & Jones, 2010; Jarzabkowski et al., 2009; Lounsbury, 2007). Common to the majority of these responses is their focus on simplifying the situation and reducing the tensions of complexity by keeping apart competing logics and the practices and people that enact them. Advocating an approach of not letting the right hand know what the left hand is doing, such studies did not fully consider that organizations that are ‘dextrous with both hands’ might not only avoid tensions, but also reap distinctive benefits.

Institutional Ambidexterity as Embracing Complexity Most recently, institutional scholars have revisited relationships between logics in a more nuanced fashion, discovering potential for more fruitful relationships between coexisting logics (e.g., Almandoz, 2012; Battilana & Dorado, 2010; Greenwood et al., 2011; Kraatz & Block, 2008; Smets & Jarzabkowski, forthcoming). Ironically, benefits of awareness and reflexivity were already highlighted by Seo and Creed (2002); indeed, Friedland and Alford pointed out that coexisting logics ‘are interdependent and yet also contradictory’ (1991, p. 250; emphasis added; see also Scott, 1991). This makes intuitive sense, as there are benefits in scientific and well-managed patient-care (Dunn & Jones, 2010; Kitchener, 2002), more client-oriented or multi-disciplinary professional services (Greenwood & Suddaby, 2006; Suddaby & Greenwood, 2005; Thornton et al., 2005), or positive feedback loops between commercialized research outputs and ideas or funding for

42

PAULA JARZABKOWSKI ET AL.

future research (D’Este & Perkmann, 2011; Perkmann & Walsh, 2009). Institutional scholars, given their predominant focus on incompatibility and tension, have largely neglected such interdependencies. Yet, interdependence indicates that truly ambidextrous approaches to complexity, in which the left hand not only knows what the right hand does, but can skilfully complement its actions, are much more apposite to address the conundrum of institutional complexity. Greenwood et al. (2011) suggest borrowing from the ambidexterity literature to provide new insights into the integration of competing logics. The basic premise of ambidexterity is the ability to simultaneously perform contradictory processes when both are critical to organizational success (March, 1991; Tushman & O’Reilly, 1996). This change in perspective is helpful for institutionalists insofar as it moves from attempting to resolve complexity to balancing its components in pursuit of distinctive outputs and identities (Goodrick & Reay, 2011; Kraatz & Block, 2008; Reay & Hinings, 2009). Based on Simsek’s (2009) distinction of ‘structural’ and ‘blended’ hybrids, two different approaches to ambidexterity have been proposed in institutional theory; neither of which, however, seems to be currently delivering on the promise of embracing institutional complexity. Structural differentiation arises from current work on compartmentalization that assigns compliance with different expectations to different organizational units. Where and how subsequent integration occurs, however, remains unclear to date. Blended hybrids sidestep this problem by allowing different logics to pervade the organization and relying on individuals to strike an appropriate balance (Battilana & Dorado, 2010; D’Aunno et al., 1991; Smets et al., 2012). So far, this appears to be the ‘gold standard’ of institutional ambidexterity. However, several problems remain. For example, suggestions for facilitating logic blending are currently limited to recommendations of using human resource policies to create an open and receptive context (Battilana & Dorado, 2010; Smets et al., 2012). How this is to be done and what individuals inhabiting these contexts do to enact different logics, however, remains obscure. Furthermore, it is questionable that an indistinguishable blend of competing logics and the practices that enact them would always be desirable. As Battilana and Dorado (2010) foreshadow, blended hybridization may produce ‘slippage’ towards either logic. From a theoretical perspective, slippage may herald institutional change, rather than balanced complexity with distinct but coexistent logics. Empirically, failure to discriminate between, and sustain, competing logics may herald the neglect of critical standards, as experienced in the case of Enron and other corporate scandals in which commercial goals superseded

Leveraging Institutional Complexity in Practice

43

professional standards (Gabbioneta, Greenwood, Mazzola, & Minoja, forthcoming; Grey, 2003). Alternatives that can integrate practices governed by competing logics on an ongoing, practical basis in order to sustain them as interdependent yet separate have thus rarely been explored in the (institutional) ambidexterity literature. We consider this neglect both theoretically and empirically problematic. We now briefly visit the foundations of ambidexterity in the strategy literature, before developing a practice-theoretical variant that we consider more suitable for addressing remaining blind spots in the study of institutional complexity and the attendant concept of institutional ambidexterity.

AMBIDEXTERITY IN THE STRATEGY LITERATURE: A FALSE PROMISE The ambidexterity literature suggests that successful firms are ambidextrous; ‘aligned and efficient with today’s business demands [exploitation] while simultaneously adaptive to changes in the environment [exploration]’ (March, 1991; Raisch & Birkinshaw, 2008, p. 685). As discussed above, Greenwood et al. (2011) link this concept to institutional theory through labelling ‘exploitation’ and ‘exploration’ as different logics of learning. Other scholars argue that, as ambidexterity can been used to signify an organization’s ability to do two different (apparently conflicting) things at the same time more generally, it is applicable to logics other than exploitation and exploration (Gibson & Birkinshaw, 2004). Ambidexterity, the ability to excel at two contradictory things simultaneously, is posited as both possible and desirable for organizations. It is thus portrayed as an organizational resource that links to organizational performance, which is the central theoretical proposition and empirically explored relationship in the ambidexterity literature (Gibson & Birkinshaw, 2004; He & Wong, 2004; Simsek, 2009; Tushman & O’Reilly, 1996). Achieving ambidexterity is not, however, assumed to be easy, as the simultaneous fulfilment of competing demands is filled with potential tensions and trade-offs (Andriopoulos & Lewis, 2009; Raisch & Birkinshaw, 2008; Smith & Tushman, 2005). The literature has explored three main ways of achieving organizational ambidexterity: structure, organizational context (culture) and leadership (Raisch & Birkinshaw, 2008; Simsek, 2009). Initially, structural solutions proposed that distinct organizational units should pursue either exploration or exploitation (Benner & Tushman, 2003; Burgelman, 2002; Gupta et al.,

44

PAULA JARZABKOWSKI ET AL.

2006; O’Reilly & Tushman, 2004; Smith & Tushman, 2005; Tushman & O’Reilly, 1996). Then, contextual ambidexterity focused on an organization’s culture, suggesting that if management created a suitable context, for example by engendering trust, ambidexterity would be apparent at all levels of the organizations (Birkinshaw & Gibson, 2004). Finally, given that both these bodies of research focus on the upper echelons of management that are either the strategic integrating apex (Smith & Tushman, 2005); or responsible for creating a supportive business-unit context (Birkinshaw & Gibson, 2004), a large body of work has focused on leadership and leaders’ characteristics that enable organizational ambidexterity (Carmeli & Halevi, 2009; Jansen et al., 2009; Lubatkin, Simsek, Yan, & Veiga, 2006; Siggelkow & Levinthal, 2003; Smith, Binns, & Tushman, 2010; Smith & Tushman, 2005). In considering the ambidexterity literature, issues emerge regarding its capacity to provide a theoretical foundation for managing institutional complexity and the desire to embrace interdependence and integration between divergent logics. First, there is a surprising lack of integration in the ambidexterity literature (Simsek, 2009). Indeed, ‘organizational ambidexterity’ was initially treated as ‘virtually synonymous with structurally differentiated hybrids’ (Greenwood et al., 2011, p. 355) and any integration that does occur is usually located at the apex of an organization rather than something that is imbued at all organizational levels (Smith & Tushman, 2005). Second, ambidexterity remains an explicitly organizational construct, which partly explains the lack of attention to the practical doing of ambidexterity by managers. The literature lacks integrative models that span multiple levels of analysis, such as the individual and institutional levels (Gupta et al., 2006; Raisch, Birkinshaw, Probst, & Tushman, 2009; Simsek, 2009). In particular, contextual and structural approaches exclude a micro focus on how individuals perform ambidexterity or macro perspectives on the institutional ramifications of ambidexterity (Simsek, 2009). While the lineage of the term ‘ambidextrous’ is explicitly individual – the capacity of an individual to do two things at once – as a metaphor in organization studies, ambidexterity has not focused on the individual per se or how actors do ambidexterity (Andriopoulos & Lewis, 2009; Gupta et al., 2006). This is true even when authors argue that ambidexterity resides at the individual level, such as Gibson and Birkinshaw (2004), who propose that individuals should use their own judgement to achieve ambidexterity, but focus on the characteristics of their context rather than on how the individuals might do that. Tellingly, in painting a multi-level picture of

Leveraging Institutional Complexity in Practice

45

ambidexterity, Bledow, Frese, Anderson, Erez, and Farr (2009) include only one explicit ambidexterity citation in their summary of the individual level (Mom, Van Den Bosch, & Volberda, 2007), drawing the rest of their literature from other research domains. Bledow et al.’s (2009) model is also explicitly intra-organizational, with no discussion of the broader organizational environment, leading to our second point that the ambidexterity literature is almost exclusively internally focused on organizational capabilities. It is not concerned with legitimacy and adherence to externally imposed rules of action (Greenwood et al., 2011). When the inter-organizational environment is considered, it is a secondary focus behind intra-organizational performance and resources, such as how alliances may be a source of intra-organizational ambidexterity (Rothaermel & Deeds, 2006), or how particular characteristics of an organizations’ immediate, local environment may impact the ambidexterityperformance link (Jansen, van de Bosch, & Volberda, 2005, 2006; Levinthal & March, 1993). Consequently, ambidexterity does not provide institutional theorists with the means to link the ambidextrous doing of institutional complexity at the individual level, with the organizational level and the broader institutional field. In its current conceptual state, it is a false promise for studying the problems and benefits of institutional complexity.

FULFILLING THE PROMISE: INSTITUTIONAL AMBIDEXTERITY IN PRACTICE In this section we draw together the concept of institutional ambidexterity, the shortcomings that currently prevent it from fulfilling its promise and our suggestion of practice theory as a way to address those shortcomings in a conceptual model of our argument. Ambidexterity presents a potential solution to how organizations can reap the benefits of interdependent logics, by suggesting that they may embrace two potentially contradictory courses of action simultaneously. However, as shown in Fig. 1, it is primarily an organizational level construct that does not adequately address the issues of logics in action. That is, ambidexterity neither examines the institutional logics within which potential contradictions are embedded, nor the actions that people take in coping with coexistence in more or less complementary ways. Hence, we need an underpinning theoretical approach that can flesh out institutional ambidexterity as a concept that embraces the potential benefits of institutional complexity and also examines how actors work

46

PAULA JARZABKOWSKI ET AL.

Fig. 1.

A Practice Approach to Institutional Ambidexterity.

within such complexity to perform ambidextrous associations between logics. Practice theories span multiple levels from institutional to individual within which institutional complexity is socially accomplished. From a practice perspective, all action is situated within, produces and reproduces the dynamics of its wider social context. Thus, by looking at the actions of individuals at the nexus of different institutional logics, such as health care, educational, arts and regulated organizations, we can see how actions that are situated within multiple logics construct ambidexterity. Institutions themselves cannot be ambidextrous; rather, people do institutional ambidexterity in their everyday actions and interactions as they work within and enact multiple logics. Practice theory (e.g. Jarzabkowski, 2005; Jarzabkowski, Balogun, & Seidl, 2007; Schatzki, 2001) directs attention to, ‘how people engage in the doing of ‘‘real work’’’ (Cook & Brown, 1999, p. 387) but also the ‘shared practical understanding’ which gives it meaning and makes it robust (Schatzki, 2001,

Leveraging Institutional Complexity in Practice

47

p. 2). The focus is on how actors interact with, construct and draw upon the social and physical features of their context in the everyday activities that constitute practice. Practice theorists have thus sought to go beyond the dualism of institutions and action that still permeates much institutional theory (Barley, 2008; Hallett, 2010), by demonstrating that institutions are constructed by and, in turn, construct action (e.g. Bourdieu, 1990; Giddens, 1984; Sztompka, 1991; Turner, 1994). For these theorists, the collective practice that constitutes institutional order exists only through the way that it is enacted in the mundane, practical actions of actors, as depicted by the double-headed ‘practice’ arrow in Fig. 1. Practice theory thus takes actions, interactions and negotiations between multiple individuals as the core unit of analysis (Jarzabkowski et al., 2007). In these actions and interactions, actors instantiate, reproduce and modify the shared or collective practice (Schatzki, 2002) that is the focus of institutional theory (Jarzabkowski, 2004; Whittington, 2006). A practice lens can thus provide insight into how actors perform institutional ambidexterity within their work; enacting competing understandings about ‘how to go on’ (Chia & Holt, 2009; Giddens, 1984) and coping with the resultant tensions in practical ways that perform the ambidexterity of working within and between multiple logics (Denis, Langley, & Rouleau, 2007). Recent developments in institutional theory show sensitivity to the practice approach. For example, a recent book exhorts institutional theorists to examine logics in action, devoting an entire paper to practices as the material enactments of institutional logics (Thornton et al., 2012), while other studies explicitly propose a practice approach as a means of understanding how actors cope with and respond to institutional complexity (e.g. Jarzabkowski et al., 2009; Smets et al., 2012). In contrast to existing accounts that locate institutional complexity in field-level ‘contradictions’ (Seo & Creed, 2002), such studies propose that institutional complexity is experienced as ‘part of the ordinary, everyday nature of work, rather than exceptional phenomena’ (Jarzabkowski et al., 2009, p. 289). Practice theoretical approaches thus address the gap in Fig. 1, by showing how we may understand institutional ambidexterity at the collective practice level through a focus on people’s practical coping with institutional complexity. The majority of such studies furnish insights into practical actions in situations of novel institutional complexity, such as changes in the juxtaposition of logics within an organization. Examples include Jarzabkowski et al.’s (2009) study of a shift in the association between market and regulatory logics in a telecoms company, Smets et al.’s (2012) study of German and English professional logics colliding in a law-firm merger, or

48

PAULA JARZABKOWSKI ET AL.

Zilber’s (2002) study of the way that practices become imbued with different meanings during a transition from feminist to clinical institutional logics. However, the practice theory focus on everyday, practical coping indicates that there may also be significant value in moving beyond novelty to also examine cases of long-standing or routine institutional complexity.

Individual Level: Action and Interaction in Doing Ambidexterity We now examine some of the main findings from such practice-oriented studies of institutional complexity, in order to develop the final section of our conceptual framework about the practical understandings through which actors perform institutional ambidexterity. ‘Practical understanding’ is a practice term that refers to the actions – the know-how and embodied repertoires – that compose a practice (Schatzki, 2002, pp. 77–78). While practical understandings are firmly grounded in actors and their actions, they encompass the collective practice within which people tacitly understand or have a ‘feel for’ how to perform the social order (see also Bourdieu, 1990; Giddens, 1984). They thus provide a useful unit of analysis for practice-based approaches to institutional theory. While not intended to be exhaustive, this section illustrates some insights that a focus on the practical, mundane doing of institutional ambidexterity in action can provide. First, is an expanded practice repertoire whereby individuals incorporate both traditional local practices, for example, English- and German logics of legal practice, and new hybridized practices in their work (Smets & Jarzabkowski, forthcoming). As these authors showed, German lawyers could undertake cross-border financing for international clients according to international standards, whilst also retaining those practices specific to their domestic markets, to be activated when required in local transactions. The finding of an expanded practice repertoire gives empirical credence to theories about the potential benefits of institutional complexity (Greenwood et al., 2011) beyond suggestions that they expand individuals’ cultural tool kit (Swidler, 1986). Specifically, they show that the expanded portfolio need not be cultural, but may also be practical. We are thus motivated to think not only of ‘strategies to respond’ to complexity, but also of the generative capacity of complexity (Kraatz & Block, 2008) to expand the practical toolkit of actors. A second, related form of practical understanding is ‘situated improvising’. Smets et al. (2012) show how urgency and high stakes in meeting client expectations encouraged lawyers to experiment and respond in a localized

Leveraging Institutional Complexity in Practice

49

fashion to institutional complexity. These actions were simply actors’ practical efforts to find ways around obstacles to getting the job done in an institutionally complex environment. Intriguingly, such situated improvising can unfold effects at the organizational and institutional levels. It may broaden a practitioners zone of competence and facilitate consolidation of an expanded practice repertoire (Smets & Jarzabkowski, forthcoming), or radiate to the level of the field and, under specific circumstances, rearrange dominant field-level logics (Smets et al., 2012). A third practical understanding is mutual adjustment (Jarzabkowski et al., 2009; Lindblom, 1965), in which individuals adjust in various ways to the logics governing each other’s actions, without the need for an overarching coordinating device or purpose. Jarzabkowski et al. (2009, p. 300) highlight mutual adjustment in a utilities company facing contradictory market and regulatory logics. In their study, actors working to different logics recognized their interdependence and therefore tried ‘to accommodate the other, advocating tolerance of the other’s position in relation to their own logic’ (ibid., p. 300). They show that pragmatic collaboration and recognition of interdependence between activities governed by divergent logics can guide practice as individuals adjust, in various ways and to varying degrees, in relation to one another. Their practice study thus complements and extends Reay and Hinings’ (2009) inter-organizational level findings about how truces between logics are constructed. A fourth practical understanding is switching. With the idea of switching resonates with ideas of compartmentalization (Kraatz & Block, 2008; Meyer & Rowan, 1977), in so much as different logics are separated into different divisions and actors. However, it is distinct insofar as it allows a single actor to shift between logics by enacting them at different times or in different spaces, so switching between compartmentalized logics. This practical understanding draws from DiMaggio’s (1997, p. 268) theorization of the capacity of individuals ‘to participate in multiple [inconsistent] cultural traditions’, and Delmestri’s (2006, p. 1518) depiction of individuals as ‘possible bearers of multiple institutionalized identities.’ Delmestri shows the capacity of managers to switch between different institutionalized modes of practice according to the way that different logics underpin their roles and identities. For example, managers that were exposed to greater institutional complexity were able to maintain multiple professional modes of practice and apply them in different contexts. Zilber (2011) provides another example of how this occurs in practice, indicating how the same actors used language and situational cues to switch between logics, and participate in and promulgate different discourses, at an industry

50

PAULA JARZABKOWSKI ET AL.

conference. Such ‘cues’ of language and meeting format connect to DiMaggio’s (1997) argument that switching is usually connected to a circumstantial trigger, such as language in Zilber’s case. In sum, this ability for an individual to switch between multiple logics, and thus participate in their associated practices, enables them to maintain what might otherwise be considered inconsistent modes of action.

Individuals in Action and Interaction While practice theorists have always focused on individuals through their actions and interactions (Jarzabkowski et al., 2007), institutional theory also suggests other, more cognitive and identity-based approaches to the individual that may also provide insights into institutional ambidexterity (Creed, Scully, & Austin, 2002; Greenwood et al., 2011; Lok, 2010; Rao et al., 2003). Such approaches suggest that personal identity and experiencebase shape how practitioners cope with institutional complexity, and that organizations can skilfully harness these personal characteristics. For instance, Delmestri (2006) identified middle managers’ personal self as a critical intervening variable in their engagement with local and foreign practices in multinational corporations. Smets et al. (2012) found that international law firms preferred more ‘cosmopolitan’ recruits who were multi-lingual, had lived or studied abroad and, as they were less wedded to a particular way of ‘doing things’, appeared to be more flexible in improvising around institutional contradictions. Likewise, Battilana and Dorado (2010) show that prior experience with – and attachment to – a particular logic hampered organizational hybridization in a microfinance bank. Here, recruits that arrived as a blank sheet, rather than with a fully formed ‘commercial’ or ‘social’ professional identity were considered more likely to develop ambidextrous capabilities to support organizational hybridization. Individual characteristics and experiences are thus important in the context of institutional ambidexterity. They shape individuals’ responses to specific logics and the complexities that arise at their interstices, and thereby help or hinder organizational engagement with coexisting and potentially contradictory logics. However, from a practice perspective such identity and experience are only antecedents to practical understandings and potential ambidextrous capacities. The individuals whose identities are being considered still have to be seen as individuals-in-interaction. To focus on them as individuals per se, in terms of their cognitive and person-centred characteristics, without

Leveraging Institutional Complexity in Practice

51

considering how these characteristics are situated within and shape the wider collective practice would be reductionist (Schatzki, 2002; Turner, 1994), taking institutionalists back to overly individualistic, disembedded or cognitive views of institutions and their complexities. This would not only constitute a step backwards in the evolution of institutional theory (Delbridge & Edwards, 2008; Delmestri, 2006; Hirsch & Lounsbury, 1997), but also fail to harness practice theory’s distinct ability to link individual-level action and interaction to the organizational and institutional dynamics that are critical for understanding institutional ambidexterity.

Organizational Level: Novel and Routine Complexity As shown in Fig. 1, the opportunity for ambidexterity arises when contradictory logics come together in an organizational context, generating practices that enact and construct institutional complexity. However, we still know little about different characteristics or forms of such ambidexterity, and their connection to different individual practices. Pache and Santos (2010) thus outline the need for increased understanding of when particular strategies may or may not be used in response to institutional complexity. Such studies could focus on a number of triggering elements at the organizational level, such as urgency (Smets et al., 2012), the extent to which complexity is internally or externally imposed (Pache & Santos, 2010) and the level of scrutiny (Aurini, 2006; Meyer & Rowan, 1977). However, these elements all indicate a focus on conditions of novel complexity that trigger a change in practical understanding. We argue, therefore, that there is a particular need for studies of organizational context that embody routine complexity, and comparisons with contexts of novel complexity. The majority of the institutional literature has focused on novel complexity, including those few studies that have focused on individual practice (Jarzabkowski et al., 2009; Smets & Jarzabkowski, forthcoming; Smets et al., 2012). In general, institutionalists have been alerted to institutional complexity by relatively recent clashes between commercial and professional logics. Consequently, their primary focus has been on moments of flux and crisis in which competing logics collide (e.g., Battilana & Dorado, 2010; Greenwood & Suddaby, 2006; Jarzabkowski et al., 2009; Reay & Hinings, 2009; Smets et al., 2012). While insightful, this may predispose a focus on urgency, contestation, crisis, and problematic understandings of complexity (Goodrick & Reay, 2011; Lawrence, Suddaby, & Leca, 2009). Instead, the call for greater attention to the

52

PAULA JARZABKOWSKI ET AL.

day-to-day actions of individuals at work (Powell & Colyvas, 2008) can help institutional theory as a whole resist the ‘search for drama’ (March, 1981, p. 564). While we urge further practice studies of the novel we therefore highlight the need for a sustained focus on the routine. In settings that have been characterized by institutional complexity for a sustained period of time, solutions may become ‘settled’ into everyday, taken-for-granted practical understandings of ambidexterity, characterized by an explicit lack of struggle and noise (Smets et al., 2012). As such, institutional complexities that form a well-rehearsed part of everyday practice may become unremarkable and mundane (Chia & Holt, 2009). Nonetheless, these routine practical understandings remain effortful accomplishments (Giddens, 1984) and clashes over the appropriate enactment of different logics can still occur. However, actors in such contexts may have a different set of practical understandings through which they both enact ambidexterity and also cope with the inevitable clashes that occur. To understand institutional ambidexterity as a routinized phenomenon, we thus need to study how institutional complexity is managed on an ongoing basis in organizations that have practiced and retained multiple, potentially competing logics over a long period of time. A practice approach is particularly suited for such a research agenda, because it is attuned to what people do when institutional complexity is settled into routinely enacted patterns of everyday working practice and, hence, less visible, than when they craft overtly strategic responses to novel complexity or institutional crisis (Chia & Holt, 2009).

Power and Politics Institutional theory has long acknowledged the importance of power and politics in organizational dynamics (DiMaggio, 1988; Jarzabkowski et al., 2009; Kostova & Zaheer, 1999). Indeed the concept of institutional complexity helped researchers highlight these issues (Greenwood et al., 2011). Despite the somewhat sanitized picture of routinized practice described above, we are usefully reminded that all contexts that generate ambidexterity involve power and politics in determining how and when to work within and between potential contradictions. It is apparent that the novel furnishes greater visibility of these concepts than the routine (e.g., Jarzabkowski et al., 2009). Nonetheless, one role of practice studies could be to surface the hidden, often forgotten, sources of power and politics at play, highlighting the persistence of such dynamics even when they are not readily

Leveraging Institutional Complexity in Practice

53

apparent. Concepts such as expanded repertoires and routinized practices are ‘power-laden’, even if they do not require the same degree of overt political negotiation as open conflict (Clegg, Courpasson, & Phillips, 2006). We thus expect that studies of ambidexterity might focus on the way that concepts of power and politics play out at the practical level of individuals in action and interaction and at the organizational level.

Institutional Complexity in Context: Characteristics of Collective Practice A practice approach to institutional complexity has the ability to connect the individual, organizational and institutional levels of analysis (Lounsbury & Crumley, 2007; Smets et al., 2012). We further argue that this practice approach can illuminate the nature of institutional complexity in its specific context. Specifically, we suggest that the characteristics of the collective practice in which logics come together influence how the nature of a particular institutional complexity is constructed. Smets and Jarzabkowski (forthcoming) show that actors, through their practice, can construct the same two logics and their associated practices as strange, contradictory, commensurable and complementary. A focus on practice thus transcends simplistic ideas that certain logics are compatible or contradictory per se. Rather, we argue that (in)compatibility is always conditioned by the specific situated practice in which those logics come together. For example, professional and commercial or managerial logics are likely to be more compatible in consulting or business service firms (Smets et al., 2012) than in healthcare (Reay & Hinings, 2009; Ruef & Scott, 1998). Similarly, professional and commercial logics are easier to hybridize within the freedom of practice inherent in private educational contexts (Aurini, 2006) than in the more restrictive and regulated practice of public education contexts (Hallett, 2010). In other examples, medical practitioners may have established practices and experiences that provide cost-effective social good, so that actors can affiliate with and move between the multiple logics that govern their organization. By contrast, a microfinance organization interested in blending commercial and community logics might need to employ practitioners with no prior affiliation with either logic (Battilana & Dorado, 2010). Hence, logics such as ‘community’, ‘commercial’ and ‘professional’ are not absolute in being either conflicting or complementary. Rather, they are interdependent in particular ways or constellations according to the particular collective and situated practice in which they are enacted.

54

PAULA JARZABKOWSKI ET AL.

CONCLUSION This paper has outlined a practice research agenda for studying institutional ambidexterity that links the individual, organizational and institutional levels within which ambidexterity is enacted. We have developed Fig. 1 as a conceptual model of our argument that is grounded in institutional theory, the literature on ambidexterity and the benefits of a practice approach in fulfilling the promise of institutional ambidexterity. In particular, we suggest that scholars should focus on individuals in action and interaction, as they develop practical understandings about how to cope with institutional ambidexterity. Furthermore, we highlight the benefits of comparing and contrasting contexts of novel and routinized organizational ambidexterity; in particular proposing that routinized ambidexterity provides a fertile avenue for future practice research. Finally, we propose that the collective practice that constitutes institutional complexity is particular to its specific context, as logics may be more or less contradictory or interdependent. Hence, actors will generate varying practices according to their particular contexts, in order to cope with and construct institutional ambidexterity in ways that go beyond our existing insights into institutional complexity. We therefore highlight the practice approach as a rich avenue for future research that can fulfil the promise of institutional ambidexterity as a solution to the problems, and enabler of the benefits, of institutional complexity. Future research is unlikely to discover much novel insight if it continues to do ‘more of the same’. Given the shift in focus that comes with a practice approach to institutional ambidexterity, commensurate methodological adjustments should be made so that a new research agenda is accompanied by a new methodological agenda. Recent practice-theoretical insights into institutional dynamics have already been derived from an increasing focus on micro-dynamics and the use of qualitative and ethnographic methods (Jarzabkowski et al., 2009; Smets et al., 2012; Zilber, 2002, 2011). We expect this trend to continue and intensify, making in-depth cases studies and real time ethnographies the methods of choice for the research agenda laid out here. As institutional theory is also concerned with embedded and longstanding patterns of social order, these real-time studies may need to be considered as critical incidents for drilling deep into institutional phenomena in action, even as they are, ideally, combined with more mainstream, large-scale institutional methods (Dunn & Jones, 2010; Greenwood et al., 2010; Ruef & Scott, 1998; Thornton, 2002). This combination holds the promise of providing the kind of multi-level data that is needed to bring a practice approach to institutional ambidexterity into its own by linking

Leveraging Institutional Complexity in Practice

55

individual, organizational and institutional levels of analysis (e.g. Lounsbury & Crumley, 2007; Smets et al., 2012). Further potential lies in the refinement of qualitative methods to make them more acutely sensitive to different aspects of practice. Promising avenues include a greater focus on the minutiae of practical actions and interactions, as advocated by micro ethnographers (Streeck & Mehus, 2005) or video ethnographers (LeBaron, 2005, 2008; LeBaron & Streeck, 1997). If the practices by which logics are enacted, rather than the logics themselves take the foreground, then the embodied interactions within which identities are formed and institutional divides bridged yield pertinent data (LeBaron, Glenn, & Thompson, 2009; Streeck, Goodwin, & LeBaron, 2011). Further, the contexts in which practices unfold are not only institutional but also mundanely material. People do things with ‘things’ in different places and spaces. While the role of boundary objects in bridging organizational divides has long been recognized (Carlile, 2002) and the affordances of technologies and materials has been recognized in practice studies (Leonardi, 2011; Orlikowski & Scott, 2008; Orlikowski, 2010), materials and spaces, their impact on interaction ,and on the instantiation of different logics has so far not been studied from an institutional perspective. Methodologies that look beyond the allocation of spaces (e.g., Kellogg, 2009) to consider their features and impacts on human interaction hold great promise for advancing a practice approach to institutional ambidexterity. Finally, the new research agenda not only opens opportunities for new methodologies, but also holds a new challenge. Studying established complexities and ‘settled’ responses is predicated on the researcher’s ability to observe the nexus of divergent logics as they come together in practice. Finding that ‘nexus’ – that point in an organization, or in time – at which the routinized coming together of contradictory logics can be studied in real time will be a critical methodological challenge. Especially where organizational responses to institutional complexity are settled and routinized, no ‘drama’ will signal a complexity of note that may appear worthy of study. By its routinized nature, it will be less visible and less immediately interesting and yet it is in precisely such everyday, settled enactment that we may gain the deepest insights into how actors perform institutional ambidexterity. It is thus vital that scholars generate the in-depth knowledge of an empirical context and the openness and interest to explore whatever ‘goes on’ quietly at the nexus of institutional logics if they are to uncover how institutional ambidexterity is accomplished in practice, particularly in contexts of routine complexity.

56

PAULA JARZABKOWSKI ET AL.

REFERENCES Almandoz, J. (2012). Arriving at the starting line: The impact of community and financial logics on new banking ventures. Academy of Management Journal, 55(6), 1381–1406 Andriopoulos, C., & Lewis, M. W. (2009). Exploitation-exploration tensions and organizational ambidexterity: Managing paradoxes of innovation. Organization Science, 20(4), 696–717. Aurini, J. (2006). Crafting legitimation projects: An institutional analysis of private education businesses. Sociological Forum, 21(1), 83–111. Barley, S. R. (2008). Coalface institutionalism. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The SAGE handbook of organizational institutionalism (pp. 490–515). London: Sage. Battilana, J., & Dorado, S. (2010). Building sustainable hybrid organizations: The case of commercial microfinance organizations. Academy of Management Journal, 53(6), 1419–1440. Benner, M. J., & Tushman, M. L. (2003). Exploitation, exploration, and process management: The productivity dilemma revisited. Academy of Management Review, 28(2), 238–256. Binder, A. (2007). For love and money: Organizations’ creative responses to multiple environmental logics. Theory & Society, 36(6), 547–571. Birkinshaw, J., & Gibson, C. B. (2004). Building ambidexterity into an organization. MIT Sloan Management Review, 45(4), 47–55. Bledow, R., Frese, M., Anderson, N., Erez, M., & Farr, J. (2009). A dialectic perspective on innovation: Conflicting demands, multiple pathways, and ambidexterity. Industrial and Organizational Psychology, 2(3), 305–337. Bourdieu, P. (1990). The logic of practice. Cambridge: Polity. Burgelman, R. A. (2002). Strategy as vector and inertia as coevolutionary lock-in. Administrative Science Quarterly, 47, 325–357. Carlile, P. R. (2002). A pragmatic view of knowledge and boundaries: Boundary objects in new product development. Organization Science, 13(4), 442–455. Carmeli, A., & Halevi, M. Y. (2009). How top management team behavioral integration and behavioural complexity enable organizational ambidexterity: The moderating role of contextual ambidexterity. The Leadership Quarterly, 20(2), 207–218. Chia, R. C. H., & Holt, R. (2009). Strategy without design: The silent efficacy of indirect action. Cambridge: Cambridge University Press. Clegg, S. R., Courpasson, D., & Phillips, N. (2006). Power and organizations. Thousand Oaks, CA: Sage. Cook, S., & Brown, J. (1999). Bridging epistemologies: The generative dance between organizational knowledge and organizational knowing. Organization Science, 10(4), 381–400. Creed, W. E. D., Scully, M. A., & Austin, J. R. (2002). Clothes make the person? The tailoring of legitimating accounts and the social construction of identity. Organizational Science, 13(5), 475–496. Delbridge, R., & Edwards, T. (2008). Challenging conventions: Roles and processes during nonisomorphic institutional change. Human Relations, 61(3), 299–325. Delmestri, G. (2006). Streams of inconsistent institutional influences: Middle managers as carriers of multiple identities. Human Relations, 59(11), 1515–1541. Denis, J. L., Langley, A., & Rouleau, L. (2007). Strategizing in pluralistic contexts: Rethinking theoretical frames. Human Relations, 60(1), 179–215.

Leveraging Institutional Complexity in Practice

57

DiMaggio, P. (1997). Culture and cognition. Annual Review of Sociology, 23(1), 263. DiMaggio, P. J. (1988). Interest and agency in institutional theory. In L. G. Zucker (Ed.), Institutional patterns and organizations: Culture and environment. Cambridge, MA: Ballinger. Djelic, M.-L., & Quack, S. (2003). Conclusion: Globalization as a double process of institutional change and institution building. In M. L. Djelic & S. Quack (Eds.), Globalization and institutions (pp. 302–334). Cheltenham: Edward Elgar. Dunn, M. B., & Jones, C. (2010). Institutional logics and institutional pluralism: The contestation of care and science logics in medical education, 1967–2005. Administrative Science Quarterly, 55(1), 114–149. D’Aunno, T., Sutton, R. I., & Price, R. N. (1991). Isomorphism and external support in conflicting institutional environments: The case of drug abuse treatment units. Academy of Management Journal, 34(3), 636–661. D’Este, P., & Perkmann, M. (2011). Why do academics engage with industry? The entrepreneurial university and individual motivations. Journal of Technology Transfer, 36(3), 316–339. Edelman, L. B. (1992). Legal ambiguity and symbolic structures: Organizational mediation of civil rights law. American Journal of Sociology, 97(6), 1531–1576. Friedland, R., & Alford, R. R. (1991). Bringing society back in: Symbols, practices and institutional contradictions. In W. W. Powell & P. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 232–263). Chicago, IL: University of Chicago Press. Gabbioneta, C., Greenwood, R., Mazzola, P., & Minoja, M. (forthcoming). The influence of the institutional context on corporate illegality. Accounting, Organizations & Society. Garud, R., Jain, S., & Kumaraswamy, A. (2002). Institutional entrepreneurship in the sponsorship of common technological standards: The case of Sun Microsystems and Java. Academy of Management Journal, 45(1), 196–214. Gibson, C. B., & Birkinshaw, J. (2004). The antecedents, consequences, and mediating role of organizational ambidexterity. The Academy of Management Journal, 47(2), 209–226. Giddens, A. (1984). The constitution of society: Outline of the theory of structuration. Cambridge: Polity. Goodrick, E., & Reay, T. (2011). Constellations of institutional logics. Work and Occupations, 38(3), 372–416. Greenwood, R., Maga`n Diaz, A., Li, S., & Ce´spedes Lorente, J. (2010). The multiplicity of institutional logics and the heterogeneity of organizational responses. Organization Science, 21(2), 521–539. Greenwood, R., Raynard, M., Kodeih, F., Micellota, E., & Lounsbury, M. (2011). Institutional complexity and organizational responses. Annals of the Academy of Management, 5(1), 1–55. Greenwood, R., & Suddaby, R. (2006). Institutional entrepreneurship in mature fields: The big five accounting firms. [Article]. Academy of Management Journal, 49(1), 27–48. Grey, C. (2003). The real world of Enron’s auditors. Organization, 10(3), 572–576. Gupta, A. K., Smith, K. G., & Shalley, C. E. (2006). The interplay between exploration and exploitation. Academy of Management Journal, 49(6), 693–706. Hallett, T. (2010). The myth incarnate: Recoupling processes, turmoil, and inhabited institutions in an urban elementary school. American Sociological Review, 75(1), 52–74. He, Z., & Wong, P. K. (2004). Exploration vs. exploitation: An empirical test of the ambidexterity hypothesis. Organization Science, 15, 481–494.

58

PAULA JARZABKOWSKI ET AL.

Heimer, C. A. (1999). Competing institutions: Law, medicine, and family in neonatal intensive care. Law & Society Review, 33(1), 17–66. Hirsch, P. M., & Lounsbury, M. (1997). Ending the family quarrel: Toward a reconciliation of ‘‘old’’ and ‘‘new’’ institutionalisms. The American Behavioral Scientist, 40(4), 406–418. Jansen, J. J., Tempelaar, M. P., van den Bosch, F. A. J., & Volberda, H. W. (2009). Structural differentiation and ambidexterity: The mediating role of integration mechanisms. Organization Science, 20(4), 797–811. Jansen, J. J. P., van de Bosch, F. A. J., & Volberda, H. W. (2005). Exploratory innovation, exploitative innovation and ambidexterity: The impact of environmental and organizational antecedents. Schmalenbach Business Review, 57, 351–363. Jansen, J. J. P., van de Bosch, F. A. J., & Volberda, H. W. (2006). Exploratory innovation, exploitative innovation, and performance: Effects of organizational antecedents and environmental moderators. Management Science, 52(11), 1661–1674. Jarzabkowski, P. (2004). Strategy as practice: Recursive, adaptive and practices-in-use. Organization Studies, 25(4), 529–560. Jarzabkowski, P. (2005). Strategy as practice: An activity-based approach. London: Sage. Jarzabkowski, P., Balogun, J., & Seidl, D. (2007). Strategizing: The challenges of a practice perspective. Human Relations, 60(1), 5–27. Jarzabkowski, P., Matthiesen, J., & Van de Ven, A. (2009). Doing which work? A practice approach to institutional pluralism. In T. B. Lawrence, R. Suddaby & B. Leca (Eds.), Institutional work: Actors and agency in institutional studies of organizations (pp. 284– 316). Cambridge: Cambridge University Press. Kellogg, K. C. (2009). Operating room: Relational spaces and micro institutional change in surgery. American Journal of Sociology, 115(3), 657–711. Kitchener, M. (2002). Mobilizing the logic of managerialism in professional fields: The case of academic health centre mergers. Organization Studies, 23(3), 391–420. Kostova, T., & Zaheer, S. (1999). Organizational legitimacy under conditions of complexity: The case of the multinational enterprise. Academy of Management Review, 24(1), 64–81. Kraatz, M. S., & Block, E. (2008). Organizational implications of institutional pluralism. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The SAGE handbook of organizational institutionalism (pp. 243–275). London: Sage. Lawrence, T. B., Suddaby, R., & Leca, B. (2009). Introduction: Theorizing and studying institutional work. In T. Lawrence, R. Suddaby & B. Leca (Eds.), Institutional work: Actors and agency in institutional studies of organizations (pp. 1–28). Cambridge: Cambridge University Press. LeBaron, C. (2005). Considering the social and material surround: Toward microethnographic understandings of nonverbal behavior. In V. Manusov (Ed.), The sourcebook of nonverbal measures (pp. 493–506). Mahwah, NJ: Lawrence Erlbaum Associates. LeBaron, C. (2008). Video-based methods for research on strategy as practice: Looking at people, places and things. Key note presented at the professional development workshop: Strategy as practice: Methodological challenges. Paper presented at the Academy of Management Meeting, Anaheim. LeBaron, C., Glenn, P., & Thompson, M. P. (2009). Identity work during boundary moments: Managing positive identities through talk and embodied interaction. In L. M. Roberts & J. E. Dutton (Eds.), Exploring positive identities and organizations: Building a theoretical and research foundations (pp. 191–215). New York, NY: Psychology Press.

Leveraging Institutional Complexity in Practice

59

LeBaron, C., & Streeck, J. (1997). Built space and the interactional framing of experience during a murder interrogation. Human Studies, 20, 1–25. Leonardi, P. (2011). When flexible routines meet flexible technologies: Affordance, constraint, and the imbrications of human and material agencies. Management Information Systems Quarterly, 35(1), 147–167. Levinthal, D., & March, J. (1993). Myopia of learning. Strategic Management Journal, 14, 95–112. Lindblom, C. E. (1965). The intelligence of democracy: Decision making through mutual adjustment. New York, NY: The Free Press. Lok, J. (2010). Institutional logics as identity projects. Academy of Management Journal, 53(6), 1305–1335. Lounsbury, M. (2007). A tale of two cities: Competing logics and practice variation in the professionalizing of mutual funds. Academy of Management Journal, 50(2), 289–307. Lounsbury, M., & Crumley, E. T. (2007). New practice creation: An institutional perspective on innovation. Organization Studies, 28(7), 993–1012. Lubatkin, A. S., Simsek, Z., Yan, L., & Veiga, J. F. (2006). Ambidexterity and performance in small- to medium-sized firms: The pivotal role of top management team behavioral integration. Journal of Management Inquiry, 32, 626–672. March, J. (1991). Exploration and exploitation in organizational learning. Organization Science, 2(1), 71–87. March, J. G. (1981). Footnotes to organizational change. Administrative Science Quarterly, 26(4), 563–577. Marquis, C., & Lounsbury, M. (2007). Vive la resistance: Competing logics and the consolidation of US community banking. Academy of Management Journal, 50(4), 799–820. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363. Mom, T. J. M., Van Den Bosch, F. A. J., & Volberda, H. W. (2007). Investigating managers’ exploration and exploitation activities: The influence of top-down, bottom-up, and horizontal knowledge inflows. Journal of Management Studies, 44(6), 910–931. Oliver, C. (1991). Strategic responses to institutional processes. Academy of Management Review, 16(1), 145–179. O’Reilly, C. A., & Tushman, M. L. (2004). The ambidextrous organization. Harvard Business Review, 84(4), 74–81. Orlikowski, W., & Scott, S. (2008). Sociomateriality: Challenging the separation of technology, work and organization. Academy of Management Annals, 2(1), 433–474. Orlikowski, W. J. (2010). The sociomateriality of organisational life: Considering technology in management research. Cambridge Journal of Economics, 34, 125–141. Pache, A., & Santos, F. (2010). When worlds collide: The internal dynamics of organizational responses to conflicting institutional demands. Academy of Management Journal, 35(3), 455–476. Perkmann, M., & Walsh, K. (2009). The two faces of collaboration: Impacts of universityindustry relations on public research. Industrial and Corporate Change, 18(6), 1033–1065. Powell, W. W., & Colyvas, J. A. (2008). Microfoundations of institutional theory. In R. Greenwood, C. Oliver, K. Sahlin-Andersson & R. Suddaby (Eds.), The SAGE handbook of organizational institutionalism (pp. 276–298). London: Sage.

60

PAULA JARZABKOWSKI ET AL.

Raisch, S., & Birkinshaw, J. (2008). Organizational ambidexterity: Antecedents, outcomes and moderators. Journal of Management, 34(3), 375–409. Raisch, S., Birkinshaw, J., Probst, G., & Tushman, M. L. (2009). Organizational ambidexterity: Balancing exploitation and exploration for sustained performance. Organization Science, 20(4), 685–695. Rao, H., Monin, P., & Durand, R. (2003). Institutional change in Toque Ville: Nouvelle cuisine as an identity movement in French gastronomy. American Journal of Sociology, 108(4), 795–843. Reay, T., & Hinings, C. R. (2005). The recomposition of an organizational field: Health care in Alberta. Organization Studies, 26(3), 351–384. Reay, T., & Hinings, C. R. (2009). Managing the rivalry of competing institutional logics. Organization Studies, 30(6), 629–652. Rothaermel, F. T., & Deeds, D. L. (2006). Alliance type, alliance experience and alliance management capability in high-technology ventures. Journal of Business Venturing, 21(4), 429–460. Ruef, M., & Scott, W. R. (1998). A multidimensional model of organizational legitimacy: Hospital survival in changing institutional environments. Administrative Science Quarterly, 43(4), 877. Schatzki, T. (2002). The site of the social: A philosophical account of the constitution of social life and change. University Park, PA: Penn State University Press. Schatzki, T. R. (2001). Introduction: Practice theory. In T. R. Schatzki, K. Knorr-Cetina & E. V. Savigny (Eds.), The practice turn in contemporary theory (pp. 1–14). London: Routledge. Scott, W. R. (1991). Unpacking institutional arguments. In W. W. Powell & P. J. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 164–182). Chicago, IL: University of Chicago Press. Seo, M.-G., & Creed, W. E. D. (2002). Institutional contradictions, praxis, and institutional change: A dialectical perspective. Academy of Management Review, 27(2), 222–247. Siggelkow, N., & Levinthal, D. (2003). Temporarily divide to conquer: Centralized, decentralized, and reintegrated organizational approach to exploration and adaptation. Organization Science, 14(6), 650–669. Simsek, Z. (2009). Organizational ambidexterity: Towards a multilevel understanding. Journal of Management Studies, 46, 597–624. Smets, M., & Jarzabkowski, P. (forthcoming). (Re)-constructing institutional complexity in practice: A relational model of institutional work and complexity. Human Relations. Smets, M., Morris, T., & Greenwood, R. (2012). From practice to field: A multi-level model of practice-driven institutional change. Academy of Management Journal, 55(4), 877–904. Smith, W. K., Binns, A., & Tushman, M. L. (2010). Complex business models: Managing strategic paradoxes simultaneously. Long Range Planning, 43, 448–461. Smith, W. K., & Tushman, M. L. (2005). Managing strategic contradictions: A top management model for managing innovation streams. Organization Science, 16(5), 522–536. Streeck, J., Goodwin, C., & LeBaron, C. (2011). Embodied interaction: Language and body in the material world. Cambridge: Cambridge University Press. Streeck, J., & Mehus, S. (2005). Microethnography: The study of practices. In K. L. Fitch & R. S. Sanders (Eds.), Handbook of language and social interaction (pp. 381–404). Mahwah, NJ: Lawrence Erlbaum Associates.

Leveraging Institutional Complexity in Practice

61

Suddaby, R., & Greenwood, R. (2005). Rhetorical strategies of legitimacy. Administrative Science Quarterly, 50(1), 35–67. Swidler, A. (1986). Culture in action: Symbols and strategies. [Article]. American Sociological Review, 51(2), 273–286. Sztompka, P. (1991). Society in action: The theory of social becoming. Chicago, IL: University of Chicago Press. Thornton, P. H. (2002). The rise of the corporation in a craft industry: Conflict and conformity In institutional logics. Academy of Management Journal, 45(1), 81–101. Thornton, P. H., Jones, C., & Kury, K. (2005). Institutional logics and institutional change in organizations: Transformation in accounting, architecture, and publishing. In M. Lounsbury (Ed.), Transformation in cultural industries (pp. 125–170). Amsterdam: Elsevier. Thornton, P. H., & Ocasio, W. (2008). Institutional logics. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The SAGE handbook of organizational institutionalism (pp. 99–129). Singapore: Sage. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure and process. Oxford: Oxford University Press. Townley, B. (1997). The institutional logic of performance appraisal. Organization Studies, 18(2), 261–285. doi: 10.1177/017084069701800204 Townley, B. (2002). The role of competing rationalities in institutional change. Academy of Management Journal, 45(1), 163–179. Turner, S. (1994). The social theory of practices. Cambridge, UK: Polity Press. Tushman, M. L., & O’Reilly, C. A. (1996). Ambidextrous organizations: Managing evolutionary and revolutionary change. California Management Review, 38(4), 8–30. Whittington, R. (2006). Completing the practice turn in strategy research. Organization Studies, 27(27), 613–634. Zilber, T. B. (2002). Institutionalization as an interplay between actions, meanings, and actors: The case of a rape crisis center in Israel. Academy of Management Journal, 45(1), 234–254. Zilber, T. B. (2011). Institutional multiplicity in practice: A tale of two high-tech conferences in Israel. Organization Science, 22(6), 1539–1559.

BEYOND THE FAMILY FIRM: REASSERTING THE INFLUENCE OF THE FAMILY INSTITUTIONAL LOGIC ACROSS ORGANIZATIONS Samantha Fairclough and Evelyn R. Micelotta ABSTRACT In this paper, we draw attention to the influence of the institutional logic of family upon a broad range of organizations, and argue that their significance has not been fully realized within the realm of institutional theory and research. Drawing on extant literature, publicly available documents, and interview data, we highlight the prominence of the family logic in the Italian legal sector, where there is a dearth of family firms and the existence of a competing professional logic that interacts with the familial logic. Our research suggests that, even in a setting where logics of capitalism and profession dominate, the organizational form and practices of Italian law firms are significantly influenced by the family logic: firms remain small, resistant to mergers and forms of internationalization, and have successfully resisted the encroachment of invading foreign legal practices. We discuss the significance of the family logic and its

Institutional Logics in Action, Part B Research in the Sociology of Organizations, Volume 39B, 63–98 Copyright r 2013 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0733-558X/doi:10.1108/S0733-558X(2013)0039A&B016

63

64

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

manifestations in organizations, and map out future research directions about how multiple logics interact and reinforce each other in national settings. Keywords: Institutional logics; Italy; logics of family; professional service firms

INTRODUCTION Since Friedland and Alford (1991) introduced the notion of logic to describe the socially shared beliefs, practices, and rules which constitute the institutional foundations of societies, the institutional logics perspective has become an important focal point for scholarly development in sociology and organization theory (Greenwood, Oliver, Sahlin-Andersson, & Suddaby, 2008; Thornton & Ocasio, 2008; Thornton, Ocasio, & Lounsbury, 2012). Drawing upon the idea that logics shape the interests, values, identities, and behaviors of both individuals and organizations, an increasing body of research has highlighted how potentially incompatible social prescriptions and assumptions exert a significant influence upon the activities, structures, and decision-making processes of organizations (see reviews by Thornton & Ocasio, 2008 and Greenwood, Raynard, Kodeih, Micelotta, & Lounsbury, 2011). How organizations make sense of, and react to, the complexity arising from multiple logics has become a key strand in the development of the institutional logics perspective (Greenwood et al., 2011; Thornton et al., 2012). While the growing body of institutional logics scholarship has unquestionably illuminated the profound effects of logics on organizational action (e.g., Battilana & Dorado, 2010; Dunn & Jones, 2010; Purdy & Gray, 2009; Reay & Hinings, 2009; Thornton & Ocasio, 1999), the interests of institutional scholars have so far been selective with regard to the examination of certain types of logics. The logics of professions, corporations and markets have attracted both theoretical attention and empirical investigation, yet scant consideration has been paid to the influence of nonmarket logics – such as those of the family, state, and religion – upon organizational forms and practices. Scholars’ inattention to the institutional logic of family and its influence upon organizational structures, practices, and strategies is particularly notable. Of the few studies which have considered the family logic (e.g., Greenwood, Magan, Li, & Lorente, 2010a;

Beyond the Family Firm

65

Miller, Le Breton-Miller, & Lester, 2011), these have mostly been confined to an examination of how it affects family-owned firms, thus failing to consider how the family logic shapes the behavior of a wider array of organizations within society, including those which are not organized and managed according to principles of kin ownership. In this paper, we seek to address this oversight by theorizing how the societal-level family logic affects organizations in industry sectors where there are very few family firms. Specifically, we present an illustrative case of the legal sector in a country where the logic of family is particularly strong – Italy. We describe how the organizational form and strategic practices of Italian law firms are impacted by a cultural setting in which the familial logic is strongly endorsed by the state and underpinned by the dominant Catholic religious tradition. This logic, which emphasizes norms and values, such as loyalty, reciprocation, altruism, close personal relationships, and the cultivation of trust-based networks, operates at a societal level (Miller et al., 2011; Thornton et al., 2012). Given that organizations in the Italian legal sector are rarely family-owned or managed, we argue that the logic of family has a broader application than currently acknowledged by institutional researchers. That is, a variety of organizations – not just family firms – will be influenced by the family logic, including those where familial altruism and kin succession are restricted by professional values which stipulate merit-based promotion and exacting educational and licensing requirements. Drawing upon recent interest in the coexistence of multiple, potentially incompatible, logics (Dunn & Jones, 2010; Greenwood et al., 2011; Kraatz & Block, 2008; Purdy & Gray, 2009; Zilber, 2011), we also suggest that the historical coevolution of family, professional and national logics has enabled law firms in Italy to preserve a number of specific national organizational ‘‘traits,’’ despite the pressure for change arising from international market forces. These include limitations on their growth, resulting in a population of small, domestically focused firms, whose preference is to resist mergers or other forms of joint enterprise with both domestic and international legal practices outside of their established networks. Although embedded in market and professional institutional logics, Italian law firms’ organizational responses are influenced by prevailing societal notions relating to the importance of family ties, reciprocity between kin and kin-like relations, and mutuality between trusted associates. In other words, our suggestion is that multiple institutional logics interact in complex ways and that their influence upon organizational structure and practice will be reflective of the varying degrees

66

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

to which several logics reciprocally influence each other. Our aim in highlighting the effect of the family logic on organizational behavior – and its interaction with market and professional logics – is to encourage researchers to take account of the profound influence of this and other nonmarket institutions upon the ecology of organizations and national markets. Our paper is organized as follows. We begin by reviewing the literature on institutional logics, with a particular focus upon extant conceptualizations of the logic of ‘‘family.’’ We then describe the circumstances of our case and explain how the family logic in Italy shapes the social and market context in which law firms operate. In particular, we elaborate on how the family logic has significantly influenced Italian lawyers’ conceptions of appropriateness in terms of the structure of professional legal organizations, the behavior and culture of professionals, and the legitimacy of their firms’ internationalization strategies. We conclude with a discussion of the implications of our analysis for institutional research. We draw attention to the significance of the family logic to comprehensive institutional analyses of organizations and to considerations of the effects of the interplay of multiple institutional logics in national and global contexts.

THEORETICAL OVERVIEW Institutional Logics Friedland and Alford (1991) introduced the concept of institutional logics to reveal fundamental organizing principles of societies, where institutions ‘‘have an order and they are organized quite specifically by what they term institutional orders, Christian religions, the nuclear family, the bureaucratic state, the capitalist market, and democracy’’ (Thornton et al., 2012, p. 43). Institutional logics are the symbolic elements and associated material practices that are used to construct the social and institutional context for behavior within and across multiple levels of analysis, from the individual to the organizational field, and in society more broadly (Thornton & Ocasio, 2008; Thornton et al., 2012). While organizations, fields and industries may have specific manifestations of logics that are salient at those levels (Spicer & Sewell, 2010; Thornton, 2004; Thornton & Ocasio, 2008), the higher order nature of logics implies that their influence should find expression across multiple industries and fields.

Beyond the Family Firm

67

Although an insightful and influential piece of theory building, Friedland and Alford’s conceptualization of an interinstitutional system of societal logics had limited application to the work of organizational theorists until Thornton (2004) elaborated its dimensions and identified the categorical elements of each logic. Importantly, Thornton reconfigured the original typology to reduce its abstractness, thus enabling scholars to more easily investigate empirically the core insights of Friedland and Alford’s piece. The revised typology included the logic of the profession and of the corporation as central organizing principles within society, and was used to empirically demonstrate how shifts in institutional logics explain patterns of change in organizational form and practice (Thornton & Ocasio, 1999). Since then, further empirical studies have successfully unpacked the effects of institutional logics across multiple levels of analysis (for reviews, see Greenwood et al., 2011; Thornton & Ocasio, 2008; Thornton et al., 2012). For example, institutional logics have been shown to influence the transformation of institutional fields (e.g., Marquis & Lounsbury, 2007; Scott, Ruef, Mendel, & Caronna, 2000), affect the emergence and selection of structures at the organizational level (e.g., Haveman & Rao, 1997), and sustain the development of distinctive identities and practices (Lok, 2010; Smets, Morris, & Greenwood, 2012). However, organizational scholars have tended to focus their attention upon the logics of the market, corporation and profession because these have become established as the underlying ‘‘motors’’ of profound institutional change in fields, industries, and upon single organizations. In effect, the institutional logics perspective has given organizational scholars and economic sociologists a set of powerful theoretical tools to generate alternative explanations to traditional economically derived accounts of the functioning of organizations (Lounsbury, 2008). In particular, a number of studies have examined the pressure of new market logics in environments traditionally dominated by other forms of logic, which have enabled scholars to provide an explanation for, inter alia, institutional changes in US health care (Scott et al., 2000), the repositioning of managerial attention in the US higher education publishing industry (Thornton, 2004), and resistance to institutional change in the US field of finance (Marquis & Lounsbury, 2007). Additionally, the institutional logics perspective has rekindled interest in the professions, whose theorists have posited the existence and influence of professional norms and mutual understandings, which include established patterns of professional authority and autonomy, and jurisdictional claims about professionals’ rights to apply expert and esoteric knowledge to

68

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

specific types of work (Abbott, 1988; Brint, 1996; Lounsbury, 2002). As Friedson (2001, p. 12) notes, ‘‘the word ‘professionalism’yrefer[s] to the institutional circumstances in which members of occupations rather than consumers or managers control work.’’ As a set of higher order principles guiding social action, professional logic is purported to exert a profound influence upon the organizational behavior of professionals, and upon the structures and practices of professional service firms (Cooper, Hinings, Greenwood, & Brown, 1996; Goodrick & Reay, 2011; Kitchener, 2002; Rao, Monin, & Durand, 2003; Reay & Hinings, 2005, 2009). Given the popularity of both theoretical and empirical inquiry into the nature and effects of the market and the professions, other core institutional logics such as the family, the state, and religion have been somewhat overlooked. This critical inattention has been recently acknowledged by institutional scholars. Greenwood and his colleagues, for example, bemoan the relative neglect afforded to nonmarket institutional logics: ‘‘ywe know relatively little about the influence of institutions such as the familyyon contemporary organizations’’ (Greenwood et al., 2010a, p. 521). Likewise, Miller et al. (2011) maintain that the study of logics has underplayed the institution of the family as one of the constitutive forces shaping individual, organizational and societal behavior. More generally, a number of scholars have called for more studies on nonmarket institutions within an appropriate historical and cultural framework in order to understand their broader impact upon local and national contexts (e.g., Colli, Fernandez Perez, & Rose, 2003; James, 2006). Clearly, there is a need to understand when and how culturally rooted institutional logics shape the form and practices of organizations and fields. In this regard, the recent elaboration of ‘‘community’’ as a component of the interinstitutional system (Thornton et al., 2012) and researchers’ burgeoning interest in the coexistence of potentially incompatible logics – referred to variously as institutional pluralism (Kraatz & Block, 2008), institutional complexity (Greenwood et al., 2011), and institutional multiplicity (Zilber, 2011) – provide a fertile theoretical milieu in which to explore not just family as a singular institutional logic, but also the interaction between logics. In this paper we dive into this milieu to examine the impact of the family logic upon the organizational forms, practices, and strategies of organizations within a professional field – law – in a national ‘‘community’’ – Italy. In the following section, we review how extant research has examined the logic of family, and argue that these accounts provide only a partial and unsatisfactory statement of their widespread influence.

Beyond the Family Firm

69

Where is the Logic of Family in Institutional Theory? Friedland and Alford (1991, p. 248) identified the institutional logic of the family as constituting ‘‘a set of cultural rules and assumptions associated with notions of community and unconditional loyalty to family members and their reproductive needs.’’ This familial logic has been characterized as one of ‘‘nurturing, generativity, and loyalty to the family’’ (Miller et al., 2011, p. 4). Although the existence of the institutional logic of family is posited at a supra-organizational level and its function in shaping the structure of institutional fields and in determining actors’ attention to issues and solutions is established (Thornton et al., 2012), investigations of the impact of family logic upon organizations has been narrowly focused and selective. That is, the logic of family has been consistently associated with organizational orientations and behaviors which emphasize family-owned enterprises, the provision of family members with job security and control within a firm, and other forms of familial loyalty in business relationships (e.g., Karra, Tracey, & Phillips, 2006; Lubatkin, Ling, & Schulze, 2007). To illustrate, Thornton (2004) linked the logic of family to the establishment of family enterprises as both an organizational form and an identity, and proposed that familial reputation, honor and security are the basis of these firms’ organizational norms and strategies. More recently, Miller et al. (2011) empirically demonstrated how familial contexts, role identities and institutional logics elicit a ‘‘conservation’’ strategy within family firms, which seek to stabilize cash flows and limit risk by minimizing investments in research and development, promote cash holdings, and distribute profits to family members rather than reinvest them. They found support for the hypothesis that family firms’ strategies are shaped by values and goals favoring the provision of stable, secure incomes and careers to family members, and the preservation of familial control. In the same vein, Greenwood et al. (2010a) find that family firms in Spain are motivated by values different to those of nonfamily firms in terms of their commitment to providing job security for their members, and are less likely to engage in practices such as downsizing, which could jeopardize valuable familial relationships. These studies are not only consistent in their results – that family logics influence strategic and organizational practices – but also support similar assumptions in research emerging from the rapidly growing academic field of family business studies (Chrisman, Chua, & Sharma, 2005; Sharma, Chrisman, & Gersick, 2012; Zahra & Sharma, 2004): that the primary – or only – context in which to find and illuminate the institutional logic of

70

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

family is the examination of family-owned businesses. Institutional researchers have followed in the footsteps of family business research by implicitly reinforcing the notion that familial logic is integral to these firms’ practices and that it shapes only family firms’ behaviors and strategic choices. In particular, the institutional lens has provided an intuitive and convenient description of how family firms attempt to preserve social and emotional capital ahead of economic or monetary concerns (Berrone, Cruz, & Gomez-Mejia, 2012; Gomez-Mejia, Cruz, Berrone, & De Castro, 2011), and how family business behavior is characterized by a high degree of altruism (e.g., Schulze, Lubatkin, & Dino, 2003). Although enlightening, these investigations of the institutional logic of the family within family firms neglect the potential of family logic to affect the behavior of organizations more broadly, and in particular where there is no family involvement. Miller et al.’s (2011) research examined firms in which family members were the largest shareholders or served as Chief Executive Officer, thus saying nothing about the effects of family logic on nonfamily controlled firms. Similarly, Greenwood et al.’s (2010a) study claims to elucidate the effects of higher-level family logic on a broad range of firms in Spain, yet their hypotheses and findings highlight the fact that it is familymanaged firms who are less likely to engage in practices legitimated by the state but in opposition to familial logics, such as downsizing. Their findings do not explicitly draw attention to the effect of family logics on organizations which are not family managed. Research examining the influence that familial logic exerts upon nonfamily firms is indeed sparse. A research ‘‘note’’ by Bhappu (2000) is the clearest example we have found, which describes an anthropological study of how an ancient Japanese feudal system connects notions of family with the institutional logics underpinning the operation of distinctive corporate networks and management practices in modern-day Japan. Bhappu investigates how the legacy of an ancient familial form has affected the development of corporate social structures and employees’ psychological attachments to their work organizations. In particular, the study shows how Japanese keiretsu networks are still based on social ties that were established among individuals in a core group of family firms and their subsidiaries (Gerlach, 1992), even though members of these firms are no longer related by blood or marriage. To us, this is an important indication that although a familial logic finds typical expression in family-owned firms, its influence is exerted upon a wider array of organizations, in some cases even after families and family firms have long since vanished. As Bhappu (2000) reveals, the family logic lingers in Japanese society and continues to have a

Beyond the Family Firm

71

significant influence upon the coordination and relationships of corporate networks, which ‘‘have come to represent the way Japanese organizations should be structured’’ (Bhappu, 2000, p. 413). In addition to supporting our argument that research examining the institutional logic of family should not be confined to the domain of familyowned or family-managed businesses, Bhappu’s (2000) study further suggests that family logic finds different expression across different societies. Understanding the cultural heterogeneity of institutional contexts has been suggested as an important area of research that should integrate an institutional logics perspective with the historical and comparative analysis of institutions (Thornton et al., 2012; see also Djelic, 2010 for a review). This is unquestionably an intersection of institutional themes which requires examination in greater detail because family logic in some settings  particularly within non-Anglo-Saxon countries, and those with a strong Catholic tradition  appears to be fertile ground for strong familial influences and may even explain the historical foundations of countryspecific organizational governance structures, practices and, more generally, societal institutional arrangements. Our paper also heeds calls to examine the ‘‘cultural’’ contingency of institutional logics and to extend their investigation within contexts beyond the Anglo-American world (e.g., Biggart & Hamilton, 1990; Guille´n, 2001; Orru, 1991; Thornton & Ocasio, 2008). Institutional research is often criticized for its distinctly US-centric perspective (e.g., Scott, 2005) and, while recent work has attended to this concern (e.g., Greenwood et al., 2010a), there is still much to discover about how culturally inspired institutional logics underpin the appropriateness of organizational behaviors, structures, and practices.

THE ILLUSTRATIVE CASE OF ITALY AND THE ITALIAN LEGAL SECTOR We illustrate how a society-wide institutional logic of family affects organizations beyond those managed or owned by family members using the case of a specific category of professional service firms within a distinctive national context  law firms in Italy. Italy is an exemplar of an institutional environment where family values are prominent, and where the legal profession is not typically constituted by family-owned firms. The following discussion is based upon our previous work and research on

72

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

professional service firms (e.g., Greenwood, Morris, Fairclough, & Boussebaa, 2010b), and investigations of Italian professions (Micelotta, 2010; Micelotta & Washington, 2011), as well as our personal experiences of working and living in Italy. We undertook a wide-ranging review of relevant academic literature, media sources, and publicly available documents such as economic and cultural surveys (e.g., The Economist; European Commission Reports, 2006, and OECD publications), in order to understand the potential role played by family logics in Italian society. In particular, we gathered data relating to the Italian legal sector by searching for relevant articles on news databases, such as Factiva, and collecting archival material from Italian newspapers and magazines, for example, Il Sole 24 Ore, Il Mondo. We also conducted interviews with senior members of four Italian law firms, five Anglo-Saxon law firms operating in Italy, and four multinational organizations with experience as clients of both Italian and international law firms.

The Logic of Family in Italian Society The significant influence of family logic upon the economic development, structures and processes of business activity in Italy has been discussed extensively in several studies by Italian economists and sociologists (Ardigo & Donati, 1976; Bianchi, Garofoli, & Mazzoni, 1994; Compagno, 2011; Paci, 1980). Among the multiple logics which exist in the Italian institutional environment, those which embody notions of family – in particular, the continuity of the familial tradition via direct influence and kinship succession, and the nurturing of benefits and trust among relatives and other trusted affiliates – have been shown to significantly shape patterns of individual and organizational behavior. These familial norms, values, and beliefs may arise from a number of sources, but we note the importance accorded to the family in Italian religious beliefs, and concomitant cultural and moral support for the traditional nuclear family unit, as well as to extended family members. According to recent surveys, approximately 77% of the Italian population identify themselves as adherents to the Catholic religious faith (e.g., Eurispes Report, 2011). Although only 24.4% claim to be active churchgoers, respect for the doctrine and teachings of the Catholic religion remains strong, with 80% of Italian parents committed to raising their children according to Catholic principles. In addition, as the ‘‘home’’ of the Roman Catholic Church (The Vatican), Italy is strongly influenced by a

Beyond the Family Firm

73

religio-cultural value system which promotes traditional conceptions of family based upon heterosexual marriage, love, loyalty, trust, the nurturing of offspring, and paternalism. Papal teachings, for example, have condemned the rise of alternative family structures and lifestyles associated with nonmarital heterosexual cohabitation, and relationships between lesbian, gay, bisexual, and transgender (LGBT) persons. Such principles find expression in both policy and social attitudes toward sexual relationships. For example, the Italian Government does not recognize any type of same-sex union and this perspective is seemingly shared by the Italian population. A Eurobarometer survey published in December 2006 showed that support for ‘‘alternative’’ types of families in Italy was consistently lower that the European average: 31% of Italians support same-sex marriage and 24% recognize same-sex couples’ rights to adopt children, whereas the EU-wide average was respectively 44% and 33% (European Commission Report, 2006). The importance of traditional family arrangements and moral imperatives in Italian society suggest that societallevel logics importantly shape shared understandings about legitimate behaviors and structures, as well as norms and values. The cultural support for familial logic as a universal model for organizing interpersonal relationships has, we suggest, established Italy as an interesting case, where the prevailing effects of the institutional logic of family are also expressed in the organizational behavior of both family and nonfamily firms. Family Firms The most obvious manifestation of the prevalence of the logic of family in Italy is that business activities are dominated by family ownership and management. The Italian economy is characterized by numerous small and medium-sized enterprises (SMEs with less than 10 employees) with flexible specialization capabilities (Sforzi, 2009). Many of these firms are family owned and managed. According to recent statistics, family firms account for 73% of all firms in Italy and account for the majority (52%) of total employment (FBN International, 2008). The cultural norms and values of Italian social life, which support the notion of family as the predominant organizing unit and a strong preference for complete control over business activities by an owner, have created a small firm economy which is dominated by networks of family run businesses connected by patterns of kinship and historical bonds of trust. As a result, Italians extol the benefits of smallness, as exemplified in the commonly used phrase, nella botte piccola c’e’ il vino buono, which translates as ‘‘in the small barrel there is the good wine.’’

74

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

The logic of family is also supported by the Italian regulative and legal environment, which encourages – both directly and indirectly – the survival of small family firms. For example, SMEs benefit from state-provided benefits such as cheap government loans, loan guarantees, lower tax and employee contributions, welfare benefits at reduced premiums, and exemptions relating to book-keeping and bankruptcy (Quintieri, 2006; Weiss, 1984). Moreover, organizations of modest size benefit from financial and technological infrastructures established in the past by both regional and local government (Nanetti, 1988), and certain variants of political ideology in Italy – particularly from the Christian Democrat party – have historically been associated with small business ownership (Weiss, 1984). Despite these benefits however, few family-owned businesses have more than 10 employees and they remain resolutely domestic, with few able to successfully grow and internationalize. To illustrate, the percentage of family businesses with a turnover greater than 2 million Euros is particularly high in Italy (18% compared to the European average – 8%), which suggests that, in contrast to Western Europe and the United States where the relative power of family businesses has declined, Italian dynasties still play a major role in the economy. Companies such as FIAT,1 Pirelli, and Fininvest (controlled by the family of the former Prime Minister, Silvio Berlusconi) have continued to grow domestically, and to make inroads into international markets (Colli, 2003; The Economist, 2009). Moreover, recent privatizations of formerly state-owned industrial enterprises have allowed family-owned corporations such as Benetton, Del Vecchio, and Lucchini to acquire an even greater proportion of Italy’s industrial capital (Aganin & Volpin, 2003; Colli & Rose, 1999; Colli et al., 2003). Additionally, there is a strong tendency among Italian family firms to perpetuate familial governance structures and retain ownership and management of the business within the family. The majority of family firms are still controlled by their first generation (81%), and in a recent survey 78% of business owners indicated that succession will transfer ownership to family members rather than to outsiders (FBN International, 2008). The Expression of Family Logics in Familial Networks and Trust-Based Relationships While familial relationships are the primary source of trusted business interactions, Italians also extend their association to ‘‘people recommended by other firms and friends’’ (Lazerson, 1988, p. 338). Thus, the prevailing logic of family is made manifest not only in the significance of genetic and marital relationships, but also in the Italian preference for ‘‘quasi-family’’

Beyond the Family Firm

75

business relationships with wider audiences, which may include clients, suppliers and members of one’s profession, industry, or union (Casson, 1999; Karra et al., 2006). As Colli et al. (2003, p. 32) comment, ‘‘the family firm has increasingly been interpreted as a network of trust that is, in turn, embedded in a wider locus of connections often centered on the local business community.’’ Moreover, Italy’s industrial districts – often referred to as the ‘‘backbone of the Italian industrial system’’ (Cainelli & Zoboli, 2004, p. 3) – are paradigmatic of how social practices and conventions infuse business relationships and create unique conditions for economic development. These industrial districts operate as local clusters of SMEs, cooperatives and microenterprises, and are frequently associated with a narrow specialization profile, for example, Prato for wool, Carpi for knitwear, and Piacenza for machine tools. They have become important centers of economic power, and are characterized by a strong social adhesion based on reciprocal trust and a sense of belonging (Cainelli & Zoboli, 2004). Organizations in these districts nurture an entrepreneurial spirit and a system of mutual cooperation, as well as a reliance on extended family labour and a widespread belief in the value of work, professionalism and product quality (Amin, 2000; Bianchi et al., 1994). This strong tradition of family capitalism is underpinned by the use of personal relations as the structuring principle of organization and leadership succession and has resulted in a system structured and operated to perpetuate the influence of a restricted group of relatives, known associates, or community. The shared culture and values embodied within the boundaries of family firms and family-centric networks has unquestionably contributed to the creation of a distinctive business environment, whose strengths for many decades have been associated with a small-firm economy, with the prevalence and power of family firms, and in the trust-based operation of business relationships. The familial logic, however, has also maintained and legitimized ‘‘conservative’’ organizational strategies among SMEs, often at the expense of growth, financial gain, or the benefits of external resources (Miller et al., 2011). For example, family firms are often growth-constrained by their founder-entrepreneur (Azzariti, 2002; Bracci & Vagnoni, 2007; Colli et al., 2003). The prevalence of SMEs in Italy is at least partially accounted for by the reluctance of business owners to grow and internationalize their companies at the expense of the complete or partial loss of control of their organizations. Because growth requires capital and the expansion of operations, expansion increases the likelihood that nonfamily members

76

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

would tend to become involved in the business, for example, as employees, investors, or managers of increasingly complex activities. In particular, the unwillingness of Italian organizations to go public in order to finance expansion has been regarded as a factor hindering their growth, and leaving many SMEs vulnerable to acquisition by larger foreign firms wishing to obtain a foothold in the domestic market (The Economist, 2009). Some studies suggest that Italian businesses’ conservative strategies have resulted in their relative neglect of investment in research and development, and in international markets (Colli et al., 2003; The Economist, 2009). In sum, we propose that the institutional logic of family permeates Italian society and shapes organizational action by providing a set of prescriptions and proscriptions about legitimate behaviors and structures. We now extend our discussion to consider how the institutional logic of family may exert influence upon organizations that are supposedly embedded in, and guided by, a different logic – that of a profession. In particular, we present the case of the Italian legal sector and suggest that the interplay between institutional logics of family and profession has given Italian law firms a distinctive set of shared understandings, values, and beliefs that provide mutual support for their organizational form and strategies.

The Interplay of Institutional Logics in the Italian Legal Sector Unlike most other Italian industries, the legal profession is not dominated by firms owned or managed by groups related by birth or marriage. Although some Italian professional service firms are family firms, they are the exception. Even among criminal lawyers – who are traditionally one of the most conservative groups in the Italian legal sector and formerly exemplified by a dynastic model (studios passed on from father to son) – only 9.1% claim a familial connection with the law, and this percentage is steadily decreasing among younger generations (Guastella, 2009). A similar survey of 874 lawyers conducted in 2007 found that only 11.6% were the sons or daughters of lawyers (CENSIS, 2007). An influential factor which discourages the establishment of familymanaged enterprises in the Italian legal sector is the existence of an alternative institutional logic – of profession – that espouses autonomy, objectivity and the dispassionate rendering of professional advice. Although professional logic favors a collegiality among fellow professionals, such relationships are for the higher purpose of providing excellence and expertise in client service rather than to engender reciprocity or personal

Beyond the Family Firm

77

interests between close relations (as a family logic might prescribe). The professional logic might also contradict the influence of familial institutions such as trust-based personal networks, loyalty and the preferential treatment of close associates and kin because it promotes notions of fairness and meritocracy above nepotism, thus constraining the development of family-owned professional service firms. Indeed, lawyers and legal firms are disadvantaged in any attempt to cultivate familial ownership and/or management because substantial barriers to individual entry into the professions frequently preclude relatives from accession. Family firms require a succession of capable heirs in order to survive across generations, yet any heirs to a professional firm must meet stringent educational and licensing requirements, which guarantee the quality and integrity of their professional services. Moreover, firms are subject to strong regulatory and normative pressures (Abbott, 1988; Malhotra, Morris, & Hinings, 2006), and a professional logic dictates that they employ a highly expert and qualified workforce, a requirement which may not always converge with the aspirations and abilities of family members. Nevertheless we suggest that, even in this most detached and dispassionate of professional settings, a familial logic will be at play and can shape the form and practices of Italian law firms. In particular, our interviews with members of Italian law firms, and evidence drawn from extant research, public documents and media sources, suggest that Italian lawyers have been largely able to resist external pressures for growth arising from an increasingly globalized industry which is becoming dominated by multinational law firms. Italian firms have been relatively impervious to the courtship offers of foreign firms and able to maintain their distinctive organizational features and strategies, that is, they remain relatively small, they adopt a conservative approach to internationalization, and they recognize the benefits which result from a collective commitment to domestic power and unity. We elaborate on these characteristics below and argue that they are the result of the particular configuration, or ‘‘constellation’’ (Goodrick & Reay, 2011) of logics – including the family logic – which exist in the Italian institutional context. Size The institutional logic of family, which encourages a culture of familial entrepreneurism among Italy’s myriad SMEs, is also evident in the profile of the country’s law firms. Industry data reveal that Italy has the third largest number of lawyers in Europe (213,081), ahead of Germany (146,910) and

78

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

the United Kingdom (140,685), and has five times as many lawyers as France (Guastella, 2009). However, only 4.7% of Italian lawyers work in firms of more than 10 people: the majority of lawyers are members of small studios of 6–8 members (Guastella, 2009). Even among the largest Italian firms, these are largely ‘‘independents’’  relatively small, domestic, national firms which dominate their local markets and often compete with international players (see Table 1). Domestic firms are bigger in size and revenues than their foreign ‘‘mega-firm’’ rivals in this market but, as Table 2 illustrates, the largest Italian firms are significantly smaller than their contemporaries in Europe and the United States. Studies of professional service firms (e.g., Brock & Alon, 2009; Hitt, Bierman, Uhlenbruck, & Shimizu, 2006) suggest that law firms have been subject to internal and external forces for both domestic and international growth, including competitive pressures arising from technological advances and deregulation. This stream of literature has highlighted how, in the Anglo-Saxon context, intense competition has forced professional service firms to engage in a merger ‘‘frenzy’’ to diversify their practices and Table 1. Largest Law Firms in Italy by Revenues. Rank

Law Firm

Total Revenues (Million Euro)

Country of Origin

140 m

Italy

3:5

244:300

125 m 98 m

Italy Italy

3:9 5:8

259:311 280:288

4 5

Bonelli Erede Pappalardo Chiomenti Gianni Origoni Grippo Pirola Pennuto Zei Clifford Chance

9:11 2:30

140:160 80:3,600

6

Freshfields

51 m

2:27

47:2,400

7

Allen and Overy

50 m

3:31

130:2,700

8 9

NCTM Legance Studio Associato CBA studio legale e tributario

3:5 2:2

200:300 120:120

4:5

170:180

1 2 3

10

90 m 53 m

42 m 35–40 m 35 m

Italy United Kingdom United Kingdom United Kingdom– Italy Italy Italy Italy

No. Local: No. Local: Total Total Offices Lawyers

Sources: Top Legal 2008 (revenues); firms’ web sites; various legal directories, 2012 (number of offices).

79

Beyond the Family Firm

Table 2.

Comparison of the Largest Law Firms in Eight Countries.

United States United Kingdom Australia Canada Germany Spain France Italy

No. of Lawyers in the Largest National Law Firm

Average No. of Lawyers (Top 10 Largest Firms)

3,300 2,654 987 704 562 1823 700 246

2,000 1,580 928 560 337 380 227 187

Sources: Lexpert; Legal Business, Top Legal; American Lawyer.

locations and to increase their size (e.g., Brock, Powell, & Hinings, 1999). Similarly, the classic work of Maister (1997) identifies the connection between growth and profitability, based on the implementation of governance and leverage structures which rely on a philosophy of naked ambition. Yet Italian law firms have shown an extreme reluctance to modify their practices and professional norms to accommodate demands for bigger, more diversified and structured practices. One of our interviewees – a senior partner of a highly prestigious Italian law firm – said that small Italian studios (also known as ‘‘boutiques’’) have been criticized by their own clients for the limited size of their practices. In particular, international clients have become concerned that the small size of Italian practices could signal their inability to provide a timely and/or consistent service. Although the demand for growth and an increased international presence had become increasingly pressing, Italian law firms have maintained their small size in the face of such pressures because the family logic encourages the development and maintenance of a network of established and trusted affiliates. To become larger implies the necessity to open the ranks of partnership to new members, greater formalization of law firms’ evaluation systems and, eventually, to a contemplation of mergers with other law firms to diversify their portfolio of practices – an option that many independent studios have found particularly unattractive. As a result, some law firms have seen their partners engaged in internal conflicts and disagreements on the strategic direction of their firms, particularly the appropriateness of merger with a foreign firm, and the decision to embrace more structured governance systems. Internal conflicts and mass departures have splintered some of Italy’s largest firms. For

80

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

example, Pavia and Ansaldo lost more than 60 lawyers from its Milan and Rome offices, as well as its entire staff in Turin and Padua, who together formed a new firm, Agnoli Bernardi & Associati. The split was the result of a series of disputes among partners about the strategic development of the firm, exacerbated by the collapse of previous attempts to establish alliances with international partners (Collins, 2005). In essence, the logics of family and of market collided, creating conflicts within the firm about whether growth should be pursued at all and, if so, the appropriate way in which it should occur and how professional relationships should be altered accordingly. Interestingly, this and many other similar episodes provide counterexamples to the prevailing view in the professional service firm literature, which argues for the strategic hegemony of international growth (e.g., Alon & McKee, 1999; Brock & Alon, 2009; Segal-Horn & Dean, 2009). The Italian case directs attention to the role of the broader institutional context in shaping responses to competitive pressures. We suggest that the behavior of Italian law firms can be better understood if the pervasive, society-wide family logic of Italian life – which nurtures robust networks of colleagues, clients, and institutional and regulatory contacts – is taken into account. Just as familial social networks support the operation of Italian business relationships more generally, partners of Italian law firms develop a close ‘‘family’’ of trusted peers and associates, underpinned by personal trust and knowledge among the partners. Our investigation reveals that these relationships are regarded as crucial to ensure a high-quality client base and to preserve the reputation of the firm. As a result, partners are reluctant to increase the size and complexity of professional partnerships, fearing that this would entail relinquishing control (perhaps to a foreign firm) and undermining the ability of the partners to cooperate in a collegial manner (Brock et al., 1999; Greenwood & Empson, 2003). Similarly, the development of ‘‘quasi-family’’ relationships – arguably the very essence of Italian business relations – between Italian lawyers and their clients seemingly reduces the ‘‘imperative’’ for Italian firms to increase their size. That is, clients also favor a form of lawyering in which their advisers are personally known, networks are tightly knit, and norms of trust and reciprocity underpin business practices. An example of this form of ‘‘stewardship’’ relationship is demonstrated by Grande Stevens: Studio Legale Associazione Professionale, a Turin-based law firm of 22 lawyers that is well known as the legal advisor of FIAT, the Italian car manufacturer owned by the Agnelli family. The law firm’s founder, Franzo Grande Stevens, is a trusted and longstanding adviser to the Agnelli family and has

Beyond the Family Firm

81

been dubbed ‘‘l’avvocato del’Avvocato’’ – the lawyer’s lawyer (Winfield, 2010). The law firm of Grande Stevens has worked extensively on the restructuring of both FIAT and the Agnelli-owned football club, Juventus, during a series of crises over the last few years, yet from an Anglo-American perspective it is surprising to find a multinational company associated with a tiny professional firm. But the relationship is not extraordinary in the Italian context, where the trust-based, long-lived and personal nature of the tie exemplifies the way in which Italians’ conception of ‘‘family’’ goes beyond genetic and matrimonial links to embrace a limited set of close, personal contacts. Moreover, the exclusivity of this boutique law firm, and Grande Stevens’ image of resolute independence, is an attractive source of status by association for clients such as the Agnelli family. We suggest that the family and professional logics underpinning the relationship between clients and trusted advisors are likely to mutually reinforce each other. That is, clients perceive the nature of legal advice and assistance to be an examination of their personal affairs, which therefore requires the engagement of a close, trusted advisor with whom they have an established relationship, or that the advisor be recommended by reliable sources. As part of an extended ‘‘family’’ of intimate business contacts, advisors share their clients’ emotive, personal concerns and develop, as in the case of Gianni Agnelli and Franzo Grande Stevens, affective commitment and friendship. Internationalization Strategy Allied with a reluctance to grow beyond a size at which individual employees or network partners become unknowable, we also observe that Italian law firms, relative to European law firms in general, have not made significant inroads into international legal markets. Table 1 illustrates that the largest domestic Italian firms have very few overseas offices (average = 6), compared with the global ‘‘magic circle’’ law firms such as Clifford Chance (30) and Allen & Overy (31). True to their ‘‘familial’’ network approach, those Italian firms which have developed international capabilities have done so only by way of relatively informal ‘‘best friend’’ linkages with foreign law firms. For example, Bonelli Erede Pappalardo, Italy’s largest law firm, requests assistance from similar stand-alone, mergerresistant law firms overseas, such as Slaughter and May in the United Kingdom, and Hengeler Mu¨ller in Germany. A reciprocal arrangement exists, where these correspondent foreign firms will engage Bonelli to deal with the Italian aspects of their own clients’ legal cases (Legal Week, 2009). By operating according to these kinds of ‘‘gentlemen’s agreements,’’ Italian

82

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

firms have avoided more formal strategies, such as mergers or alliances with international or transnational practices, or the creation of their own overseas offices by transferring partners, hiring local lawyers, or merging with local firms. Although some research has considered how and why professional service firms decide to pursue specific internationalization strategies (e.g., Aharoni, 1999; Morgan & Quack, 2005), their focus has been the organization and practices of the largest US and UK firms facing pressures to globalize their services and corporatize their operations to meet the demands of clients who are themselves becoming more diversified geographically and functionally (e.g., Alon & McKee, 1999; Hitt et al., 2006; Lowendahl, 2000). The largest firms in a domestic market have been shown to respond to these pressures via mergers and acquisitions, licensing, or ‘‘greenfield’’ investments, and that internationally oriented firms succeed on the basis of their scale and international experience, inevitably emerging to dominate smaller, domestic competitors (e.g., Aharoni, 1999; Brock & Alon, 2009; Segal-Horn & Dean, 2007, 2009). The assumption of these studies is that the growing complexity of client transactions demands increasingly specialized professional advice that can only be provided most effectively by professional service firms with the critical mass to nurture both deep specialization and assemble teams of multiple experts to handle sophisticated projects (Aharoni, 1999; Brock et al., 1999; Empson, 2007; Greenwood et al., 2010b). In effect, this stream of literature has documented the progressive pervasiveness of a market logic into professional fields, and attempts to reconcile market demands with a professional logic. Critically, this body of work has largely ignored professional organizations outside of the United States and United Kingdom, with the exception of some studies on Germany (e.g., Morgan & Quack, 2005, 2006). Little is known about how national contexts and locally instantiated institutional logics influence internationalization strategies in non-Anglo Saxon markets. For example, the most successful Italian law firms, unlike those in the United Kingdom or Germany, typically avoid merging with global ‘‘megafirms’’ or, indeed, other foreign practices. Moreover, they have not established an international presence in the wake of the increasing internationalization of their clients. Where an international connection is required, Italian lawyers subcontract their business to an elite group of foreign firms, which results in them having barely any direct involvement in international client matters. We suggest that Italian firms’ preference for a limited form of internationalization is a further indication of the preeminence of their

Beyond the Family Firm

83

insular, trust-based business networks based on the family logic. Given that entry into a foreign market requires that a firm establish a system of relationships which are crucial to its operations and success in a new market, Italian law firms’ ‘‘conservative’’ approach to overseas business is an expression of a familial logic which promotes the maximization of direct control and the minimization of exposure to firms or persons unproven as trustworthy. This logic is also expressed in their commitment to a deep embeddedness in local networks of trust and social capital, which keeps them focused on domestic issues for domestic clients; meanwhile, they cultivate a small network of foreign ‘‘best friends’’ in order to service the needs of their international clients. In both cases, Italian firms have managed to ‘‘insulate’’ themselves from the pressures of the market logic by working inside relatively restricted social systems, where ‘‘insiders’’ enjoy privileged access to resources and structural and strategic change is likely to occur only slowly, if at all. Culture of Domestic Power and Unity As we noted earlier, the Italian legal profession is embedded within a broader institutional context which supports business networks with secure, binding relationships and a mixture of competitive and cooperative behaviors among its members. Italian law firms, we suggest, have been able to foster a culture of domestic power and unity, a collective stance motivated by the threat of overseas law firms trying to penetrate their market. Since their entrance into the Italian market in the early 1990s, foreign law firms (first from the United Kingdom, with US firms arriving in a second wave) have had difficulty acquiring local clients and gaining ground on established domestic players. A strong family logic which secures the ongoing and intensely loyal relationship between Italian lawyers and their clients means that the latter are resistant to attempts by foreign law firms to establish commercial relationships with them. Evidence from the legal media suggests that foreign legal practitioners face difficulties in establishing the intimate, personal bonds which are customary and long-standing in the Italian context. As Leah Dunlop, the Italian managing partner of the UKheadquartered law firm Lovells commented, ‘‘Success is really dependent on having that core of local clients. The independents [Italian firms] have really consolidated their contacts and done an excellent job in keeping close to their clients’’ (Legal Week, 2009). Even some of the world’s largest and most powerful law firms, such as those in the UK’s ‘‘magic circle,’’ have been unable to establish themselves

84

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

in Italy with the same speed and success as they have done in other foreign jurisdictions. As Table 1 illustrates, three of the magic circle law firms (Clifford Chance, Freshfields Bruckhaus Deringer, and Allen & Overy) are among the top 10 largest firms in Italy; their revenue figures, however, are significantly less than those of the largest Italian firms. In addition, the Italian offices of these large foreign firms are much smaller than their other correspondent offices in European countries of similar size, such as Spain and France. Allied with their reluctance to operate beyond their own borders, Italian firms have maintained their domestic independence, despite intensive courtship by large UK and US mega-firms intent on merging with a local player. There are many examples of failed mergers or broken alliances which pepper the history of foreign law firms’ Italian ventures. For example, Clifford Chance, the world’s largest law firm, went through a highly public and acrimonious divorce with Studio Legale Grimaldi e Associati in 2002, just two years after merging (O’Brian, 2006). Magic circle firm Linklaters terminated its four-year alliance with Gianni, Origoni, Grippo, and Partner by mutual consent in April 2004, leaving the City firm with a small US and UK law practice in Italy, but no local law capability (Eaglesham, 2001). Freshfields Bruckhaus Deringer also began merger talks with Chiomenti in 2000, but the union was never finalized because of the resistance of the latter (The Lawyer, 2001). In all these cases, ‘‘cultural differences’’ were blamed for the failure of the alliances or mergers. We concur, insofar as environment-specific institutional logics describe societal and professional cultural differences which can impede the brokerage of business relationships and the development of shared strategic goals. In particular, the logic of family in the Italian legal environment influences lawyers’ perspectives on the kinds of behavior which are considered appropriate. Lawyers nurture a profound attachment to their professional practices and trusted network of close contacts, and distrust arrangements that imply losing control over crucial decisions or may introduce strangers into their business relationships. As a senior lawyer of an Italian law firm which had just rejected a merger proposal said to us, ‘‘We did not want to be reduced to the ranks of the ‘‘arm’’ of this company and leave the ‘‘brain’’ to go abroad.’’ We suggest that professional, family and national community logics are deeply intertwined in the Italian context and provide cultural understandings and norms that profoundly shape the cognitions and behavior of actors. As our investigation suggests, when foreign organizations attempt to penetrate and disrupt an established, collectively created system, these collective frames are mobilized and therefore tend to be more clearly visible.

85

Beyond the Family Firm

That is, institutional actors draw upon these logics as symbolic and material tools in the mobilization efforts of incumbent organizations to resist pressures from contradictory, external logics in order to protect the boundaries of their community. In this regard, our case is similar to Marquis and Lounsbury’s (2007) account of resistance by community banks to the spread of national banking logics in the United States. Unlike professional bankers, however, who defended an extant logic of community by founding new organizations, Italian lawyers have been successful in maintaining, protecting and reinforcing the boundaries of the Italian legal ‘‘community.’’ Critically, this resilience has been facilitated by the configuration of professional, family, and national logics that provides these law firms with a cohesive set of cultural understandings, values, and behaviors which have supported a cohesive response to the ‘‘attacks’’ of foreign firms. In sum, we have highlighted how the family logic may find expression in a nonfamily setting. The case of the Italian legal market suggests that lawyers’ embeddedness in significant networks of clients, contacts and fellow professionals – characterized by close, trusted relationships – is underpinned by a strong family logic of cohesion and reciprocity between network members. The strength of these ‘‘quasi-family’’ networks, we suggest, has resulted in specific organizational outcomes. In particular, Italian law firms remain resolutely small, are reluctant or resistant to merger with both domestic or international partners, and tend to resist internationalization (except in a weak form) by remaining steadfast in the face of attempts by large, foreign law firms to enter the Italian market.

DISCUSSION The Influence of the Family Logic on Professional Service Firms Our paper calls attention to the need for more sophisticated and explicit attention to nonmarket logics – and in particular the family logic – in institutional theorizing and research. Previous research has narrowly focused on the institutions of the market and the professions, and of familial logics upon family firms, but a more comprehensive application of the institutional logics thesis is required in order to understand how the perceptions and behavior of organizations are influenced by prevailing societal principles and values (Thornton et al., 2012). We focus on an institutional context – Italy  in which a logic of family is established at a

86

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

national level as an important social and cultural influence, and we suggest that such institutional assumptions and values create a distinctive pattern of organizational practices and structure across organizations. Our illustrative case of the Italian legal sector is notable because it is one in which logics other than the family, such as logics of capitalism and the profession, would be expected to dominate. While we acknowledge that our case setting is a pluralistic one, where multiple institutions provide a complex set of demands, we highlight that a familial logic interacts with other logics, exerting influence upon the organizational characteristics and cultural profile of firms neither owned nor managed by families. We have emphasized the manifestation of the institutional logic of family in terms of the importance of maintaining ‘‘patriarchal’’ control over business activities and preserving networks of, inter alia, within-firm colleagues, clients, international ‘‘best friend’’ law firms, and fellow industry professionals. To varying degrees, these trusted network relationships within and across organizations are valued and nurtured because they represent an investment in dependence and reciprocity between individuals and organizations, which mirrors that which is taken for granted among relatives and close family associates. As ‘‘quasi-families,’’ professional partnerships and networks represent secure, binding relationships between their members, and ensure a degree of unanimity and cooperative behavior among them. However, the close and personal nature of these network associations means that there may be limits to the number of relationships that can be maintained; hence law firms remain relatively small, and attempts to increase size may generate instability. The influence of the family logic is also a potential barrier to the establishment of relationships with organizations or individuals who are not within the familial ‘‘circle of trust’’ of the partnership or network. That is, the family logic interacts with the professional logic by shaping growth and internationalization strategies of local firms, as well as maintaining their resistance to foreign competitors’ attempts to penetrate a local market. Foreign firms from the United States and United Kingdom face particular difficulties in establishing intimate, personal bonds with potential legal partners, as well as with clients whose relationships are customary and longstanding in the Italian context. These difficulties arise because foreigners are unable to win the trust of local partners, who perceive the Anglo-Saxon ideology of unfettered individualistic competition as alien to Italian values and beliefs. Italian law firms resist long-term relationships with foreign enterprises because such liaisons would necessarily mean that firms become both larger and exposed to persons unfamiliar and, by implication, more

Beyond the Family Firm

87

impersonal. Increasing size and personal disconnection would therefore undermine shared local understandings of how law firms should be organized and operated. As a result, clients would see the alliance or merger with a foreign partner as a weakening of their own trust-based relationships with Italian firms, which depend on long-standing bonds with ‘‘known’’ advisors. Whether consciously or not, by rebuffing the courtship attempts of foreign merger partners Italian law firms have collectively contributed to maintaining the integrity of their national market, where domestic firms still dominate. Indeed, our case suggests that the ‘‘constellation’’ of logics (Goodrick & Reay, 2011) that underpin cultural understandings and social interactions have an important influence upon the strategy and form of domestic and foreign firms. Importantly, institutional logics can be the basis upon which professions temper forces for globalization, and can even successfully resist them.

Interplay and Historical Co-evolution of Institutional Logics Our appeal to include the logic of family in broader conceptions of institutional theorizing resonates with recent scholarly interest in the coexistence and effect of multiple, potentially incompatible, logics upon fields and organizations (Kraatz & Block, 2008; Greenwood et al., 2011; Zilber, 2011). We have highlighted how multiple logics are at play in a specific context; in particular, we argue that the logic of family exerts a profound influence upon both family and nonfamily firms, even in an institutional setting where logics of market and profession seemingly dominate. In our setting, close-knit, quasi-family relations between professionals, and between professionals and clients, are embedded in collective understandings of professionalism in Italy and are therefore critical to the success of business relations. Our case suggests that the logics of family, market and profession interact in complex ways. That is, the logic of family supports the delivery and execution of professional work within Italian law firms by promoting the importance of quasi-familial networks, intimate relationships between associates, and trust between lawyer and client. In turn, the professional logic – which favors the establishment of a trustee relationship between professional and client in order to provide efficient and expert knowledge – is facilitated by values founded on a familial logic which privileges trusted personal contacts.

88

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

In addition, we have highlighted how the mutual reinforcement of family and professional logics may be further fortified by the establishment and continuance of distinctive national forms of network or regional/community embeddedness, not just in the creation of an unyielding community of domestic law firms, but in wider Italian society, such as in the industrial districts (Cainelli & Zoboli, 2004; Marquis, Lounsbury, & Greenwood, 2011; Thornton et al., 2012). In essence, repetitive market relations create close, cooperative, long-term social and business relationships, which generate trust and reciprocity and can create social and economic benefits for those involved (e.g., Gulati & Gargiulo, 1999; Ingram & Roberts, 2000; Uzzi, 1997). However, we argue that logics of profession and market combine with the family logic in terms of the specific behaviors, practices and beliefs which will be nurtured in the Italian legal sector. Market logic, for example, dictates that firms pursue profits by being competitive in terms of their organization and practice; yet this logic is supported in the Italian case by a familial logic which provides favorable terms and repeat business transactions to kin and kin-like associates, which may result in competitive advantages (or at least competitive parity) in a market in which transactional favoritism is both expected and appropriate. Thus, our paper speaks to an emerging body of scholarship that explores the interplay of multiple institutional logics in pluralistic environments and the consequences of such complexity for organizations (Greenwood et al., 2011; Zilber, 2011). Importantly, we also highlight that the relationship between multiple logics is rooted in the historical co-evolution of national institutions. Our findings suggest that the mutual interplay between logics and their organizational effects can only be fully understood by looking at the historical development of their relationships over time. For example, informal networks of family and trusted affiliates, which encourage nepotism and cronyism in terms of appointments and patronage, have long been a feature of Italian society (e.g., Padgett & Ansell, 1993) and, as we have illustrated, find expression in some of the distinctive features of contemporary Italian society. Similarly, scholars emphasize the distinctive character of modern professionalism in Italy, rooted in a history of individual professional independence and close relationships with clients, which may have contributed to the insulation of some professions from competitive market dynamics (Pammolli, Cambini, & Giannaccari, 2007). The historical study by Padgett and Ansell (1993) of Cosimo di Medici’s rise to power in 15th-century Florence is a particularly germane illustration of the power of family in Italian society, a power which still resonates today. The study reveals that familial values of loyalty, gratitude, dependence, and

Beyond the Family Firm

89

altruism within a network of marriage and business alliances were key to Medici’s acquisition of power. We are also aware of studies which describe Sicilian and Italian–American mafia organizations as founded on family and friendship (e.g., Anderson, 1979; Dickie, 2005; Walker Fields, 2004), and we have already noted research which explains the success of Italian familycentered industrial districts (e.g., Sforzi, 2009). This body of work adds weight to our suggestion that the historical co-evolution of constellations of logics – in our case family, professional and market logics – may be a particularly fruitful direction for future research that seeks to understand how long-established multiple logics interact in shaping national organizational structures and practices. By adopting a perspective which considers the ways in which historically contingent and co-evolved sets of logics interact, opportunities also arise to gain a deeper understanding of professional behavior, by elucidating how locally instantiated cultural beliefs and values profoundly shape the development of professional service firms. While our case describes the Italian legal sector, we acknowledge that the effect of family and other logics in other professional fields in Italy may exhibit a different set of interactions. For example, some of the structural and behavioral characteristics of domestic Italian audit firms provide support for our general thesis  that the family logic has a significant effect on the form and practices of nonfamily-owned organizations  but with notable differences to the pattern we observed in the legal sector. Like their legal counterparts, audit firms are also small. In general, Italian accounting professionals (‘‘dottore commercialista o ragioniere’’) work in small ‘‘studios,’’ that is, firms of independent sole practitioners with personal liability, who account for 70% of the industry’s personnel. In addition, they have a strong incentive to develop long term, trusting relationships, and are frequently located close to their clients (Zambon, 2003). Just as in the legal profession, the logic of family appears to influence the provision and organization of accounting services, and supplies a continued emphasis upon the importance of established networks and embedded relationships with trusted individuals who form part of an ‘‘extended’’ family. However, we note a major difference in the profile of the accountancy profession in Italy, as compared with law. Notably, the market is dominated by the ‘‘Big Four’’ transnational accounting firms, who have been able to enter and dominate the market without finding strong resistance from national players (Zambon, 2003). This example reveals how some firms or industries have been able to break away from a powerful familial logic

90

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

which sets the tone for much of Italian societal behavior. Although family logic still finds expression in some aspects of the organization of the audit industry, competing logics of market and of profession have become much more influential in that setting. By examining this, and other, instances of pluralistic institutional environments, we encourage researchers to unpick the threads which constitute the interface between multiple logics in order to reveal insights into practice variation within and across fields. Additionally, the existence of multiple logics, and their intricate mutual interplay, presents an opportunity for future research which examines the nature of such interactions in terms of their strength and effect.

Studying Institutional Logics in a Global Context We have already described the specific cultural characteristics of the Italian institutional context and we suggest that a logics perspective is an apposite refinement to current institutional theorizing because of its ability to take into account the cultural heterogeneity of societies. In this regard, we support recent calls for an integration of institutional logics with the historical and comparative analysis of institutions (Thornton et al., 2012; Djelic, 2010). The ‘‘varieties of capitalism’’ approach has revealed how organizations in different countries are shaped by legal, economic, and political institutions (e.g., Gertler, 2004; Hall & Soskice, 2001; Teubner, 2001; Whitley, 1999, 2005). In particular, this work has begun to uncover the role of national institutional frameworks in shaping organizational responses and in understanding the adoption of particular forms and practices (e.g., Morgan & Quack, 2005, 2006). An institutional logics perspective adds richness to this theorizing by acknowledging that political and economic conditions of domestic environments are themselves shaped by higher-order logics. We have sought to comment on the application of institutional logics to an international context by elaborating upon the influence of societal-level family logic on the formation of economic structures and organizational behaviors. Our conclusions are based upon the specific case of Italy. However, they might also be relevant for future investigations in other countries. For example, previous research has already identified Spain as ‘‘significantly shaped by its Catholic tradition and its emphasis on paternalistic family values’’ (Greenwood et al., 2010a, p. 522) and thus may exhibit a similar pattern of organizational behaviour across multiple fields and industries to that which we have identified in Italy. Given that

91

Beyond the Family Firm

both Italy and Spain are predominantly Catholic countries, we wonder whether organizations in South America, Ireland and the Philippines – where the Catholic faith is still strong – are similarly immersed in a powerful familial logic which provides actors with taken-for-granted assumptions, values and organizing principles. Similarly, our focus on social structures based on extended kinship networks and strong personal relationships may be equally applicable to other societies. For example, personal ties combined with normative obligation, or guanxi, are a distinctive feature of Chinese trading networks (So & Walker, 2006). Trust-based networks have been demonstrated as a significant source of business resources and markets for Asian and South American communities (e.g., Bagwell, 2008; Portes, Haller, & Guarnizo, 2002; Sher, 2003). Given the extent to which previous work has shown how kin and kin-like relationships can provide a value code for the structuring and tone of business relations, we propose that, in certain ethnic and community settings, a familial logic may be especially significant to the establishment of embedded network structures. From such a perspective it would be interesting to consider the extent to which the logic of family influences the formation and operation of ethnically distinct business elites or other forms of patterned business relationship. Moreover, theories which explain organizations’ adoption of particular practices, and their propensity to enter into joint ventures and alliances, might also be enhanced if researchers were to take account of the supportive or constraining effect of familial institutions. The effect of these logics may be wide ranging, even extending to such diverse examples of altruism and loyalty among quasi-family members as alumni associations, criminal gangs, nonprofit organizations, trade unions, and online social communities.

CONCLUSIONS In this paper we have emphasized the importance of developing a more comprehensive understanding of the inter-institutional system and advocated for more attention to be paid to nonmarket logics – and specifically the family logic – in institutional theorizing. Given the primacy of market and professional logics in institutional scholarship, we encourage researchers to examine the interaction between logics at different levels to create distinctive institutional environments. The specificity of our research context and the illustrative nature of our case signal an opportunity to design

92

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

rigorous empirical studies to examine institutional logics of family across countries and industries. Much remains to be explored about the way in which organizations are shaped by the historical development of constellations of logics. By highlighting how societal assumptions and values based on notions such as family loyalty, altruism and reciprocity can be a significant influence upon organizational forms and practices, our hope is to contribute to a more sophisticated understanding of the complex relationship between logics, institutions and organizations.

NOTE 1. FIAT is an acronym for Fabbrica Italiana Automobili Torino S.p.A., an Italian automobile, engine manufacturer, financial, and industrial group based in Turin and controlled by the Agnelli family.

REFERENCES Abbott, A. (1988). The system of professions: An essay on the division of expert labor. Chicago, IL: University of Chicago Press. Aganin, A., & Volpin, P. (2003). History of corporate ownership in Italy. Finance Working Paper No. 17. Retrieved from http://www.ecgi.org/wp/f1703.pdf. Accessed on April 20, 2010. Aharoni, Y. (1999). Internationalization of professional services: Implications for accounting firms. In D. Brock, M. Powell & C. Hinings (Eds.), Restructuring the professional organization: Accounting, health care and law (pp. 20–40). London: Routledge. Alon, I., & McKee, D. (1999). The internationalization of professional business service franchises. Journal of Consumer Marketing, 16, 74–85. Amin, A. (2000). Industrial districts. In E. Sheppard & T. Barnes (Eds.), A Companion to Economic Geography (pp. 149–168). Oxford: Blackwell. Anderson, A. G. (1979). The business of organized crime: A Cosa Nostra family. Stanford, CA: Hoover Institution Press. Ardigo’, A., & Donati, P. (1976). Famiglia e industrializzazione; continuita` e discontinuita` negli orientamenti di valore in una comunita` in forte sviluppo endogeno. Milano: Franco Angeli. Azzariti, F. (2002). I percorsi di crescita delle piccole e medie imprese. Teorie, modelli e casi aziendali. Milan: Franco Angeli. Bagwell, S. (2008). Transnational family networks and ethnic minority business development: The case of Vietnamese nail shops in the UK. International Journal of Entrepreneurial Behaviour & Research, 14, 377–394. Battilana, J., & Dorado, S. (2010). Organization building amid multiple institutional logics: The case of commercial microfinance organizations. Academy of Management Journal, 53(6), 1419–1440.

Beyond the Family Firm

93

Berrone, P., Cruz, C., & Gomez-Mejia, L. R. (2012). Socioemotional wealth in family firms theoretical dimensions, assessment approaches, and agenda for future research. Family Business Review, 25(3), 258–279. Bhappu, A. D. (2000). The Japanese family: An institutional logic for Japanese corporate networks and Japanese management. Academy of Management Review, 25, 409–415. Bianchi, G., Garofoli, G., & Mazzoni, R. (1994). Requiem per la Terza Italia? Sistemi territoriali di piccola impresa e transizione postindustriale. In G. Bianchi (Ed.), Appunti di un programmatore. IRPET Report. Retrieved from http://www.irpet.it/storage/ pubblicazioneallegato/236_Giuliano%20Bianchi.pdf. Accessed on August 25, 2012. Biggart, N. W., & Hamilton, G. G. (1990). Explaining Asian business success: Theory no. 4. Business and Economics Review, 5, 11–16. Bracci, E., & Vagnoni, E. (2007). Le piccole imprese familiari. Il capitale intellettuale nella gestione del ricambio generazionale. Milan: Franco Angeli. Brint, S. (1996). In an age of experts: The changing role of professionals in politics and public life. Princeton, NJ: Princeton University Press. Brock, D., & Alon, I. (2009). Internationalization of professional service firms. International Business: Research Teaching and Practice, 3, 52–70. Brock, D., Powell, M., & Hinings, C. R. (1999). The restructured professional organization: Corporate, cobwebs and cowboys. In D. Brock, M. Powell & C. Hinings (Eds.), Restructuring the professional organization: Accounting, health care and law (pp. 215– 229). London: Routledge. Cainelli, G., & Zoboli, R. (2004). The evolution of industrial districts: Changing governance, innovation and internationalization of local capitalism in Italy. Heidelberg: PhysicaVerlag. Casson, M. C. (1999). ‘Entrepreneurship and the theory of the firm’. In Z. A. Acs, C. Carlsson & C. Karlsson (Eds.), Entrepreneurship, small and medium-sized enterprises and the macroeconomy (pp. 45–78). Cambridge: Cambridge University Press. CENSIS. (2007). Contrastare la crescita.Promuovere la mobilita`. Indagine sugli avvocati italiani. Roma: Associazione Italiana Giovani Avvocati (AIGA). Chrisman, J. J., Chua, J. H., & Sharma, P. (2005). Trends and directions in the development of a strategic management theory of the family firm. Entrepreneurship Theory and Practice, 29(5), 555–576. Colli, A. (2003). The history of family business, 1850–2000. Cambridge: Cambridge University Press. Colli, A., Fernandez Perez, P., & Rose, M. (2003). National determinants of family firm development? Family firms in Britain, Spain, and Italy in the nineteenth and twentieth Centuries. Enterprise & Society, 4, 28–64. Colli, A., & Rose, M. B. (1999). Families and firms: The culture and evolution of family firms in Britain and Italy in the nineteenth and twentieth centuries. Scandinavian Economic History Review, 47, 24–47. Collins, A. (2005). Stemming the decline. Legal Week, June 30, 2005. Compagno, C. (2011). Assetti di governance e processi di internazionalizzazione nelle PMI. Sinergie rivista di studi e ricerche. Available at http://www.sinergiejournal.it/rivista/ index.php/sinergie/article/view/79. Accessed on August 25, 2012. Cooper, D. J., Hinings, C. R., Greenwood, R., & Brown, J. L. (1996). Sedimentation and transformation in organizational change: The case of Canadian law firms. Organization Studies, 17, 623–647.

94

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

Dickie, J. (2005). Cosa Nostra: A history of the Sicilian mafia. New York, NY: Palgrave Macmillan. Djelic, M. L. (2010). Institutional perspectives – Working towards coherence or irreconcilable diversity? In G. Morgan, J. Campbell, C. Crouch, O. K. Pedersen & R. Whitley (Eds.), The Oxford handbook of comparative institutional analysis (pp. 15–40). Oxford: Oxford University Press. Dunn, M. B., & Jones, C. (2010). Institutional logics and institutional pluralism: The contestation of care and science logics in medical education, 1967–2005. Administrative Science Quarterly, 55, 114–149. Eaglesham, J. (2001). Wall St. law firm forms Italian alliance. Financial Times, June 12, 2001. The Economist. (2009). Country briefing: Italy. The Economist. Retrieved from http:// www.economist.com/countries/Italy/. Accessed on January 12, 2010. Empson, L. (2007). Managing the modern law firm: New challenges, new perspectives. Oxford: Oxford University Press. Eurispes Report. (2011). Report on Italian families. Retrieved from http://www.eurispes.it/ index.php?option=com_content&view=article&id=2462:documento-di-presentazioneosservatorio-sulle-famiglie&catid=66:famiglia&Itemid=333. Accessed on August 25, 2012. European Commission Report. (2006). Standard Eurobarometer 66: Public opinion in the European Union. Retrieved from http://ec.europa.eu/public_opinion/archives/eb/eb66/ eb66_en.htm. Accessed on May 15, 2010. FBN International. (2008). Family business international monitor report. Retrieved from http:// www.fbn.org/fbn/web.nsf/LibraryLU2/AA3786AA36C250F9872575210070073E/$file/ Monitor2008.pdf. Accessed on August 25, 2012. Friedland, R., & Alford, R. R. (1991). Bringing society back in: Symbols, practices and institutional contradictions. In W. Powell & P. DiMaggio (Eds.), The New Institutionalism in Organizational Analysis (pp. 232–265). Chicago, IL: University of Chicago Press. Friedson, E. (2001). Professionalism: The third logic. Cambridge: Polity. Gerlach, M. L. (1992). Alliance capitalism: The social organization of Japanese business. Berkeley, CA: University of California Press. Gertler, M. S. (2004). Manufacturing culture. Oxford: Oxford University Press. Goodrick, E., & Reay, T. (2011). Constellations constellations of institutional logics. Changes in the professional work of pharmacists. Work and Occupations, 38(3), 372–416. Gomez-Mejia, L. R., Cruz, C., Berrone, P., & De Castro, J. (2011). The bind that ties: Socioemotional wealth preservation in family firms. Academy of Management Annals, 5(1), 653–707. Greenwood, R., & Empson, L. (2003). The professional partnership: Relic or exemplary form of governance? Organization Studies, 24, 909–933. Greenwood, R., Magan, A. D., Li, X., & Lorente, J. (2010). The multiplicity of institutional logics and the heterogeneity of organizational responses. Organization Science, 21, 521–539. Greenwood, R., Morris, T., Fairclough, S., & Boussebaa, M. (2010a). The organizational design of transnational professional service firms. Organizational Dynamics, 39(2), 173–183. Greenwood, R., Oliver, C., Sahlin-Andersson, K., & Suddaby, R. (2008). The Sage handbook of organizational institutionalism. Los Angeles, CA: Sage.

Beyond the Family Firm

95

Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E. R., & Lounsbury, M. (2011). Institutional complexity and organizational responses. Academy of Management Annals, 5, 317–371. Guastella, G. (2009). Avvocati, un futuro di piccoli studi. Corriere della Sera, April 29, 2009. Guille´n, M. F. (2001). The limits of convergence. Princeton, NJ: Princeton University Press. Gulati, R., & Gargiulo, M. (1999). Where do interorganizational networks come from? American Journal of Sociology, 104, 1439–1493. Hall, P. A., & Soskice, D. (2001). Varieties of capitalism: The institutional foundations of comparative advantage. Oxford: Oxford University Press. Haveman, H. A., & Rao, H. (1997). Structuring a theory of moral sentiments: Institutional and organizational coevolution in the early thrift industry. American Journal of Sociology, 102, 1606–1651. Hitt, M., Bierman, L., Uhlenbruck, K., & Shimizu, K. (2006). The importance of resources in the internationalization of professional service firms: The good, the bad, and the ugly. Academy of Management Journal, 49, 1137–1157. Ingram, P., & Roberts, P. W. (2000). Friendships among competitors in the Sydney hotel industry. American Journal of Sociology, 106, 387–423. James, H. (2006). Family capitalism, wendels, haniels, falks and the continental european model. Cambridge: Belnap Press of Harvard University Press. Karra, N., Tracey, P., & Phillips, N. (2006). Altruism and agency in the family firm: Exploring the role of family, kinship and ethnicity. Entrepreneurship Theory and Practice, 30, 861–877. Kitchener, M. (2002). Mobilizing the logic of managerialism in professional fields: The case of academic health care centre mergers. Organization Studies, 23, 391–420. Kraatz, M. S., & Block, E. (2008). Organizational implications of institutional pluralism. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 243–275). London: Sage. Lazerson, M. (1988). Organizational growth of small firms. American Sociological Review, 53, 330–342. Legal Week. (2009, Autumn). Independent law firms in Europe. Retrieved from http:// www.legalweek.com/digital_assets/427/LW_ILFIE_2009.pdf. Accessed on June 1, 2010. Lok, J. (2010). Institutional logics as identity projects. Academy of Management Journal, 53(6), 1305–1335. Lounsbury, M. (2002). Institutional transformation and status mobility: The professionalization of the field of finance. Academy of Management Journal, 45, 255–266. Lounsbury, M. (2008). Institutional rationality and practice variation: New directions in the institutional analysis of practice. Accounting, Organizations and Society, 33(4), 349–361. Lowendahl, B. R. (2000). The globalization of professional business service firms: Fad or genuine source of competitive advantage. In Y. Aharoni & L. Nachum (Eds.), Globalization of services: Some implications for theory and practice (pp. 142–162). London: Routledge. Lubatkin, M. H., Ling, Y., & Schulze, W. S. (2007). An organizational justice-based view of self-control and agency costs in family firms. Journal of Management Studies, 44, 955–971. Maister, D. (1997). Managing the professional service firm. New York, NY: Free Press.

96

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

Malhotra, N., Morris, T., & Hinings, C. R. (2006). Variation in organizational form among professional service organizations. Research in the Sociology of Organizations, 24, 171–202. Marquis, C., & Lounsbury, M. (2007). Vive la resistance: Competing logics and the consolidation of US community banking. Academy of Management Journal, 50, 799–820. Micelotta, E. R. (2010). Professional service firms between resistance and change. Saarbru¨cken, Germany: LAP Lambert Academic Publishing. Micelotta, E. R., & Washington, M. (2011). Defending your institutions? The case of Italian lawyers. Best paper proceedings academy of management conference, San Antonio. Miller, D., Le Breton-Miller, I., & Lester, R. H. (2011). Family and lone founder ownership and strategic behaviour: Social context, identity and institutional logics. Journal of Management Studies, 48(1), 1–25. Morgan, G., & Quack, S. (2005). Institutional legacies and firm dynamics: The growth and internationalization of UK and German law firms. Organization Studies, 26, 1765–1785. Morgan, G., & Quack, S. (2006). The internationalization of professional service firms: Global convergence, national path-dependency or cross-border hybridisation? Research in the Sociology of Organizations, 24, 403–431. Nanetti, R. Y. (1988). Growth and territorial factors. The Italian model of social capitalism. London: Pinter Publishers. O’Brian, H. (2006). Italy feature; divorce, Italian style. The American Lawyer, January 1, 2006. Orru, M. (1991). The institutional logic of small firm economies in Italy and Taiwan. Studies in Comparative International Development, 26(Spring), 3–28. Pammolli, F., Cambini, C., & Giannaccari, A. (2007). Introduzione: Liberalizzazioni e concorrenza in Italia. In F. Pammolli, C. Cambini & A. Giannaccari (Eds.), Politiche di liberalizzazione e concorrenza in Italia (pp. 7–26). Rome: Il Mulino. Paci, M. (1980). Struttura e funzioni della famiglia nello sviluppo industriale e periferico. In M. Paci (Ed.), Famiglia e mercato del lavoro in un’economiaperiferica (pp. 9–70). Milano: Franco Angeli. Padgett, J. F., & Ansell, C. K. (1993). Robust action and the rise of the Medici, 1400–1434. American Journal of Sociology, 98, 1259–1319. Portes, A., Haller, W., & Guarnizo, L. (2002). Transnational entrepreneurs: An alternative form of immigrant economic adaptation. American Sociological Review, 67, 278–298. Purdy, J. M., & Gray, B. (2009). Conflicting logics, mechanisms of diffusion, and multilevel dynamics in emerging institutional fields. Academy of Management Journal, 52(2), 355–380. Quintieri, B. (2006). I distretti industriali: dal local al globale. Roma: Fondazione Manlio Masi. Rao, H., Monin, F., & Durand, R. (2003). Institutional change in Toque Ville: Nouvelle cuisine as an identity movement in French gastronomy. The American Journal of Sociology, 108, 795–843. Reay, T., & Hinings, C. R. (2005). The recomposition of an organizational field: Health care in Alberta. Organization Studies, 26, 349–382. Reay, T., & Hinings, C. R. (2009). Managing the rivalry of competing institutional logics. Organization Studies, 30, 629–652. Schulze, W. S., Lubatkin, M. H., & Dino, R. N. (2003). Toward a theory of agency and altruism in family firms. Journal of Business Venturing, 18, 473–490.

Beyond the Family Firm

97

Scott, W. R. (2005). Institutional theory: Contributing to a theoretical research program. In K. G. Smith & M. A. Hitt (Eds.), Great minds in management (pp. 460–485). Oxford: Oxford University Press. Scott, W. R., Ruef, M., Mendel, P., & Caronna, C. (2000). Institutional change and health care organizations: From professional dominance to managed care. Chicago, IL: University of Chicago Press. Segal-Horn, S., & Dean, A. (2007). The globalization of law firms: Managerial issues. International Journal of Service Industry Management, 18, 206–219. Segal-Horn, S., & Dean, A. (2009). Delivering ‘‘effortless experience’’ across borders: Managing internal consistency in professional service firms. Journal of World Business, 44, 41–50. Sforzi, F. (2009). Section 6: Empirical evidence. In G. Becattini, M. Bellandi & L. De Propis (Eds.), A handbook of industrial districts (pp. 323–342). Cheltenham: Edward Elgar Publishing. Sharma, P., Chrisman, J. J., & Gersick, K. E. (2012). 25 Years of family business review: Reflections on the past and perspectives for the future. Family Business Review, 25(1), 5. Sher, P. (2003). Ethnic innovation networks: A strategic linkage explanation for intellectual advantage in international business. Taiwan Academy of Management Journal, 3, 41–58. Smets, M., Morris, T., & Greenwood, R. (2012). From practice to field: A multilevel model of practice-driven institutional change. Academy of Management Journal, 55(4), 877–904. So, Y., & Walker, A. (2006). Explaining Guanxi: The Chinese business network. London: Routledge. Spicer, A., & Sewell, G. (2010). From national service to global player: Transforming the organizational logic of a public broadcaster. Journal of Management Studies, 47, 913–943. Teubner, G. (2001). Legal irritants: How unifying law ends up in new divergences. In P. Hall & D. Soskice (Eds.), Varieties of capitalism (pp. 417–441). Oxford: Oxford University Press. The Lawyer. (2001). Freshfields-Chiomenti merger collapses. The Lawyer, January 15. Thornton, P. (2004). Markets from culture: Institutional logics and organizational decisions in higher education publishing. Stanford, CA: Stanford University Press. Thornton, P., & Ocasio, W. (1999). Institutional logics and the historical contingency of power in organizations: executive succession in the higher education publishing industry, 1958–1990. American Journal of Sociology, 105, 801–843. Thornton, P. H., & Ocasio, W. (2008). Institutional logics. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 99–129). London: Sage. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure, and process. Oxford: Oxford University Press. Uzzi, B. (1997). Social structure and competition in inter-firm networks: the paradox of embeddedness. Administrative Science Quarterly, 42, 35–67. Walker Fields, I. (2004). Family values and feudal codes: the social politics of America’s twentyfirst century gangster. The Journal of Popular Culture, 37, 611–633. Weiss, L. (1984). The Italian state and small business. Archives Europeenes de Sociologie, 25, 214–241. Whitley, R. (1999). Divergent capitalisms: The social structuring and change of business systems. Oxford: Oxford University Press.

98

SAMANTHA FAIRCLOUGH AND EVELYN R. MICELOTTA

Whitley, R. (2005). Changing capitalisms? Internationalization, institutional change, and systems of economic organization. New York, NY: University Press Inc. Winfield, N. (2010). How the pope got his US lawyer. Associated Press, April 17, 2010. Retrieved from http://tricountyi.net/newap/D9F58IJ00.html. Accessed on May 27, 2010). Zahra, S. A., & Sharma, P. (2004). Family business research: A strategic reflection. Family Business Review, 17(4), 331–346. Zambon, S. (2003). Accounting in Italy. In P. Walton, A. Haller & B. Raffournier (Eds.), International accounting. London: Thompson Learning. Zilber, T. B. (2011). Insitutional multiplicity in practice: A tale of two high-tech conferences in Israel. Organization Science, 22(6), 1539–1559.

PUTTING NEW WINE IN OLD BOTTLES: UTILIZING RHETORICAL HISTORY TO OVERCOME STIGMA ASSOCIATED WITH A PREVIOUSLY DOMINANT LOGIC Shilo Hills, Maxim Voronov and C. R. (Bob) Hinings ABSTRACT In this paper, we seek to highlight how adherence to a dominant logic is an effortful activity. Using rhetorical analysis, we show that the use of rhetorical history provides a key mechanism by which organizations may convince audiences of adherence to a dominant logic, while also subverting or obscuring past adherence to a (currently) subordinate logic. We illustrate such use of rhetorical history by drawing on the case study of Ontario wine industry, where wineries use rhetorical history to demonstrate adherence to the logic of fine winemaking, while obscuring the industry’s past adherence to the now-subordinate and stigmatized logic of alcohol making. Implications for future research on institutional logics are discussed. Keywords: Institutional logics; institutional complexity; practice; institution

Institutional Logics in Action, Part B Research in the Sociology of Organizations, Volume 39B, 99–137 Copyright r 2013 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0733-558X/doi:10.1108/S0733-558X(2013)0039A&B017

99

100

SHILO HILLS ET AL.

The institutional logics perspective has emerged as a leading body of research in organization theory. The notion of institutional logics (Friedland & Alford, 1991; Thornton & Ocasio, 2008; Thornton, Ocasio, & Lounsbury, 2012) refers to the ‘‘the socially constructed, historical patterns of material practices, assumptions, values, beliefs, and rules by which individuals produce and reproduce their material subsistence, organize time and space, and provide meaning to their social reality’’ (Thornton & Ocasio, 1999, p. 804). A key contribution of this perspective has been to enable scholars to more fully acknowledge the institutional contradictions and contestations that permeate many organizational fields, impacting both people and organizations (e.g., Creed, Dejordy, & Lok, 2010; Lounsbury, 2007), as actors attempt to either displace one logic with another, defend their preferred logic or navigate multiple logics that coexist within a field on an ongoing basis (Pache & Santos, 2010; Reay & Hinings, 2009). Whereas researchers have begun to focus attention on understanding actors’ efforts involved in dealing with conflicting institutional logics and the conflicting pressures resulting from such conflicts and contradictions (Greenwood, Raynard, Kodeih, Micelotta, & Lounsbury, 2011), most research on situations where there is a dominant logic has neglected the study of processes by which actors try to adhere to dominant logics, and glossed over variation in actor adherence implying that adherence is automatic and effortless. However, adherence to a logic that is currently dominant might involve a great deal of effort that requires actors to not only utilize particular kinds of knowledge and skills (Emirbayer & Mische, 1998) but also engage in a social performance that is convincing to varied audiences (Alexander, 2004). We argue that a key aspect of the actor’s or the group’s adherence to a currently dominant institutional logic involves strategic construction and reconstruction of the past to simultaneously reconstruct actors’ identities and project the associated images convincingly in accordance with the dominant logic (Wry, Lounsbury, & Glynn, 2011), while erasing the memories of past adherence with a previously dominant logic (Zerubavel, 2003). Such use of rhetorical history (Suddaby, Foster, & Trank, 2010) extends beyond decoupling (e.g., Meyer & Rowan, 1977) because it utilizes rhetorical constructions of history to demonstrate actors’ ongoing commitment to the dominant institutional logic and obscure their past adherence to the now-subordinated logic. In other words, the past is reinterpreted to suit the needs of the present. This might be especially salient in fields where remnants of the previously dominant logic are now not only subordinate but also illegitimate and perhaps even a source of collective stigma for actors

101

Putting New Wine in Old Bottles

(Hudson, 2008). Examples of manifestations of previously dominant but now stigmatized logics might range from bureaucratic control systems (Gouldner, 1954) in present day corporations (Hamel & Prahalad, 1996) and government agencies (Du Gay, 2008) to more extreme examples, like slavery in the American South (Clarke & Fine, 2010). We illustrate the important use of rhetorical history as a mechanism for demonstrating adherence to a dominant logic by drawing on a case study of the Ontario wine industry where the hierarchy of the logic of alcohol making that dominated until 1988 has since been rebalanced with the logic of fine winemaking. Table 1 outlines the differences between these two sets of logics

Table 1.

Two Logics of Winemaking.

Commercial Logic of Alcohol Making Characteristics Organizational identity

Efficiency over quality Commercial winemaking for ad hoc purposes

Legitimacy

Whatever market will bear

Authority structures

Government and Liquor Control Board of Ontario (LCBO)

Mission

Creating inexpensive wine for a local commercial market

Focus of attention

North American grapes (labrusca varietals)

Strategy

Creation of wines for purposes of wide alcohol consumption and intoxication Capital committed to high quantity production of wines Local market Frequent changes in ownership

Logics of investment Consumers Rules of succession

Aesthetic Logic of Fine Winemaking

Viticulture and viniculture Unique combinations of techniques and terroir Artisan acclaim and authenticity Internationally accepted quality standards Regulated geographic areas, natural sugar levels, grape growing techniques, and production methods Old World winemaking techniques Vintage variation Vitners Quality Alliance (VQA) Government and Liquor Control Board of Ontario (LCBO) Grape growers marketing board Creating aesthetically pleasing, premium wines that can compete with the best wines produced globally European grapes (vinifera varietals) Viticultural areas or appellation systems Effective grape stewardship to express local terroir Capital committed to high quality production of wines Domestic and international markets Family estate firms

102

SHILO HILLS ET AL.

by using the framework provided by Thornton and Ocasio (1999). A shift in dominance of logics occurred that resulted in such a clear demarcation between these plural logics, that the logic of alcohol making became marginalized, and the practices associated with it became normatively perceived as illegitimate. By examining the routine communication in which wineries engage with their clients through newsletters and web sites, we illustrate that a significant component of demonstrating adherence to the dominant logic of fine winemaking involves attempts to import and connect to European winemaking history while omitting and repressing the history of cultural beliefs and material practices associated with the previously dominant logic of alcohol making.

THEORETICAL BACKGROUND Challenges of Adherence Institutional logics aid in understanding contradictions between beliefs and practices within and across institutions (Alford & Friedland, 1985) by recognizing interinstitutional connections occurring between society, organizations and individuals (Friedland & Alford, 1991) that both regulate, and provide opportunities for, agency (Friedland & Alford, 1991; Thornton & Ocasio, 1999). The notion of institutional logics was initially defined by Friedland and Alford (1991, p. 243) as ‘‘supraorganizational patterns of human activity by which individuals and organizations produce and reproduce their material subsistence and organize time and space.’’ These scholars conceptualized organizational actors as embedded within situational contexts of various social locations within broader interinstitutional systems. Thornton and Ocasio (1999) further integrated structural, normative and symbolic dimensions of institutions that complement one another by arguing that the values, interests and assumptions of both individuals and organizations are embedded within the institutional logics dominating in particular contexts. Various plural institutional logics are at play within different social orders, and these logics influence how reasoning and rationality occur and are perceived (Thornton et al., 2012). An important contribution of these authors is to emphasize the importance of cultural and symbolic structures in diffusion processes, and to emphasize the societal contextuality of actorhood. Institutional logics link material practices and the symbolic to explain contradictions between rituals, mental schemas, and behavior (DiMaggio,

Putting New Wine in Old Bottles

103

1997; Friedland & Alford, 1991). Contestations and contradictions in institutional environments allow actors to strategically exploit inconsistent and alternate logics to advocate for and influence institutional change (Rao & Giorgi, 2006; Seo & Creed, 2002) and exercise agency (Battilana, Leca, & Boxenbaum, 2009; Lounsbury, Ventresca, & Hirsch, 2003). Institutional logics researchers see organizations as involved with multiple institutional spheres (Thornton et al., 2012), and the fact that organizations encounter institutional complexity due to the multiplicity of logics that exist within any field or industry is primary (Greenwood et al., 2011), often with inconsistent and heterogeneous effects (Dunn & Jones, 2010). It has been acknowledged thoroughly that institutional complexity requires a great deal of work by actors to navigate their institutional environments effectively (Greenwood et al., 2011; Kraatz & Block, 2008; Pache & Santos, 2010). Scholars have dedicated considerable attention to the examination of actors’ effort and challenges in navigating such complexities. For example, some researchers have investigated the organizational arrangements needed to address multiple logics that apply conflicting pressures to organizations (Reay & Hinings, 2009). Others have attended to the theorization efforts undertaken by actors to navigate such contradictions (Lounsbury & Crumley, 2007) and to champion a new logic (Lounsbury, 2002; Suddaby & Greenwood, 2005; Wry et al., 2011). Yet, others attended to how even the most private aspects of people’s selfhood – their identities – get mobilized in the process of inter-logic contestation (Creed et al., 2010). However, the efforts of actors to adhere with a new, prevailing logic have not been paid adequate attention, implying that such adherence is unproblematic. Our main argument is that adherence with prevailing institutional logics is effortful, potentially difficult, and deserving of research attention in its own right, especially under conditions where actors have been historically situated in an environment that has valorized a different, opposing institutional logic. Even in environments where the hierarchy of logics is clearly established, actors may face difficulties in adherence (e.g., Lok & de Rond, 2013). Although we are not suggesting that dealing with conflicting logics is unimportant, we think that focusing on how actors adhere to a dominant logic is as important as understanding the efforts actors make to navigate institutional contradictions. As Emirbayer and Mische (1998) observe, even putatively routine and habitual behavior is skillful and requires selective attention to the environment, and being able to deploy from the appropriate repertoire of actions and maintaining expectations. It is, at least in part, a creative and improvisational activity

104

SHILO HILLS ET AL.

(Bourdieu & Wacquant, 1992; Emirbayer & Mische, 1998) that requires intricate interplay between particular actors and audiences (Alexander, 2004). Fields vary in terms of the salience of logic contradictions and the intensity of contestation among proponents of different logics. Some conflicts may appear to be settled (Helms, Oliver, & Webb, 2012) – at least temporarily – and actors then must focus their attention and effort on adhering to the logic that is currently the dominant one. Some logics might become so entrenched (e.g., Du Gay, 2008; Hoffman, 2001) as to require at least ceremonial adherence from all actors operating in a particular field. Taking seriously the effort involved in adherence to a dominant logic and the processes through which such adherence is perceived by others to be genuine is important to more fully explicate processes of mollifying complexity. Quite simply, navigating multiple logics and adhering to a clearly dominant logic are challenging, albeit in different ways, and both deserving of research attention.

Rhetorical History: Avoiding Stigma and the Work of Adherence Even in environments where there is a clear hierarchy between dominant and subordinate logics, efforts to demonstrate adherence to a prevalent logic involve strategically aligning actors’ identities and images with the dominant logic while at the same time demonstrating discontinuity and distancing the identities and images from the subordinate logic. In cases where organizations or industries have historically espoused subordinate logics, as in our case study of the Ontario wine industry, it may be necessary to forget (Anteby & Molna´r, 2012) this history by downplaying or obscuring it, and instead demonstrate continuity with legitimate logics. In this way, organizations engage storytelling efforts (Lounsbury & Glynn, 2001) that strategically utilize historical constructions (Foster, Suddaby, Minkus, & Wiebe, 2011; Suddaby et al., 2010), to both convince audiences of adherence, as well as to distance organizations from the stigma (Hudson, 2008; Rivera, 2008) of past association with a now illegitimate logic. Within the institutional entrepreneurship literature, the role of storytelling in the symbolic importation of alternate institutional logics has been highlighted (Lounsbury & Glynn, 2001; Vaara & Tienari, 2011; Weber & Dacin, 2011; Wry et al., 2011). According to Boje (1991, p. 106), storytelling constitutes ‘‘the institutional memory of the organization’’ and operates as a method by which actors can promote preferred logics. Collective activity in telling and retelling stories bridge past, present and future activities,

Putting New Wine in Old Bottles

105

allowing individuals to make sense of environments and events via supplementing individual with institutional memories (Boje, 1991, 1995; Gephart, 1991). Organizational legitimacy can be enhanced by expression of a coherent defining collective identity story (Wry et al., 2011) by making identity understandable to audiences and appropriately positioning identity to determine membership boundaries. In response to isomorphic pressures (Meyer & Rowan, 1977), stories often come to resemble those deemed predominately appropriate and that conform to normative logics and audience expectations (Golant & Sillince, 2007; Polletta, Chen, Gardner, & Motes, 2011). As a form of storytelling, history operates as an important resource from which organizations can frame identity and images and enhance audience perceptions of legitimacy (Anteby & Molna´r, 2012; Foster et al., 2011; Suddaby et al., 2010). Although often viewed as neither malleable nor manageable due to its existence or lack of existence (Ostrom, 1991), the notion of history has been extended to include a rhetorical dimension, whereby actors can construct particular versions of history that tap into acceptable values and norms (Anteby & Molna´r, 2012; Bates, 2011; MacMillan, 2008). The notion of rhetorical history is defined by Suddaby et al. (2010, p. 14) as a ‘‘means through which organizations can strategically mediate between their material and symbolic environments,’’ that can be used to manage organizational identity and image as a strategic resource. Generally, rhetorical history, and other forms of storytelling denote the strategic persuasion of audiences, using history, to shape perceptions of legitimacy and appropriateness (Lounsbury & Glynn, 2001). In this paper, we argue that rhetorical history works to facilitate organizations’ efforts to demonstrate adherence to a dominant institutional logic and obscure past adherence to a now-subordinate and illegitimate institutional logic. Dialogic in nature, history is a form of collective memory and storytelling that informs actors’ interpretations of the past. Foster et al. (2011) describe such discursive narratives as taking on two forms: the specific narratives linking events along plotlines grounded in temporal order and geographical space, and the schematic narratives that utilize meta-narratives of cultural tradition to associate significance and meaning that extends beyond the activity itself. This, we argue, involves tapping into institutional logics because they provide the toolkits from which these narratives can be constructed (Thornton et al., 2012). In contrasting historical narratives, eras are presented as eventful or uneventful, continuous or discontinuous, and take on different shapes and

106

SHILO HILLS ET AL.

spaces occupying our memories (Zerubavel, 2003). Mnemonic bridging and pasting, involved in linking present and past times, depicts historical continuities that connect traditions and current activities. On the other hand, historical discontinuity can be depicted through social and mental partitioning of the past into separate and distinct periods. Zerubavel’s (2003) distinction between continuous and discontinuous historical narratives provides a useful framework from which to understand the efforts of actors as they employ rhetorical history to demonstrate adherence to a dominant institutional logic, while suppressing historical alignment with subordinate logics. We illustrate the use of such strategic portrayals of history in demonstrating adherence to a dominant and simultaneous rejection of subordinate institutional logics. We do so by drawing on the case of Ontario wine industry to illustrate how organizations can utilize rhetorical history effectively in order to demonstrate their adherence to a dominant logic.

METHODS Research Setting We draw upon our case study of the Ontario wine industry, where the logic of alcohol making predominated until circa 1988. Since that time, great efforts on the behalf of actors within the industry have worked to replace this logic with the more aesthetic logic of fine winemaking. The subordinate logic is now widely perceived as unacceptable and illegitimate, and the historical association with it has been a source of stigma for the industry as a whole. Although winemaking in Ontario dates back to the middle of the 19th century, many attribute the birth of ‘‘serious’’ winemaking in Ontario to the founding of Inniskillin Winery in 1975 (Frank, 2008), because it was the first to adhere to the logic of fine winemaking that was a dominant logic in many established regions and would become an insurgent logic (Rao, Monin, & Durand, 2003) in Ontario, whose wine industry adhered to the logic of alcohol making (Bramble, 2009; Phillips, 2004). The distinction between the two logics can be made as follows: whereas the logic of fine winemaking constructs wine as an essentially artistic product that is appreciated, consumed, critiqued and priced based on its aesthetic qualities (Beverland, 2005; Colman, 2008; McCoy, 2005; Roberts & Reagans, 2007; Zhao, 2005); the logic of alcohol making constructs wine in a more banal way – as

Putting New Wine in Old Bottles

107

essentially an alcoholic beverage. The former logic is codified through a variety of norms (about grape growing, winemaking, labeling, marketing, etc.) that elevate wine into the realm of art appreciation. In contrast, the latter logic is less specific with respect to most aspects of wine production, and the various laws and regulations are typically limited to safety and retailing (Colman, 2008). In 1988, the Free Trade Agreement with the United States threatened to destroy the Canadian wine industry because it exposed Canadian wineries to increased international competition from both industrial wine producers as well as fine wine producers. The jolt boosted Ontario wineries adhering to the logic of fine winemaking, and eventually the government supported the efforts to institutionalize the logic. Thus, in the case of the Ontario wine industry, the Free Trade Agreement acted as a shock that, while nearly decimating the industry, also motivated the wineries to transform the way they operated and their identities, and the images they projected. It incentivized the industry to uproot the cold hardy labrusca grapes, such as Concord, that dominated Ontario vineyards due to ease of cultivation. Such grapes were frequently used to produce sweet, foxy wines that are illegitimate in accordance with the more aesthetic logic of fine winemaking. Instead, Ontario wineries planted vinefera grapes, such as Riesling, Chardonnay, and Pinot Noir, associated with high-quality wines as defined by the logic of fine winemaking. Since then, the industry has focused on producing high-quality wines in accordance with the logic of fine winemaking, and the logic of alcohol making has been marginalized, so much so, that even the wineries adhering to it seek to decouple their actions from their public images. Several Ontario wineries have since earned major international awards and garnered critical acclaim from such sources as the noted British critic Jancis Robinson, American critic Matt Kramer and the prestigious US-based Wine Spectator magazine, among others, and the region’s profile and prestige has been steadily increasing. All this indicates increasingly successful adherence to the logic of fine winemaking. In 2009, the industry produced 13 million liters of wine with a retail value of $210 million. The most famous and commonly exported Ontario wine product is Icewine, which in 2009 comprised 550,000 liters. The main export destinations for Ontario wine include United States, China, South Korea, United Kingdom, and Hong Kong, among others. One of the challenges encountered by the Ontario wine industry has been to erase the poor reputation that the pre-1988 wines had built among many domestic consumers. This has made it necessary for wineries and other actors interested in supporting the industry (e.g., government official, local

108

SHILO HILLS ET AL.

food activists, some wine critics, etc.) to continuously rationalize what has been done and explain why Ontario wine should be seen as ‘‘world class.’’ In other words, they must continuously remind audiences that they now adhere to the dominant logic of fine winemaking, while at the same time distancing themselves from their past association with the logic of alcohol making. Due to the valuing of tradition and nostalgia in the wine industry in general (e.g., Beverland, 2005; Colman, 2008; Ulin, 1995), the wine industry offers excellent opportunities to study rhetorical history, as wineries routinely attempt to associate themselves and their activities with images of a mythologized premodern past. Yet, as we argue below, our case offers instances of particularly interesting usages of rhetorical history for two reasons. First, it is very salient that rhetorical history is used not in idiosyncratic ways by particular wineries to differentiate or brand themselves. Rather, there is remarkable consistency across wineries with respect to the rhetorical devices being used as a means of demonstrating adherence to a dominant logic. This logic is somewhat foreign to the region because it lacks its own winemaking traditions that can be readily classified as adhering to a dominant logic of fine winemaking. Second, and related, unlike some regions that draw upon local historical and cultural events (e.g., Ulin, 1995), Ontario wineries’ rhetorical strategies are drawn primarily from foreign historical and cultural resources.

Data Collection The data sources used in this paper are excerpted from a broader study of the organization of Ontario wine industry wherein qualitative data was collected in various forms (interviews, observations, and documents) over a period of four years. This study consists of the analysis of 326 electronic newsletters from 40 wineries and web sites of 15 wineries. Our sampling was driven by the desire to incorporate wineries exhibiting a range of variables, such as size, age, commercial success, quality reputation, ownership structure, location, product range and pricing, among other factors, and to understand overall how the Ontario wine industry might be collectively utilizing different forms of rhetorical history. Newsletters and web sites offer a unique opportunity for researchers to understand the ongoing use of rhetorical history by organizations and the implicit rules at the field level that guide the selection and deployment of such techniques. We were especially interested in understanding how wineries tapped into various historical resources in order to demonstrate

109

Putting New Wine in Old Bottles

their adherence to the dominant logic of fine winemaking, and to essentially rebalance the hierarchy of dominance between the plural logics of alcohol and fine winemaking. While newsletters typically target loyal customers and other individuals that have ongoing relationships with particular wineries, web sites represent wineries’ attempts to represent themselves in a desirable manner to a more general audience. Thus, the two data sources offer useful complements to each other.

Analysis Rhetorical analysis offers a method for both the strategies employed and the broader context of relevant societal discourses and their links (Heracleous & Barrett, 2001; Sillince & Brown, 2009; Suddaby & Greenwood, 2005). Because newsletters and web sites have a persuasive intent and are designed purposefully to influence audience perceptions, attending to the rhetorical devices used in these narratives enhanced our understanding of the ways in which wineries are persuasively using history. Repeated readings of the web sites and newsletters, in consultation with the literature on rhetoric, allowed us to identify various rhetorical strategies identified in prior rhetoric research that could be used as sensitizing devices in our data analysis. Although a great many such devices are available, we focused on the ones that were used most commonly in the context of Ontario wineries’ deployment of rhetorical history, and we used them as a starting point to sift through the plethora of texts generated by the wineries. This decision was mainly pragmatic and intended to avoid ‘‘getting overwhelmed’’ by the extensive and varied usage of rhetorical history that was apparent in the data. Please see the appendix for descriptions of the key terms we used in the rhetorical analysis. Analytical Strategy The stages of analysis were similar for both the newsletters and the web sites. Yet similarities and differences between the two genres of texts were examined throughout the research process because they are directed at somewhat different audiences and might use rhetorical devices in different ways to theorize practices and activities. The first stage of the analysis involved reading over the newsletters and web sites to understand their general content and format. While reading, general impressions were noted, possible themes for coding identified, and notes taken on how and why these

110

SHILO HILLS ET AL.

themes were emerging. For example, the fact that wineries often use historical depictions of the land upon which their grapes are grown was noted as a potential theme. The coding stage of analysis entailed looking for words and phrases that were frequently used within the newsletters and web sites, primed by our research interest in rhetorical history. For example, phrases involving the notion of terroir1 were quite prevalent, and were easily coded under ‘‘terroir.’’ Themes within the web sites and newsletters were identified, as were broader ideas, concepts, activities, relations and terminology used. Because rhetorical analysis requires grasping the significance of a particular utterance in its context (Heracleous & Barrett, 2001), we designated the units of analysis to range from a paragraph to a whole newsletter – whatever smallest passage could be interpreted meaningfully. When a rhetorical strategy could not be identified or no link to rhetorical history could be made, no further analysis on a particular passage or even a document was undertaken. At this stage we began to use the aforementioned rhetorical devices to identify various discursive themes and interpretive repertoires that emerged from the body of texts as a collection. For example, we intended to understand what role actors’ use of terroir plays in the newsletters and web sites, and how and why they act rhetorically to persuade audiences. The notion of terroir here presented an interesting role, as it is used by wineries to both conform to broader, traditional Old World values while at the same time to confer a sense of distinctiveness and uniqueness to wineries by appealing to a more local history. As the newsletters and web sites were further categorized, other themes were identified and served as subcategories. For example, the notion of terroir was sometimes used in reference to soil composition as being similar to that of more legitimized wine regions, while at other times it was used to highlight the uniqueness of local plots of land. We attempted to substantiate the existence of shared discourses by comparing and contrasting these themes and categories, and by moving back and forth between broader, more general societal discourses and the specific utterances. For example, the use of terroir may be related to the broader societal construct of geological history, which details the origins and history of the development of earth, and essentially describes time as sequential in attempting to make sense of the events that have taken place in a region. As the analysis continued, and definitions of those categories were refined, new categories were identified that did not fit with existing categories of rhetorical strategies. Thus, it became necessary to analyze this manifestation of rhetorical history as a new category of strategies. The similarities,

111

Putting New Wine in Old Bottles

differences and patterns within and between subcategories were compared, contrasted and noted. The process of category building continued, in order to examine and interpret the newsletters and web sites, until no new themes were found and a point of saturation was reached. Most of the analysis was done by the first author, who had no access to the field and relied purely on the newsletters and web sites to do the coding. She periodically engaged the second author who had conducted the interviews and observations that were a part of the larger study of the organization of the Ontario wine industry to assess the similarities and differences of the interpretations resulting from the researchers’ differences and the access to different data. As the study progressed, the second author also independently analyzed a sample of the newsletters and web sites previously analyzed by the first author. Disagreements were discussed, and the coding was refined. This allowed the study to benefit from both detached analysis by the first author and in-depth industry insight acquired by the second author (Pratt, 2009).

FINDINGS The findings of this study demonstrate the importance of rhetorical history as a mechanism by which adherence to a dominant logic is achieved. In the routine communication attempts of web sites and newsletters, a significant aspect of demonstrating adherence to a dominant logic involves attempts to import and connect to traditional European winemaking history while repressing and even omitting the history of association with the previously dominant, now subordinate logic. Continuity with the Logic of Fine Winemaking Because the logic of fine winemaking is rooted primarily in European – especially French – winemaking conventions (Robinson, 2006), a very salient theme found in our data features Ontario wineries characterizing themselves as strict adherents of European traditions and techniques of winemaking. Traditional European winemaking privileges techniques and methods that aid the slow process of winemaking and embraces the influences of land, soil and climate, or terroir, in wines. The main themes were Following Old World Traditions, Terroir-Driven Winemaking, Traditional Farming Methods, and Claiming Geographical Similarity (see Table 2).

112

Table 2.

SHILO HILLS ET AL.

Additional Quotes Illustrating Continuity with Logic of Fine Winemaking.

Theme Following Old World Traditions

Illustrative Quotes and Rhetorical Devices Used Through centuries of winemaking history in the Old World, the art of the blend was always something that was given much attention and appreciation. [y] Most people are familiar with the blends of Bordeaux – probably some of the greatest in the world – which consist of primarily Cabernet Sauvignon, Merlot, and Cabernet Franc with smaller proportions of Petit Verdot and Malbec. Outside of Bordeaux, these blends are sometimes called ‘‘Meritage’’ blends, or in the case of Pillitteri, we simply call it a ‘‘Cabernet Merlot’’ blend – an assemblage consisting of the two Cabernet grapes, Franc and Sauvignon, and Merlot. Like our winemaking forefathers, Pillitteri Estates takes the art of blending very seriously as well. [characterization, metaphor, denotation, expertise] (Pillitteri newsletter) ‘‘All grapes are hand-picked and crafted into lively genuine wines using only gentle traditional winemaking techniques.’’ [characterization] (Frogpond website) ‘‘Jeff has a dedicated passion for producing fine wines in Prince Edward County. He holds a great respect for traditional, oldworld styles of wines and a passion for excellence.’’ [characterization] (Great Estates newsletter) ‘‘We are traditionalists and find the process of pulling a cork from a bottle a bit more sensuous than twisting off a cap. Saying this we purchase only high quality corks. Ken Douglas’s son David who works as assistant winemaker for the Murphy Good Winery in California sourced our most recent corks. David has, over the last several years, developed considerable expertise in selecting corks and conducts laboratory testing of each batch of corks offered for sale to his winery to assess its TCA content. He is now selecting our corks. Of our most recent batch, less than 1 cork out of 100 corks had detectable TCA. The industry norm is greater than 5 per 100 corks.’’ [characterization, expertise] (13th Street newsletter) ‘‘This wine was produced in the traditional method with the fermentation occurring in the bottle. It was disgorged after resting on the lees for three years.’’ [characterization, imagery] (13th Street newsletter) ‘‘Our Winemaster, Marc Bradshaw, was able to draw on some Old World techniques from France and Italy to help create this superb wine.’’ [characterization] (Pillitteri newsletter) ‘‘A Chianti-style red with bold, upfront fruit for lovers of traditional Italian cuisine. Serve it ever so slightly chilled to

113

Putting New Wine in Old Bottles

Table 2. (Continued ) Theme

Terroir-based Wines

Illustrative Quotes and Rhetorical Devices Used bring the evening to life. Enjoy with your grilled meat and veggie favourites.’’ [diatyposis, imagery] (Magonotta newsletter) ‘‘Slightly off dry, showing a beautiful balance of floral sweetness and citrus fruit along with the typical minerality of our terroir. Their birthplace, the Moira Vineyard, appears from above as an island nestled into an angle of the Niagara Escarpment. From its well-drained, mineral-infused soil comes wines whose trademark richness is interwoven with an exotic smokiness.’’ [imagery, denotation, personification] (Malivoire newsletter) ‘‘Wine has been associated with its point of origin for hundreds of years and is often referred to as an ‘expression of place’. Recognizing that the character of a wine is directly influenced by where the grapes are grown, European countries with a long history of winemaking slowly evolved appellation systems to identify different wines for consumers. The ‘terroir’ – the combination of location, soil, topography and climate – is an important indicator of the character of a wine and in many cases, its quality.’’ [metaphor, denotation] (13th Street newsletter) ‘‘With his knowledge and passion for winemaking, his style of traditional winemaking focuses on creating wines which showcase the purity of fruit and exhibit the unique terroir, or environment, in which the grapes are grown.’’ [characterization] (Cattail Creek newsletter) ‘‘The return is a wine that expresses the diversity and age of our vines, complexity in our soils and the vagaries of Mother Nature – all adding up to be a very fine wine!’’ [personification, imagery] (Stratus newsletter) ‘‘The Niagara Escarpment Bench provides drainage for air, frost and water. All of these can have a negative effect on growing grapes. They drain down the Escarpment to the flats below and particularly in the case of frost, are eventually warmed by the waters of Lake Ontario and return to moderate the cold frosty nights in Spring and Fall or freezing nights during the winter. Grapes do not like wet feet and the acreage is under-drained to follow the natural slope. The Syrah planted ravine acts as a heat sink for this particular heat loving variety.’’ [characterization, denotation, imagery] (Kacaba website) ‘‘The roots of a romance with the land run deep in the Lizak family. In 1946, matriarch Sophie Lizak and husband John planted the seeds of enterprise on a tender fruit farm of a dozen acres. The fertile soil and micro-climate of the Niagara peninsula fed his passion and as time passed, the customs of

114

SHILO HILLS ET AL.

Table 2. (Continued ) Theme

Traditional Farming Methods Used

Illustrative Quotes and Rhetorical Devices Used working the land were handed down from father, to son Ted Lizak to grandsony From the first vintage in 2000, it became clear Legends Estates Winery had been graced with excellent terroir. The individuality of each varietal, the nuance of the land, the essence of every fruit is evident in each bottle.’’ [characterization, metaphor] (Legends Estates newsletter) ‘‘When we first planted to vinifera grapes we knew that no mechanical harvesters could pick these sensitive grapes and harvest would have to be manual. Then, during our first harvest in 1985 new harvesting machines from Europe came to Niagara and mechanical harvesting of even vinifera varieties became the norm. Over the years we’ve done both, and find the quality from manual harvests is far superior to mechanical harvesting. Today all our grapes are harvested by hand.’’ [problemsolution, characterization] (13th Street newsletter) ‘‘By November our vineyards can be strange to behold, completely buried in soil to protect the buds, with only trellising wires and posts protruding. This means tying down 40,000 canes onto a low wire while on bended knee, ploughing soil over the canes with a tractor, and then reversing the process in the spring. The payoff, we hope, will continue to be dramatically higher bud survival rates that produce grapes with an enhanced minerality in the wines.’’ [characterization, imagery, denotation] (Chadsey’s Cairns website) ‘‘Other ‘tried and true’ measures we practice every year include carrying a small crop, which provides the vine with more carbohydrates and improves its winter-hardness, and hilling up the soil over the lower trunk of the vine. Hilling the soil provides protection of basal buds but provides no protection to the buds located higher on the vine.’’ [denotation, characterization] (13th Street newsletter) ‘‘This lovely wine was hand picked and sorted to perfection.’’ [characterization] (Cornerstone newsletter)

Following Old World Traditions At times, continuity with the Old World winemaking tradition is expressed in most general terms by asserting that a winery operates by strictly adhering to Old World norms and practices. One such effort to maintain these traditions is found in a newsletter, and again in a web site passage, of Foreign Affair Winery as they rely upon the Amarone-style of drying grapes prior to making wine from them: ‘‘We are excited to report [y] the long anticipated 2007

Putting New Wine in Old Bottles

115

100% Amarone-style Cabernet Franc [y] We are confident that this latest vintage is the ultimate expression of what our wines are all about and for what we have continuously and patiently been striving!’’ The straightforward message from this statement relates to the traditional methods and the logic of fine winemaking used to create this wine. The notion of terroir is less obviously insinuated from the pronouncement that this wine is the ‘‘ultimate expression’’ of what the winery stands and works hard for. What is left unsaid, but is still nonetheless bundled into the interpretation is the role that terroir, the earth, weather, and sun, play in the expression of character in wine. Thus, a commitment to terroir is, in effect, communicated by metaphorically linking the wine and the ultimate expression of the winery, the land upon which the grapes are grown, and thus, the local history that contributed to the characteristics of the land expressed in the wine. Furthermore, the fact that this wine is a vintage that expresses this metaphor links to global and traditional values of age and slowness in winemaking that are consistent with the fine winemaking logic. This passage involves expression of the logic of fine winemaking, in both the Amarone method of winemaking, as well as communicating a dedication to the expression of terroir. As a whole, this passage indicates a commitment to traditional methods and values in winemaking, and links to the logic of fine winemaking, through its use of the Old World mythology of terroir to characterize this wine as a vintage expression of local history. In this newsletter passage, the Amarone-style is merely mentioned. This passage works to characterize this winery as traditional and as adhering to the logic of fine winemaking in their approach to creating this wine, and as committed and dedicated to producing quality wines that are an ‘‘expression of what our wines are all about.’’ In a web site narrative, the story behind the commitment to Amaronestyle winemaking is explicated in much more detail: yItaly introduced us to a whole new world of fine food, wine, and the simple joys of everyday life. We discovered the wonders of Amarone styled wines in the northern part of the country. The more we learned about them, the more excited we became at the prospect of bringing this incredible craftsmanship back with us to Ontario [y] And so we didy. [y] In the time-held tradition of Amarone, we then proceeded to delicately dry our grape stock in a barn until each of the varietals were perfect for winemakingyWe are very proud of the fact that our wines are Canadian and home grown. But [y] we were greatly influenced by the rich Italian (Veneto Region) history of appassimento (drying process). It was in Veneto where it all started and reference to amarone or recioto belong only to those winemaking visionaries. They were and still are the legendary pioneers for whom we have huge respect and admiration. Our Canadian

116

SHILO HILLS ET AL.

stylistic interpretation with locally grown vinifera varietals are referred to as appassimento (grape drying or Reductio Method) rather than the uniquely owned names found in Veneto. A glass of our wine tells the rest of the story.

The most explicit references to history in this selection involve connections to Old World Traditions and local history, which are both aspects involving a globally accepted logic of fine winemaking. The Amarone method, itself somewhat controversial with respect to the extent that it is truly ‘‘traditional’’ even in the context of European French-centric tradition (Robinson, 2006), is ascribed to the traditional wine craftsmen of Amarone and used rhetorically to indicate ‘‘tradition’’ that the winery seeks to carry on. Because the techniques are imitated locally, in local soils with their own characteristics, rather than in Amarone, they are referred to as appassimento, or Amarone-style. The reference to local terroir, and thus the local history and geology of the land that allow for this ‘‘locally grown vinifera,’’ is implicit, and it is integrated into the broader narrative of demonstrating continuity with mythologized Old World history. The family connection to northern Italian wines is explained in more detail on the web site, as is Amarone-style winemaking. We learn the history of the method, and how this family winery adapted these putatively traditional Italian methods to Canadian winemaking. Wine is personified at the end of this passage, as it takes on the ability to tell a story. Such a personification of wines represents a traditional construction of wine as having the characteristics of living entities (Kramer, 2004; McCoy, 2005); by personifying wine, the wine is given an individual identity which essentially brings the wine to life in the mind of a reader. By rhetorically tapping into this historical construction, terroir and widely accepted repertoire, this passage indicates alignment with the dominant logic of fine winemaking. Terroir-Driven Winemaking The concept of terroir is heavily relied upon to demonstrate the continuity with the Old World history of winemaking in a more specific way, and Ontario wineries overwhelming rely upon descriptions of terroir, which indicates their commitment to Old World winemaking traditions and the logic of fine winemaking. For example, Henry of Pelham seeks to define this notion using the rhetorical device of denotation in one passage on its web site: ‘‘The origin of the grapes used to make wine has long been important in traditional winemaking regions to both winemakers and wine drinkers. The combination of location, soil, topography and climate – the terroir – is an important factor in determining the character of a wine and, in many cases, its quality.’’

Putting New Wine in Old Bottles

117

The traditional ideology of terroir is very clearly discussed in this passage, and by tapping into this ideology, this passage works to demonstrate a commitment to the traditional logic of fine winemaking. In this particular excerpt, the notion of traditional winemaking, and thus the connection to this dominant logic is made very explicit; however, the local aspect of this notion is somewhat downplayed. Here, as in many other narratives that describe wines and their terroir, wine is personified as having character, and the land is deemed responsible for imbuing wine with particular characteristics. Additionally, this passage, much like the narratives of other wineries, provides a definition of terroir that strategically links the land of the winery to the quality wines it produces. The explicit overtone is that of traditional winemaking and continuity with the logic of fine winemaking, while local terroir and the history of the land must be inferred from the definition provided. A passage on the Cave Springs web site offers a further explanation of the concept of terroir: ‘‘Great wines come from great vineyards. As winegrowers, our goal is to honor the rich, expressive character of our vineyards, and to craft wines that capture the essential essence – the terroir – that defines our land.’’ The key theme of this passage again relates to traditional values of terroir and indicates that the role of the winegrower is to ‘‘capture the essential essence’’ of the vineyard in the wines that are produced. Further, we see the notion of the ‘‘craft’’ of winemaking. Such references to winemaking as a craft, and especially to wines as handcrafted, are highly prevalent throughout the newsletters and web sites of Ontario wineries, and they highlight traditional conceptions of winemakers as artisans who, through laborious methods, create quality wines by hand. However, the role of terroir is not diminished by the efforts of the winemaker; indeed, the winemaker is often seen, as here, merely as bringing out the qualities of the land that are already inherent within the grapes. In evoking terroir in this manner, wineries seek to demonstrate adherence to the logic of fine winemaking by framing their activities and practices in light of the age-old notion that harks back to the mythologyrich and putatively glorious past of French winemaking (Kramer, 2004; Robinson, 2006). Traditional Farming Methods Frequently, wineries elaborate their claims of terroir-based winemaking by emphasizing the Traditional Farming Methods being used in their winemaking, thereby highlighting the primacy of the land and vineyards over the winemaking that is a hallmark of the terroir philosophy

118

SHILO HILLS ET AL.

(Robinson, 2006). A passage on the web site of Sprucewood Shores also defines terroir: The creation of excellent wine starts in the vineyard. The French have coined the term ‘terroir’ to define the factors that have an impact on the unique taste of each wine. These factors are the specific compilation of soil conditions, the trellising method used, the amount of precipitation, the proximity to the moderating influence of bodies of water, the intensity of heat units and the length of time for growing and maturing the grapes.

Through a form of denotation, the notion of terroir is seen as involving the methods of growing grapes and proximity to water, as well as the climate and soil. Thus, this message pertains explicitly to global traditions that value terroir based, and fine, winemaking. There is also an explicit reference to the ‘‘French’’ aspect of terroir, which further indicates continuity with the traditions of this Old World wine producing region, also indicating commitment to the logic of fine winemaking. However, less explicit is the reliance upon local geological history for providing the ‘‘bodies of water’’ and ‘‘length of time for growing’’ that influence grape growing. Both in newsletters and on web sites, wineries tend to detail the laborious, hands on methods of growing and harvesting grapes. Such references implicitly rely upon Old World notions of terroir and commitment to the expression of the land in the creation of quality wines, which are all concepts that are bundled together under the logic of fine winemaking. While some Ontario wineries, particularly in earlier newsletters, describe the work associated with winemaking as grueling drudgery, most wineries describe the work associated with traditional methods, including hand harvesting, as motivated by passion and a dedication to quality and the production of fine wines. For example, an older newsletter of 13th Street described grape growing as follows: First of all, let me be blunt, there is no romance in growing grapes, just a lot of hard work consisting of a series of repetitive tasks that typically have to be completed within a narrow window of opportunity. Why do we do it? I guess because it is very satisfying at the end of the day to look back on what you have achieved, whether it is pruning a row of grapes or cutting the grass in the vineyard. We view the vineyard as an extension of our garden and continue to feel challenged by the prospect of growing the best grapes we can.

While describing the process of growing grapes as lacking in romance, this passage still portrays Old World values, and the logic of fine winemaking, implicitly as the owners are depicted as satisfied with their work. The explicit message of this selection is that grape growing is hard work, while the more implicit idea is that this hard work is pleasurable. Despite the depiction of this

Putting New Wine in Old Bottles

119

work as laborious and less than romantic, this passage raises and answers a question that readers will perhaps ask, and at the same time characterizes the winemakers as dedicated to growing quality grapes. The motivation behind partaking in the ‘‘series of repetitive tasks’’ is the satisfaction achieved after labouring to grow the ‘‘best grapes we can.’’ Most other wineries tend to instead portray the more romantic side of grape growing and winemaking. On many occasions, readers of newsletters are invited to get involved with the work that goes into harvesting and are urged to take part in exciting opportunities to experience the traditions involved in fine winemaking. In one newsletter from Flat Rock Cellars, readers are invited to partake in the exciting activities of the Grape Stomp: ‘‘Harvest is a busy and exciting time, and this is your chance to experience it! Roll up your pants and make wine the old fashioned way. Book now, spaces are filling up fasty’’ In this passage, identification is used to offer readers a chance to take part in the creation of a wine by performing the traditional method of grape stomping. The winery is characterized as excited at the prospects of the harvest and to share this busy time with readers, ‘‘the old fashioned way,’’ clearly indicating to audience members that this winery is devoted to the principles of the logic of fine winemaking. Invitations like this one extend the enjoyment of hard work to readers by offering the chance to partake in various events around the winery. Flat Rock also invites their newsletter audience to join in similar activities: ‘‘As with every fall there are always many things to experience at Flat Rock Cellars. Please take the time to truly immerse yourself in the joys of harvest. It’s what truly motivates all of us at Flat Rock Cellars.’’ Together, these passages characterize the wineries as enjoying the hard work associated with the harvest season and as faithful to the age-old principles of terroir winemaking, so much so that they want readers to join in the enjoyment and excitement.

Discontinuity with Logic of Alcohol Making We noted concerted efforts to obscure or downplay the local winemaking history (see Table 3) prior to 1970s (or 1980s, in some cases), because early winemaking adhered to the now-marginalized logic of alcohol making. Narratives strive to emphasize the newer, more legitimate winemaking history in Ontario, while marginalizing or even disparaging the more distant past, characterized by adherence to the logic of alcohol making.

120

Table 3.

SHILO HILLS ET AL.

Additional Quotes Illustrating Discontinuity with Logic of Alcohol Making.

Young industry

Disparaging local winemaking history

‘‘Made from our youngest vines (still 20+ years old, the same age as most ‘Old Vines’ in Ontario), it represents Chardonnay at its purest.’’ [characterization, denotation, metaphor] (Daniel Lenko newsletter) ‘‘There is a widespread impression that Canadian viticulture and viniculture did not exist in a meaningful way before the late 1900s. True, early Canadian wine had little or no impact (or appeal) beyond its domestic markety Vineyards planted to native hybrids spread across southern Ontario through the mid1800s but the cultivation of traditional wine grape varieties was not yet possible. North American pests and the Canadian climate’s legendary adversity thwarted early efforts.’’ [aphorism] (Malivoire website) ‘‘On July 31, 1975, Inniskillin Wines incorporated and its founders Karl J. Kaiser and Donald J.P. Ziraldo were granted the first winery license in Ontario, Canada since prohibitiony Then in 1984, Kaiser made the most of the extreme Canadian winter to produce his first Icewine – a pivotal point for Inniskillin. The world began to take notice.’’ [characterization] (Inniskillin website)’’ ‘‘So much happens when a wine region is born. The landscape is altered. Long-time crop fields are replaced with row upon row of unfamiliar cedar posts and wires, virtual factories of growth sprouting overnight. Laconic farmers with squint-lined eyes, well schooled in the ebbs and flows of agriculture, politely listen to lectures on grape growing delivered by the new kids in the fields, winegrowers, if you please. A buzz quickly fills the air: will the vineyards usher in a new era of economic prosperity and sophistication, or unexpected tax hikes for traditional farmers startled to discover they’re now living on prime grape terroir, for heaven’s sake. As for the pioneer grape growers, we are like characters in a black and white movie who have entered one of those goofy car races across treacherous terrain, careening from madcap adventures to tragic mishaps.’’ [denotation, metaphor, characterization] (Chadsey’s Cairns website), ‘‘As the native labrusca varieties used for Canadian ports and sherries were not suited to table wines, hardy French hybrids seemed to be the answer. Marechal Foch vines were planted widely across the Niagara Peninsula during the third quarter of the 20th century, although the increasing demand for vinifera acreage ensured that few of these plantings survived into the 1990s.’’ [problem-solution] (Malivoire newsletter) ‘‘At the time vinifera grapes were not yet as popular as they are today and so the decision to grow only vinifera grapes was a

Putting New Wine in Old Bottles

121

Table 3. (Continued ) difficult one for us to make, but we were encouraged by our mentor Paul Bosc at Chateau des Charmes.’’ [problem-solution] (13th Street newsletter) ‘‘Both Karl Kaiser and Donald Ziraldo believed their future in the wine business was dependent on using the Vitis vinifera grapes, the preferred family of grapes used to produce the fine wines in the great wine regions of the world. When sourcing these limited grapes became a challenge in the early 70’s, Ziraldo took the lead and planted a vinifera vineyard which included Riesling, Chardonnay and Gamay and formed the quality base for Kaiser to work with.’’ [characterization, denotation] (Inniskillin website) ‘‘The existing vineyard was very run down and after one year of picking the Fredonia, Concord and Niagara grapes; it was obvious that the whole farm needed to be replanted.’’ [problemsolution] (13th Street)

Young Industry Although winemaking in Ontario dates back to the mid-1800s, the industry is often characterized as young. One example that obscures the history of winemaking, as it adhered to the subordinate logic of alcohol making can be found in a Pillitteri newsletter, as it describes the industry as new: When compared to the likes of France, Italy, and Greece, all of which have hundreds or even thousands of years of wine making history under their belts, Canada has a mere 40 or so years of solid wine making roots to draw from. Where Niagara lacks in experience, we make up in creativity and flexibility in our wine making practices.

This passage works to characterize the industry as young, new and lacking in experience. Further, it describes Canada as not having a history that it can draw upon, but what must be inferred from this description is that Canada does not have a legitimate history or acceptable experiences in winemaking, as can be read from the lack of ‘‘solid winemaking roots.’’ This phrase also uses metaphor, as winemaking history acts as the roots from which present practices may or may not be drawn upon. Niagara is characterized, and personified, as creative and flexible, as a result of not having these historical roots to build from (a paradoxical contrast to the oft repeated claim that Ontario winemakers follow the tradition rigidly). In essence, this passage privileges the widely accepted and legitimated historical winemaking practices of European wine regions, and obscures the fact that Ontario has a long history of winemaking, but that this history involved unacceptable practices and techniques in their production of wine that used

122

SHILO HILLS ET AL.

to comply with a logic of alcohol making, rather than those practices consistent with the dominant and normatively proper logic of fine winemaking. The historical adherence of Ontario winemakers with this subordinate logic is suppressed by describing the industry as only existing within the last 40 years, rather than bringing to light the existence of an industry in the region for two hundred years. This relatively young portrayal of the history of wine production in Canada can be seen as a way of eradicating collective memories of poor quality wine production in Canada prior to the 1980s. Even the oldest wineries only make reference to the part of their winemaking history in Canada that falls within this timeframe, despite the fact that many draw on a family heritage of winemaking dating sometimes centuries prior in more established, European wine regions. Another depiction of Ontario as a newer wine growing region can be found in a 13th Street newsletter: ‘‘Ontario is a much younger wine growing area and has so far identified three primary viticultural areas and 12 subappellations.’’ What needs to be implied from the phrase ‘‘much younger’’ is the idea that Ontario is much younger than various established wine growing regions, and this works to characterize the industry as such. This comparison is implicit, but is more easily seen as the number of sub appellations of Ontario is identified. Appellation systems have historically been very important to quality wine production in Europe, particularly in France, where the appellation system was initially introduced (Colman, 2008). Often, wineries in Europe only identify their wines in terms of these appellations or Chateau locations. This reference to the twelve subappellations of Ontario alludes to the adherence to the logic of fine winemaking. At the same time, the reference to the young age of the industry is downplaying, and even completely obscuring the historical existence of an industry that was less than desirable within Ontario. A web site selection from Featherstone demonstrates one way of minimizing poor local history and distancing the region from historical commitment to the subordinate logic of alcohol making by highlighting instead the fact that their winery holds some of the oldest vines of a particular variety in the region: ‘‘The Chardonnay and Cabernet Franc, planted in 1986, were left to thrive as was the Riesling, which was planted in 1978 and is some of the oldest Riesling in Niagara. Today the entire grape crop is bottled to produce Featherstone wines, with approximately 5,000 cases a year being vinified.’’ This passage utilizes ideas of traditional values and local history as the noble European vinifera varietals that have been grown on the land of the winery since the late 1970s are highlighted. Of note is the phrase ‘‘were left

Putting New Wine in Old Bottles

123

to thrive,’’ as it may be assumed that other varietals historically were grown and removed from the property, and that those varietals were likely the undesirable native labrusca varietals. This omission is important, because it indicates a repression of local history, wherein it was uncommon prior to these dates for grape growers to cultivate the traditionally honored varietals of grapes involved in fine winemaking. The Riesling vines are described as some of the oldest in the region, but upon looking at the date of its planting, we can see that these vines are much younger than many vines in Europe, which are often hundreds of years old. In characterizing these Riesling vines as old, we understand that they are only old in relation to the noble grape vines currently existing in Ontario, but are actually significantly younger than those of Europe or other world class wine regions. Thus, in a roundabout way, and more implicitly, this passage too describes the Ontario wine region, and its vines, as young, when compared to Old World wine regions. Disparaging Local Winemaking History Another way in which the historical association of the region with the logic of alcohol making is concealed is by explicitly disparaging the local winemaking prior to the diffusion of the logic of fine winemaking in the region. One of the most explicit illustrations of attempts to disparage local history can be found on the web site for Henry of Pelham, which ties its founding story to the uprooting of the then-popular labrusca grapes, associated with the logic of alcohol making: ‘‘In 1984 when our father, along with each of us (Matthew, Daniel and Paul), tore out the Concord and Niagara grapes at our own expense, people thought we were crazy. When we planted Riesling and Chardonnay under newly contoured hills and under drainage, many said they would never grow.’’ While this passage itself includes a rare reference to a time where labrusca grapes were ordinarily grown, and the planting of vinifera varietals was seen as out of the ordinary, this passage does not explicitly detail the routine practices of most as such. Instead, this passage refers to the fact that people thought that attempts to plant vinifera grape varietals were ‘‘crazy,’’ thus characterizing this winery as pioneering the development of fine winemaking in the region because these unusual activities were undertaken, going against the normative expectations that fell under adherence to the logic of alcohol making. What is left unsaid is that most grape growers were largely unwilling to undertake such activity, as their investment in and adherence to the logic of alcohol making took priority, at that time, over attempts to

124

SHILO HILLS ET AL.

comply with the global standards and values of fine winemaking. In this manner, the winery also characterizes itself as a pioneer of the diffusion of the logic of fine winemaking. Similar characterization is found in the passage on the Inniskillin web site, which also indicates, but does not explicitly mention, the poor state of the industry in its reference to a gap in the Ontario wine market: While tasting Ontario Wines in the early 70s, they [the founders] realized a gap existed in the premium market segment. They seized the opportunity to fill that gap and set out to break new ground to produce premium varietal wines from premium grapes grown in the Niagara Peninsula. Both [y] believed their future in the wine business was dependent on using the Vitis vinifera grapes, the preferred family of grapes used to produce the fine wines in the great wine regions of the world. [y] Ziraldo took the lead and planted a vinifera vineyard which included Riesling, Chardonnay and Gamay and formed the quality base for Kaiser to work with.

The founders of this winery are characterized as playing a legendary role in Niagara history; they noticed ‘‘a gap existed’’ and, in looking to fill this gap, they decided to plant the European vinifera varietals that are consistent with the logic of fine winemaking. We can surmise that this gap refers to the poor winemaking history, and historical accommodation of the alcohol making logic in Ontario and that existed because grape growers were not cultivating acceptable wine varietals, and were instead growing the native labrusca varietals. We see that the innovative thinking of this team, in realizing that the future of winemaking was in vinifera varietals and pioneering their entry to the region, and acceptance of the fine winemaking logic has been integral to the success of the winery. We also see a connection between the family histories of the founders and the Old World winemaking regions of Austria and Italy, which indicates a link to innovation in that this heritage helped establish them as the ‘‘founding fathers of Canadian wine.’’ Yet another such reference is made by Cave Spring on its web site, as it describes the movement toward the growth of vinifera varietals: From his early days as one of the first farmers to plant the noble grape varietals of Europe in Ontario, to his forward-thinking integration of Niagara wine and tourism, Len has worked to define the future of the Niagara Peninsula. Growing up working on his family’s small vineyard, Len learned the meticulous art of viticulture, pruning and tying the vines alongside his father and grandfather, both hobbyist winemakers. Seeing his son’s interest in viticulture, Len’s father John, Sr. decided that the Pennachetti family would invest in better vineyards. Together, he and Len scouted the benchlands of the Niagara Escarpment by plane and discovered the historic Cave Spring Farm, with its hillside location, clay-limestone soils and ideal proximity to Lake Ontario. Situated on one of the finest slopes of the Beamsville Bench, Len and his father founded Cave Spring Vineyard in 1974, and in 1978 made their first vinifera plantings of Riesling and Chardonnay.

Putting New Wine in Old Bottles

125

Here, we see the references to global traditions, in the ‘‘meticulous art of viticulture,’’ and the laborious methods used to realize this art. This family is thus characterized as pioneers and as hard working and committed to these traditional values of winemaking. When they mention the decision to ‘‘invest in better vineyards’’ they are making an implicit reference to the fact that many vineyards at the time were subpar, and not up to the standards of appropriate viticulture. This notion of the poor suitability of vineyards is again implicitly referred to when the efforts to plant vinifera varietals, indicating that these had to be planted, and did not previously exist, as most vineyards held plantings of the native labrusca varietals. The fact that this land is suited for prime grape growing due to its unique location indicates a reference to terroir, implying both traditional values and geological history of the land. In looking at the dates that are mentioned and the characterization of the family as pioneers, we also note that this family was planting vinifera varietals at a time when most other wineries were not doing so, or at the beginning of the era when native varietals were beginning to be uprooted in favor of vinifera varietals across the region. Interestingly, as is the case in many newsletters and web sites, the widespread grape growing practices of planting and harvesting labrusca varietals is not explicit, and the mention of this history is avoided or refrained from. Frequently we found, such as in a newsletter from Angels Gate, that the acceptability of growing vinifera varietals is more taken for granted, and assumed: ‘‘Mountainview speaks to our location which is nestled against the escarpment while having a commanding view of Lake Ontario in the distance. These two geographical features create the ideal conditions for growing premium vinifera grapes.’’ The fact that this parcel of land is ‘‘ideal’’ for growing vinifera varietals is not even of argument in this passage, which taps into the idea that this fact is socially and normatively accepted by audiences as true, and that the logic of fine winemaking is now dominant. This is noteworthy because as little as 30 years previously, it was commonly thought that vinifera varietals absolutely could not be grown in the region, and certainly were unable to thrive due to a number of reasons including temperature fluctuations, molds, and various pests. These beliefs required much action and effort on the behalf of those willing to counter them, but the way in which this passage describes the region as not only suitable, but ideal, for growing noble varietals heavily represses these efforts, as well as the historical adherence to an alternate logic that prevailed in the region. In all, this passage works to characterize the winery as legitimate producers of premium wines, and uses metaphor to

126

SHILO HILLS ET AL.

link the land to premium grapes, as only these varietals adhering to the dominant logic of fine winemaking are cultivated.

DISCUSSION Not all fields are characterized by ongoing and unresolvable tensions between multiple logics. Some may have reached settlements (Helms et al., 2012), at least temporarily, whereby the hierarchy between logics is clearly established. Our intention in this paper is to highlight the effort involved in complying with the dominant logic and to illustrate the use of rhetorical history in demonstrating such adherence and to avoid stigma related to past associations. In sum, our findings illustrate that adherence to a dominant logic entails effortful work by actors to indicate continuity with this legitimate, dominant logic and simultaneous rejection and obscuration of historical adherence to a subordinate logic widely deemed unacceptable and inappropriate. The attention is directed toward the current state of the industry as being world class. Yet, these efforts to distance current practices from this past also involve rhetorically obscuring or minimizing prior adherence to the logic of alcohol making. We think that the acknowledgment of the effortful nature of adherence to a dominant logic and of the use of rhetorical history in doing so, illustrated in our study, offers a number of important contributions to the study of institutional logics and prompts some interesting research questions.

Micro-Foundations of Institutional Logics Perhaps our most obvious contribution is to respond to Thornton et al.’s (2012) call for greater attention to the micro-foundations of institutional logics and a closer integration with the research on institutional work (Lawrence & Suddaby, 2006). Institutional work researchers have made considerable effort to understand the processes and practices through which people and organizations create, maintain and/or disrupt institutions, and effort has been a defining feature of this research program (Lawrence, Suddaby, & Leca, 2011). Both research areas share the interest in embedded agency (Battilana & D’Aunno, 2009; Thornton & Ocasio, 2008); however, when effort is addressed, it tends to be done in the context of attempts to transform or challenge institutional logics, and the efforts of

Putting New Wine in Old Bottles

127

actors to cope with logics on a day-to-day basis, without necessarily trying to bring about change in the field remain poorly understood (Greenwood et al., 2011). Thus, we aim to demonstrate that adhering to a particular logic is effortful, even when in circumstances where the distinction between plural legitimate and illegitimate logics is unambiguous and seemingly straightforward. Effort then is expanded not only in the service of either promoting or stifling institutional change (Lawrence et al., 2011), but also in the service of coping with day-to-day challenges of adherence – that can be daunting at times. Adhering to a particular institutional logic is by no means free of challenges. It requires a great deal of situational awareness in order to ensure that one’s adherence to a particular institutional logic is credible to varied audiences (Alexander, 2004). After all, intended adherence to a particular logic that is not acknowledged or understood as such by audiences would likely result in social sanctions, similar to those for willful nonadherence. Thus, in our research on the Ontario wine industry it is salient that wineries not only try to adhere to the dominant logic of fine winemaking, but importantly, they try to communicate and demonstrate this adherence to various audiences. To do so they draw on historical constructions in a strategic manner, which indicates awareness that private adherence to a particular logic is distinct from demonstrating such adherence publically, and ensuring that the displays of adherence resonate with audiences. Given that institutional change is not always intentional (Barley, 1986), future research may examine to what extent rhetorical history may, over time, reconstitute relations between logics in the field. Understanding the use of rhetoric in demonstrating adherence is important. Prior research has attended to the role of rhetoric in motivating or promoting institutional change (Greenwood, Suddaby, & Hinings, 2002; Suddaby & Greenwood, 2005), and we find that it is also important in the context of more mundane demonstration of adherence to a dominant logic – where the role of such a logic is commonly coherent and pervasively recognized as acceptable and legitimate. An interesting finding in our data is the duality of actors insisting that they adhere to the dominant logic and that they do not adhere to the subordinate logic. In other words, they attempt to strengthen the impression of adherence to the dominant logic by casting the subordinate logic as the antagonist (Ruebottom, 2013). Historical constructions become malleable resources in this endeavor, whereby actors attach themselves to the putatively glorious pre-modern past of French winemaking while distancing themselves from the stigma-inducing local winemaking history. All of this implies an important emotive component,

128

SHILO HILLS ET AL.

whereby rhetorical constructions of history are meant to appeal not only to reason, but perhaps more importantly to audiences’ emotions (Alexander, 2004). Thus, we found wineries often attempting to ‘‘have it both ways’’ by repudiating certain elements of local history in culture, while attempting to leverage others in order to construct narratives that are less about logic and more about emotional resonance (Green, 2004). Thus, another potential contribution of greater attention to the use of rhetoric in in adherence to a dominant institutional logic might be shedding more light on affective components of institutional logics that have received less attention, relative to the cognitive ones (Voronov & Vince, 2012). Examining how organizations use rhetorical history to comply with institutional logics is but one of many ways to examine the effortful nature of adherence, and it requires researchers to pay attention to the nuances of rhetoric use and the societal context within which such rhetoric is rendered meaningful (Heracleous & Barrett, 2001). It forces researchers to retain the logics as well as the people and organizations that need to navigate them and to constantly confront the reality that neither is meaningful without the other (Friedland & Alford, 1991). Naturally, we would argue that other forms of effort are worthy of investigation. In particular, more material aspects of adherence, ranging from practices, physical arrangements (of rooms, buildings, etc.) and other artifacts to mundane nonverbal behavior are worthy of investigation. Cultivating an understanding of adherence effort requires closeness and fine-grained analyses (Barley, 1986, 2008), and in our aim to do so, we find that the use of rhetorical history acts as an important mechanism in such processes. However, there is a need for scholars to study this along with other material practices to understand the nature of decoupling and other processes associated with organizational attempts of logic suppression and adherence. Another avenue for future research, which due to the extent and type of data that was used in this study, we were unable to address in detail, would be to examine more thoroughly the variation that exists between the types of rhetorical history used by different firms, and how this affects public perceptions as well as firm reputation and performance.

Rootedness of Logics in Time and Place Our study also illustrates the rootedness of logics in a time and a place (e.g., Lounsbury, 2007). The logic of fine winemaking, the dominant logic in our study, is rooted heavily in the winemaking conventions of France and, more

Putting New Wine in Old Bottles

129

recently, of other more established wine regions (Colman, 2008; Robinson, 2006). In the absence of recognized past adherence to this logic by Ontario wineries, prior to 1970s (or 1980s on any kind of a significant scale), the logic then is essentially foreign to Ontario, and rhetorical history is used to link Ontario winemaking of the present to the logic of fine winemaking rooted in the past and in a foreign country. Our study then speaks to the nestedness of institutional logics within the broader societal forces (Greenwood et al., 2011). In addition, our study links to the institutional studies of globalization (Drori, 2008). The logic of fine winemaking entered Ontario on large scale through the institutional arrangements that constitute globalization (i.e., Free Trade and NAFTA). Importantly, however, whereas institutional researchers of globalization tend to cast globalization as either benign or neutral, it is apparent in our study that the importation of the logic of fine winemaking to Ontario constitutes a force of domination, because it has redefined what winemaking is all about and altered ‘‘the range of options available’’ (Lawrence, 2008, p. 178) to the local wineries. In our case study, however, this domination served to provide the impetus for beneficial change in the identity of the Ontario wine industry, insofar as reputation and legitimacy are concerned. This was accomplished through the altering of actors’ mindsets, and their efforts toward redefining both ends and means of what is desirable in winemaking – so much so, that the winemaking of old is now cast as ‘‘bad,’’ but also provides a template of illegitimate wine production, from which more legitimated identities can be realized and constructed.

Logics and Domination Related to the above, it is interesting that although concern with domination was quite salient in Friedland and Alford’s seminal chapter on institutional logics, and they took neo-institutionalists to task for offering up ‘‘an institution-free conception of interest and power’’ and assuming ‘‘objective interests that can be understood independently of the actors’ understanding’’ (Friedland & Alford, 1991, p. 244), the topic of domination has been virtually absent from institutional logics research (see Thornton et al., 2012 for an exception).2 We think domination needs to be taken more seriously in the research on institutional logics, because without accounting for domination, scholars may overestimate the likelihood of successful institutional transformation.

130

SHILO HILLS ET AL.

In our study, it was apparent, for example, that Ontario wineries were more preoccupied with demonstrating faithful adherence to the dominant institutional logic than with finding ways to differentiate themselves from their foreign competitors. This suggests that once invested in a particular logic (Voronov & Vince, 2012), actors may have difficulty reflecting on the extent to which this logic is suitable to advancing their interests – quite simply because the logic also defines what those interests are (Friedland & Alford, 1991), a point that seems to have been underemphasized by institutional logics researchers. We would suggest that one way to avoid ‘‘losing’’ the focus on domination might involve incorporating Bourdieu’s notion of habitus, which refers to the actors’ field prescribed dispositions (Golsorkhi, Leca, Lounsbury, & Ramirez, 2009; Voronov & Vince, 2012), into the research on institutional logics. The notion would be helpful not only in sensitizing scholars to the pervasive role of domination in various fields, but it would also offer a useful meso-link between macrolevel structures and individuallevel actions. Attending to actors’ habitus would enable researchers to better assess the extent to which particular logics are internalized and become ‘‘second nature,’’ while others are adhered to more ceremonially. Such variance in the level of internalization of institutional logics into actors’ thoughts, feelings and actions is an important measure of the relative dominance of particular logics, with those that are incorporated into particular actors’ habitus being more dominant than those that provoke ceremonial adherence. Furthermore, because different actors’ habitus within a field is likely to vary, these differences might also help researchers grasp the extent to which different logics are more dominant in some segments of a field as compared to others.

CONCLUSION We welcome the increased interest in micro-foundations among institutional logics researchers. In this paper, we argued that adherence to the dominant logic should not be taken-for-granted or treated as automatic. One way to capture this effort is by attending to rhetorical history. We hope that the study of effort, such as of the use of rhetoric, can facilitate a more epistemologically and methodologically open research program and allow institutional logics research to continue offering important insights into a variety of organizational and societal issues.

131

Putting New Wine in Old Bottles

NOTES 1. This notion refers to ‘‘the relationship between the characteristics of an agricultural product (quality, taste, style) and its geographic origin, which might influence these characteristics’’ (Van Leeuwen & Seguin, 2006, p. 1). 2. Even these authors, though, in our view underemphasize the role of domination and utilize social psychological theories that do not incorporate concern with domination, when theorizing individual level behavior (see Cooper, Ezzamel, & Willmott, 2008 for more expanded discussion of neoinstitutional theory’s persistant inability to accomodate issues of domination).

ACKNOWLEDGMENT This research was supported, in part, by a grant from the Social Sciences and Humanities Research Council of Canada.

REFERENCES Alexander, J. C. (2004). Cultural pragmatics: Social performance between ritual and strategy. Sociological Theory, 22(4), 527–573. Alford, R. R., & Friedland, R. (1985). Powers of theory: Capitalism, the state, and democracy. New York, NY: Cambridge University Press. Anteby, M., & Molna´r, V. (2012). Collective memory meets organizational identity: Remembering to forget in a firm’s rhetorical history. Academy of Management Journal, 55(3), 515–540. Baldick, C. (2009). The oxford dictionary of literary terms. Oxford: Oxford University Press. Barley, S. R. (1986). Technology as an occasion for structuring: Evidence from observations of ct scanners and the social order of radiology departments. Administrative Science Quarterly, 31(1), 78–108. Barley, S. R. (2008). Rejoinder. Journal of Management Inquiry, 17, 168–171. Bates, T. G. (2011). The reagan rhetoric: History and memory in 1980s america. DeKalb: Northern Illinois University Press. Battilana, J., & D’Aunno, T. (2009). Institutional work and the paradox of embedded agency. In T. B. Lawrence, R. Suddaby & B. Leca (Eds.), Institutional work: Actors and agency in institutional studies of organizations (pp. 31–58). Cambridge: Cambridge University Press. Battilana, J., Leca, B., & Boxenbaum, E. (2009). How actors change institutions: Towards a theory of institutional entrepreneurship. Academy of Management Annals, 3, 65–107. doi: 10.1080/19416520903053598 Beverland, M. B. (2005). Crafting brand authenticity: The case of luxury wines. Journal of Management Studies, 42(5), 1003–1029. Boje, D. (1991). The storytelling organization: A study of storytelling performance in an office supply firm. Administrative Science Quarterly, 36, 106–126.

132

SHILO HILLS ET AL.

Boje, D. (1995). Stories of the storytelling organization: A postmodern analysis of Disney as ‘Tamara-land’. Academy of Management Journal, 38(4), 997–1035. Bourdieu, P., & Wacquant, L. J. D. (1992). An invitation to reflexive sociology. Chicago, IL: University of Chicago Press. Bramble, L. (2009). Niagara’s wine visionaries: Profiles of the pioneering winemakers. Toronto: Lorimer. Burke, K. (1969). A grammar of motives. Berekeley, CA: University of California Press. Clarke, M. A. X., & Fine, G. A. (2010). ‘‘A’’ for apology: Slavery and the discourse of remonstrance in two american universities. [Article]. History & Memory, 22(1), 81–112. Colman, T. (2008). Wine politics: How governments, environmentalists, mobsters, and critics influence the wines we drink. Berkeley, CA: University of California Press. Cooper, D. J., Ezzamel, M., & Willmott, H. (2008). Examining institutionalization: A critical theoretic perspective. In R. Greenwood, C. Oliver, R. Suddaby & K. Sahlin (Eds.), Handbook of organizational institutionalism (pp. 673–701). London: Sage. Creed, W. E. D., Dejordy, R., & Lok, J. (2010). Being the change: Resolving institutional contradiction through identity work. Academy of Management Journal, 53(6), 1336–1364. Cuddon, J. A., & Preston, C. (1998). A dictionary of literary terms and literary theory. Oxford: Blackwell. DiMaggio, P. J. (1997). Culture and cognition. Annual Review of Sociology, 23, 263–287. Drori, G. S. (2008). Institutionalism and globalization studies. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The sage handbook of organizational institutionalism (pp. 449–472). Los Angeles, CA: Sage. Du Gay, P. (2008). ‘Without affection or enthusiasm’ problems of involvement and attachment in ‘responsive’ public management. Organization, 15(3), 335–353. doi: 10.1177/ 1350508408088533 Dunn, M. B., & Jones, C. (2010). Institutional logics and institutional pluralism: The contestation of care and science logics in medical education, 1967–2005. Administrative Science Quarterly, 55, 114–149. Emirbayer, M., & Mische, A. (1998). What is agency? American Journal of Sociology, 103(4), 962–1023. Foster, W. M., Suddaby, R., Minkus, A., & Wiebe, E. (2011). History as social memory assets: The example of tim hortons. Management & Organizational History, 6(1), 101–120. Frank, M. (2008). Niagara rises. Wine Spectator, May 15, pp. 93–101. Friedland, R., & Alford, R. R. (1991). Bringing society back in: Symbols, practices, and institutional contradictions. In W. W. Powell & P. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 232–263). Chicago, IL: University of Chicago Press. Gephart, R. P. (1991). Succession, sensemaking, and organizational change: A story of a deviant college president. Journal of Organizational Change Management, 4(3), 35–44. Golant, B., & Sillince, J. (2007). The constitution of organizational legitimacy: A narrative perspective. Organization Studies, 28(8), 1149–1167. Golsorkhi, D., Leca, B., Lounsbury, M., & Ramirez, C. (2009). Analysing, accounting for and unmasking domination: On our role as scholars of practice, practitioners of social science and public intellectuals. Organization, 16(6), 779–797. doi: 10.1177/1350508 409343400

Putting New Wine in Old Bottles

133

Gouldner, A. W. (1954). Patterns of industrial bureaucracy. New York, NY: Free Press. Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E. R., & Lounsbury, M. (2011). Institutional complexity and organizational responses. The Academy of Management Annals, 5(1), 317–371. Green, M. C. (2004). Transportation into narrative works: The role of prior knowledge and perceived realism. Discourse Processes, 38(2), 247–266. Greenwood, R., Suddaby, R., & Hinings, C. R. (2002). Theorizing change: The role of professional associations in the transformation of institutionalized fields. The Academy of Management Journal, 45(1), 58–80. Hamel, G., & Prahalad, C. K. (1996). Competing for the future. Boston, MA: Harvard Business School Press. Helms, W. S., Oliver, C., & Webb, K. (2012). Antecedents of settlement on a new institutional practice: Negotiation of the ISO 26000 standard on social responsibility. Academy of Management Journal, 55(5), 1120–1145. Heracleous, L., & Barrett, M. (2001). Organizational change as discourse: Communicative actions and deep structures in the context of information technology implementation. Academy of Management Journal, 44, 755–778. Hoffman, A. J. (2001). From Heresy to Dogma: An institutional history of corporate environmentalism. Stanford, CA: Stanford University Press. Hudson, B. A. (2008). Against all odds: A consideration of core-stigmatized organizations. Academy of Management Review, 33(1), 252–266. Kraatz, M., & Block, E. S. (2008). Organizational implications of institutional pluralism. In R. Greenwood, C. Oliver, R. Suddaby & K. Sahlin (Eds.), The sage handbook of organizational institutionalism. Los Angeles, CA: Sage. Kramer, M. (2004). Making sense of wine. Philadelphia, PA: Running Press. Lawrence, T. B., & Suddaby, R. (2006). Institutions and institutional work. In S. R. Clegg, C. Hardy, T. B. Lawrence & W. R. Nord (Eds.), The sage handbook of organization studies (2nd ed., pp. 213–254). London: Sage. Lawrence, T. B. (2008). Power, institutions and organizations. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), Sage handbook of organizational institutionalism (pp. 170–197). London: Sage. Lawrence, T. B., Suddaby, R., & Leca, B. (2011). Institutional work: Refocusing institutional studies of organization. Journal of Management Inquiry, 20(1), 52–58. Lok, J., & de Rond, M. (2013). On the plasticity of institutions: Containing and restoring practice breakdowns at the cambridge university boat club. Academy of Management Journal. DOI: 10.5465/amj.2010.0688. (published online Jul 2012; forthcoming in print). Lounsbury, M. (2002). Institutional transformation and status mobility: The professionalization of the field of finance. Academy of Management Journal, 45(1), 255–266. Lounsbury, M. (2007). A tale of two cities: Competing logics and practice variation in the professionalizing of mutual funds. Academy of Management Journal, 50(2), 289–307. Lounsbury, M., & Crumley, E. T. (2007). New practice creation: An institutional perspective on innovation. Organization Studies, 28(7), 993–1012. doi: 10.1177/0170840607078111 Lounsbury, M., & Glynn, M. A. (2001). Cultural entrepreneurship: Stories, legitimacy, and the acquisition of resources. Strategic Management Journal, 22(6/7), 545–564. Lounsbury, M., Ventresca, M., & Hirsch, P. M. (2003). Social movements, field frames and industry emergence: A cultural-political perspective on us recycling. Socio-Economic Review, 1, 71–104.

134

SHILO HILLS ET AL.

MacMillan, M. (2008). The uses and abuses of history. Toronto: Viking Canada. McCoy, E. (2005). The emperor of wine: The rise of Robert M. Parker, jr. and the reign of American taste. New York, NY: Harper Perennial. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363. Mills, M. S. (2010). Concise handbook of literary and rhetorical terms. Estep-Nichols Publishing. Ostrom, E. (1991). Rational choice theory and institutional analysis. Toward Complementarity, 85(1), 237–243. Pache, A. C., & Santos, F. (2010). When worlds collide: The internal dynamics of organizational responses to conflicting institutional demands. The Academy of Management Review, 35(3), 455–476. Phillips, R. (2004). Ontario wine country. North Vancouver: Witecap. Polletta, F., Chen, P. C., Gardner, B., & Motes, A. (2011). The sociology of storytelling. Annual Review of Sociology, 37, 109–130. Pratt, M. G. (2009). For the lack of a boilerplate: Tips on writing up (and reviewing) qualitative research. Academy of Management Journal, 52(5), 856–862. Rao, H., & Giorgi, S. (2006). Code breaking: How entrepreneurs exploit cultural logics to generate institutional change. In Research in organizational behavior (Vol. 27, pp. 269–304). doi:10.1016/s0191-3085(06)27007-2 Rao, H., Monin, P., & Durand, R. (2003). Institutional change in toque ville: Nouvelle cuisine as an identity movement in french gastronomy. American Journal of Sociology, 108(4), 795–843, i. doi:10.1086/367917 Reay, T., & Hinings, C. R. (2009). Managing the rivalry of competing institutional logics. Organization Studies, 30(6), 629. Rivera, L. A. (2008). Managing ‘‘spoiled’’ national identity: War, tourism, and memory in croatia. American Sociological Review, 73(4), 613–634. Roberts, P. W., & Reagans, R. (2007). Critical exposure and price-quality relationships for new world wines in the us market. Journal of Wine Economics, 2, 56–69. Robinson, J. (2006). The oxford companion to wine (3rd ed.). Oxford: Oxford University Press. Ruebottom, T. (2013). The microstructures of rhetorical strategy in social entrepreneurship: Building legitimacy through heroes and villains. Journal of Business Venturing, 28(1), 98–116. Seo, M. G., & Creed, W. E. D. (2002). Institutional contradictions, praxis, and institutional change: A dialectical perspective. Academy of Management Review, 27(2), 222–247. Sillince, J. A. A., & Brown, A. D. (2009). Multiple organizational identities and legitimacy: The rhetoric of police websites. Human Relations, 62(12), 1829–1856. doi: 10.1177/ 0018726709336626 Suddaby, R., Foster, W. M., & Trank, C. Q. (2010). Rhetorical history as a source of competitive advantage. Advances in Strategic Management, 27, 147–173. Suddaby, R., & Greenwood, R. (2005). Rhetorical strategies of legitimacy. Administrative Science Quarterly, 50(1), 35–67. Thornton, P. H., & Ocasio, W. (1999). Institutional logics and the historical contingency of power in organizations: Executive succession in the higher education publishing industry, 1958–1990. American Journal of Sociology, 105(3), 801–843. Thornton, P. H., & Ocasio, W. (2008). Institutional logics. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 99–129). Thousand Oaks, CA: Sage.

Putting New Wine in Old Bottles

135

Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure and process. Oxford: Oxford University Press. Ulin, R. C. (1995). Invention and representation as cultural capital: Southwest French winegrowing history. American Anthropologist, 97(3), 519–527. Van Leeuwen, C., & Seguin, G. (2006). The concept of terroir in viticulture. Journal of Wine Research, 17(1), 1–10. Vaara, E., & Tienari, J. (2011). On the narrative construction of multinational corporations: An antenarrative anlaysis of legitimation and resistance in a cross-border merger. Organization Science, 22(2), 370–390. Voronov, M., & Vince, R. (2012). Integrating emotions into the analysis of institutional work. The Academy of Management Review, 37(1), 58–81. Weber, K., & Dacin, T. (2011). The cultural construciton of organizational life: Introduction to the special issue. Organization Science 7039(1047), 1526–5455. Whately, R. (1962). Elements of rhetoric: Comprising an analysis of the laws of moral evidence and of persuasion, with rules for argumentative composition and elocution. Carbondale, IL: Southern Illinois University Press. Wry, T., Lounsbury, M., & Glynn, M. A. (2011). Legitimating nacent collective identities: Coordinating cultural entrepreneurship. Organization Science, 22(2), 449–463. Zerubavel, E. (2003). Time maps: Collective memory and the social shape of the past. Chicago, IL: University of Chicago Press. Zhao, W. (2005). Understanding classifications: Empirical evidence from the American and French wine industries. Poetics, 33(3-4), 179–200. doi: 10.1016/j.poetic.2005.09.010

136

SHILO HILLS ET AL.

APPENDIX: KEY TERMS IN RHETORICAL ANALYSIS

Characterization

refers to particularly vivid and idiosyncratic representations of persons as portrayed within texts. We took this to ‘‘include direct methods like the attribution of qualities in description or commentary, and indirect (or ‘dramatic’) methods inviting readers to infer qualities from characters’ actions, speech, or appearance’’ (Baldick, 2009, p. 52). According to Burke (1969), it allows audiences to ascribe particular moral qualities to an actor. In our context, it involves attempts to portray wineries and their representatives as having certain values, morals and characteristics. These might include, for example, accounts of a winemaker’s heroic effort to produce a superb wine and the sacrifices that had to be made to accomplish that.

Identification

we relied heavily upon Burke’s (1969, p. 580) observation that ‘‘you persuade a man only insofar as you can talk his language by speech, gesture, tonality, order, image, attitude, idea, identifying your ways with his,’’ thereby attempting to establish a shared sense of values with the audience. Thus, wineries might try to ingratiate themselves to the audience or invite the audience to participate in certain winery experiences that should be especially appealing to them.

Diatyposis

‘‘a rhetorical figure in which advice is given’’ (Mills, 2010, p. 115) is used as a way to provide instruction to audiences through rules and precepts and indicates that the wineries occupy a space of authority in their ability to provide such direction (Whately, 1962). This rhetorical device includes, for example, suggesting traditional pairings of particular food with a certain wine.

Putting New Wine in Old Bottles

137

APPENDIX (Continued ) Denotation

provides ‘‘the most literal and limited meaning of a word, regardless of what one may feel about it or the suggestions and ideas it connotates.’’ (Cuddon & Preston, 1998, p. 215). As Burke (1969, p. 24) suggests, ‘‘to tell what a thing is, you place it in terms of something else. This idea of locating, or placing, is implicit in our very word for definition itself: to define, or determine a thing, is to mark its boundaries, hence to use terms that possess, implicitly at least, contextual reference.’’ Thus, in providing literal definitions for audiences, wineries can work to define the boundaries of meaning surrounding a word, and can ensure congruity of understanding among readers. For example, wineries might explain in detail the meaning of a particular grape growing or winemaking practice.

Personification

refers to a rhetorical device that describes inanimate objects, and other nonhuman concepts as having human characteristics and qualities (Baldick, 2009; Cuddon & Preston, 1998). For example, wineries often endow wines with person-like traits, such as ‘‘personality.’’

Imagery

involves the use of language to represent various sensory experiences that extend beyond mental pictures and may appeal to senses other than sight (Baldick, 2009; Cuddon & Preston, 1998). As such, the use of imagery allows audiences to participate in the texts on a deeper, sensory level that provides a different experience as individuals can witness what the text describes. For example, wineries may attempt to use such a device in order to help the reader to imagine herself in the vineyard, experiencing the beauty, the sounds and other sensations that are associated with it.

IMAGERIES OF CORPORATE SOCIAL RESPONSIBILITY: VISUAL RECONTEXTUALIZATION AND FIELD-LEVEL MEANING Markus A. Ho¨llerer, Dennis Jancsary, Renate E. Meyer and Oliver Vettori ABSTRACT In this paper, we explore how corporations use visual artifacts to translate and recontextualize a globally theorized managerial concept (CSR) into a local setting (Austria). In our analysis of the field-level visual discourse, we analyze over 1,600 images in stand-alone CSR reports of publicly traded corporations. We borrow from framing analysis and structural linguistics to show how the meaning structure underlying a multifaceted construct like CSR is constituted by no more than a relatively small number of fundamental dimensions and rhetorical standpoints (topoi). We introduce the concept of imageries-of-practice to embrace the critical role that shared visual language plays in the construction of meaning and the emergence of field-level logics. In particular, we argue that imageriesof-practice, compared to verbal vocabularies, are just as well equipped to link locally resonating symbolic representations and globally diffusing

Institutional Logics in Action, Part B Research in the Sociology of Organizations, Volume 39B, 139–174 Copyright r 2013 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0733-558X/doi:10.1108/S0733-558X(2013)0039A&B018

139

140

MARKUS A. HO¨LLERER ET AL.

practices, thus expressing both the material and ideational dimension of institutional logics in processes of translation. We find that visual rhetoric used in the Austrian discourse emphasizes the qualities of CSR as a bridging concept, and facilitates the mediation of inconsistencies in several ways: By translating abstract global ideas into concrete local knowledge, imageries-of-practice aid in mediating spatial oppositions; by linking the past, present, and future, they bridge time; by mediating between different institutional spheres and their divergent logics, they appease ideational oppositions and reduce institutional complexity; and, finally, by connecting questionable claims with representations of authenticity, they aid in overcoming credibility gaps. Keywords: Visual artifacts; imageries-of-practice; visual rhetoric; field-level logics; translation; corporate social responsibility

In no other form of society in history has there been such a concentration of images, such a density of visual messages. One may remember or forget these messages but briefly one takes them in, and for a moment they stimulate the imagination by way of either memory or expectation. John Berger: Ways of seeing (1972, p. 129)

INTRODUCTION With the proliferating interest in the effects of multiple institutional logics coexisting in fields, meaning and the interpretive work of actors – individual and organizational alike – have recently received much attention in institutional research. Our study follows this line of scholarly work by examining how one of the most prominent globally theorized management concepts from the recent past has been recontextualized and translated into a specific local setting. As such efforts at re-localization and translation (e.g., Boxenbaum, 2006; Campbell, 2004; Czarniawska & Sevo´n, 1996; Meyer, 2004; Zilber, 2006; for an overview, see Sahlin & Wedlin, 2008) have to be understood against the backdrop of the socio-historic specificities of the adopting field – Goodrick and Reay (2011) call this the ‘‘constellation of institutional logics’’ – we examine the ways in which organizations create locally resonating interpretations that resolve, bridge, and/or conceal existing inconsistencies between different institutional spheres and their underlying logics (e.g., Friedland & Alford, 1991; Greenwood, Raynard,

Imageries of Corporate Social Responsibility

141

Kodeih, Micelotta, & Lounsbury, 2011; Thornton & Ocasio, 2008; Thornton, Ocasio, & Lounsbury, 2012). While previous research on institutional logics has well embraced the linguistic turn in neoinstitutional theory – Thornton et al. (2012, p. 150), for instance, emphasize that ‘‘[l]anguage, embodied in theories, frames, and narratives, and embedded in vocabularies of practice, provides a critical linchpin by which institutional logics are constructed and meanings and practices are brought together’’ – we argue here that this focus on verbal discourse leaves the performative power of visuals in the emergence of fieldlevel meanings and logics unaccounted for. To close this gap, we introduce the concept of ‘‘imageries-of-practice’’ in analogy to ‘‘vocabularies-ofpractice’’ (Loewenstein, Ocasio, & Jones, 2012, building on Mills, 1940). In particular, we argue that visual ‘‘text’’ is, with its more immediate and less controlled mode of meaning construction, even better suited than words to express the symbolic and material character of institutional logics. The visual provides a crucial and unique resource through which the unobservable, unknowable substances of institutional logics (Friedland, 2009) take form and become, in the literal sense of the word, visible. Surprisingly enough, the visual realm has, to date, remained largely unexplored in organization and management studies (e.g., Davison & Warren, 2009; Styhre, 2010; Meyer, Ho¨llerer, Jancsary, & van Leeuwen, 2013). Although it is generally acknowledged that organizational discourse also covers visual representations (e.g., Grant, Hardy, Oswick, & Putnam, 2004; Phillips, Lawrence, & Hardy, 2004), the majority of empirical research has focused on verbal texts (influential examples include, for instance, Cornelissen, 2005; Vaara, Tienari, & Laurila, 2006; for an overview, see Phillips & Oswick, 2012), even when analyzing genres in which visualization plays an essential role. Here, we claim that visual ‘‘language,’’ with its more plastic and ambiguous ‘‘vocabulary,’’ offers manifold and distinct opportunities for actors to locally realign theorized and decontextualized ideas and concepts. A socially shared visual vocabulary is, we argue, central to the local interpretation and sense-making/-giving as well as for the emergence of field-level logics. We explore these issues for the case of corporate social responsibility (CSR) in Austria. In more detail, we draw on concepts and methodologies from the sociology of knowledge (e.g., Meyer, 2008; Raab, 2008), visual semiotics (Kress & van Leeuwen, 2001, 2006), framing analysis (e.g., Benford & Snow, 2000; Gamson & Lasch, 1983; Gamson & Modigliani, 1989), hermeneutics (e.g., Froschauer & Lueger, 2003; Lueger, Sandner, Meyer, Hammerschmid, 2005; Mu¨ller-Doohm, 1997), and structural linguistic approaches in discourse analysis (e.g., Bublitz, 2011; Greimas, 1983; Link, 1997) to analyze images in

142

MARKUS A. HO¨LLERER ET AL.

CSR reports of Austrian publicly traded corporations. This enables us, first, to reconstruct the fundamental semantic dimensions and polar categories that open up the meaning space for CSR in Austria. Second, we identify several rhetorical standpoints – or topoi – that, by constituting different positions within this meaning space, provide a distinct perspective on the concept of CSR. Third, we specify the prevailing set of societal logics – profession, corporation, community, religion, and, to a far lesser extent, market – that shape the field-level formation of the concept’s meaning. We show that visual artifacts are particularly qualified to reconcile inconsistencies. More specifically, they work as bridging devices in various distinct ways: They align abstract, globally theorized ideas with specific, locally resonating examples and symbols; in addition, their multivocal and plastic nature allows for the simultaneous communication of potentially antagonistic ideas, thereby mediating as well as balancing divergent ideational systems. And finally, they are not bound to the comparatively strict rules and conventions governing verbal text, thus providing an opportunity to address more fully and directly issues of, for instance, emotionality and spirituality. The remainder of this paper is organized as follows: The next two sections sketch the conceptual premises of our research. We then briefly highlight the characteristic features of the phenomenon of CSR against the empirical setting of our study, followed by a section presenting the methodological framework and research design. The discussion of findings commences with an outline of the set of ‘‘discourse-carrying dimensions’’ that represent the ‘‘structural skeleton’’ of the visual CSR discourse in Austria. Subsequently, we explore our data by means of two descriptive networks. These provide further insights into the various topoi that corporations use to make sense of, and communicate, the category of the ‘‘socially responsible corporation’’; moreover, they illustrate the field-level formation of institutional logics these topoi are built upon. A concluding section highlights the core contributions of the paper.

THE ROLE OF VISUALS IN MEANING CONSTRUCTION The Performative Power of Visuals Verbal language is undoubtedly a highly relevant system of signification and ‘‘reservoir’’ of typifications and institutional knowledge (Berger & Luckmann, 1967; Schu¨tz & Luckmann, 1973), but it is by no means the only

Imageries of Corporate Social Responsibility

143

one. Indeed, from an anthropological perspective, visual forms of representation and meaning construction have always been a vital part of social and cultural life, and there is considerable evidence that the proliferation of visual artifacts1 has once again gained considerable momentum (Kress & van Leeuwen, 2006; Mitchell, 1994). The increasing amount of publicly available visual material, in combination with new information and communication technologies, has created novel opportunities for the use of visuals, and has consequently enabled social actors to move beyond the limitations of verbal language and linear writing that had dominated social life in the Western world for centuries (Kress & van Leeuwen, 2001). Elsewhere (Meyer et al., 2013), we have elaborated more extensively on how visuals, as do words, materialize, organize, communicate, store, and pass on knowledge (see also Raab, 2008) – thereby mirroring as well as constructing reality in a distinct way. Consequently, processes of theorization, institutionalization, and translation, for instance, are not only manifested in practices and verbal text but also in various visual artifacts. Likewise, institutional logics are symbolically represented by verbal and visual vocabularies. We argue that while both verbal and visual languages draw on the same fundamental systems of meaning that constitute our cultures, each does so by means of its own specific mode, and therefore requires different treatment in scholarly analysis. Various researchers point to a trend that has been labeled as a distinct ‘‘iconic’’ (e.g., Boehm, 1994; Maar & Burda, 2004), ‘‘imagic’’ (e.g., Fellmann, 1995), or ‘‘pictorial’’ (e.g., Mitchell, 1994) turn in the cultural and social sciences. While, more recently, the number of scholars who take the specific potential of visuals seriously has been on the rise – also within the domain of organization and management studies (for overviews, see, for instance, Bell, Schroeder, & Warren, 2013; Meyer et al., 2013), a broad acknowledgment of the visual mode of meaning construction and of its far-reaching consequences for theory building, as well as for the discipline’s empirical research agenda, is still missing. Interestingly, research on the role of visuals has found a particularly fertile ground in accounting research (see, for instance, the special issue in the Accounting, Auditing & Accountability Journal, 22[6]) – a discipline generally considered to be much more concerned with numbers, indicators, and facts, rather than imagery. This becomes less surprising when taking into consideration that ‘‘to account’’ means ‘‘to justify’’ and ‘‘to provide reasons,’’ and that visuals, due to their inherently iconic nature, have a similar fact-like character as have numbers. Moreover, visuals are equally qualified to disguise the ideological core they transport behind a ‘‘veil’’ of

144

MARKUS A. HO¨LLERER ET AL.

allegedly objective representation (Kress & van Leeuwen, 2006; Mitchell, 1994). Carruthers and Espeland (1991), for instance, remind us that in the 15th and 16th centuries, Italian merchants – in order to support their truth claims and prove their decency as businessmen – visually invoked God and a variety of religious figures in their account books. Similarly, Quattrone (2009) argues that by providing complex ordering instruments through minimal signs, visuals have powerfully aided the global spread of accounting practices. Until today, the various genres of corporate communication and public disclosure (for example, annual reports or corporate websites) have remained a central site of accounting for organizational activities and, therefore, for the struggle over meaning and social legitimacy. Pioneering research on the visual aspects of corporate public disclosure, while covering a broad range of topics and empirical phenomena, conceptually focused on how images and symbols had become integral elements in organizational self-depiction and corporate reporting practices (see, for instance, Benschop & Meihuizen, 2002; Breitbarth, Harris, & Insch, 2010; Campbell, McPhail, & Slack, 2009; Davison, 2008; Drori, Delmestri, & Oberg, 2013; Preston, Wright, & Young, 1996; Simpson, 2000). What unites all these endeavors is the notion that the relevance of visual artifacts goes far beyond a purely aesthetic moment. As Graves, Flesher, and Jordan (1996, p. 83) emphasize, they serve ‘‘the rhetorical purpose of arguing the truth claims of those reports and the social constructs they represent.’’ Hence, visualization in corporate documents is not at all trivial, or a simple and decoupled addon. On the contrary, it plays a decisive role not only in inspiring readers’ sense-making activities through various forms and shades of imagination, but also in the construction of social meaning and reality.

Visuals as a Means of Translation and Recontextualization Images and other visual artifacts are used to communicate complex ideas to broader audiences. As such ideas (for instance, in a managerial context, the concepts of shareholder value, total quality management, or – in our case – CSR) are mostly theorized on a global and rather abstract level, translation becomes necessary when implementing them in specific local – that is, historical, cultural, geographic, sectoral, or organizational – settings. The idea of ‘‘translation’’ (Czarniawska & Joerges, 1996) or ‘‘editing’’ (SahlinAndersson, 1996; for an overview, see Sahlin & Wedlin, 2008) has been developed in institutional theory as an extension of previous approaches

Imageries of Corporate Social Responsibility

145

that conceptualized globally diffusing ideas and practices as ‘‘ready-to-wear’’ offers. In contrast to this conventional diffusion research, more recent studies show that managerial ideas and concepts have to pass ‘‘powerful filters of local cultural and structural opportunities and constraints’’ (Meyer & Ho¨llerer, 2010, p. 1241) when being ‘‘imported’’ into local settings, and that their meaning is often altered considerably during these processes. While global ideas are decontextualized, theorized, and abstracted as ‘‘prototypes’’ or ‘‘templates’’ (Sahlin & Wedlin, 2008; Strang & Meyer, 1993), and can, in principle, be made sense of from a multiplicity of different perspectives and by use of various legitimating story lines (Meyer, 2013; Meyer & Ho¨llerer, 2010), these perspectives and story lines have to be recontextualized in order to fit the local setting and confer legitimacy to the adopting organization. Thus, any translation and re-localization process has to be understood against the backdrop of the specific socio-historical context within which it unfolds; the outcome (i.e., the re-localized, or ‘‘glocalized,’’ ideas and practices) bears the imprint of the global prototype itself as well as of the socio-historical contingencies of the local context of its adoption (Drori, Ho¨llerer, & Walgenbach, 2013). An increasing number of studies have been concerned with the way in which actors perform such translation work (e.g., Boxenbaum, 2006; Creed, Scully, & Austin, 2002; Frenkel, 2005; Hallett & Ventresca, 2006; Lawrence & Suddaby, 2006; Meyer, 2004; Powell & Colyvas, 2008; Zilber, 2006). With regard to the means of translation, the vast majority of studies have been primarily focused on how actors use and manipulate verbal discourse. In line with sociological (e.g., Raab, 2008) and social semiotic (e.g., Kress & van Leeuwen, 2006) approaches to visual studies, we argue that the visual mode of meaning construction facilitates the recontextualization of global concepts in several distinct ways: First, visual artifacts are comparatively better suited to communicate novel ideas across divergent audiences. Although often less precise than verbal text, their symbolic content is supposedly more widely understood. Machin (2004) even argues that the use of globally available image databases aids the creation of a ‘‘global visual language.’’ Visually conveyed messages, in such a way, can more easily oscillate between different symbolic realms. Second, visual artifacts are able to transport complex messages while consuming rather low amounts of space and time: They present themselves to the viewer in a much greater ‘‘immediacy’’ than verbal text does (e.g., Raab, 2008). Instead of a lengthy verbal treatise constrained by the ‘‘corset’’ of language, an image can convey its message(s) in a more holistic way and often – as, for example, in the case of photographs – also implies a greater facticity compared to

146

MARKUS A. HO¨LLERER ET AL.

words. Third, visual discourse is – still – less controlled than verbal text (Kress & van Leeuwen, 2006). Visual artifacts not only open up for imagination, they derive their persuasive effects especially from this implicitness, ambiguity, and openness (Phillips, 2000; McQuarrie & Phillips, 2005). This makes the use of visuals particularly well suited for the task of transcending dichotomies (e.g., ‘‘modernity,’’ ‘‘globalism,’’ and ‘‘traditional Islam’’ as shown by Kamla & Roberts, 2010), communicating ideas that are, for whatever reason (e.g., threats of illegitimacy, or taboos), difficult to verbalize, or addressing institutional plurality and complexity (e.g., Greenwood et al., 2011). The ability to ‘‘invoke’’ without having to provide an argument is a powerful rhetorical tool. In summary, these characteristics of the visual, together with the difficulty of unanimously tying down the meaning of an image, make the use of visual artifacts very attractive in times when corporations are increasingly being held accountable for views and values they express in public. As Styhre (2010, p. 12) points out, ‘‘[i]mages and pictures may not communicate more things or offer more accurate accounts of perceived social reality, but they certainly offer new opportunities for theorizing organizations and organizational practice.’’

THE ARCHITECTURE OF DISCOURSE The analysis of discourse has greatly contributed to research on organizations and institutions over the past decades (for an overview, see Phillips & Oswick, 2012). We extend existing research by integrating concepts from three literatures: First, in order to gain insights into the overall structure of a discourse, we draw on ideas from structural linguistics as adapted and utilized in German-speaking variants of discourse analysis (e.g., Bublitz, 2011; Diaz-Bone, 2010; Link, 1997). In more detail, we suggest that the meaning structure of a specific discourse is organized by a relatively small number of constitutive dimensions. According to Link (1997, p. 15, our translation), these discourse-carrying dimensions act as ‘‘steel beams’’ of the discourse in the sense that pulling them out ‘‘would make it tumble like a house of cards.’’ Discourse, then, appears as a comprehensive network of these dimensions and of the ‘‘polar opposites’’ that define their range (Bublitz, 2011; we henceforth refer to these polar opposites as ‘‘polar categories’’ to stress their role as building blocks of discourse).2

Imageries of Corporate Social Responsibility

147

Understanding the structures that underlie actors’ interpretations requires a reconstruction of these dimensions and of the meaning space they demarcate. Second, we argue that framing approaches (e.g., Benford & Snow, 2000; Gamson & Modigliani, 1989) provide a useful way of studying meaning and logics on the field level (Meyer, 2004; Meyer & Ho¨llerer, 2010). Frames invoke social stocks of knowledge; they have, at their core, an organizing idea to assign meaning to a diverse array of symbols or ‘‘idea elements’’ and to cluster them in a coherent pattern (e.g., Gamson & Lasch, 1983). They make sense of relevant events, suggesting ‘‘what is at issue’’ (Gamson & Modigliani, 1989, p. 3), and, therefore, indicate a specific discursive ‘‘position’’ for interpretation. With new or contested topics, a variety of such positions is available (e.g., Meyer & Ho¨llerer, 2010). While verbal framing cues have successfully been utilized to analyze how issues are framed and ideas are adapted to specific local settings, the potential of visual artifacts in this respect has yet to be explored in organizational research. Third, to describe the discursive formation on the field level and the different clusters of dimensions and polar categories that are used by fieldlevel actors to make sense of and to construct organizational practices, we borrow the notion of topos (from Ancient Greek to´poB, ‘‘place’’). Topoi resemble interpretive packages (Gamson & Modigliani, 1989; Meyer, 2004) in that they are meaningful, contextualized sets of argumentative resources that can be employed to mobilize consensus among one’s audiences. Considering their etymological origin, topoi denote places in the discursive space that provide a solid foundation; by referring to and invoking such topoi, statements and claims are located within the discourse (i.e., reflect a standpoint toward a certain issue). The analysis of topoi has its roots in linguistics (e.g., Kienpointner, 1992; Kopperschmid, 1989; Rubinelli, 2006; Wengeler, 2003); applications in organization research are rare (for exceptions, see, for instance, Grue, 2009; Jancsary, 2013). We take the assemblage of ‘‘standpoints’’ or topoi available within a particular field to constitute the discursive ‘‘landscape’’ (in a similar notion, e.g., Gamson [1992] calls this the ‘‘issue culture’’; Meyer & Ho¨llerer [2010] refer to a ‘‘topography of meaning’’). This landscape, we argue, also delineates the meaning horizon within which novel ideas, concepts, or practices may be recontextualized. How this landscape is shaped, and where its boundaries lie, is a result of the specific historical, cultural, and material contingencies and the constellation of institutional logics at work in a particular field.

148

MARKUS A. HO¨LLERER ET AL.

PHENOMENON AND EMPIRICAL CONTEXT Since the turn of the millennium, corporations have found themselves in the center of a growing public debate on the role and responsibilities of business within society (Ho¨llerer, 2012). Legal regulation and ‘‘soft law’’ on corporate governance have pressured corporations to engage proactively in this discourse, resulting in various business-led initiatives. Indeed, maintaining, substantiating, and managing societal legitimacy (e.g., Brammer & Pavelin, 2006) have become central tasks in times of crisis and worldwide corporate malfeasance – and even more so in the light of instable financial markets and a loss of confidence in executive boards. Stakeholder dialogue and communication with influential actors in a corporation’s environment have thus gained increasing relevance, and the effects thereof can be seen across various genres of corporate communication (den Hond, de Bakker, & Neergard, 2007). CSR broadly denotes social and societal challenges that come with the conduct of business (Hiss, 2009) and aligns a corporation’s economic processes with the ideas of integrity, fairness, transparency, and generally accepted social values (e.g., Matten & Moon, 2008; Thompson, 2008). As a vaguely defined umbrella concept and comprehensive label for a bundle of related sub-concepts, the notion of CSR looks, on the one hand, back on a ‘‘long and varied history’’ (Carroll, 1999, p. 268); on the other, and despite the recent hype in corporate practice and academic literature (for an overview, see Crane, McWilliams, Matten, Moon, & Siegel, 2008), its actual meaning, content, and scope have been the subject of controversial debates ever since. Depending on perspective and institutional background, different aspects and elements have been emphasized (Brammer, Jackson, & Matten, 2012; see also, for instance, Jackson & Apostolakou, 2010; Kinderman, 2009, 2012; Matten & Moon, 2008). Although in some countries – especially continental European countries characterized by coordinated market economies – the understanding of a broad social/societal responsibility of business is not at all new, the AngloAmerican coined terminology of CSR has only recently been adopted (Hiss, 2009; Ho¨llerer, forthcoming). The same applies to the highlighting of, and reporting on, corporate social performance beyond legal requirements (Ho¨llerer, 2012; Vogel, 2005). It is in this sense that the dissemination of an explicit commitment to CSR in countries like Austria mirrors the concept’s global victory march (see also Meyer & Ho¨llerer, 2011). For various reasons, Austria is an excellent setting for the empirical study of the translation of CSR in greater detail. As a country with a strong

Imageries of Corporate Social Responsibility

149

corporatist tradition (for comparative data, see Gourevitch & Shinn, 2007), social/societal responsibility of business has been firmly anchored in Austria’s institutional framework as part of ‘‘institutionalized solidarity’’ (e.g., Kinderman, 2009), but also in the often paternalistic self-understanding of the nation’s business elite (Ho¨llerer, forthcoming) – long before it was ‘‘discovered’’ as a strategic instrument to signal sound management conduct. Explicit social disclosure is, in general, a rather recent phenomenon in Austria, and had long been restricted to a small number of corporations. Before 2000, issues of CSR had altogether been addressed infrequently – and if so, only in passing. Gradually throughout the late 1990s and early 2000s, corporations learned the explicit vocabulary and rhetoric of CSR. The absence of detailed regulation or standards has, however, left considerable leeway for corporations to pick issues deliberately from the CSR agenda and disclose selected information and data to the public. Stand-alone annual CSR reports were not issued by any publicly traded corporation in Austria prior to 2001. While sometimes criticized as glossy marketing instruments or mere ‘‘talk,’’ these reports nonetheless materialize corporations’ interpretations of CSR. Apart from verbal accounts, descriptions, and explanations, these reports contain a conspicuous number of images in which corporations give insights into their interpretations and translations of CSR. These visual claims are the focus of this paper.

EMPIRICAL DESIGN Sample and Empirical Material Our empirical sample encompasses stand-alone annual CSR reports3 issued by Austrian publicly traded corporations in addition to their annual financial reports. The observation period starts in 2001, when the first reports were issued, and ends in 2008. What makes this new (sub-)genre of corporate communication particularly attractive for our study is the fact that it addresses multiple audiences and, hence, can be assumed to embrace the diverse expectations corporations are confronted with in the context of CSR. Our sample constitutes the full count4 of publicly traded corporations in Austria issuing stand-alone annual CSR reports (i.e., a total of 37 reports from 12 different corporations featuring 1,652 images). The majority of corporations included in our sample are part of the Austrian Traded Index (ATX); various industries are covered, with utilities playing an important role among our empirical sample. Due to some reporting periods comprising

150

MARKUS A. HO¨LLERER ET AL.

two fiscal years in one volume, our empirical material covers a total of 43 fiscal years.

Analytical Procedures Even though the development of visual research methods has gained momentum during the last years (e.g., Margolis & Pauwels, 2011; Ray & Smith, 2012; Rose, 2007; see also a recent special issue in Qualitative Research in Organizations and Management), scholarly work in organization and management studies has been mostly limited to hermeneutical or semiotic analyses of single pictures or fairly small sample sizes (a notable exception is, for instance, the content-analytical approach of Breitbarth et al., 2010). Studies that are interested in field-level formations, however, require a methodological foundation that is able to deal with larger amounts of visual data without, at the same time, compromising the claim to reconstruct meaning. For such analyses, existing methodologies are useful only to a limited degree. Elsewhere (Jancsary, Ho¨llerer, Meyer, & Vettori, 2011), we elaborate in more detail on an approach for the analysis of large quantities of visual data that also incorporates hermeneutical procedures of meaning reconstruction. We will briefly discuss our various analytical steps in the following. In a first step, we inductively developed codes and categories regarding the manifest (i.e., content and stylistic elements) and latent (i.e., symbols and connotations) aspects of the images included in our sample.5 The development of these detailed codes mainly served to understand better the specific visual vocabulary used (see also Kress & van Leeuwen, 2006). The result was a comprehensive visual ‘‘dictionary’’ of ‘‘symbolic devices’’ (Gamson & Lasch, 1983) referring to different aspects of the potential meaning of CSR in the Austrian corporate world. The core objective of the second step was the analysis of patterns within and across individual images, using manifest elements and latencies as sensitizing concepts. We adapted the analytical grid of Froschauer and Lueger’s hermeneutic analysis (2003; Lueger et al., 2005; Lueger, 2010) and applied it to the reconstruction of visual idea elements (i.e., of the typical ‘‘claims’’ transported by the images). At least one short paraphrase was formulated for each of the 1,652 individual images, answering the question: ‘‘What are the claims the image conveys?’’. For each such paraphrase, subsequently, potential structural conditions (‘‘Under what circumstances can such claims be perceived as reasonable and/or typical?’’) and hypothetical consequences

Imageries of Corporate Social Responsibility

151

(‘‘What effects would such a claim typically have, and how would this impact on our understanding of CSR?’’) were considered. To avoid subjectivity and an improper ‘‘narrowing’’ of interpretations, coding was carried out in a research team of up to four and included several interpretive ‘‘cycles.’’6 The results were condensed and aggregated, resulting in a set of 154 different idea elements, each denoted by a catchphrase. In order to account for the inherent multivocality of visuals, each image was coded with one to five idea elements (with single coding being the rare exception). The third step was devoted to the reconstruction of the discourse-carrying dimensions that underlie the idea elements: For each idea element, the whole research team discussed the organizing dimensions it addresses, as well as the polar opposites that are implied by these dimensions.7 This resulted in a list of 21 discourse-carrying dimensions and 42 polar categories, with each pair representing the oppositional ends of a discourse-carrying dimension. In a fourth step, we considered the more quantitative aspects of our data in order to get a comprehensive picture of the discursive formation or landscape. We plotted the polar categories into a network, with links between them denoting the typicality of their mutual co-occurrence in individual images.8 This illustrates, on the aggregate field level, which categories and dimensions occupy similar positions in the discursive space. We then applied the Newman clustering algorithm9 to the network in order to identify internally consistent, modular clusters. As these clusters consist of polar categories that are – in our sample – typically invoked together to create a visual claim, we take them to constitute seven different topoi; that is, they represent different rhetorical standpoints within the overall discursive landscape which enable specific perspectives on the issue of CSR. A topos – much like an interpretive package – has a central story line expressing its main argument (see, e.g., Meyer & Ho¨llerer, 2010). For each topos, thus, we reconstructed this story line using polar categories, as well as related idea elements, as cues. Subsequently, we assigned labels to the topoi capturing their most prominent ideas.10

THE VISUAL DISCOURSE OF CSR IN AUSTRIA: FINDINGS AND DISCUSSION Constitutive Dimensions We define discourse-carrying dimensions – building on Link (1997) – as the ‘‘steel beams’’ that hold together the discursive edifice. They are connected

MARKUS A. HO¨LLERER ET AL.

152

to each other and, through these ties, constitute the overall character of a specific discourse; through their formation, they present the ‘‘architecture’’ of discourse (Bublitz, 2011). Table 1 presents the 21 ‘‘steel beams’’ of Austrian visual CSR discourse, as well as the polar categories that delineate each dimension’s outreach. The table also shows frequencies as the number of images representing a specific dimension and its polarities.11 Among the discourse-carrying dimensions in Table 1, we find six that express divergent attitudes or orientations relevant for CSR. These comprise [value system], which is by far the most frequently invoked dimension, but also [impetus], [attitude toward change], [strategic preference], [human values], and levels of [seriousness]. Three of these dimensions are heavily lopsided in respect to the polarities invoked, implying that when referring to CSR, there is a focus on [other (noneconomic) values] over [economic value], of [innovation] over [tradition], and of [exploration] over [exploitation]. For the remaining three

Table 1. Discourse-Carrying Dimensions and Their Polarities. Discourse-Carrying Dimension

N

Polarity A

N

Polarity B

N

Value system

650

Economic value

86

585

Impetus Attitude toward change Strategic preferences Human values Seriousness Exchange Exertion of influence/ control Area of human influence Life sphere Level of abstraction Scope/sociability Variance Connectivity Familiarity Locus Timeline Development Trustworthiness Professionalism Capability

306 239 214 78 57 488 238

Rational Tradition Exploration Material Serious Giving/sharing Managed/ controlled Nature Sphere of work Typified Individuality Homogeneity Connection Familiar/close Universal/global Past Improvement Trustworthy Professional Potency

190 78 187 78 33 475 165

Other (noneconomic) values Emotional Innovation Exploitation Spiritual Playful Taking/keeping Untouched/ uncontrolled Technology Beyond work Personalized Collectivity Heterogeneity/diversity Separation Unfamiliar/alien Specific/local Future Deterioration Untrustworthy Unprofessional Impotency

592 302 380 284 31 390 165 140 272 258 640 579 270

257 257 44 1 0 390 134 80 26 254 638 579 270

123 196 33 125 33 14 74 469 583 341 283 31 2 70 68 265 5 3 0 0

Imageries of Corporate Social Responsibility

153

dimensions, both poles are depicted rather equally in the visual CSR representations. A second set of dimensions addresses different forms of behavior. On the dimension of [exchange], the polar category of [giving/ sharing] is more often implied than [taking/keeping]. A look at the dimension of [exertion of influence/control] shows that a focus on the active management of the organization’s environment ([managed/controlled]) is favored over the notion of unobtrusiveness ([untouched/uncontrolled]). A third set contains dimensions differentiating [nature] from [technology], and the [sphere of work] from personal life spheres [beyond work]. CSR is strongly related to all these domains. Much more imbalance in terms of frequencies can be found in the fourth set that deals with social dimensions. While the scales clearly tip in favor of [personalized] accounts and [heterogeneity/diversity], [collectivity] is preferred over [individuality]. With regard to spatial dimensions, we find dimensions relating to distance ([connection] vs. [separation], and [familiar/ close] vs. [unfamiliar/alien]) as well as reach ([universal/global] vs. [specific/ local]). We assess that ‘‘distances’’ are minimized, with [familiar/close] and [connection] being clearly more frequent than their respective opposites. With regard to temporal dimensions, we find an overwhelming credo for advancement. [Future], together with its developmental ‘‘companion’’ [improvement], is clearly dominant and conveys a rather optimistic view. A final set of dimensions represents different qualities in the sense of specific attributions that corporations assign to themselves: [trustworthiness], [professionalism], and [capability]. These dimensions constitute dichotomies of ‘‘either/or,’’ with the organization either having a specific quality or not. Unsurprisingly, the positively connoted pole of the respective dimensions is always dominant, with [unprofessional] and [impotency] being, effectively, absent in the discourse (i.e., they are purely hypothetical poles that are never manifested in actual images). These dimensions, and the polar categories that define their oppositional ends, are the ‘‘raw material’’ from which the field-level visual CSR discourse is built. The next section will present the visual-rhetorical positions that are constructed from this material.

Topoi: Rhetorical Standpoints Toward CSR In order to map the formation of visual CSR discourse in Austria, we plot the polar categories into a network based on their co-occurrence in individual images. We use the Newman algorithm to cluster these categories into modular sets (Fig. 1).12 Building on the conceptual framework outlined

MARKUS A. HO¨LLERER ET AL.

154

Fig. 1.

Network of Polarities (Newman Clustering Algorithm).

above, we argue that the resulting network can be interpreted in a ‘‘spatial’’ manner: Each of the clusters forms a topos – a distinct standpoint providing a specific perspective on CSR.13 Taking this spatial metaphor further, the comprehensive set of topoi within a field (i.e., our clustered network) provides a ‘‘topography of meaning’’ (Meyer & Ho¨llerer, 2010), with the individual topoi claiming different degrees of space and ‘‘visibility’’ within the whole landscape (see also the frequency of their use within the discourse in Fig. 2). Exhibit 1 provides a brief portrait of the seven central topoi we found in our analysis of ‘‘Austrian-style’’ visual CSR discourse. To characterize each topos, we interpret the specific constellation of clustered polar categories and the signature idea elements14 that are represented by these categories. These seven topoi encompass the different rhetorical standpoints that are assumed by publicly traded corporations when they visually enact CSR. Corporations use them with varying intensity. Taking into account Austria’s long history of institutionalized social solidarity and implicit CSR (e.g., Ho¨llerer, forthcoming), together with the corresponding notion of the entrepreneur as an essentially paternalistic actor, it comes as little surprise that the topos of Local Community is the most dominant one, appearing in about half of the featured images. A corporation’s

Imageries of Corporate Social Responsibility

Fig. 2.

155

Collapsed Model of Topoi and Their Interrelationships.

responsibilities, thus, seem to be directed foremost at the community in which it is embedded. However – and strongly overlapping – the topos of Values is invoked almost to the same degree. On the one hand, it complements responsibility for the local community with one for other focal internal and external stakeholder groups, such as employees and families. On the other hand, it stresses responsibility toward more abstract ‘‘stakeholders’’ like nature or future generations. Both avenues are characterized by religious and ethical ideas, using spiritual/metaphysical symbolism as a basis for legitimacy. Concern with the future, however, is paralleled by a strong rootedness in history and tradition: Progress, a third highly visible topos relating to CSR, invokes the country’s (as well as the focal corporation’s) rich history as a solid basis for a bright future – and deproblematizes technological developments. Topoi with a more direct link to the corporate/business world – Mastery, Globalism, and Enterprise – are featured to a far lesser degree, with the latter two – in terms of frequency of occurrence – bordering on insignificance. CSR in Austria – as portrayed by corporations in visual discourse – thus, is above all a strongly integrated amalgam of community values, spiritual reverence of untouched nature as divine creation and place for recreation, and commitment to past traditions and achievements ‘‘spiced’’ with references to economic and technological

156

MARKUS A. HO¨LLERER ET AL.

Exhibit 1. Topoi and Central Story Lines. Mastery. Visualizing CSR from the standpoint, or topos, of Mastery involves presenting the corporation as strong, capable, and vigorous. Through the use of polar categories such as [potency], [rational], [managed/controlled], [material], and [exploitation], the corporation claims power and control over itself and its environment. This is relevant in the context of CSR in at least two ways: On the one hand, acting socially and environmentally responsible necessitates a certain control over external factors (for instance, how could anyone be held responsible for something he/she cannot influence?). Idea elements like having an impact or measurability lie at the core of this argument. On the other hand, only an economically potent corporation will have the means to go beyond the ‘‘required’’ levels of responsibility and engage in the protection of nature or the support of society (see also the classic arguments presented by, for instance, Carroll, 1991; Drucker, 2007). Such capacity is, for instance, inherent in idea elements like demonstrating physical strength and vigor and evidence of success. The implicit story line of this topos is, thus, one of mastering challenges and managing the broader organizational environment. Progress. There are three distinct ways in which the topos of Progress supports efforts to mediate apparently incompatible ideas. First, it directly connects the polar categories of two dimensions: [timeline] by incorporating [past] as well as [future], and [attitude toward change] by linking [tradition] and [innovation]. These connections create a strong sense of future-oriented development without neglecting one’s own history. They are invoked by core idea elements such as bridging time, mediating tradition and innovation, or we have a history. Second, technology is linked to the preservation of natural resources, as illustrated in the idea element mobility and clean energy. And third, technology is linked to accessibility (easy-to-use technology, playful approach to technology). The story line of this topos is one of creating development and change that is path-dependent and hails the achievements of the past. Defining CSR from this perspective emphasizes the genuineness and casualness by which the corporation strives for technological progress. Local Community. The corporation does not exist in a social vacuum; it is physically and culturally embedded ([connection]) in a [specific, local]

Imageries of Corporate Social Responsibility

157

community of various stakeholders ([collective]) toward whom it assumes a certain responsibility. While – most of the time – this means integration, embracing the community, as well as an exchange of views and ideas, sometimes more drastic action is necessary. In the case of [serious] disasters it also means being prepared for crisis response operations in order to protect the community. Overall, the underlying story line here is one of mutual interdependence, and CSR essentially entails giving back ([giving/sharing]) to the community in which the corporation is embedded. Globalism. The topos of Globalism seems to be antagonistic to that of Local Community. It is concerned with relationships beyond the specific environment in which a corporation is situated ([universal, global]) and also includes [economic value] – a category that seems to be in contradiction to the common good-oriented tone within the Local Community cluster. Looking more closely, however, the story line stresses that an international orientation not only means novel and international business opportunities, but also increased responsibility. The signature element global orientation captures this broader understanding of a corporation’s global engagement. It expresses adherence to global standards (for instance, the UN Global Compact) as well as, more broadly, recognition of the global impact of corporate decisions and actions. We therefore interpret the topos of Globalism as an extension of the community idea that transcends a specific local context. CSR, from this perspective, emphasizes that the economic world has become global – and so has the corporation’s responsibility. Values. CSR, expressed through this very topos, emphasizes the importance of values other than purely economic ones. This topos evokes the [spiritual] and [emotional] realm, often embodied in an idealistic, or even naı¨ ve, view on [nature]. Corresponding idea elements comprise, for instance, appreciation of nature’s wealth, or sanctity/ divinity of nature, which, to a certain degree, also imply worshipping nature. The consequence of such appreciation of nature is the notion of preserving natural idylls, which is particularity well expressed through the polar category of [untouched/uncontrolled]. Even though established religions are not explicitly referred to, this topos tells a story of reverence, appreciation, and gratitude. It contains a variety of religious symbols and connotations, such as a ray of light, people in awe, or notions of infinity, fertility, and creation. Providing a counterpoint to the economic and occupational sphere, the topos of Values also

158

MARKUS A. HO¨LLERER ET AL.

addresses the human being in its everyday life-sphere [beyond work], be it as part of a family or community, or, mostly, as an individual in need of recreation. The topos extends the corporation’s domain of responsibility to include formerly private spheres of life – an extension that, from a power and control perspective, may arouse negative associations. Enterprise. Visionary ideas, however, are not necessarily restricted to the natural realm. The comparatively infrequently invoked topos of Enterprise is dedicated to looking beyond existing solutions and practices. It is concerned with [exploration] and leaving the beaten track in order to discover and face the [unfamiliar/alien]. Here, corporations present themselves as visionaries daring to transcend the taken-for-granted, with the signature idea element being looking for answers. Accordingly, the storyline of this topos is that CSR entails the quest to discover novel answers to existing or new challenges for the corporation. Credibility. This last topos is not primarily concerned with depicting certain aspects of CSR; rather, it aims at ascertaining the credibility of the focal corporation as ‘‘claims-maker.’’ Visual artifacts using this topos portray the corporation as [professional] and [trustworthy] in its core domain ([sphere of work]). Credibility is [personalized]; that is, it comes ‘‘with a face and a name.’’ Such focus on the person is further enhanced by [heterogeneity/diversity]. A closer look at the signature idea elements reveals more cues: Corporations visually establish credibility in two different – but related – ways. On the one hand, they aim at invoking ideas such as fairness in business, transparency, valuing diversity, we are responsible, or quality of services. On the other hand, they utilize the voice and testimonial of prominent and/or prestigious stakeholders by visualizing external approval, being under the scrutiny of stakeholders, and various forms of advocacy.

capability, global engagement, and visionary entrepreneurial spirit. This is highly consistent with paternalistic elements and the influence of Christian ethics on the traditional Austrian economic elite (see also Meyer & Ho¨llerer, 2011; Ho¨llerer, forthcoming). What seems noteworthy is that publicly traded corporations as well evoke the interpretive scheme of the traditional industrialist family when visualizing CSR. However, corporations construct

Imageries of Corporate Social Responsibility

159

not just the practice of CSR, but also themselves in relation to the practice (i.e., the socially responsible corporation). The high prevalence of yet another topos – Credibility – makes evident that, with regard to the symbolic representation of their interpretation of CSR, corporations are considerably concerned with broader issues of credibility and accountability – especially when it comes to the noneconomic sphere and to community claims.

Field-Level Formation and Governing Logics Topoi are field-level constructs – according to Kopperschmidt (1989), they only appear as true or legitimate within specific belief systems. As Thornton et al. (2012, p. 148) note, such field-level constructs are embedded in, and shaped by, societal-level logics; they draw on categories and schemes provided by them, but are equally subject to field-level pressures and processes that ‘‘generate distinct forms of instantiation, variation, and combination of societal logics.’’ Thus, on a higher level of abstraction, topoi have core organizing ideas or principles; these ideas, and the categories and schemes they provide, bear the imprints of the combination of societal logics that govern the field. While some discursive ‘‘zones’’ are exclusively governed by single logics, others combine different logics and, in this way, absorb institutional complexity on the field-level into meaningful interpretations of the issue. In our network, this becomes apparent in the assortment of polar categories within the clusters, but also in the existence or absence of links between the various topoi. About two thirds of our idea elements bridge topoi. Looking at individual images, this mediating effect is even more pronounced: About 75% of images within our sample invoke aspects of several topoi simultaneously. Such images provide a ‘‘blended’’ perspective on the phenomenon, enabling several standpoints at the same time. To facilitate the presentation of relationships between topoi, we collapsed the various clusters from Fig. 1 into single nodes and arranged them in a schematic manner. Fig. 2, then, represents the emerging model. The size of a node symbolizes the overall visibility of the corresponding topos. We find that images establish relationships between topoi in two different ways: First, they link topoi that are similar with regard to the underlying constellation of logics. Second, they provide bridges between topoi that represent different societal logics. In addition to established relationships, the absence or relative weakness of links also deserves attention.

160

MARKUS A. HO¨LLERER ET AL.

Concerning the first type of relationship, topoi can be related by similarity (solid lines in Fig. 2). Such relationships emerge between clusters that are close to each other in the original network (see Fig. 1) and form topoi that instantiate similar combinations of societal logics. The topoi of Mastery and Progress, for instance, both essentially claim that the corporation ‘‘knows its craft,’’ visualize organizational and managerial expertise, and thus give shape to the schemes and categories provided by a specific combination of the logics of profession and corporation. The connection between them is primarily built on the bridging power of [technology] as both a rational tool used to master ones tasks and as an area of improvement, modernization, and progress. In a similar vein, the topoi of Values and Enterprise are related insofar as their metaphysical symbolism and focus on faith and sacredness bear the imprint of the societal logic of religion. They are connected by the symbolic representation of exploring the unknown as a [spiritual] experience and [other (noneconomic) value] in its own right. Local Community very clearly activates a community logic, while the topos of Globalism extends this notion of embeddedness, reciprocity and [giving/sharing] to the entire activity space of the corporation, yet does so with a conspicuous coloring of a market logic [economic value]. Second, images and their respective idea elements constitute bridges across different institutional spheres (dotted lines in Fig. 2). For instance, Mastery – a topos anchored in a professional and corporate logic – is aligned with the spiritual sphere of Values through the symbolic representation of [nature] as a material resource for the corporation to be used and as life’s essential elements, beyond human understanding and control. Mastery and Globalism are linked by conjuring up ideas that belong to the otherwise fairly underrepresented logic of the market: [material] and [economic value], visualized especially as the corporation’s industrial and economic power, as well as the ideal of the free (capital) market. Progress is aligned with Values through framing technology in alternative ways, as in, for instance, industrial romance, but also by the visual alignment of innovation and family values. It links with Local Community through ‘‘improving everyday life,’’ as expressed in idea elements like embeddedness of technology in our lifestyle. Finally, the weakness or absence of links between two topoi is as interesting as the existence of ties, because it points to spheres that, in general, remain visually separated. In our case, the topos of Credibility provides an interesting example: It exclusively links to two other topoi: Values and Community. The appreciation of noneconomic values and the

Imageries of Corporate Social Responsibility

161

pursuit of community goals are precisely those discursive zones where corporations presumably lack credibility. Thus, images are employed not only to appeal to different audiences simultaneously, they also transport authenticity claims and allow for the bridging of ‘‘credibility gaps.’’ To sum up: The field-level topoi used by Austrian publicly traded corporations to construct visually the practice of CSR evoke the societal logics of corporation, profession, community, religion, and – in a rather vague manner and confined mainly to a global action radius of the corporation – market. What is striking is that images neither rely on single topoi nor on single underlying logics. To the contrary, we find that most images bridge topoi and logics, thereby establishing CSR as a ‘‘bridging concept’’ that overcomes various traditional dichotomies.

CONCLUSION AND CONTRIBUTION Our paper started with the question of how institutional complexity is managed during the translation and recontextualization of a prominent, globally diffusing managerial concept. While prior research has focused on verbal discourse and demonstrated how field-level logics and vocabularies emerge to guide local interpretation and sense-making, we emphasize here the performative power of the visual in making visible the invisible, unobservable, and unknowable institutional substances (Friedland, 2009). In order to tackle our initial question empirically, we reconstructed the meaning dimensions that underlie the ways in which Austrian publicly traded corporations visually make sense of, and construct, the practice of corporate social responsibility in their CSR reports. Borrowing the concepts of ‘‘discourse carrying dimensions’’ and ‘‘topoi’’ from structural linguistics and rhetorics, we reconstructed a set of meaning dimensions that organize the visual discourse via fundamental opposites. We identified consistent clusters of polar categories within the discourse, representing seven standpoints that provide quite distinct perspectives corporations use to relate to CSR. We then discussed how these topoi and the ‘‘topography’’ they amount to bear evidence of the combination of broader societal logics that govern the field: Austrian corporations primarily use categories and schemes provided by the broader logics of profession, corporation, community, and religion to depict and represent a CSR orientation. Whether the relative insignificance of the market logic is a

162

MARKUS A. HO¨LLERER ET AL.

specificity of the practice (i.e., CSR), or the corporatist context in Austria, is a question that exceeds the scope of this paper and requires further analysis. Corporate actors, in their efforts at creating a locally resonant version of the global prototype, have to address multiple audiences and resolve inconsistencies created by the specific constellation of institutional logics prevalent at the field-level. We show that – visually – CSR is a bridging concept not only in the sense that global ideas are made resonant on the local level, but also in that the images used facilitate the reconciliation of these inconsistencies. Similar to, for instance, Meyer (2004), who argues that verbal accounts that do not belong to one interpretive package exclusively but ‘‘fit’’ into multiple story lines are an important resource for ‘‘frame alignment’’ (see also Snow, Rochford, Worden, & Benford, 1986), we find that multivocal or ambiguous visual accounts can be used to profess the compatibility of divergent positions. However, such ambiguity can also be detrimental to corporate interests, as the polysemous nature of visuals exacerbates targeted communication and invites unintended and subversive readings. An emerging awareness of these issues and the increasing ‘‘visual literacy’’ both on the producer and the consumer side can be expected to lead to more hesitation regarding the non-reflective use of visualization in corporate communication. Our overall impression is that, over time and with increasing experience, corporations exhibit more compartmentalization, greater reflection and expertise (e.g., fewer ‘‘snapshot’’ images, more stylization), and more differentiated and individualized forms of claimsmaking and enactment of field-level logics. Also, attention increasingly seems to be given to other, less ambiguous forms of visualization, and thus to visual artifacts such as graphs, typography, fonts, or elaborated color schemes. Our work adds to literature on institutional complexity (e.g., Greenwood et al., 2011) and field-level institutional logics (e.g., Thornton et al., 2012) by examining how a pluralistic local environment and broader societal logics impact the translation and recontextualization of a complex managerial concept. In promoting their institutional logics perspective, Thornton et al. (2012) have called for more attention to field-level logics, in particular to how practices and symbolic representations are entwined through the emergence of field-level ‘‘vocabularies-of-practice.’’ Prior research has demonstrated that vocabularies are strongly tied to and representative of institutional logics, and has shown how their strategic use can help to reify, resist, or transform these logics (e.g., Dunn & Jones, 2010; Loewenstein et al., 2012; Nigam & Ocasio, 2010). Burke (1989) also notes that the

Imageries of Corporate Social Responsibility

163

blending of vocabularies can downplay distinctions by acting as bridges between two terminologies. Examining how imageries-of-practice provide a shared visual language around the category of the ‘‘socially responsible corporation,’’ we argue that visuals are equally central to communication, sense-making, and collective identities as are verbal vocabularies. In fact, we argue that pictures might be even better suited to imply facticity and to objectify the socially constructed categories they represent. Moreover, we contend that the holistic, immediate, and less tightly controlled visual mode of meaning construction (see Meyer et al., 2013) is particularly well equipped to address situations of institutional complexity. Visual artifacts are able to invoke different rationalities simultaneously: Happy children in front of a power plant situated in green pastures, complex technology presented as children’s toys, or an oil rig portrayed against a red sunset are all examples of such unobtrusive visual blending. In addition, visual artifacts can symbolically represent aspects that are hardly possible to articulate through more traditional means (for instance, it has been quite difficult for corporations to verbally express metaphysical and spiritual ideas in annual reports, while this is feasible through the use of imagery). This capacity to reconcile and mediate makes visual rhetoric especially useful for processes of translation and recontextualization that always have to ‘‘locate’’ a new practice in ‘‘what’s already there’’ (e.g., Boxenbaum, 2006; Campbell, 2004; Czarniawska & Sevo´n, 1996). In our study, we find images to support the mediation of oppositions in several ways: First, by translating abstract global ideas into concrete local knowledge, they aid in mediating spatial oppositions; second, by linking the past, present, and future, they bridge time; third, by mediating between different institutional spheres and their divergent logics, they appease ideational oppositions and reduce institutional complexity; and fourth, by connecting questionable claims with representations of authenticity, they aid in overcoming credibility gaps. While topoi and the related imageries-of-practice are embedded in and shaped by field-level logics and societal logics, their relationship is, diachronically, recursive. Future research is needed to analyze the dynamic dimension of the emergence of imageries-of-practice, field-level logics, and the evolution of the broader societal logics. With regard to the translation of global ideas and practices in institutionally complex local environments, our study indicates that the visual recontextualization concerns at least three different levels of abstraction: First, on the level of images, abstract global ideas are transformed into locally resonating symbols through the use of specific visual cues (e.g., depicted objects, stylistic

164

MARKUS A. HO¨LLERER ET AL.

variations, rhetorical figures). Examples are the depiction of ‘‘trust’’ through a handshake, the evocation of ‘‘community’’ by showing an idyllic village in the mountains, or of the ‘‘flow of time’’ through a series of locally resonant icons from different historic eras. Such use of rhetorical devices may be strategic, although the preference of some symbols over others might not be a conscious choice, but rather the result of the cultural socialization in a specific life-world. In documents like CSR reports, corporations also have ample opportunity to use multiple images, including their sequencing, to achieve local resonance. Second, recontextualization is realized on the level of idea elements and topoi: As rhetorical devices have to draw on the social stock of knowledge in order to be understood, the visual claims they make transcend the single image. Visual translation, thus, has to be achieved by using a ‘‘standpoint’’ and making visual claims that are regarded as legitimate for the particular claimsmaker within the local setting. Similar to verbal recontextualization, claims that are potentially problematic – in our case: corporations claiming to champion noneconomic and community values – require specific visual accounts as safe-guarding devices. Finally, on the level of discourse structure, the meaning horizon within which ideas and practices can be recontextualized is defined by the particular arrangement of fundamental discursive dimensions and the polar categories they contain. The extent to which a novel idea or practice can be meaningfully related to such dimensions, and the degree to which they resonate with the local contingencies and the particular constellation of field-level logics, defines the number and the persuasiveness of topoi; it thus expands or restricts the discursive space available for actors to recontextualize innovations. Visuals seem especially suited to enlarge this meaning space through their capacity to invoke without arguing, as well as to bridge and blend. Like any study, this study has its limitations, the most apparent of which is the lack of a comparative design. More research is needed to explore the use of visual cues across different cultural settings, but also to compare visual and verbal topoi as well as the discursive structures they create. In a sense, this study is only a first step: Longitudinal, cross-sectional, and comparative designs open up for a vast array of avenues for future research. Moreover, and despite the fact that we analyze the full sample of publicly traded corporations in Austria, we do not cover the entirety of the visual discourse on CSR in this specific empirical setting. It would be interesting to include voices from outside the world of business as well, and to compare the corporate perspective with the one of other societal actors (e.g., civil society, interest groups, or media). Finally, we did not consider the context of production of visuals in detail. We are, however, aware that a

Imageries of Corporate Social Responsibility

165

considerable part of CSR reports are produced with the professional help of public relations agencies. In closing, we return to the remark that a specific strength of visual rhetoric undoubtedly lies in enabling a particularly creative and unrestricted use of symbolic language, mainly as it is not limited by the linear and successive logic of verbal text. And while it might be true that visual artifacts do not enjoy the same ‘‘status’’ in our Western culture as verbal text does this also means that they are less institutionally regulated and still fly ‘‘under the radar of control’’ (see also Kress & van Leeuwen, 2006). One has to be aware, however, that visual rhetoric – as an effect of the very same multivocal nature that enables bridging different institutional spheres – creates ‘‘excess meaning’’ that is controllable only to a limited extent: Images may communicate more, or different, things than intended, and they are particularly prone to creating irony and subversive reading. Thus, visual rhetoric can be strategic only to a certain degree. Still, their powerful presence – even if often unobtrusive – is hard to ignore, as they immediately make an impression on their consumers (Mitchell, 1994). Or, as Berger (1972, p. 7) puts it: ‘‘Seeing comes before words.’’

NOTES 1. ‘‘Visual artifacts,’’ in general, encompass a broad array of forms, from photographs, pictures, paintings, drawings, sketches, and figures to logos and typography, or even to the ‘‘visual design’’ of social situations (for instance, in architecture). 2. A focus on oppositional structures is quite common in structuralist theory. It is, for instance, embodied in Greimas’ (e.g., 1983) idea of ‘‘semic categories.’’ Apart from linguistics, oppositions have also been used in Foucauldian discourse analysis (e.g., Link, 1997) and systems-theoretical approaches (e.g., Titscher, Wodak, Meyer, & Vetter, 1998). In organization research, Jones, Maoret, Massa, and Svejenova (2011) have recently presented a research design to study the emergence of a de novo category in architecture that points in a similar direction. 3. We also cover reports that are labeled as ‘‘sustainability reports’’ and/or use German equivalents; environmental reports were not included as they constitute yet another (sub-)genre (see also Ho¨llerer, 2012). 4. Two reports were excluded from the sample as they did not contain any visual artifacts. 5. We used NVivo to facilitate coding during all steps. 6. While two team members conducted all steps of coding and interpretation, the other two were involved especially in the initial, second, and the third step, as well as for reliability measures.

166

MARKUS A. HO¨LLERER ET AL.

7. The idea element displaying heroism, for instance, was decomposed into the dimensions [exchange], [capability], and [trustworthiness]. On the dimension of [exchange], it occupied the pole [giving/sharing] as opposed to [taking/keeping]; on the dimension of [capability], it expressed [potency] rather than [impotency]; and on the dimension of [trustworthiness], it incorporated [trustworthy] as opposed to [untrustworthy]. See also below for more details. 8. Plotting co-occurrences as absolute numbers would bias our results due to variation in the frequency of individual nodes. Instead, we opted for normalizing the co-occurrence matrix in ORA (http://www.casos.cs.cmu.edu/projects/ora). In this way, links represent the proportion of images coded with the two respective polar categories (normalized with the size of the smaller node, and resulting in a coefficient between 0 and 1). We also deleted all links with values of 0.1 and below in order to focus our discussion on typical relationships. 9. The Newman clustering algorithm attempts to optimize modularity in a particular network by creating clusters in a way that maximizes internal links between cluster members and minimizes external links (for technical details, see, for instance, Newman, 2006). 10. We refer to topoi using capitalization (idea elements in italics; discoursecarrying dimensions and polar categories in square brackets). 11. Note that the frequencies of the two polarities of a dimension do not necessarily add up to the frequency of the dimension, as individual images can contain both polarities for two reasons: First, a single idea element can integrate both polarities (e.g., future and past); second, images were usually coded with more than one idea element, and their combination could result in both polarities being represented. 12. Links to images illustrating the respective clusters can be found in the Appendix. 13. Burke (1989) presents a quite similar spatial metaphor when conceptualizing loci of motives that place the object of definition in contexts of various scopes, and that have a corresponding effect upon its interpretation. 14. We define, as a ‘‘core’’ idea element, one that expressed polar categories only within one specific topos. In the terminology of Gamson and Lasch (1983), these are referred to as ‘‘signature elements,’’ providing a ‘‘shortcut’’ to the topos (see also Meyer, 2004).

ACKNOWLEDGMENTS We wish to express our gratitude to several colleagues who provided valuable comments and helpful suggestions. We are especially indebted to Vitaliano Barberio, Eva Boxenbaum, Christof Brandtner, Maria Grafstro¨m, Johanna Hofbauer, Stefan Jonsson, Martin Kornberger, Mike Lounsbury, Manfred Lueger, Michael Meyer, Ryan Raffaeli, Vaughn Tan, Eero Vaara, Sam Warren, and Johanna Winter. Omissions and errors are entirely our own. The authors acknowledge support from the Austrian Science Fund (FWF): I 635-G17.

Imageries of Corporate Social Responsibility

167

REFERENCES Bell, E., Schroeder, J. E., & Warren, S. (Eds.). (2013). The Routledge companion to visual organization. London: Routledge. Benford, R. D., & Snow, D. A. (2000). Framing processes and social movements: An overview and assessment. Annual Review of Sociology, 26, 611–639. Benschop, Y., & Meihuizen, H. E. (2002). Keeping up gendered appearances: Representations of gender in financial annual reports. Accounting, Auditing & Accountability Journal, 27(7), 611–636. Berger, J. (1972). Ways of seeing. London: BBC and Penguin. Berger, P. L., & Luckmann, T. (1967). The social construction of reality: A treatise in the sociology of knowledge. New York, NY: Anchor Books. Boehm, G. (1994). Die Wiederkehr der Bilder. In G. Boehm (Ed.), Was ist ein Bild? (pp. 11–38). Mu¨nchen: Fink. Boxenbaum, E. (2006). Lost in translation: The making of Danish diversity management. American Behavioral Scientist, 49(7), 939–948. Brammer, S. J., Jackson, G., & Matten, D. (2012). Corporate social responsibility and institutional theory: New perspectives on private governance. Socio-Economic Review, 10(1), 3–28. Brammer, S. J., & Pavelin, S. (2006). Corporate reputation and social performance: The importance of fit. Journal of Management Studies, 43(3), 435–455. Breitbarth, T., Harris, P., & Insch, A. (2010). Pictures at an exhibition revisited: Reflections on a typology of images used in the construction of corporate social responsibility and sustainability in non-financial corporate reporting. Journal of Public Affairs, 10(4), 238–257. Bublitz, R. (2011). Differenz und Integration. Zur diskursanalytischen Rekonstruktion der Regelstrukturen sozialer Wirklichkeit. In R. Keller, A. Hirseland, W. Schneider & W. Vieho¨ver (Eds.), Handbuch sozialwissenschaftliche Diskursanalyse. Band 1: Theorien und Methoden (3rd ed., pp. 245–282). Wiesbaden: VS Verlag. Burke, K. (1989). On symbols and society. Chicago, IL: University of Chicago Press. Campbell, D., McPhail, K., & Slack, R. (2009). Face work in annual reports: A study of the management of encounter through annual reports, informed by Levinas and Bauman. Accounting, Auditing & Accountability Journal, 22(6), 907–932. Campbell, J. L. (2004). Institutional change and globalization. Princeton, NJ: Princeton University Press. Carroll, A. B. (1991). The pyramid of corporate social responsibility: Toward the moral management of organizational stakeholders. Business Horizons, July–August, 39–48. Carroll, A. B. (1999). Corporate social responsibility: Evolution of a definitional construct. Business & Society, 38(3), 268–295. Carruthers, B. G., & Espeland, W. N. (1991). Accounting for rationality: Double-entry bookkeeping and the rhetoric of economic rationality. American Journal of Sociology, 97(1), 31–69. Cornelissen, J. P. (2005). Beyond compare: Metaphor in organization theory. Academy of Management Review, 30(4), 751–764. Crane, A., McWilliams, A., Matten, D., Moon, J., & Siegel, D. S. (Eds.). (2008). The Oxford handbook of corporate social responsibility. New York, NY: Oxford University Press.

168

MARKUS A. HO¨LLERER ET AL.

Creed, W. E. D., Scully, M. A., & Austin, J. R. (2002). Clothes make the person? The tailoring of legitimating accounts and the social construction of identity. Organization Science, 13(5), 475–496. Czarniawska, B., & Joerges, B. (1996). Travel of ideas. In B. Czarniawska & G. Sevo´n (Eds.), Translating organizational change (pp. 13–48). Berlin: de Gruyter. Czarniawska, B., & Sevo´n, G. (Eds.). (1996). Translating organizational change. Berlin: de Gruyter. Davison, J. (2008). Rhetoric, repetition, reporting and the ‘‘dot.com’’ era: Words, pictures, intangibles. Accounting, Auditing & Accountability Journal, 21(6), 791–826. Davison, J., & Warren, S. (2009). Imag[in]ing accounting and accountability. Accounting, Auditing & Accountability Journal, 22(6), 845–857. den Hond, F., de Bakker, F. G. A., & Neergard, P. (Eds.). (2007). Managing corporate social responsibility in action: Talking, doing, measuring. Aldershot: Ashgate. Diaz-Bone, R. (2010). Kulturwelt, Diskurs und Lebensstil (2nd ed.). Wiesbaden: VS Verlag. Drori, G. S., Delmestri, G., & Oberg, A. (2013). Branding the university: Relational strategy of identity construction in a competitive field. In L. Engwall & P. Scott (Eds.), Trust in higher education institutions (pp. 134–147). London: Portland Press. Drori, G. S., Ho¨llerer, M. A., & Walgenbach, P. (2013). The glocalization of organization and management: Issues, dimensions, and themes. In G. S. Drori, M. A. Ho¨llerer & P. Walgenbach (Eds.), Global themes and local variations in organization and management: Perspectives on glocalization (pp. 3–23). New York, NY: Routledge. Drucker, P. F. (2007). The practice of management. Oxfort: Butterworth-Heinemann. Dunn, M. B., & Jones, C. (2010). Institutional logics and institutional pluralism: The contestation of care and science logics in medical education, 1967–2005. Administrative Science Quarterly, 55(1), 114–149. Fellmann, F. (1995). Innere Bilder im Licht des imagic turn. In K. Sachs-Hombach (Ed.), Bilder im Geiste: Zur kognitiven und erkenntnistheoretischen Funktion piktorialer Repra¨sentationen (pp. 21–38). Amsterdam: Rodopi. Frenkel, M. (2005). The politics of translation: How state level political relations affect the cross-national travel of ideas. Organization, 12(2), 275–301. Friedland, R. (2009). Institution, practice, and ontology: Toward a religious sociology. Research in the Sociology of Organizations, 27, 45–83. Friedland, R., & Alford, R. R. (1991). Bringing society back in: Symbols, practices, and institutional contradictions. In W. W. Powell & P. J. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 232–263). Chicago, IL: University of Chicago Press. Froschauer, U., & Lueger, M. (2003). Das qualitative Interview: Zur Praxis interpretativer Analyse sozialer Systeme. Wien: WUV. Gamson, W. A. (1992). Talking politics. Cambridge, MA: Cambridge University Press. Gamson, W. A., & Lasch, K. E. (1983). The political culture of social welfare policy. In S. E. Spiro & E. Yuchtman-Yaar (Eds.), Evaluating the welfare state: Social and political perspectives (pp. 397–415). New York, NY: Academic Press. Gamson, W. A., & Modigliani, A. (1989). Media discourse and public opinion on nuclear power: A constructionist approach. American Journal of Sociology, 95(1), 1–37. Goodrick, E., & Reay, T. (2011). Constellations of institutional logics. Work and Occupations, 38(3), 372–416.

Imageries of Corporate Social Responsibility

169

Gourevitch, P. A., & Shinn, J. (2007). Political power and corporate control: The new global politics of corporate governance. Princeton, NJ: Princeton University Press. Grant, D., Hardy, C., Oswick, C., & Putnam, L. (2004). Organizational discourse: Exploring the field. In D. Grant, C. Hardy, C. Oswick & L. Putnam (Eds.), The Sage handbook of organizational discourse (pp. 1–36). London: Sage. Graves, O. F., Flesher, D. L., & Jordan, R. E. (1996). Pictures and the bottom line: The television epistemology of U.S. annual reports. Accounting, Organizations and Society, 21(1), 57–88. Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E. R., & Lounsbury, M. (2011). Institutional complexity and organizational responses. Academy of Management Annals, 5, 317–371. Greimas, A. J. (1983). Structural semantics: An attempt at method. Lincoln, NE: University of Nebraska Press. Grue, J. (2009). Critical discourse analysis, topoi and mystification: Disability policy documents from a Norwegian NGO. Discourse Studies, 11(3), 305–328. Hallett, T., & Ventresca, M. J. (2006). Inhabited institutions: Social interactions and organizational forms in Gouldner’s patterns of industrial bureaucracy. Theory and Society, 35(2), 213–236. Hiss, S. (2009). From implicit to explicit corporate social responsibility: Institutional change as a fight for myths. Business Ethics Quarterly, 19(3), 433–451. Ho¨llerer, M. A. (2012). Between creed, rhetoric fac- ade, and disregard. Dissemination and theorization of corporate social responsibility in Austria. Frankfurt am Main: Peter Lang. Ho¨llerer, M. A. (forthcoming). From taken-for-granted to explicit commitment: The rise of CSR in a corporatist country. Journal of Management Studies, doi: 10.1111/joms.12029 Jackson, G., & Apostolakou, A. (2010). Corporate social responsibility in Western Europe: An institutional mirror or substitute? Journal of Business Ethics, 94(3), 371–394. Jancsary, D. (2013). Die rhetorische Konstruktion von Fu¨hrung und Steuerung: Eine argumentationsanalytische Untersuchung deutschsprachiger Fu¨hrungsgrundsa¨tze. Frankfurt am Main: Peter Lang. Jancsary, D., Ho¨llerer, M. A., Meyer, R. E., & Vettori, O. (2011). Framing images: Visually embodied meaning and its reconstruction on the field level. Working paper. WU Vienna University of Economics and Business. Jones, C., Maoret, M., Massa, F. G., & Svejenova, S. (2011). Rebels with a cause: Formation, contestation, and expansion of the de novo category ‘‘Modern Architecture,’’ 1870–1975. Organization Science, 23(6), 1523–1545. Kamla, R., & Roberts, C. (2010). The global and the local: Arabian Gulf states and imagery in annual reports. Accounting, Auditing & Accountability Journal, 23(4), 449–481. Kienpointner, M. (1992). Alltagslogik. Struktur und Funktion von Argumentationsmustern. Stuttgart: frommann-holzboog. Kinderman, D. (2009). Why do some countries get CSR sooner, and in greater quantity, than others? The political economy of corporate responsibility and the rise of market liberalism across the OECD: 1977–2007. Discussion paper. WZB Wissenschaftszentrum Berlin fu¨r Sozialforschung. Kinderman, D. (2012). ‘‘Free us up so we can be responsible!’’: The co-evolution of corporate social responsibility and neo-liberalism in the UK, 1977–2010. Socio-Economic Review, 10(1), 29–57. Kopperschmidt, J. (1989). Methodik der Argumentationsanalyse. Stuttgart: frommann-holzboog.

170

MARKUS A. HO¨LLERER ET AL.

Kress, G., & van Leeuwen, T. (2001). Multimodal discourse. The modes and media of contemporary communication. London: Hodder Education. Kress, G., & van Leeuwen, T. (2006). Reading images: The grammar of visual design (2nd ed.). London: Routledge. Lawrence, T. B., & Suddaby, R. (2006). Institutions and institutional work. In S. R. Clegg, C. Hardy, T. B. Lawrence & W. R. Nord (Eds.), The Sage Handbook of organization studies (pp. 215–254). London: Sage. Link, J. (1997). Versuch u¨ber den Normalismus. Opladen: Westdeutscher Verlag. Loewenstein, J., Ocasio, W., & Jones, C. (2012). Vocabularies and vocabulary structure: A new approach linking categories, practices, and institutions. Academy of Management Annals, 6, 1–46. Lueger, M. (2010). Interpretative Sozialforschung: Die Methoden. Wien: WUV. Lueger, M., Sandner, K., Meyer, R. E., & Hammerschmid, G. (2005). Contextualizing influence activities: An objective hermeneutical approach. Organization Studies, 26(8), 1145–1168. Maar, C., & Burda, H. (Eds.). (2004). Iconic Turn. Die neue Macht der Bilder. Ko¨ln: DuMont. Machin, D. (2004). Building the world’s visual language: The increasing global importance of image banks in corporate media. Visual Communication, 3(3), 316–336. Margolis, E., & Pauwels, L. (Eds.). (2011). The Sage handbook of visual research methods. Los Angeles, CA: Sage. Matten, D., & Moon, J. (2008). ‘‘Implicit’’ and ‘‘explicit’’ CSR: A conceptual framework for a comparative understanding of corporate social responsibility. Academy of Management Review, 33(2), 404–424. McQuarrie, E. F., & Phillips, B. J. (2005). Indirect persuasion in advertising: How consumers process metaphors presented in pictures. Journal of Advertising, 34(2), 7–20. Meyer, R. E. (2004). Globale Managementkonzepte und lokaler Kontext: Organisationale Wertorientierung im o¨sterreichischen o¨ffentlichen Diskurs. Wien: WUV. Meyer, R. E. (2008). New sociology of knowledge: Historical legacy and contributions to current debates in institutional research. In R. Greenwood, C. Oliver, K. SahlinAndersson & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 519–538). Thousand Oaks: Sage. Meyer, R. E. (2013). ‘‘Re-localization’’ as micro-mobilization of consent and legitimacy. In G. S. Drori, M. A. Ho¨llerer & P. Walgenbach (Eds.), Global themes and local variations in organization and management: Perspectives on glocalization (pp. 80–90). New York, NY: Routledge. Meyer, R. E., & Ho¨llerer, M. A. (2010). Meaning structures in a contested issue field: A topographic map of shareholder value in Austria. Academy of Management Journal, 53(6), 1241–1262. Meyer, R. E., & Ho¨llerer, M. A. (2011). Laying a smoke screen: Obfuscation and ambiguity as organizational responses to institutional complexity. Working paper. WU Vienna University of Economics and Business. Meyer, R. E., Ho¨llerer, M. A., Jancsary, D., & van Leeuwen, T. (2013). The visual dimension in organizing, organization, and organization research: Core ideas, current developments, and promising avenues. Academy of Management Annals, 7, 487–553. Mills, C. W. (1940). Situated actions and vocabularies of motive. American Sociological Review, 5(6), 904–913. Mitchell, W. J. T. (1994). Picture theory. Essays on verbal and visual representation. Chicago, IL: University of Chicago Press.

Imageries of Corporate Social Responsibility

171

Mu¨ller-Doohm, S. (1997). Bildinterpretation als struktural-hermeneutische Symbolanalyse. In R. Hitzler & A. Honer (Eds.), Sozialwissenschaftliche Hermeneutik (pp. 81–108). Opladen: Leske + Budrich. Newman, M. E. J. (2006). Modularity and community structure in networks. Proceedings of the National Academy of Sciences, 103(23), 8577–8582. Nigam, A., & Ocasio, W. (2010). Event attention, environmental sensemaking, and change in institutional logics: An inductive analysis of the effects of public attention to Clinton’s health care reform initiative. Organization Science, 21(4), 823–841. Phillips, B. J. (2000). The impact of verbal anchoring on consumer response to image ads. Journal of Advertising, 29(1), 15–24. Phillips, N., Lawrence, T. B., & Hardy, C. (2004). Discourse and institutions. Academy of Management Review, 29(4), 635–652. Phillips, N., & Oswick, C. (2012). Organizational discourse: Domains, debates, and directions. Academy of Management Annals, 6, 435–481. Powell, W. W., & Colyvas, J. A. (2008). The microfoundations of institutions. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 276–298). Thousand Oaks: Sage. Preston, A. M., Wright, C., & Young, J. J. (1996). Imag[in]ing annual reports. Accounting, Organizations and Society, 21(1), 113–137. Quattrone, P. (2009). Books to be practiced: Memory, the power of the visual, and the success of accounting. Accounting, Organizations and Society, 34(1), 85–118. Raab, J. (2008). Visuelle Wissenssoziologie: Theoretische Konzeption und materiale Analysen. Konstanz: UVK. Ray, J. L., & Smith, A. D. (2012). Using photographs to research organizations: Evidence, considerations, and application in a field study. Organizational Research Methods, 15(2), 288–315. Rose, G. (2007). Visual methodologies: An introduction to the interpretation of visual materials (2nd ed.). Los Angeles: Sage. Rubinelli, S. (2006). The ancient argumentative game: To´poı and loci in action. Argumentation, 20(3), 253–272. Sahlin, K., & Wedlin, L. (2008). Circulating ideas: Imitation, translation and editing. In R. Greenwood, C. Oliver, K. Sahlin-Andersson & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 218–242). Thousand Oaks: Sage. Sahlin-Andersson, K. (1996). Imitating by editing success: The construction of organizational fields. In B. Czarniawska & G. Sevo´n (Eds.), Translating organizational change (pp. 69–92). Berlin: de Gruyter. Schu¨tz, A., & Luckmann, T. (1973). The structures of the life-world. Evanston, IL: Northwestern University Press. Simpson, L. L. (2000). The annual report: An exercise in ignorance? Accounting Forum, 24(3), 231–247. Snow, D. A., Rochford, E. B., Jr., Worden, S. K., & Benford, R. D. (1986). Frame alignment processes, micromobilization and movement participation. American Sociological Review, 51(4), 464–481. Strang, D., & Meyer, J. W. (1993). Institutional conditions for diffusion. Theory and Society, 22(4), 487–511. Styhre, A. (2010). Visual culture in organizations: Theory and cases. London: Routledge.

172

MARKUS A. HO¨LLERER ET AL.

Thompson, L. J. (2008). Gender equity and corporate social responsibility in a post-feminist era. Business Ethics: A European Review, 17(1), 87–106. Thornton, P. H., & Ocasio, W. (2008). Institutional logics. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 99–129). Thousand Oaks: Sage. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure and process. Oxford: Oxford University Press. Titscher, S., Wodak, R., Meyer, M., & Vetter, E. (1998). Methoden der Textanalyse: Leitfaden und U¨berblick. Opladen: Westdeutscher Verlag. Vaara, E., Tienari, J., & Laurila, J. (2006). Pulp and paper fiction: On the discursive legitimation of global industrial restructuring. Organization Studies, 27(6), 789–813. Vogel, D. (2005). The market for virtue: The potential and limits of corporate social responsibility. Washington, DC: Brookings Institution Press. Wengeler, M. (2003). Topos und Diskurs: Begru¨ndung einer argumentationsanalytischen Methode und ihre Anwendung auf den Migrationsdiskurs (1960–1985). Tu¨bingen: Niemeyer. Zilber, T. B. (2006). The work of the symbolic in institutional processes: Translations of rational myths in Israeli high tech. Academy of Management Journal, 49(2), 281–303.

173

Imageries of Corporate Social Responsibility

APPENDIX: Topoi and Illustrative Images. Topos

Image and Source

Description of Image

OMV, Sustainability Report 2007/08: 33 http://www.omv.com/portal/01/com/omv/ OMVgroup/sustainability/reports OMV, Performance Report 2003/04: 83 http://www.omv.com/portal/01/com/omv/ OMVgroup/sustainability/reports

Petroleum refinery at night

Progress

VKW, Nachhaltigkeitsbericht 2008: 56 http://www.vkw.at/downloads/at/ nachhaltigkeitsbericht_illwerkevkw_2008.pdf EVN, Nachhaltigkeitsbericht 2006/07: 38 http://www.evn.at/getattachment/cd31f8a89d2d-4205-9088-df7b26c1b936/nhb-06_07.aspx

Traditional and innovative forms of hydro power User-friendly and easyto-use technology

Local Community

Lenzing, Nachhaltigkeit in der Lenzing Gruppe 2003: 4 http://www.lenzing.com/fileadmin/template/ flashbooks/Nachhaltigkeit_Broschuere_ 2003_DE/pubData/source/Nachhaltigkeit_ Broschuere_2003_DE.pdf Telekom Austria, Nachhaltigkeitsbericht 2004: 28–29 http://www.telekomaustria.com/verantwortung/ Archiv/nachhaltigkeitsbericht-2004.pdf

The board of directors in front of the headquarters and production site in rural Austria

OMV, Performance Report 2001/02: 30–31 http://www.omv.com/portal/01/com/omv/ OMVgroup/sustainability/reports

Human rights as a global responsibility of the multinational corporation An OMV employee salutes a local in Pakistan

Mastery

Globalism

OMV, Sustainability Report 2007/08: 1 http://www.omv.com/portal/01/com/omv/ OMVgroup/sustainability/reports Values

Enterprise

Engineer checking gauges

The company and its products as an integral part of local communities

Lenzing, Nachhaltigkeit in der Lenzing Gruppe 2008: 38 http://www.lenzing.com/fileadmin/template/ flashbooks/Nachhaltigkeit_Broschuere_2008_ DE/pubData/source/Nachhaltigkeit_ Broschuere_2008_DE.pdf Telekom Austria, Nachhaltigkeitsbericht 2002: 11 http://www.telekomaustria.com/verantwortung/ Archiv/nachhaltigkeitsbericht-2002.pdf

Appreciation of natural beauty and sanctity

OMV, Performance Report 2005/06: 38–39 http://www.omv.com/portal/01/com/omv/ OMVgroup/sustainability/reports

R&D as exploring unknown shores

Kids at play – ‘‘nostalgic optimism’’

174

MARKUS A. HO¨LLERER ET AL.

APPENDIX (Continued ) Topos

Credibility

Image and Source

Description of Image

Verbund, Nachhaltigkeitsbericht 2006: 63 http://www.verbund.com/B/media/ 0BD4E117AA334735ABE8358C9F6D3BC1.pdf

Taking unusual/ alternative perspectives

Verbund, Nachhaltigkeitsbericht 2008: 2 http://www.verbund.com/B/media/ D2F86D6686B04810A769E2115830B613.pdf

The board of directors as the ‘‘institutional face’’ of the corporation The CEO being interviewed by Paul Lendvai, a renowned Austrian journalist

OMV, Performance Report 2001/02: 4 http://www.omv.com/portal/01/com/omv/ OMVgroup/sustainability/reports

LOGIC PLURALISM, ORGANIZATIONAL DESIGN, AND PRACTICE ADOPTION: THE STRUCTURAL EMBEDDEDNESS OF CSR PROGRAMS Mary Ann Glynn and Ryan Raffaelli ABSTRACT The institutional logics perspective highlights how organizations are embedded within broader systems of meaning and how this embeddedness activates salient institutional logics in organizations that can enable or constrain organizational decisions, practices, and actions. We investigate a core premise of the institutional logics perspective, that of the alignment of institutional logics and organizational practices and design, in the organizational adoption of Corporate Social Responsibility (CSR) practices. We hypothesize that, in the adoption of practices, organizations will house those practices in structural units that align with the logic emphasized by the practice: when adopting practices reflecting a market logic, organizations will locate them in mainline business units, such as marketing; conversely, when adopting practices reflecting a community

Institutional Logics in Action, Part B Research in the Sociology of Organizations, Volume 39B, 175–197 Copyright r 2013 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0733-558X/doi:10.1108/S0733-558X(2013)0039A&B019

175

176

MARY ANN GLYNN AND RYAN RAFFAELLI

logic, organizations will locate them in non-mainline business units, such as corporate or philanthropic foundations. Using survey and archival data from 161 Fortune 500 (F500) firms, we find support for our hypotheses. Our findings reveal how institutional logics serve as underlying lynchpins, connecting organizational practices to organizational design so as to reinforce and enable each other. Keywords: Institutional logics; practice adoption; organizational design; corporate social responsibility

The institutional logics perspective is important for illuminating how organizations are embedded within broader systems of meaning and how this impacts organizational decisions, practices, and actions. This perspective is necessarily cross-level, highlighting the interplay between institutions (at a more macro level) and organizations (at a more micro level) and, particularly, how institutional logics penetrate organizations to enable or constrain their actions. Institutional logics arise in institutional orders at higher levels of analysis beyond the individual organization, but can infiltrate the individual organization through a variety of cognitive, normative, or structural mechanisms (Thornton, Ocasio, & Lounsbury, 2012). We investigate one of these mechanisms, the structuration of institutional logics in organizational design, an area in need of empirical inquiry, as Thornton, Ocasio and Lounsbury point out (2012, p. 181), ‘‘it would be interesting to study how institutional logics in a field become variously embedded in organizations via design.’’ Our empirical investigation examines how the adoption of organizational practices reflecting different institutional logics – the market logic or the community logic – are implanted in different units of the organizational design that reflect these logics. We investigate this linkage in the context of Fortune 500 (F500) firms adopting corporate social responsibility (CSR) practices. Simply put, ‘‘institutional logics represent frames of reference that condition actors’ choices for sense-making, the vocabulary they use to motivate action, and their sense of self and identity’’ (Thornton et al., 2012, p. 2). As cultural resources for organizations (Friedland & Alford, 1991), organizations ‘‘pull down’’ (Powell & Colyvas, 2008) institutional logics and embed them in their structures, which can become conduits for the adoption and implementation of organizational practices (Thornton et al., 2012, p. 81). Thus, organizational practices can render institutional logics transparent, and organizational structures that align with logic-based

The Structural Embeddedness of CSR Programs

177

practices can enable the implementation of these practices in organizations. Thornton et al. (2012) laid the groundwork for our investigation, by theorizing the linkages among institutional logics, organizational structures, and organizational practices. Our contribution is to delve into these relationships more deeply and put them on empirical footing. In our study, we show empirically how institutional logics align with organizational structures so as to enable the adoption of new practices within organizations and how this occurs even when multiple institutional logics are present. We investigate the organizational adoption of two prevalent CSR practices that differ in the extent to which they emphasize either a market logic or a community logic. A market logic, the most commonplace logic in for-profit organizations, emphasizes business ideals of profitability, efficiency, stakeholder accountability and economically motivated transactional behavior in a system of ‘‘market capitalism’’ (Friedland & Alford, 1991). By contrast, a community logic, a logic variant introduced by Thornton et al. (2012, pp. 68–72), emphasizes how organizations are more than economic actors; the community logic recognizes that organizations are also members of local, geographically bounded collectives that order activities by ideologies and value systems that are shared (Marquis, Glynn, & Davis, 2007) as ‘‘cooperative capitalism’’ (Thornton et al., 2012, p. 73). CSR practices can be based on either type of logic, but will differ in the extent to which they emphasize the rationale or benefits of CSR practices: Market-based practices emphasize the business benefits that can accrue to organizations for engaging in CSR practices, while community-based practices emphasize the social benefits that can accrue to organizations. We propose that, when organizations adopt CSR practices, organizations will seek structural alignment consistent with the institutional logic emphasized. Organizations will house CSR practices in structural units that align with the logic emphasized by the practice: those CSR practices that emphasize a market logic will be housed in organizational structures that are concerned with mainline business functions such as marketing, finance, strategy, or human resources; conversely, those practices that emphasize a community logic will be housed in organizational structures that are not aligned with a market logic but, rather, with the community logic of philanthropy or corporate foundations. Using survey and archival data from 161 F500 firms, we find empirical support for our proposition of logic structuration in organizational design: When adopting CSR practices, organizations align the logic emphasized by

178

MARY ANN GLYNN AND RYAN RAFFAELLI

the practice in structures that emphasized that same logic. Our findings reveal how institutional logics serve as underlying lynchpins, connecting organizational practices to organizational designs so as to align and thereby reinforce each other; in other words, we find that there is congruence between institutional logics and the organizational structures consistent with those practices. Our research affirms and exposes a key tenet of the institutional perspective, that of the macro (societal) and micro (organizational) linkages (Powell & Colyvas, 2008). We show how macro institutional logics furnish an appropriate setting within organizations that can supply ‘‘rules and tools’’ (Thornton et al., 2012, p. 81) that shape structures and practices. We extend this model to show how structures and practices themselves are linked logically and suggest that practices, once implemented in organizational structures that support them, are likely to feedback and reinforce existing logics. We begin with a theoretical overview of the institutional logics perspective, focusing on the interrelationships among logics, organizational design, and organizational practices. From this, we draw out hypotheses which we subsequently test empirically, finding support for our hypotheses. Finally, we discuss our findings, drawing out implications for organizational theory and management.

THEORETICAL FRAMEWORK: RELATING INSTITUTIONAL LOGICS TO ORGANIZATIONAL DESIGN AND PRACTICES Logics, Institutions, and Organizational Appropriation Drawing on a long line of thought in the social sciences, and particularly, neoinstitutional theory (e.g., DiMaggio & Powell, 1983; Friedland & Alford, 1991; Meyer & Rowan, 1977), the notion of institutional logics gained traction with empirical inquiry (Dunn & Jones, 2010; Glynn & Lounsbury, 2005; Lounsbury, 2007; Marquis & Lounsbury, 2007; Thornton, 2004; Thornton & Ocasio, 1999) and theoretical elaboration (Greenwood, Raynard, Kodeih, Micelotta, & Lounsbury, 2011; Thornton et al., 2012). Broadly speaking, Thornton et al. (2012, p. 2) view the institutional logics perspective as a meta-theoretical framework for ‘‘analyzing the interrelationships among institutions, individuals, and organizations in social

The Structural Embeddedness of CSR Programs

179

systems.’’ Importantly, their perspective on institutional logics redirects attention away from the core tenet of institutional theory – the organizational drive for isomorphism – and, instead, to the possibilities of organizational agency in appropriating and implementing different types of logics. Thus, the perspective allows for logic pluralism, that is, the coexisting of multiple, different logics within an organization, and the management of these by the organization. A key task, then, for organizations adopting new practices, is to align the logics of those practices with the logics of the organizational structures that implement then. Institutional logics capture wider belief systems and material practices in society that are associated with key institutions in society (Friedland & Alford, 1991; Thornton & Ocasio, 1999). Institutional logics can vary significantly between different societal sectors, because each sector has a central logic and organizing principles that affects the focus of cognition, social relations, material practices, and symbolic meanings. Thus, because of their association with different institutional orders, institutional logics can vary considerably across these sectors and in their implications for actors. Actors, including organizations, draw on these institutional logics as frames of reference for sense-making, vocabularies of motive and action, and a sense of self and identity (Thornton et al., 2012). Organizational scholars have shown that institutional logics operate in organizational contexts as diverse as symphony orchestras (Glynn & Lounsbury, 2005) and food societies (Rao, Monin, & Durand, 2003), thereby suggesting the suppleness of logics in organizational appropriation and the variance in their application in organizations. Friedland and Alford (1991) identified five institutions with distinctive and associated logics: family, religion, state, market, and corporation. To this, Thornton et al. (2012) add two: community and profession. Although organizations may appropriate any of these ideal types of logics, we focus on two logics – the market logic and the community logic – because both are important to, and definitive of, the practice we study, CSR programs. Marquis et al. (2007) define CSR (or, corporate social action, as they label it) as: ‘‘those firm behaviors and practices that extend beyond immediate profit maximization goals and are intended to increase social benefits or mitigate social problems for constituencies external to the firm.’’ Thus, CSR practices seem attuned to both a business and social concern, particularly for those external to the firm. Evidence has shown that a firm’s local geographic communities, where important external constituencies reside, are important influences on CSR practices, a notion reinforced by other

180

MARY ANN GLYNN AND RYAN RAFFAELLI

researchers (Galaskiewicz, 1991, 1997; Galaskiewicz & Burt, 1991; Guthrie, 2003). In general, then, CSR practices seem to invoke both market-based logics (‘‘profit maximization goals’’) and community-based logics (‘‘social benefits’’). The market logic is concerned with the profitability and efficiency of the business; competitive advantage and stakeholder responsibility; and the transactional basis of organizational behavior; simply put, it is described as ‘‘market capitalism’’ (Thornton et al., 2012, p. 73). The community logic is focused less on the economic health of the firm and more on its social fitness, as a member of a geographically based locale that emphasizes the ideology and value systems shared by the community and order economic activity (Marquis et al., 2007); simply put, it is described as ‘‘cooperative capitalism’’ (Thornton et al., 2012, p. 73). In many ways, these two ideal types of logics – market and community – seem orthogonal and yet, organizational practices that are socially responsible necessarily incorporate both. This is evident, for instance, in corporations’ effects on nonprofit growth in American communities (Navis & Glynn, 2011) and in hybrid firms that combine both business principles and ideological values (Battilana & Dorado, 2010). More generally, the coexistence of multiple logics in organizations seems to parallel concerns raised by researchers investigating multiple or hybrid identities in organizations (e.g., Glynn, 2000; Golden-Biddle & Rao, 1997; Pedersen & Dobbin, 2006; Pratt & Foreman, 2000). This is not surprising, for logics link not only to material practices but also to organizational identities (Thornton et al., 2012). And yet, although all CSR practices may incorporate both market and community logics, they can vary in the extent to which they emphasize each of these, with some practice variants focusing on business benefits and others on social benefits. Next, we turn to understanding the relationship between institutional logics and organizational practices broadly, and then the case of CSR practices more specifically.

Institutional Logics, Organizational Practices, and Identity Thornton et al. (2012, p. 128) observe that ‘‘from the inception of the institutional logics perspective, the concepts of practice and identity have been integral y (but) most research to date has not effectively analyzed how institutional logics shape and are shaped by the material instantiations of logics – the practices and identities of concrete actors.’’ This is the gap we seek to fill with our empirical investigation.

The Structural Embeddedness of CSR Programs

181

Practice ‘‘refers to forms or constellations of socially meaningful activity that are relatively coherent and established y [and] that are informed by wider cultural beliefs’’ (Thornton et al., 2012, p. 128). Logics find their expression in practices, which provide a link to identities, ‘‘since many social actors strive for consistency with regard to how practices articulate core organizational identity beliefs’’ (Thornton et al., 2012, p. 136). Glynn (2000) presents an illustration of this practice-identity coupling in her study of the musicians’ strike at the Atlanta Symphony Orchestra. She notes how, during the strike when orchestral performances had been cancelled, musicians took to the streets of the city, spontaneously playing concerts for free. The musicians explained: ‘‘We’re musicians and we need to play’’ (Glynn, 2000, p. 290). The practice of playing helped the musicians to maintain and reinforce their professional identities when that identity was threatened; implicitly, too, continuing the practice of symphonic performance also reinforced the identity of the orchestra when it was nearly invisible. Glynn’s study also illustrates how different practices – musical and administrative – were held by the different roles organizational members enacted (musicians and administrative) and housed in different elements of the organizational design (front stage and back stage). Practices make evident the features of institutional logics, as they link organizational action to institutional fields (Greenwood et al., 2011; Guler, Guille´n, & Macpherson, 2002; Westphal, Gulati, & Shortell, 1997). Likewise, CSR practices activate the salience of multiple institutional logics and can highlight the tensions between business and society (Margolis & Walsh, 2001; Margolis & Walsh, 2003). However, organizations engaged in CSR seem to be coalescing around several practices and activities to display the earmarks of such an institutional field (Aguilera, Rupp, Williams, & Ganapathi, 2007), including: professionalization in standards (Pinney & Lee, 2010), CSR company rankings (Corporate Responsibility Magazine, 2010), CSR training and development curricula (The Aspen Institute, 2012), global certification standards such as the ISO 26000 (International Organization for Standardization, 2010), and partnerships with academic centers, think tanks, and consultancies devoted to studying the role of corporations in society (for an overview of the emergence of CSR standards and practices, see Waddock, 2006). Importantly, institutional logics are intertwined not only with organizational practices, but also with organizational identity. Institutional logics furnish a frame for organizations to explain and justify two critical questions: ‘‘what we do’’ (practices) and ‘‘who we are’’ (identity). The answers to both questions define the organization’s identity (Navis & Glynn,

182

MARY ANN GLYNN AND RYAN RAFFAELLI

2010). Logics are critical in linking practices to identity to give a coherent understanding of the organization’s identity. Thus, while logics may shape an organization’s identity, the organization’s identity also shapes the institutional logics it appropriates and the design instantiated to implement the practices associated with those logics.

Institutional Logics and Organizational Design Thornton et al. (2012, p. 181) explain the critical role organizational design plays in the organizational appropriation of logics and associated practices: y Organizational design is important because it filters how institutional logics reach an organization and shapes whether pressures and motivations associated with different logics become encoded in diverse coalitions within an organization, creating or inhibiting conflict over goals and strategies.

In effect, organizational design consists of ‘‘the structures of accountability and responsibility used to develop and implement strategies, and the human resource practices and information and business processes that activate those structures’’ (Greenwood & Miller, 2010, p. 78). Organizational structures are themselves associated with different institutional logics. Market logics find a home in those structural units that are most concerned with efficiencies and the profit maximization goals of the firm; these include marketing and communications, human resources, and other business units charged with carrying out activities that impact profit and loss within the firm. Community logics find a home in those structural units that are most concerned with the local environment’s social needs or issues. One site that is especially and uniquely attuned to these concerns is that of corporate foundations. A corporate foundation is ‘‘a legally separate entity that exists to pursue the philanthropic activities of its parent company’’ (Marquis & Lee, 2013, p. 7). Foundations are typically structured as separate and independent entities, often with a unique board of governors, charters, and a group of professionals with specific skills suited for managing philanthropic endeavors. While the independent nature of corporate foundation creates clear structural differentiation from the corporate parent, ‘‘corporate foundations depend almost fully on the firm itself for future resources and therefore remain heavily influenced by it’’ (Marquis & Lee, 2013, p. 8).

The Structural Embeddedness of CSR Programs

183

Following this line of reasoning, we propose linkages among institutional logics and organizational practices such that both align in the design of the organizational structure. Thus, we hypothesize: H1. When CSR practices associated with a specific logic are adopted by an organization, they will be housed in those organizational infrastructures that are most closely associated with that logic. More specifically: H1a. When CSR practices perceived to have primarily a business benefit are adopted, organizations will house them in structures associated with core business functions (e.g., human resources, marketing, and communications). H1b. When CSR practices perceived to have primarily a social benefit are adopted, organizations will house them in structures outside the core business functions (e.g., corporate foundation).

RESEARCH METHODS Data Sources We gathered data from the administration of a survey and from public archives. These data are part of a larger data collection effort (Raffaelli & Glynn, forthcoming). Our survey was administered by the Boston College Center for Corporate Citizenship, a university center dedicated to CSR research and practice. The questionnaire was designed to gather information from community involvement professionals in organizations who were engaged in developing and implementing CSR practices. The particular CSR practice on which we focus is the employee volunteering practices (EVP), which is designed to encourage employees to volunteer their time and skills in service to the community (Wild, 1993). The survey data used herein focuses on the different types of employee volunteering practices adopted by the organizations, as well as where the practice is housed and managed within the organizational structure, the levels of integration with the business, and resource allocation. Respondents received a benchmarking report for their participation. Surveys were sent to the organization’s senior executive responsible for CSR initiatives. This survey is unique in querying companies from a wide range of industries, all of whom are F500 firms, and in offering a view of organizational adoption, implementation, and management of EVPs across

184

MARY ANN GLYNN AND RYAN RAFFAELLI

companies. The survey was initially piloted on fifteen F500 firms via phone interviews. We tried to ensure a high rate of response by contacting every firm via e-mail inviting their participation and following up by phone with those who did not respond initially. Our final sample consisted of 161 F500 firms, which reflected a 33% response rate. About two-thirds of the organizations in our sample (67%, n=107) reported that they housed the EVP in a mainline business department (e.g., marketing and communications, human resources, public relations), while about one-eighth housed EVP in the corporate foundation unit (12%, n=20). Less than one-tenth of the firms (8%, n=14) housed one or more components of EVP in both the mainline business and foundation units. The firms were from a diverse set of industries, including: services (16%), retail (16%), financial (15%), utilities and transportation (14%), manufacturing (12%), wholesale trade (7%), and other (20%). Initial analyses did not detect a significant association between industries and decisions about housing EVPs. In addition to the survey data, we used several archival sources to develop our empirical measures, as we describe below.

Dependent Variables We created two dummy variables, to capture whether the practice was housed in the firm’s (1) mainline division or (2) corporate foundation. The former was assessed by combining responses to separate questions that asked respondents whether the EVP was housed in: human resources, marketing and communications, communication relations, or other business unit. Affirmative responses were coded 1 and 0 otherwise. The latter was assessed by responses indicating that the practice was housed in the corporate foundation (1=yes, 0=no).

Independent Variables We developed two independent variables to measure whether CSR practices were perceived by the organization as contributing to a business benefit (consistent with a market logic) or a social benefit (consistent with a community logic). To measure organizational perceptions of a business benefit, and an emphasis on the market logic, a survey item asked if the CSR practice ‘‘had an explicit business connection’’ (1= yes, 0=no). To measure

The Structural Embeddedness of CSR Programs

185

organizational perceptions of a social benefit, we used a less explicit indicator because, by definition, CSR practices are intended to have social benefits. Based on the literature associating institutional logics with organizational identity (e.g., Glynn, 2000; Glynn & Abzug, 2002; Thornton et al., 2012), we chose to use an independent indicator of a firm’s identity which reflects an orientation toward the community logic and associated social benefits: organizational completion of the global reporting initiative (GRI) report (Etzion & Ferraro, 2010; Nikolaeva & Bicho, 2011). Using such a behaviorally anchored measure of organizational identity affirms the organization’s claim of, and commitment to, an identity that is community oriented. An organization’s identity makes certain types of logics more appealing and legitimate (Glynn, 2008) and sensitizes organizations to assimilate some types of logics over others (Greenwood et al., 2011, p. 319). As well, GRI reporting by an organization is recognized as an organizational identity claim to being a member of the community of socially responsible organizations (Nikolaeva & Bicho, 2011). We coded 1 for GRI reporting firms and 0 otherwise, using the online list of all companies that adopted the GRI (www.globalreporting.org), the most widely used reporting framework in the world. The GRI requires companies to demonstrate commitment to sustainability, community impact, human rights, and gender (GRI, 2011). Completing the GRI reflects support for ‘‘a sustainable global economy where organizations manage their economic, environmental, social and governance performance and impacts responsibly and report transparently’’ (GRI, 2012). Thus, it is a robust indicator of an organizational identity that was oriented to social benefits and functioned as a filter for practice adoption. Almost one-third of our sample completed a GRI report (29%, n=47).

Control Variables We included four control variables. First, regardless of where a practice is housed in an organization, we accounted for organizational infrastructure, which affects practice implementation (e.g., Cohen & Levinthal, 1990; Volberda, Foss, & Lyles, 2010). We developed a scale (Cronbach’s a = 0.72) consisting of four survey items asking respondents about the existence of written processes and standards for: (a) employees to contribute to CSR activities, (b) nonprofit board involvement, (c) managing asset allocation to nonprofits, and (d) facilitating agreements with nonprofits. Responses to each item were coded 1 (yes) or 0 (no) and then summed to index the extent

186

MARY ANN GLYNN AND RYAN RAFFAELLI

of the organizational infrastructure in place to implement an EVP practice; higher values indicated a more extensive infrastructure to implement CSR practices. Second, we assessed an organization’s status or elite standing in its field using the Fortune online database, coding 1 if the firm was a member of the Fortune 100 in 2008 (the time of the survey) and 0 otherwise. We included this control because it has been theorized to affect practice adoption (Greenwood et al., 2011). The third control is the organization’s geographic location. Because geography may play a role how CSR is perceived, implemented, and managed (e.g., Galaskiewicz, 1991; Marquis et al., 2007; Matten & Moon, 2008), and because geography may be associated with a particular CSR model (e.g., Anglo-American in the United States and Continental in Europe and Asia (Aguilera et al., 2007)), we coded 1 if the firm’s EVP practices were based only within the United States and 0 otherwise. Our fourth variable controls for industry; we coded each of firms in our sample using two-digit SIC codes (identified using the NAICS online database).

Analyses To test our hypotheses, we used STATA 12 and ran a series of logit regression analyses, separately modeling the two outcome variables (mainline and foundation). We ran several diagnostic tests to check the robustness of our models and to ensure that multicollinearity was not a threat. The mean variance inflation factor (VIF) across both models was less than 1.20, indicating that multicollinearity was not a serious concern.

RESULTS Table 1 presents the descriptive statistics and correlations for the variables in our models. Tables 2 and 3 present the tests of hypothesis using our two dependent variables: mainline and foundation, respectively. Each dependent variable has 4 models, with model 1 consisting only of the control variables, models 2–4 presenting the main effects for the variables. Models 2–3 examine each of the perceived benefits (business and social) individually (H); these are presented for informational purposes. Model 4 is the full model with all the main effects, which we use to test H1a and H1b.

187

The Structural Embeddedness of CSR Programs

Descriptive Statistics and Correlations.

Table 1. Variable 1 2 3 4 5 6 7 8

Mainline business unit Corporate foundation Business benefit Social benefit Industry Geography Infrastructure Elite field position

Table 2.

Mean

SD

1

2

3

4

5

6

7

0.67 0.12 0.32 0.29 5.35 0.34 0.82 0.11

0.47 0.33 0.47 0.46 3.10 0.47 1.19 0.31

1.00 0.03 0.38 0.20 0.13 0.45 0.36 0.12

1.00 0.06 0.17 0.10 0.05 0.04 0.18

1.00 0.23 0.03 0.21 0.50 0.11

1.00 0.14 0.04 0.15 0.14

1.00 0.10 0.08 –0.08

1.00 0.02 0.06

1.00 0.19

Logit Regression Models Predicting EVP Housed in Mainline Business.

Variable

(1)

(2)

Perceived benefit of practice Social Benefit

0.84 (0.50)

Business Benefit Controls Industry Geography Infrastructure Elite field position Constant Observations LR ratio w2 ProbW w2 Pseudo R2

(3)

(4)

1.44 (0.64)

0.72 (0.52) 1.33 (0.65)

0.03 (0.07) 3.36 (0.76) 1.03 (0.27) 0.15 (0.78) 0.79 (0.41)

0.04 (0.07) 3.27 (0.77) 0.83 (0.28) 0.07 (0.82) 0.95 (0.43)

0.03 (0.07) 3.37 (0.77) 1 01 (0.28) 0.03 (0.80) 0.94 (0.43)

0.03 (0.07) 3.27 (0.77) 0.84 (0.28) 0.19 (0.83) 1.07 (0.44)

161.00 67.66 0.00 0.33

161.00 73.24 0.00 0.36

161.00 70.53 0.00 0.34

161.00 75.19 0.00 0.37

Coefficients are reported with standard errors in parentheses. Tests are two-tailed. po0.05; po0.01; po0.001.

Tests of Hypothesis Our hypothesis, predicting that practices associated with a specific logic, will be housed in organizational structures that align with that logic, was

188

Table 3.

MARY ANN GLYNN AND RYAN RAFFAELLI

Logit Regression Models Predicting EVP Housed in Corporate Foundation.

Variable

(1)

Perceived benefit of practice Social benefit

(2)

0.89 (0.66)

Business benefit Controls Industry Geography Infrastructure Elite field position Constant Observations LR ratio w2 ProbW w2 Pseudo R2

0.09 (0.08) 0.34 (0.51) 0.03 (0.21) 1.16 (0.63) 1.83 (0.50) 161 5.69 0.22 0.05

0.10 (0.08) 0.48 (0.53) 0.20 (0.24) 1.15 (0.64) 1.73 (0.50) 161 7.64 0.18 0.06

(3)

1.11 (0.52) 0.13 (0.08) 0.35 (0.52) 0.01 (0.21) 1.00 (0.64) 2.03 (0.51) 161 10.21 0.07 0.08

(4) 1.31 (0.54) 1.21 (0.70) 0.14 (0.08) 0.54 (0.54) 0.26 (0.25) 0.99 (0.65) 1.98 (0.52) 161 13.56 0.04 0.11

Coefficients are reported with standard errors in parentheses. Tests are two-tailed. po0.05; po0.001.

supported. H1a, predicting that practices perceived to carry a market logic are more likely to be housed in organizational structures related to the mainline business was supported (po0.05); see Model 4, the full model, in Table 2. H1b, predicting that practices perceived as carrying a community logic are more likely to be housed in structures related to the company’s foundation was also supported (po0.01); see Model 4, the full model in Table 3. Thus, we find general support for our hypothesis, predicting an alignment between the institutional logics associated with organizational practices adopted and the organizational units that house them. In order to investigate if our findings generalized beyond a single CSR practice, that is, EVPs, we conducted post hoc analyses. We repeated our analyses (described above) on a different type of CSR practice, employee giving programs (EGVs). Like EVPs, these programs provide opportunities for employees to contribute to the community; they have also been

The Structural Embeddedness of CSR Programs

189

shown to have potential positive effects for the company itself, such as access to new clients who may serve on nonprofit boards, positive exposure and reputation management, or increased brand awareness (Hess, Rogovsky, & Dunfee, 2002). EGVs typically take the form of charitable drives that allow employees to donate to one or more specific charities, often through payroll deduction (Weinger, 2012). Within our sample of 161 F500 firms, 58% (n=94) housed their giving practices in a mainline business unit and 24% (n=38) housed them in their corporate foundation. We found the same patterns for EGPs as we did for EVPs (regression analyses not reported herein, but available from the authors): EGVs perceived as carrying a business benefit (i.e., associated with the market logic) are more likely to be housed in structures relating to the mainline business units) (po0.05); alternatively, EGVs perceived as carrying a social benefit (i.e., associated with the community logic) are more likely to be housed in structures relating to a community orientation (e.g., corporate foundation) (po0.01). These findings suggest that the link among practices, structures, and logics extends beyond those practices that require employee engagement (i.e., EVPs); to those that are less demanding of employee time (EGPs); EGPs typically require only a once-a-year commitment to support a charity (e.g., United Way, Combined Federal Campaign). Overall, we found support for our core hypothesis: Organizational CSR practices are likely to be housed in those organizational structures that align most closely with the logic primarily associated with the practice. We also note an interesting pattern of findings associated with the four control variables (see Model 1 in Tables 2 and 3): Although an organization’s industry and elite standing was not significant for either dependent variable, two control variables – geography and infrastructure – were significant in predicting whether EVP practices would be housed in the mainline business, but not in the corporate foundation. Having EVP practices based only in the United States, and having a more extensive organizational infrastructure in place to implement an EVP practice, make it more likely that EVP practices with perceived business benefits will be housed in a mainline business unit in the organization. These results seem to suggest a strong sense of alignment among a number of organizational attributes – organizational infrastructure, geographic orientation, and mainline business units – in implementing an EVP practice perceived to have business benefits (and not simply social ones).

190

MARY ANN GLYNN AND RYAN RAFFAELLI

DISCUSSION AND CONCLUSIONS Our investigation into the association among institutional logics, organizational practices, and organizational design revealed an alignment among these constructs that has been theorized but unexamined empirically; our contribution is to offer evidence that affirm the significant relationships previously theorized. More generally, our findings attest to the role of institutional logics – and, especially the market logic and community logic, in the case of CSR practices – in affecting the adoption and implementation of new organizational practices. Our analyses of survey and archival data of 161 F500 firms revealed that, in adopting employee volunteering programs (EVPs), organizations tended to house the practices in the organizational design units perceived to be consistent with the institutional logics primarily associated with the practice. Our core hypothesis found support: When EVPs were seen as having a business benefit (associated with the market logic), organizations tended to house these practices in mainline business units, such as human resources, marketing and communications, communication relations or another business unit; conversely, however, when EVPs were seen as having a social benefit (associated with the community logic), organizations tended to house these practices in the corporate foundation unit of the organization. Broadly speaking, our findings point to how institutional logics serve as critical lynchpins around which organizational structures and practices cohere and align. Logics, it seems function as a central source of support and stability and a touchstone for relating practices to design. We depict these relationships in Fig. 1. Fig. 1 displays the conceptual model that is suggested by our findings and that we advance. We posit that institutional logics filter organizational perceptions of practices that they adopt in terms of the ideal types implied by the logics; for the CSR practices we studied, organizational perceptions differentially filtered the EVP practices in terms of a market logic (business benefit) or community logic (social benefit). In turn, institutional logics embed practices in the organizational structures most closely associated with them: mainline business functions for practices perceived to be marketoriented and corporate foundations for the practices perceived to be community oriented. As a result, institutional logics align organizational practices and structures in synergy. Interestingly, our qualitative data hint at a possibility of a more subtle alignment among practices, logics, and structure. For example, a respondent to our survey wrote that his firm had adopted volunteering and giving

191

The Structural Embeddedness of CSR Programs

align… Practices

Structures

Institutional Logics

Fig. 1.

Relationships among Institutional Logics, Practices, and Organizational Structures.

programs, noting ‘‘They are managed by different individuals who report to different supervisors, but are in the same department.’’ His comment reflects how, in this specific case, dual reporting structures housed within the same unit allowed for multiple logics to influence the practice. Our qualitative data allude to the possibility that some structures, in effect, may still allow for the comingling of logics within the framework we introduce in Fig. 1. Reinforcing this view of the logics-practice-structure coupling are the findings that two of the control variables – organizational infrastructure and geographic location – were significant for practices associated with the market logic but not the community logic. Consistent with other researchers (e.g., Galaskiewicz, 1991; Marquis et al., 2007; Matten & Moon, 2008), we found that geography played a role in how CSR practices are perceived, implemented, and managed, but only for practices perceived to have a business benefit. We can puzzle over a number of possible reasons for this. First, an organization whose CSR practices are housed only in the United States (versus global) may have a more singular view of CSR, that is, an Anglo-American CSR model (Aguilera et al., 2007); this homogeneity in perspective may have made alignment with the mainline business more straightforward or coherent. Corporations in the global context may be facing diverse CSR models that do not point to a clear structural placement for CSR practices. Second, institutional pressures arising from the geographic community in which the organization is embedded shape corporate social action such that organizations come to resemble each other

192

MARY ANN GLYNN AND RYAN RAFFAELLI

(Marquis et al., 2007). For our study, community isomorphism may be occurring on a national scale, such that, in the United States, corporations tend to locate EVP practices within mainline business. While our study did not set out to test these constructs directly, our findings suggest that it may be useful to examine the role of geography and infrastructure in future research. Similarly, we found that having an organizational structure with written processes and standards for employee engagement in CSR practices increased the likelihood that organizations would house CSR practices in their mainline business units. This extends previous research on practice implementation (e.g., Cohen & Levinthal, 1990; Volberda et al., 2010) but adds nuance. Moreover, the findings suggest the possibility that it may be an organization’s overall orientation to implementing CSR practices in design and structures that is what makes the infrastructure variable significant. We can speculate that institutional logics may permeate the organization to shape its structure so as to make it more receptive to practices that align with that logic; herein, we saw this alignment occurring for the market logic. Like all studies, and especially those initiating explorations into new theoretical frontiers, ours was limited in terms of the types of practices studied (CSR), the types of institutional logics studied (market vs community), and the kinds of firms sampled (the F500, which tend to be large and high status). Moreover, we modeled the organizational adoption of an EVP practice in terms of reflecting either the market or community logic primarily. Clearly, it is possible for organizations to adopt practices from both kinds of logics and to house them in both kinds of structures. In our sample, this was a relatively rare occurrence, as only 8% of our 161 F500 firms (n=14) housed one or more components of EVP in both the mainline business and foundation units. This might reflect the complexity of organizational design required for implementation or the hybrid orientation of the organization to both the market and community logics. Nonetheless our results would suggest that organizations are most likely to segregate practices structurally, partitioning them by design to align with the logics they represent. Our study makes several contributions to the existing literature on institutional logics and to the advancement of this literature going forward. First, we empirically demonstrate that institutional logics penetrate organizations through firm perceptions of practices and their adoption in organizational structures perceived to align with their ideal type. We show how different types of organizational mechanisms – cognitive filters and structural

The Structural Embeddedness of CSR Programs

193

implementation – can function as important filters on logics and as significant carriers of logics in the organizational adoption of CSR practices. Although we cannot make any causal inferences from our data, we can however speculate about the relationships between institutional logics and perception formation, organizational identity, and decision-making. In some ways, it may seem like a fairly straightforward relationship, that is, that institutional logics embedded in an organization may affect perceptions formed (e.g., seeing a practice as oriented to a market or community logic), the identities claimed (e.g., as a CSR firm), and the decisions made (e.g., implementing and funding a particular practice). But, we believe it is considerably more complicated. For one thing, institutional logics are often pluralistic in organizations, and different types of logics – market and community, for instance – can exist side by side. Similarly, organizations often have multiple identities (e.g., Pratt & Foreman, 2000) that may coexist synergistically or competitively and a pattern of decision-making over time that may or may not be clear in its rationale or implications. Like the hybridization of organizational identities (Albert & Whetten, 1985), logics too may be hybridized; as such, they are likely segmented by business units, professions or strategic intent within the organization. When problems, practices or decisions have a strong connection (or contestation) with a particular logic, that logic is likely activated and used to frame the concern. Second, our work suggests that although institutional logics can penetrate organizations to significant effect, organizations are not simply passive ‘‘dopes’’ in incorporating logics; rather, organizations were agentic in choosing how they construed practices in terms of logics and how they implemented them structurally. Our findings show that, although it was the same CSR practice adopted by organizations (i.e., EVPs or EGVs), these practices were perceived differentially by organizations in ways that related to their cognitive orientations and identity. Thus, our work suggests a reciprocity between organizations and institutional orders within the context of CSR, and builds upon previous work that has explored the interplay between organizational and field identity in settings such as elite French cuisine (Rao et al., 2003), orchestra strikes (Glynn, 2000), and institutional investors (Lok, 2010). Thus, as scholars have observed, identity serves as a salient filter for institutional complexity, especially when organizations select a structures to support practices imbued with multiple (and possibly conflicting) institutional logics (Glynn, 2000; Greenwood et al., 2011; Raffaelli & Glynn, forthcoming). Third, our findings, and their graphic depiction in Fig. 1, suggest several useful pathways forward in the study of institutional logics. The alignment

194

MARY ANN GLYNN AND RYAN RAFFAELLI

among logics, practices and structures raises the idea that logics may be configural in shaping organizational actions. Our work suggests a tight coupling among logics and their manifestation in practices and embodiment in structures that was not random; as such, perhaps logics can be thought of not as an independent vector of beliefs, norms or values but a configuring trigger that orients an organization in predictable ways. Finally, we suggest that the institutional logics perspective might fruitfully draw on other theoretical perspectives in the organizational literature to examine logics conjointly with other theoretical perspectives. Our work indicates some associations: logics and theories of organizational practice or materiality; logics and theories of organizational design; and logics and theories of organizational identity. We believe that there is probably a large universe of theoretical perspectives that might inform the institutional logics perspective and, in turn, which might be informed by the institutional logics perspective. Our work is a small step forward in this research agenda that, we hope, stimulates future thinking on institutional logics.

ACKNOWLEDGMENTS Direct all correspondence to Ryan Raffaelli, Harvard Business School, Soldiers Field Road, Boston, Massachusetts 02263, [email protected]. We are extremely grateful to the editors, Michael Lounsbury and Eva Boxembaum, as well as to the participants at the ABC Organizing Institutions Conference, for their comments. We also wish to thank the Boston College Center for Corporate Citizenship, especially Katherine Smith and Bea Boccalandro. This research has benefitted from the generous financial support of the Boston College Winston Center for Leadership and Ethics and the Joseph F. Cotter Professorship.

REFERENCES Aguilera, R., Rupp, D., Williams, C., & Ganapathi, J. (2007). Putting the S back in corporate social responsibility: A multilevel theory of social change in organizations. Academy of Management Review, 32(3), 836–863. Albert, S., & Whetten, D. A. (1985). Organizational identity. Research in Organizational Behavior, 7, 263–295. Battilana, J., & Dorado, S. (2010). Building sustainable hybrid organizations: The case of commercial microfinance organizations. The Academy of Management Journal, 53(6), 1419–1440.

The Structural Embeddedness of CSR Programs

195

Cohen, W. M., & Levinthal, D. A. (1990). Absorptive Capacity: A New Perspective on Learning and Innovation. Administrative Science Quarterly, 35(1), 128–152. Corporate Responsibility Magazine. (2010). CR’s 100 best corporate citizens 2011, Corporate Responsibility Magazine. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48(2), 147–160. Dunn, M. B., & Jones, C. (2010). Institutional logics and institutional pluralism: The contestation of care and science logics in medical education, 1967–2005. Administrative Science Quarterly, 55(1), 114–149. Etzion, D., & Ferraro, F. (2010). The role of analogy in the institutionalization of sustainability reporting. Organization Science, 21(5), 1092–1107. Friedland, R., & Alford, R. R. (1991). Bringing society back in: Symbols, practices and institutional contradictions. In W. W. Powell & P. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 232–263). Chicago, IL: University of Chicago Press. Galaskiewicz, J. (1991). Making corporate actors accountable: Institution-building in Minneapolis-St, Paul. In W. W. Powell & P. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 293–310). Chicago, IL: University of Chicago Press. Galaskiewicz, J. (1997). An urban grants economy revisited: Corporate charitable contributions in the Twin Cities, 1979–81, 1987–89. Administrative Science Quarterly, 42(3), 445–471. Galaskiewicz, J., & Burt, R. S. (1991). Interorganization contagion in corporate philanthropy. Administrative Science Quarterly, 36(1). Global Reporting Initiative. (2011). What is GRI? Retrieved from http://www.globalreporting. org/AboutGRI/WhatIsGRI/. Accessed on September 20, 2012. Global Reporting Initiative. (2012). About GRI. Retrieved from https://www.globalreporting. org/information/about-gri/Pages/default.aspx. Accessed on September 20, 2012. Glynn, M. A. (2000). When cymbals become symbols: Conflict over organizational identity within a symphony Orchestra. Organization Science, 11(3), 285–298. Glynn, M. A. (2008). Beyond constraint: How institutions enable identities. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), Sage handbook of organizational institutionalism. Thousand Oaks, CA: Sage. Glynn, M. A., & Abzug, R. (2002). Institutionalizing identity: Symbolic isomorphism and organizational names. Academy of Management Journal, 45(1), 267–280. Glynn, M. A., & Lounsbury, M. (2005). From the critics’ corner: Logic blending, discursive change and authenticity in a cultural production system. Journal of Management Studies, 42(5), 1031–1055. Golden-Biddle, K., & Rao, H. (1997). Breaches in the boardroom: Organizational identity and conflicts of commitment in a nonprofit organization. Organization Science, 8(6), 593–611. Greenwood, R., & Miller, D. (2010). Tackling design anew: Getting back to the heart of organizational theory. The Academy of Management Perspectives, 24(4), 78–88. Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E. R., & Lounsbury, M. (2011). Institutional complexity and organizational responses. The Academy of Management Annals, 5(1), 317–371. Guler, I., Guille´n, M. F., & Macpherson, J. M. (2002). Global competition, institutions, and the diffusion of organizational practices: The international spread of ISO 9000 quality certificates. Administrative science quarterly, 47(2), 207–232.

196

MARY ANN GLYNN AND RYAN RAFFAELLI

Guthrie, D. (2003). Survey on corporate-community relations. New York, NY: Social Sciences Research Council. Hess, D., Rogovsky, N., & Dunfee, T. W. (2002). The next wave of corporate community involvement: Corporate social initiatives. California Management Review, 44(2), 110–125. International Organization for Standardization. (2010). ISO 26000 social responsibility project overview. Geneve. Lok, J. (2010). Institutional logics as identity projects. Academy of Management Journal, 53(6), 1305–1335. Lounsbury, M. (2007). A tale of two cities: Competing logics and practice variation in the professionalizing of mutual funds. Academy of Management Journal, 50(2), 289–307. Margolis, J., & Walsh, J. (2001). People and profits? The search for a link between a company’s social and financial performance. Psychology Press. Margolis, J. D., & Walsh, J. P. (2003). Misery loves companies: Rethinking social initiatives by business. Administrative Science Quarterly, 48(2), 268–305. Marquis, C., Glynn, M. A., & Davis, G. F. (2007). Community isomorphism and corporate social action. Academy of Management Review, 32(3), 925–945. Marquis, C., & Lounsbury, M. (2007). Vive la re´sistance: Competing logics and the consolidation of U.S. community banking. Academy of Management Journal, 50(4), 799–820. Marquis, C., & Lee, M. (2013). Who is governing whom? Executives, governance, and the structure of generosity in large U.S. firms. Strategic Management Journal, 34(4), 483–497. Matten, D., & Moon, J. (2008). ‘‘Implicit’’ and ‘‘explicit’’ CSR: A conceptual framework for a comparative understanding of corporate social responsibility. Academy of Management Review, 33(2), 404–424. Meyer, J. P., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363. Navis, C., & Glynn, M. A. (2010). How new market categories emerge: Temporal dynamics of legitimacy, identity, and entrepreneurship in satellite radio, 1990–2005. Administrative Science Quarterly, 55(3), 439–471. Navis, C., & Glynn, M. A. (2011). Legitimate distinctiveness and the entrepreneurial identity: Influence on Investor judgments of new venture plausibility. Academy of Management Review, 36(3). Nikolaeva, R., & Bicho, M. (2011). The role of institutional and reputational factors in the voluntary adoption of corporate social responsibility reporting standards. Journal of the Academy of Marketing Science, 39(1), 136–157. Pedersen, J. S., & Dobbin, F. (2006). In search of identity and legitimation. American Behavioral Scientist, 49(7), 897–907. Pinney, C., & Lee, A. (2010). Profile of the profession 2010: Corporate citizenship leaders for today and tomorrow: 26. Chestnut Hill, MA: Boston College Center for Corporate Citizenship. Powell, W., & Colyvas, J. (2008). Microfoundations of institutional theory. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 276–298). London: Sage. Pratt, M. G., & Foreman, P. O. (2000). Classifying managerial responses to multiple organizational identities. Academy of Management Review, 25(1), 18–42.

The Structural Embeddedness of CSR Programs

197

Raffaelli, R., & Glynn, M. A. (forthcoming). Turnkey or tailored? Relational pluralism, institutional complexity, and the organizational adoption of more or less customized practices. Academy of Management Journal. Rao, H., Monin, P., & Durand, R. (2003). Institutional change in toque ville: Nouvelle cuisine as an identity movement in French gastronomy. American Journal of Sociology, 108(4), 795–843. The Aspen Institute. (2012). Beyond grey pinstripes: Top 100 MBA Programs. Retrieved from http://www.beyondgreypinstripes.orgwww.beyondgreypinstripes.org. Accessed on December 15, 2012. Thornton, P. H. (2004). Markets from culture: Institutional logics and organization decisions in higher education publishing. Stanford, CA: Stanford University Press. Thornton, P. H., & Ocasio, W. (1999). Institutional logics and the historical contingency of power in organizations: Executive succession in the higher education publishing industry, 1958–1990. American Journal of Sociology, 105(3), 801–843. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: Foundations, research, and theoretical elaboration. Oxford: Oxford University Press. Volberda, H. W., Foss, N. J., & Lyles, M. A. (2010). Absorbing the concept of absorptive capacity: How to realize its potential in the organization field. Organization Science, 21(4), 931–951. Waddock, S. A. (2006). Building the institutional infrastructure for corporate social responsibility. Corporate social responsibility initiative. Working Paper No. 32. John F. Kennedy School of Government, Harvard University, Cambridge, MA. Weinger, A. (2012). Two types of corporate employee giving programs. Philanthropy Journal. Retrieved from http://www.philanthropyjournal.org/resources/fundraisinggiving/twotypes-corporate-employee-giving-programs. Accessed on September 20, 2012. Westphal, J., Gulati, R., & Shortell, S. (1997). Customization or conformity? An institutional and network perspective on the content and consequences of TQM adoption. Administrative Science Quarterly, 42(2). Wild, C. (1993). Corporate volunteer programs: Benefits to business. New York, NY: Conference Board.

STRANGE BREW: BRIDGING LOGICS VIA INSTITUTIONAL BRICOLAGE AND THE RECONSTITUTION OF ORGANIZATIONAL IDENTITY Lærke Højgaard Christiansen and Michael Lounsbury ABSTRACT How do organizations manage multiple logics in response to institutional complexity? In this paper, we explore how intraorganizational problems related to multiple logics may be addressed via the mechanism of institutional bricolage – where actors inside an organization act as ‘‘bricoleurs’’ to creatively combine elements from different logics into newly designed artifacts. An illustrative case study of a global brewery group’s development of such an artifact – a Responsible Drinking Guide Book – is outlined. We argue that intraorganizational institutional bricolage first requires the problematization of organizational identity followed by a social process involving efforts to renegotiate the organization’s identity in relation to the logics being integrated. We show that in response to growing pressures to be more ‘‘responsible,’’ a group of organizational actors creatively tinkered with and combined elements from social

Institutional Logics in Action, Part B Research in the Sociology of Organizations, Volume 39B, 199–232 Copyright r 2013 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0733-558X/doi:10.1108/S0733-558X(2013)0039A&B020

199

200 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

responsibility and market logics by drawing upon extant organizational resources from different times and spaces in an effort to reconstitute their collective organizational identity. .

Keywords: Institutional logics; institutional complexity; practice; institution

Over the past decade, research on institutional logics and complexity has exploded (e.g., Greenwood, Raynard, Kodeih, Micelotta, & Lounsbury, 2011; Lounsbury, 2007; Thornton & Ocasio, 1999, 2008; Thornton, Ocasio, & Lounsbury, 2012). The institutional logics perspective allows for a much needed integration of the top-down effects of institutional logics with the bottom-up effects of situational features within individual organizations (Ocasio, 2011; Thornton et al., 2012). As such, and unlike institutional theory’s mainline focus on stability and isomorphism (DiMaggio & Powell, 1983; Scott, 2008), the institutional logics perspective provides new theoretical leverage to understand and explain variation across different organizations. Further, both local situational factors and wider institutional pressures and influences are combined to explain the reproduction and transformation of organizational structures (Thornton et al., 2012, p. 101). Despite these developments, research on institutional logics is still in the early stages, and many open questions and opportunities for scholarly development remain. One such opportunity is the study of intraorganizational dynamics, especially under conditions of institutional complexity where organizations are exposed to pressures from multiple institutional logics. The institutional logics perspective provides theory and method for understanding the institutional influences emanating from societal-level institutional orders that shape the action and interactions of organizations and individuals (Thornton, Jones, & Kury, 2005; Thornton & Ocasio, 2008). Institutional logics embody supraorganizational patterns of activity (Friedland & Alford, 1991) – belief systems and material practices that represent particular worldviews, valuable ends, and the appropriate means to achieve these ends. Early research on institutional logics focused on shifts in logics and its consequences (e.g., Christensen & Molin, 1995; Thornton & Ocasio, 1999). More recently, as attention has shifted to understanding how organizations engage more pluralistic environments (e.g., Kraatz & Block, 2008), scholars have begun to focus on the interactions between and mixing of institutional logics – logics in action.

Strange Brew

201

For instance, Glynn and Lounsbury (2005) studied how the Atlanta Symphony Orchestra tried to increase attendance rates by blending their traditional aesthetic logic with a market logic when faced with growing resource constraints and a musicians’ strike. Reay and Hinings (2009) investigated how a multiplicity of logics persisted and proposed that the interaction between logics is managed through collaborative relationships between field-level actors. Battilana and Dorado (2010) explored how microfinance organizations aimed to hybridize development and banking logics in a way that created balance between the logics. Yet, we still know relatively little about the intraorganizational dynamics involved in managing multiple institutional logics and how logics get blended as an outcome of an organization’s response to institutional complexity (Greenwood et al., 2011; Pache & Santos, 2010). There have been several recent efforts to theorize these issues, setting the stage for systematic empirical inquiry. Greenwood et al. (2011) provide a theoretical model to guide such analysis, and Pache and Santos (2010) elaborate on Oliver’s (1991) theoretical typology to propose a general model of responses to conflicting institutional demands – provocatively suggesting that organizations respond to competing logics by selectively hybridizing logics in some practices but not others. This attention to the level of practice inside organizations is also focalized by Thornton et al. (2012, pp. 133–141). They outline an institutional logics approach to the dynamics of practices and identities in organizations that explores both the internal and external aspects of organizational behavior and practice (see also Jarzabkowski, 2005; Lounsbury, 2001; Lounsbury & Crumley, 2007; Smets, Morris, & Greenwood, 2012; Smets & Jarzabkowski, forthcoming). We seek to extend these insights and calls by exploring how conflict between multiple logics may be addressed via the mechanism of bricolage. That is, we argue that when institutional complexity creates tangible pressures related to conflicting logics, organizational actors may act as ‘‘bricoleurs’’ – a form of cultural entrepreneurship (Lounsbury & Glynn, 2001) – to creatively combine elements from different logics into newly designed artifacts. This is in line with the more dynamic conceptualization of institutional logics (Thornton et al., 2012, pp. 57–61) that theorizes logics as partially autonomous and nearly decomposable – where categorical ‘‘Y-axis’’ elements of logics may be mixed to construct new hybrid practices and symbolic systems. The concept of bricolage (Le´vi-Strauss, 1962/66) has been increasingly adopted by researchers within organization studies (Boxenbaum & Rouleau, 2011; Duymedjian & Ru¨ling, 2010), but it is yet to be fully incorporated into

202 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

the institutional logics perspective. The notion of bricolage refers to the way a ‘‘bricoleur’’ builds an artifact by using whatever is available, within a restricted environment, to get the job done. The ‘‘bricoleur’’ uses the remains and debris of events, fossilized evidence of the history of an individual or society, to construct something new – that is, for which the individual parts were not originally intended (Levi-Strauss, 1962/66). More recently, Baker and Nelson (2005) elaborate on the characteristics of bricolage. Based on their broad reading of multidisciplinary literature, they provide an integrative definition of bricolage as: ‘‘making do by applying combinations of the resources at hand to new problems and opportunities’’ (Baker & Nelson, 2005, p. 333). The result of bricolage is often a tangible artifact that has a unique configuration of symbolic and material properties. The notion of bricolage is in line with more recent work at the interface of institutions and identity (i.e., Glynn, 2008; Glynn & Abzug, 2002; Pedersen & Dobbin, 2006; Rao, Monin, & Durand, 2005) that emphasizes how organizations tend to be constructed by ‘‘bricks’’ that are available in their institutional environment. This stream of research focuses on the way in which organizations are constituted by different institutionally derived elements or bricks that are then combined by a ‘‘bricoleur’’ to create something that is both legitimate within a given institutional field, but at the same time can be assembled to foster organizational distinctiveness. For instance, Glynn (2008) suggests that organizational identity construction is a process of institutional bricolage, where organizational actors incorporate different cultural meanings, sentiments, rules, and material artifacts into their identity claims and displays (Glynn, 2008). Building on this line of work, we argue that at the core of institutional bricolage inside an organization are the efforts to reconstitute an organization’s identity. We illustrate this process through a case in which an organization created an important ‘‘boundary object’’ – a flexible artifact that incorporated elements from several interacting social worlds (Bechky, 2003; Star & Griesemer, 1989). More specifically, we detail how a group of organizational actors in the headquarters of a global brewery firm created a ‘‘Guidebook’’ that reflected a mixing of social responsibility and market logics as well as an effort to reconstitute organizational identity. In line with bricolage research on the production of novel artifacts, we trace the development of the Guidebook as a tangible artifact produced as a result of intraorganizational negotiations regarding the two logics and how they relate to the identity of the organization. We argue that intraorganizational institutional bricolage first requires the problematization of organizational identity followed by a social process involving the renegotiation of an

Strange Brew

203

organization’s identity related to the logics being integrated. We show that in their efforts to reconstitute identity, a group of organizational actors creatively tinkered with and combined elements from the social responsibility and market logics, but to do so importantly drew upon extant organizational resources from different times and spaces. We begin by reviewing the literature on bricolage, focusing on its conceptualization and use within institutionally oriented scholarship. Then, we explore the associations to research on organizational identity construction as well as the links between institutional logics and organizational identity. Thereafter, we situate our illustrative case study and briefly outline data collection and sources. We then explore the way in which conflicting institutional demands are experienced within a large international brewery, and describe the process through which a group of organizational actors within the headquarters developed the Guidebook in their aspiration to bridge logics and reconstitute identity. We conclude with a discussion of the implications of our research for future scholarship.

INSTITUTIONAL BRICOLAGE AND IDENTITY RECONSTITUTION The concept of bricolage (Le´vi-Strauss, 1962/66) has increasingly been adopted by researchers within organizational studies (e.g., Boxenbaum & Rouleau, 2011; Duymedjian & Ru¨ling, 2010). It has been applied within at least three areas: improvisation and sensemaking (e.g., Weick, 1998), entrepreneurship (e.g., Baker & Nelson, 2005) and to describe organizational identity construction (e.g., Dobbin & Strandgaard-Pedersen, 2006; see Duymedjian & Ru¨ling, 2010 for a thorough review of the concept of bricolage within management and organizational studies). By forwarding the notion of institutional bricolage, we mean to focus attention on the creation of new assemblages of institutional elements such as those associated with logics. The notion of bricolage refers to the way a ‘‘bricoleur’’ constructs an artifact by using whatever is available, within a restricted environment, to get the job done. The ‘‘bricoleur’’ creatively assembles new things by using the historical remains and debris of events or structures (Levi-Strauss, 1962/66). Schneiberg (2007, p. 48) has referred to these as flotsam and jetsam – elements of alternative orders and abandoned or partly materialized institutional projects. From an institutional perspective, bricolage may be conceptualized as a mechanism related to institutional and organizational

204 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

change where solutions to problems involve the recombination of available and accessible institutional elements (e.g., logics). The concept of bricolage is an appealing concept to institutionalists because of its innate roots in the idea of historical contingency and path dependence, portraying the nature of change and innovation as being evolutionary rather than revolutionary since novel artifacts embody remnants from the past (Borum & Westenholz, 1995; Campbell, 2004, p. 69; Douglas, 1986, pp. 66–68; Schneiberg, 2007). Over the past decade, bricolage has increasingly been used to describe the construction of organizations and organizational identities (Glynn, 2008; Lounsbury & Glynn, 2001; Strandgaard-Pedersen & Dobbin, 2006). For instance, several studies have examined processes of identity reinterpretation via a kind of temporal bricolage (i.e., Gioia & Thomas, 1996; Ravasi & Schultz, 2006; Schultz & Hernes, 2012) that links past, present, and future identities. Ravasi and Schultz (2006) illustrated how a Danish audiovisual manufacturer, Bang & Olufsen, revisited past cultures in response to different external identity threats during three different periods of identity reconstruction. Similarly, a recent study by Schultz and Hernes (2012) offers an ongoing temporal perspective on organizational identity and show how change and stabilization of identity arises from the work of organizational actors as they use the resources of the present to invoke and transform past organizational experience into ambitions for the future. Based on their longitudinal study of the toy manufacturer, LEGO, Schultz and Hernes developed a conceptual framework where they focused on the influence of different memory forms: textual, material, and oral. Their findings suggest that variations in use and combination of different memory forms influence the scope and depth of identity claims for the future, and that differences in time span echo in relation to the organization’s past and future. In a review of the literature at the interface of institutions and identity, Glynn (2008, p. 424) draws on Swidler (1986) to depict identity construction as a process of institutional bricolage where different cultural meanings, values, and rules are incorporated into the organization’s identity claims. One of her key points is that since organizations draw from the same pool of cultural elements and resources, they come to resemble one another; yet the way that organizations combine different cultural elements are at the same time relatively unique (Lounsbury & Glynn, 2001). The distinctiveness of organizations and their identities stems from the fact that institutional elements are often assembled by actors in a way that is tailored to their specific organizational context and associated cultural repertoires (Glynn, 2008; Pedersen & Dobbin, 2006; Rindova, Dalpiaz, & Ravasi, 2011; Westenholz, Strandgaard Pedersen & Dobbin 2006). Further, when

Strange Brew

205

organizations are imprinted with several different institutional models, each of these represents both possibilities and constraints to entrepreneurial efforts toward change (Borum & Westenholz, 1995). The intraorganizational process of assemblage, as we highlight in this paper, can therefore often involve conflict and negotiation, especially when the bricolage process involves multiple institutional logics. As Friedland and Alford (1991) propose, there is no change without actors and there is no way to account for change without multiple institutional logics available to provide alternative meanings as the sources for change. They note that ‘‘under some conditions, they (actors) are artful in the mobilization of different institutional logics to serve their purposes’’ (Friedland & Alford, 1991, p. 254). Their invocation of ‘‘artful’’ closely resembles the spirit of bricolage, especially in relation to the Levi-Strauss idea of ‘‘making do’’ by relying on the means that are available. Since organizations often operate in complex environments, they typically have to respond and adapt to a multitude of institutional pressures. As a result, many organizations comprise a constellation of different institutionally derived elements. A good deal of recent work has focused on how organizational hybridity linked to multiple logics is achieved and managed (i.e., Binder, 2007; Battilana & Dorado, 2010; Glynn & Lounsbury, 2005; Pache & Santos, 2010). For instance, in a study of commercial microfinance organizations, Battilana and Dorado (2010) explore how social responsibility and banking logics can be successfully hybridized. Their findings show that organizations can sustain their hybridity if they develop a common organizational identity that strikes a balance between the logics that constitute the organization. Pache and Santos (2010) argue that hybridity might also occur at the level of practice where manifestations of hybridized logics could coexist alongside un-hybridized practices. One could imagine that this could be the case in commercially driven corporations that have CSR departments or non-profits seeking out a social goal that must be sustained via different commercially driven revenue-generating practices. This can be understood as a type of compartmentalization (Kraatz & Block, 2008; Pratt & Foreman, 2000) in which two (or more) important but antithetical identities (and logics) are maintained separately and operate relatively independent of each other. Such a separation, however, might become a focal point for scrutiny if organizational audiences find that there is an illegitimate decoupling of organizational practices. Of course, organizational identity is constantly made and remade as a result of intraorganizational dynamics as well as ongoing societal shifts (Navis & Glynn, 2010; Pedersen & Dobbin, 2006).

206 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

Since we argue that the process of institutional bricolage involves identity reconstitution, the problematization of the organizational identity by organizational members that perceive a threat or opportunity related to institutional complexity is a necessary trigger. While not focused on institutional complexity, there is a literature on how organizations defend or remake identity due to external threats (e.g., Dutton & Dukerich, 1991; Elsbach & Kramer, 1996; Ravasi & Schultz, 2006). For instance, Elsbach and Kramer’s (1996) study of university responses to identity threats suggests that managers may strategically attempt to reinvent or rediscover organizational identity after periods of identity dissonance. Below, we document how threats related to institutional complexity triggered efforts to reconstitute identity via institutional bricolage of logics, thus bridging disparate research streams on institutional logics, bricolage, and organizational identity.

THE SETTING J.C. Jacobsen founded the Carlsberg brewery in 1847 and in 1876 he established the Carlsberg Foundation, which took ownership of the company after his death in 1886. Since then, Carlsberg has grown to be a large multinational corporation, active in 150 markets, operating 85 breweries across 46 countries and employing 41,000 people. In recent years, Carlsberg has grown very aggressively through mergers and acquisitions; it has grown from being a local brewery to a regional one (started exporting in 1868 to Scotland) to, after the acquisition of Scottish & Newcastle in 2008,1 being the fourth largest brewery group in the world. In 2008, Carlsberg also signed the UN Global Compact to symbolize the beginning of a Carlsberg Group-wide commitment to corporate social responsibility (CSR). Since then, they have been formalizing their approach to CSR, formulating and implementing what they refer to as their ‘‘glocal approach’’ – emphasizing the balance between operating globally and locally in different CSR focus areas. This balance is especially challenging in a global brewery group comprising a wide range of companies that have been united through mergers and acquisitions. The Corporate Communication department has been the driver of the CSR implementation, although a separate functional subunit of CSR was installed in 2008 along with a CSR governance structure that was to underpin the implementation of CSR across the entire organization. A number of CSR policies and guidelines have been developed in recent years, covering areas such as labor and human rights, business ethics,

Strange Brew

207

environment and marketing communication. In some of these areas, different organizational actors involved in policy development are informed by different institutional logics. We study the role of competing institutional logics in Carlsberg’s development of CSR by focusing on initiatives related to responsible drinking (RD) – one of the most prominent focal points for CSR development within Carlsberg and across the brewery (and alcohol) industry. The concept of RD is commonly associated with activities aimed at preventing the misuse of alcohol (e.g., binge drinking, youth drinking, and drunk driving). In the development of a new and more strategic groupwide approach to RD, a group of people at Carlsberg headquarters engaged in what we consider to be an institutional bricolage as they combined elements of different logics to construct an ‘‘RD Guide Book.’’ This book was intended to forge a linkage between the communication and CSR employees, who are advocates of a social responsibility logic, and the marketing and sales people who are primarily carriers of a market logic, thus providing an overarching corporate approach to RD. Consequently, embracing the RD concept entails that organizational actors find ways to manage the inherent institutional complexities (i.e., multiple institutional logics) related to the issue. Table 1 provides a comparative overview of the market and social responsibility logics as they are manifest in Carlsberg Group’s headquarters. In daily operation and practice, Carlsberg Group is dominated by a ‘‘market logic,’’ characterized by profit maximization and a set of overarching goals that are pursued via classic textbook focus areas such as marketing, brand building, return on investment, management control, market expansion, global standardization, and economies of scale. This list reflects the very dominant and broad societal logic of the market (in contemporary western society) as theorized by Friedland and Alford (1991). The Group’s ambition to be ‘‘the fastest growing global beer company’’ is a vision and a strategy that dominates and is very much present in their daily operations at the headquarters. The rules for achievement are to increase market share and operational efficiency; this is especially evident in the Group’s current efforts to centralize and optimize their supply chain. As some interviewees noted, the company is increasingly moving toward an ‘‘FMCG model’’ (fast moving consumer goods). In contradistinction to the market logic, the issues surrounding RD invoke a social responsibility logic – a belief system that focuses on caring about how an organization’s practices impact broader society. A socially responsible brewery, for instance, might work hard to try to minimize societal problems related to the use of their products. This may lead to

208 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

Table 1.

Comparative Overview of the Logics Represented in the Responsibe Drinking Project.a Social Responsibility Logic

Market Logic

Belief systems (what goals are pursued within the field)

Live up to ethical responsibility toward stakeholders/society

Build competitive position Increase profits

Focus

Stakeholder relation Environmental impact Societal concerns, that is, health issues and public opinion in general

Return on investment Brand positioning Global standardization

Rules for achievement

Live up to ethical standards

Efficiency Increased market share Clear positioning

Associated practices (means for pursuing the goals and values)

Not encourage more than moderate drinking

Sell the beer, as much possible, no matter what it takes

Contributing to minimizing the downsides of the product

Run campaigns that get the products sold and claim distinction Invent new products that satisfy consumers’ needs for alcohol/ beer products – ‘‘getting drunk’’

Encouraging and helping consumers with public transportation to and from concerts a

Source: Thornton and Ocasio (1999) and Reay and Hinings (2005).

educational and marketing campaigns that aim to encourage people to drink in moderation or even not to drink in specific contexts such as when pregnant or driving. Concrete efforts in Carlsberg include initiatives to prevent underage drinking at concerts they sponsor, and/or to provide free water to encourage consumers to supplement their alcohol intake. However, the market and social responsibility logics often come into conflict since such ‘‘responsibility’’ initiatives often come at the cost of restricting efforts to maximize profits and shareholder value. Carlsberg Group is best understood as a late adopter of CSR, as they did not create their global CSR organization until 2008 – well after most other major global brewers created their CSR initiatives. While the social responsibility logic does not yet have an overtly strong influence on Carlsberg organizational practice given that its embrace is so recent, when visiting the

Strange Brew

209

corporate headquarters in Carlsberg City, it is evident that there is a strong symbolic and material imprinting of the philanthropic founder.2 Carlsberg has always had a strong philanthropic heritage – the founder J.C. Jacobsen (1811–1887) actively gave to the arts and science, in addition to running his own laboratory where clean yeast was first invented and shared with the rest of the brewing community. He also donated to the city of Copenhagen its first fire engine after his own brewery burned down in 1867. Further, he established the Natural History Museum of Denmark and the Botanical Garden in Copenhagen.3 Lastly, he set up the first foundation in the world to have ownership of a company – the Carlsberg Foundation (1875). The foundation still owns 25 percent of MNC Carlsberg Group today and runs the Carlsberg Laboratory (focused on research in Chemistry and Physiology) where all scientific advancements must be shared according to founder’s original charter. In addition, the Carlsberg Foundation supports fundamental research in Denmark within the social sciences, humanities, and natural sciences. The Carlsberg Group headquarters is located in the old Carlsberg City in Copenhagen where brewing used to take place (today the operating brewery in Denmark is in Frederica), where the old home of the founder still stands, and where there is a company museum, the Carlsberg archives, and the Carlsberg Laboratory. Hence, the logic of social responsibility is available and accessible to the people working at headquarters. Most employees will at least at one point during the day see an artifact symbolizing the founder and or the foundation. Indeed, many of the interviewees actually mentioned the importance of ‘‘social heritage’’ explicitly during interviews: At headquarters, we often discuss that CSR is part of our heritage, and that there is almost like a red thread leading back to our founders, that they acted as socially responsible citizens before the CSR term was invented, of course in a more traditional and philanthropic way. When the company was doing great, then the society should feel it too, leading to a close interaction with the community. Further, part of the overhead was donated to society via the Carlsberg Foundations. It was clear for me when I was hired in to Carlsberg (in 2004), that ‘‘Oh wow, there are so many practices in our history that points to the possibility of CSR once again becoming an important part of the way we do business, but as it was, CSR was not at all part of the Group’s strategic agenda.’’ (Top executive within communication, 2011)

Toward the end of the 1990s, brewers became aware of a new form of scrutiny within and across Europe – it revolved around the fact that brewers and other alcohol producers were seen to be at the core of one of society’s main health challenges. Beer and alcohol were seen to be particularly problematic in relation to young people. Some industry stakeholders

210 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

thought that the alcohol industry was targeting its products toward young people as well as pushing boundaries in their commercial communication (e.g., by making provocative advertisements with sexual images to distinguish their products). The industry was criticized for setting aside societal considerations in the pursuit of market share, wanting to sell the product no matter what the social implications might be for broader society. Consequently, the alcohol and brewery industry came under attack, and health organizations such as the World Health Organization amplified the push for more regulation and higher taxation. Consequently, brewers experienced increased restrictions on their advertising across Europe, as well as higher tax and duty rates. In countries like Finland, the taxation on beer is 55–80%, and in 2010 Russia experienced a 200% beer duty increase due to a change in the categorization of beer as an alcohol (Carlsberg Group, 2010; Panimoliitto,4 2010; Brewers of Europe, 2010). Further, in some European markets, advertising of alcohol is prohibited in all media while other markets impose restrictions on television commercials and billboards. These institutional changes have spurred different responses across the brewery community, and in Carlsberg Group, people within corporate communications and public affairs were the first to react to such shifts in the political environment. Communication department employees started by framing these changes as something that should be of concern to the organization and that a behavioral change would have to be made if the company wanted to ‘‘maintain its commercial platform’’ or have a ‘‘license to operate’’ long term. By the mid-2000, industry associations such as the Brewers of Europe indicated that inappropriate marketing activities were the main cause for the change in public perception. They argued that people marketing beer had been too shortsighted and irresponsible in their eagerness to sell the product. Thus, marketeers were framed as the culprits, or at least accomplices, and the ones that needed to change behavior and make sure they did not urge any inappropriate use of beer in their marketing communications. Correspondingly, some Carlsberg employees also expressed the view that sales and marketing people did not understand their own role in fomenting increased regulatory attention and pressure to change. Instead of radically altering their established practices, they invented new tactical operating procedures – for example, they categorized different markets into grey and dark markets5 to indicate the level of regulation in different markets, and created manuals and guidelines for operating in these regulated markets. On more than one occasion, people involved with the implementation of social responsibility in the Sales and Marketing department expressed that the

211

Strange Brew

challenge and the root of the problem was that: ‘‘the accountability for the long term maintenance of our marketing platform does not lie in Marketing, it is not there today!’’ (Manager within communications, 2011). The identified problem was that marketing did not take ownership because they considered it to be irrelevant in relation to their goals and objectives, which are thoroughly rooted in a market logic. Instead they wittily re-labeled the CSR unit the ‘‘de-sales department’’ thereby signifying that the goal when implementing CSR is not only irrelevant to the people within marketing and sales, but actually in direct conflict with the market logic. This is the conflict which CSR advocates attempted to overcome via the development of an RD Guide Book.

Data Collection The case that we use to illustrate our theoretical argument and conceptualization of institutional bricolage and identity reconstitution is extracted from a larger ongoing inductive study of institutional complexity and logic combination efforts within the Carlsberg Group. The first author has been collecting data real time in the organization since 2010, both at headquarters and in local subsidiaries6 to investigate how organizational actors proceed in a situation where noticeably distinct institutional logics interact and are combined into a new constellation. In Carlsberg’s conceptualization of RD, the organizational actors deal with different logics and the controversy and immediacy of RD to Carlsberg Group’s product, beer, is intrinsic and closely linked to the legitimacy of the organization. In headquarters, the development of a concrete artifact has been followed – the development of the RD Guide Book’’ – that has been initiated to instigate change in organizational identity and behavior. First, the Guide Book is meant to inspire cross-functional collaboration or ‘‘alignment’’ between the marketing, sales, communication, and public affairs departments. Second, the aim is to influence the organizational structure and make ‘‘social responsibility’’ a more ingrained part of Carlsberg Group’s organizational identity and practice. In the present paper we use elements of this ongoing research project for illustrative purposes.

Data Sources The Guide Book has been under development since September 2010 and was finalized in June 2011 in a 1.0 version. The following data sources have

212 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

informed the case writing: 13 interviews with key RD stakeholders, the RD Guide Book at different stages of development, over 150 pages of organizational/industry documents and artifacts, as well as 25 hours of observations of meetings and presentations and 11 planning and feedback meetings with the organization.7 The interviews were conducted on the topic of CSR in Carlsberg, the RD project, CSR policies, and meetings on RD that had been observed. The informants were selected based on their association with the area of RD as important input givers and/or decision makers. The informants were managers within the Communications, CSR, and Marketing departments. Organizational documents have been collected from the time of establishment of the CSR department in 2008 up until 2012. The documents include all CSR policies and guidelines, presentations of the CSR department, including those related to strategy and competitor analysis that had been made for the Carlsberg Group executive committee (from 2008 until 2010) as well as more specific documents related to RD and the development process (e.g., materials from industry associations and subsidiaries). The documents contributed to a greater understanding of Carlsberg, its history, the organizational actors, the RD project, and its methods and aims. The first author also attended seven meetings at global headquarters regarding the development of the ‘‘RD Guide book’’ in the period from December 2010 to May 2011. The meetings provided valuable information about the RD project process, organizational stakeholders, organizational structure, past events and practices, strategies for implementation, and upcoming events. Eleven feedback meetings with the organization were conducted over the course of the research process, both with people from within the CSR unit, but also with members of the executive committee. Lastly, the first author also attended senior management presentations made in relation to the global relaunch of the Carlsberg brand in 2011as well as public presentations made at Copenhagen Business School. Together, these sources have given a great insight into the different logics represented within the RD project and the organizational actors engaging in a process of institutional bricolage to get people ‘‘on board.’’

INSTITUTIONAL BRICOLAGE: BRIDGING LOGICS AND RECONSTITUTING IDENTITY INSIDE AN ORGANIZATION We conceptualize the construction of the RD Guide Book8 via logic bridging as a kind of bricolage process, and aim to illustrate different

213

Strange Brew

mechanisms related to what we label institutional bricolage. We begin by outlining the outcome of the process – the RD Guide Book and show that it is an assembly of bits and pieces from two different logics that have been brought together to create an altered balance between the two logics. The Guide Book may be understood as a kind of ‘‘boundary object’’ (Bechky, 2003; Star & Griesemer, 1989), an artifact that combines the social responsibility and market logics by integrating elements from the past and present, as well as from different geographical places. At the core of our argument is that institutional bricolage inside an organization fundamentally involves the reconstitution of identity. Greenwood et al. (2011) conceptualized organizational identity as a filter that shapes how organizations react to institutional complexity, but we believe that acts of combining logics often involve efforts to alter core aspects of an organization’s identity. In our case, we emphasize how institutional bricolage may involve recovering aspects of the history of the organization to give tangible life to a marginalized or latent logic (the social responsibility logic in our case), and to engender a conversation about organizational identity. Schultz and Hernes (2012) argue that the past is often evoked in processes of identity reconstruction as well as the articulation of claims for future identity. We argue that the construction of the Guide Book and the artifact itself involve these fundamental processes. The case of the development of a guidebook illustrates that institutional bricolage requires that actors within the organization mobilize to successfully frame key strategic issues related to marginalized logics as fundamentally related to organizational identity, and to persuasively convince others of the importance of the melding logics. In our case, the market logic was dominant, and our efforts focus on how actors inside the organization tried to convince a broader array of actors of the importance of the more marginalized social responsibility logic by emphasizing its rootedness in the identity of the organization across time and space.

The Responsible Drinking Guide Book Fieldwork indicated that the communication and CSR people at headquarters were intent on making the RD Guide Book function as a boundary object (Bechky, 2003; Star & Griesemer, 1989) that would be a centerpiece in redefining organizational identity, enabling interaction across different organizational divisions and geographically dispersed subsidiaries. The Guide Book was constructed to provide a rationale to encourage marketing

214 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

people and senior management across the organization to effectively and in a coordinated way engage in RD initiatives: Aligning RD with sales & marketing process (headline) Performing Responsible Drinking initiatives often requires input from different functions within the local company. It is recommended that alignment is created across Marketing & Sales, Communications, External Affairs, CSR and PR. When you are considering doing large scale Responsible Drinking initiatives you may even want to consider setting up a cross-functional task force to drive them. (RD Guide Book, 2011, p. 9)

Through the Guide Book construction process, different cross-functional resources were combined, consisting of a mix of locally constituted elements as well as those drawn from wider discourses. The organizational elements included both resources from headquarters and subsidiaries as well as elements representing the organizational past and present. An overview of these elements is provided in Table 2. The first two columns portray the presence of the two logics, while the third column illustrates instances where the two logics are combined. The third row also includes elements of time and space, two features that were activated to enable institutional bricolage and the reconstitution of the organizational identity – ‘‘who we are’’ and ‘‘what we do,’’ essentially producing an organization-wide commitment to the RD issue. Social Responsibility Logic In the Guide Book, the social responsibility logic informs the overarching argument that the local subsidiaries should engage in solving societal challenges in relation to alcohol. It also signals that the RD Guide Book is the authorized Carlsberg Group tool to engage with the issue of alcohol misuse. In other words, the Guide Book presents the official Group response to a challenge that the CSR unit in headquarters considers to be widespread in societies in which the Group operates. This argument is supported by the introduction of four areas that are considered to be standard in relation to preventive alcohol action: ‘‘drunk driving,’’ ‘‘under-age drinking,’’ ‘‘binge drinking,’’ and ‘‘moderate consumption.’’ These areas reflect the societal problems related to beer and alcohol consumption that have been constituted and conceptualized in a wider alcohol industry context, for example, through debates in local and regional industry associations such as British Beer and Pubs Association, the Portman Group (a UK alcohol CSR advocate and consultancy), and Brewers of Europe. Hence the four areas are presented as generic and fairly standardized RD focus areas applied across the industry, which in and of itself signifies ‘‘that the alcohol prevention program works’’ and again authorizes the overall Guide Book initiative.

Group Sales, Marketing, and Innovation.

Transportation to prevent drink/ drunk driving

Cases

a

CSR, communication, and public affairs

Elephant gate signifies the philanthropic heritage

Organizational members

Pictures

Improved reputation

Marketing and sales

Brand building and social impact Current practices within the represented subunits and competitors (global industry collection) – ‘‘best practices’’ (space and time)

Alignment across functions and/or cross-functional task forces Guide future interaction and ‘‘alignment’’ of practices (time and space)

Pictures from RD campaigns and events where the local units have executed RD campaigns with a social and commercial impact Elephant gate signifies heritage and headquarter (time and space) Campaign pictures signify the subunits (space)

Drink in moderation (time – future)

SWOT analysis, KPIs, ROI, brand building, evaluating performance The brands (corporate of product) visibility in the RD communication – the linkage to brand building

Concepts

Responsible drinking, drink driving, underage drinking, etc. Positive pictures signifying the concept of RD

Unification of past and present (time) Elephants signifies HQ (space)

Modeled after GSMIa process for doing projects

Layout

Combined ‘‘The purpose of Responsible Drinking initiatives is primarily to do something good for the benefit of consumers and society, and secondarily To build the reputation of our business and our brands’’ Drawing on heritage and local best practices from key markets in the different regions: Northern & Western Europe, eastern Europe and Asia (time and space)

Engage in the solving societal challenges in relation to alcohol

Market Logic Maintain product legitimacy – avoid stigmatization, risk avoidance (costs) Increase or maintain market shares

Arguments

Social Responsibility Logic

Table 2. Exemplary Overview of the Elements United in the RD Guide Book.

Strange Brew 215

216 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

The pictures in the book signify that RD has a positive impact. This impact is especially evident in relation to the case examples used. The case examples have been collected from a range of local subsidiaries; they have been carefully selected in view of their reflections of local experience with RD, their size in the Carlsberg portfolio, and their relationship to headquarters. Many cases illustrate instances where local subsidiaries have engaged in creating a solution to the issue of RD – for example, providing transportation to and from concerts to prevent drunk driving or what they call ‘‘Drink Driving.’’ Market Logic In the Guide Book, the market logic informs the argument that RD will enable the organization to maintain product legitimacy while limiting the stigmatization of beer as a product category (Vergne, 2012). The line of argument was primarily posed within the organization as a threat or a risk, but also as something that could be turned into an organizational opportunity if the local RD initiators ‘‘align with the sales & marketing process’’ (p. 10) by finding synergies with the tools and approaches that sales and marketing staff use in their interaction with consumers. The importance of this ‘‘alignment’’ with sales and marketing is also supported in the layout of the Guide Book which is modeled after the official sales and marketing communication material sent to subsidiaries – the latest iteration branded as the ‘‘Group Way.’’ The ‘‘Group Way’’ concept was developed to spread ideas, tools, and best practice across the group to create more consistency and efficiency in daily operations. Five stages are singled out: ‘‘Opportunity identification,’’ ‘‘Strategize,’’ ‘‘Plan,’’ ‘‘Execute,’’ and ‘‘Evaluate’’; the Guide Book is structured accordingly. In addition, the Guide Book contains different generic marketing concepts such as a SWOT analysis, key performance indicators, return on investment, and brand building. Pictures similarly illustrate a product brand dimension, and it is suggested that the visibility of RD communications can contribute to brand building and improved image for the company and its products. Overall, the role of the Guide Book is to provide tangible and strategic organization-wide communication directives for staff in subsidiary units. It is emphasized throughout that the message should be kept positive! For instance, this point is clearly emphasized in a fact-box in the Guide Book: IMPORTANT: Avoid ‘‘negative’’ key messages (e.g. ‘‘don’t binge drink’’). The risk is that your initiative becomes associated with the problem (e.g. ‘‘beer causes unsafe roads’’) rather than offering solutions to consumers and to the general public (e.g. ‘‘beer provides safe transport back home from happy evenings out’’). (RD Guide Book: 17).

Strange Brew

217

This extract illustrates the conscious effort to create frames that combine RD and beers in a way that supports a renewed branding effort. It is not only about doing ‘‘good,’’ but also about seizing the opportunity to promote the company and/or the product brand. In a similar vein, the Guide Book strategically avoids related topics such as drinking during ‘‘pregnancy’’: You may also have seen some stakeholders mentioning ‘‘pregnancy’’ as a subject of Responsible Drinking. However, since ‘‘pregnancy’’ is best addressed through brewers associations or even better, social aspects organisations, it is been given less attention in this booklet. (sic)

Thus, our case of institutional bricolage of the social responsibility and market logics became highly strategic – social elements that were considered to be irrelevant or too risky in relation to the company’s market goals were edited out of the RD Guide Book. While the Guide Book project aimed to bridge logics, it is also clear that it required a committed effort on the part of social responsibility advocates since Carlsberg still prioritized profits over social welfare. Alongside the framing of the RD issue as being ‘‘aligned’’ with the market logic described above, Guide Book development also entailed a subtle framing of RD as being an issue of significance to the organization’s identity – ‘‘what we do determines who we are.’’ The claim that is put forward in the Guide Book is that social responsibility is not something new, it is actually an ingrained element in our heritage (historical and material) and ‘‘who we ARE.’’ In this way, time and space elements become resources in support of the overall agenda to combine different logics and convey to the recipient that the social responsibility logic should be a more salient component of a reconstituted organizational identity. For instance, when entering Carlsberg City, one goes through the main gate: the elephant gate, the most famous building at the Carlsberg site, which symbolizes the founder and his commitment to brewing and society. Above the gate is the founders motto: laboremus pro patria – let us work for our country. Thus, for organizational actors in headquarters, the organization’s heritage is present symbolically as well as substantively in the buildings and materials that surround their workplace. These images were prominent in the RD Guide Book which has the elephant gate on the front page and at least five elephants included in the layout as ‘‘place holders’’ (see Fig. 1). Hence, the elephant was used to frame the message and create the argument that ‘‘social responsibility’’ is not something new, it is already part of ‘‘who we are.’’

218 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

Fig. 1.

The Front Page of the RD Guide Book (Left) and A ‘‘Clean’’ Picture of the Elephant Gate (Right).

Nonetheless, fieldwork indicated that the organizational members working on the RD Guide Book in headquarters experienced a rather close and perhaps unquestioned tie to the social responsibility logic, yet the market logic remained dominant. The whole book and its supplement was framed as an argument for the ‘‘business case of responsible drinking’’ and the Guide Book was consistently discussed by sales and marketing personnel in terms of related initiatives such as the ‘‘Group Way.’’ Thus, the Guide Book was importantly shaped to fit into the sales and marketing toolkit; the aim was that the Guide Book would become an integrated part of the communications that subsidiaries receive from headquarters.

The Process of Institutional Bricolage Below, we describe the process of institutional bricolage – the bridging of market and social responsibility logics inside Carlsberg. Our focus is on institutional bricolage inside an organization, although we acknowledge the role of external influences and the dynamic relationship and mutually constitutive nature of intra- and extraorganizational processes. We argue that at the core of intraorganizational institutional bricolage are efforts by actors to reconstitute organizational identity in ways they assume to be most appropriate or that support their own interests (Borum & Westenholz, 1995). This process first requires the problematization of organizational identity followed by a negotiated social process of transforming an organization’s identity to reflect the bridging of logics. In the case of Carlsberg, we highlight how marginalized actors (in the communications and CSR

Strange Brew

219

departments) attempt to reconstitute organizational identity to account for their interests (social responsibility) by redefining brand image and developing a new integrated, firm-wide communications platform – the Guide Book plays a key role in relation to both.

Triggering Institutional Bricolage: Problematizing Organizational Identity As in most organizational change efforts, a trigger is required – some sort of problematization of the current state of affairs, and the emergence of leaders who propose solutions. What is different here is that when external logics threaten the dominant logic of a firm, the problematization of organizational identity is typically required. This problematization results from complicated internal negotiation processes that center around how to bridge or find a settlement between the dominant and insurgent logics. The Carlsberg CSR department has a broad agenda of changing extant organizational practices and identity to become more socially responsible. In many of the targeted areas they work with (e.g., labor and environment), their goals and aims are compatible with the dominant market logic; in fact, using less energy or water to produce beer may often lead to reductions in expenditures. However, in relation to the area of responsible drinking (RD), the contradictions between the market and the social responsibility logics are vivid; selling as much as possible to increase market share (market logic) while not encouraging excessive drinking (social responsibility logic) is especially challenging since the target audience includes young people just above the legal drinking age (varying across the product portfolio). Not surprisingly, however, organizational actors informed by the social responsibility logic are a minority within the organization. In relation to the specific area of RD, marketing employees are generally resistant to social responsibility ideas and practices – for instance, they pejoratively labeled the CSR unit as the ‘‘de-sales department.’’ This led to overt conflict and frustration on the part of activist-employees in the CSR unit who believed that there were latent links between the two logics that could be exploited. Of course, they were also painfully aware that their interests were marginalized in the organization. Thus, CSR unit employees faced a situation common to other intraorganizational activists (e.g., see Creed & Scully, 2000 on gay and lesbian employees), often requiring astute efforts of tempered radicals (Meyerson & Scully, 1995) to foment change. In order to convince organizational audiences (e.g., top management) that the social responsibility and market logics could be bridged, the CSR unit

220 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

drew upon wider efforts across the brewing industry as evidence that Carlsberg’s closest competitors (e.g., ABInbev, Heineken, and SABMiller) were all engaged in such efforts already and that Carlsberg was a laggard. In a sense, the CSR unit argued that there was a ‘‘legitimacy imperative’’ – RD simply was ‘‘a given’’ in the industry. For instance, a typical argument made internally by advocates of logic bridging was that: You cannot be a brewer, selling an alcohol product without working with responsible drinking , you have to do it. You can not be a brand company like ours that sell alcohol products to young people, without having a marketing policy that addresses how to act responsibly in the commercial arena, that is a given. Otherwise you might risk loosing your license to operate. (Top executive within communication, 2011)

The emerging perception of Carlsberg Group as lagging behind in this area became increasingly problematic for organizational actors. In many ways, the socially responsible aspect of Carlsberg’s organizational identity was to some degree compartmentalized (Kraatz & Block, 2008; Pratt & Foreman, 2000) in the Carlsberg Foundation; social responsibility apparently had little impact on day–to-day practice. In addition, pressure for an RD approach from subsidiary units began to mount. Many of the subsidiary companies had been working with the concept on a local scale individually and through industry associations, and they began making inquiries about a more strategic group way of approaching RD. As the project manager later expressed: ‘‘Today we are just sitting there and (re-) inventing the wheel all the time’’ (RD project manager, 2011). Further, the idea of moving toward a more global approach to RD coheres with other simultaneous agendas within the group such as the centralization of various parts of the supply chain and the initiative to unite the entire organization under a common corporate brand.

Institutional Bricolage and the Reconstitution of Organizational Identity The notion of bricolage is often invoked to conceptualize how a particular actor or entrepreneur brings different elements together to create something novel. However, in the case of intraorganizational institutional bricolage, it is necessarily a group process that requires negotiation among a variety of actors. That is, institutional bricolage must be understood as a social process where actors create a ‘‘solution’’ to a problem related to conflicting logics by ‘‘combining’’ existing elements from more than one extant logic to guide future behavior and practice – thus reconstituting organizational

Strange Brew

221

identity (Schultz & Hernes, 2012). However, it cannot be a group of actors working in solitude or isolation. They must interact and debate with different members across the organization. In our case, the bricoleurs are all organizational actors who actively take part in constructing a response to institutional complexity and who are intent on redefining Carlsberg’s identity in a way that blends social responsibility and market logics. As a result, bricoleurs had to develop an understanding of, and navigate, different logics, as well as enroll other key actors who were committed (wittingly or unwittingly) to one logic or another as this extract from the project manager early in the development process clearly articulates: I can’t change the world on my own just by standing on a beer crate telling the glad tidings. There need to be others who also address this (responsible drinking) in their own words, and in their own way, who also have this as a conviction, and as a value, and as a goal. So every time we succeed in getting someone like Camilla ((pseudonym) manager in Group Sales, Marketing, and Innovation) to say: ‘‘this is something that I want to work with and for,’’ we are one step closer to the goal. (RD project manager, 2011)

The constellation of bricoleurs may even change during the process as the direction and the outcome start to crystallize. In the case of Carlsberg, the initial part of the development process was dominated by actors who were primarily influenced by, or embedded in, the social responsibility logic, while latter parts of the process involved actors associated with both logics. The key element in this regard is that the constellation of bricoleurs represent the different logics that are being ‘‘put together,’’ and it is through their dynamic social interactions and negotiations that institutional logics get combined in one way rather than another. Although the group of bricoleurs may be dynamic, there may still be a few bricoleurs who are active throughout the process, adding some level of continuity to the overall process. At Carlsberg, the hiring of a best practice manager, the RD project manager (lead bricoleur), marked the beginning of the development of the new RD strategy. The RD project manager had been working for a hard liquor company, which has distinguished itself within the alcohol industry by successfully linking responsible drinking and corporate branding. The spirits company had a global campaign urging people to ‘‘enjoy their brands responsibly,’’ with the corporate brand as the sender of the message. Duymedjian and Ru¨ling (2010, p. 148) proposed that the ‘‘manager-bricoleur holds intimate knowledge of the human, material, and symbolic resources of their organization,’’ but Carlsberg’s lead bricoleur had intimate knowledge of a specialized area or problem with which the organization needed to deal. However, he did not have an intimate

222 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

knowledge of the organization, instead he relied on other bricoleurs and their knowledge of the organization, and constantly interacted with different organizational members across headquarters to get a sense of the latent connections (Duymedjian & Ru¨ling, 2010) between the two logics. ‘‘Making do’’: Reforming Brand Image In the current literature, ‘‘Making do’’ (Le´vi-Strauss, 1962) is understood to imply that bricoleurs are more prone to act and actively engage in addressing problems or seizing opportunities rather than deliberating about whether the resources at hand can create a workable outcome (Baker and Nelson, 2005, p. 334). This implies a more dynamic and interactive creative process and an outcome that relies on prior approaches and artifacts (Garud & Karnøe, 2003) as well as existing practices. In our case, the process of figuring out how to respond to RD, and negotiate between market and social responsibility logics led to a coupling of RD efforts to broader issues related to brand image. This was also necessitated by the fact that the response to RD and the development of the Guide Book became animated by broader efforts to reconstitute the identity of the organization as a whole. The resources put aside for the Group’s RD initiative were at first very limited. In the hiring process, while looking for the right candidate to lead the RD project, Carlsberg strategically reviewed the future challenges of their business. Through this review process, the senior management of the group came to realize that they had a large challenge in relation to the image of beer as a category, a tendency that they saw in Northern and Western Europe and emerging in Eastern Europe as well. From their market statistics, they could see a change: previously, beer was understood to be a good wholesome product, but today young people consider hard liquor to be more ‘‘cool,’’ while mature males tend to change their preference to wine as they get older. The senior management decided that they had to revitalize the image of beer. They discerned that the image of beer needed to consist of a range of aspects including social responsibility. At this point, there were limited resources to deal with the image issue; therefore, management decided to upgrade the ‘‘Responsible Drinking’’ position in the CSR department to include a wider business development scope, the ultimate aim being to improve the image of beer. Hence, the idea of RD and the very development of the RD Guide Book were imagined as a component of a larger project of transforming the image of beer. Case examples and practices from around the organization were applied as ‘‘evidence’’ that RD can be done. Part of the work of collecting the cases had already been undertaken earlier by one of the associate bricoleurs as it

Strange Brew

223

was noted ‘‘his drawer is full of cases.’’ These cases had been collected in relation to an earlier development of a ‘‘marketing communication policy’’ and a guidebook intended for the same audience – marketers within and across the group. Importantly for research on bricolage, ‘‘making do’’ in the context of institutional bricolage involves the pursuit of ‘‘new’’ (in the given context) or unorthodox ideas (see also Baker and Nelson, 2005, p. 357). Co-Shaping: Uniting the Organization The element of bricolage that has been adopted most widely is the idea of combining resources for new purposes (e.g., Boxenbaum & Rouleau, 2011; Strandgaard-Pedersen & Dobbin, 2006; Weick, 1998). Building on this work, we have emphasized that institutional bricolage importantly involves the reconstitution of organizational identity. In the Guide Book development process, the organizational actors at headquarters combined existing resources in two distinct ways: By combining cross-functional resources, models, and agendas that represent the two distinct logics and by drawing on the existing practices of subunits. As the efforts unfolded, interactions across the organization proved crucial not only to the Guide Book development but also, perhaps more importantly, as a key resource activated by the bricoleurs in their attempt to reconstitute organizational identity. After all, the transformation of organizational identity requires the building of solidarity and commitment to new ideas and practices throughout an organization. As the Guide Book developed, organizational actors carefully reviewed existing organizational practices within the area of RD. In this process, they relied heavily on interaction with local CSR champions across different subunits to provide concrete examples of RD initiatives. Some local subsidiaries provided information and examples for a ‘‘best practice catalogue,’’ one of the Guide Book’s supporting documents. They provided background information on RD practices, details of the actual initiatives, their impact, as well as contact information of the people with experience in carrying out RD in practice. Further, regional communication and CSR managers were asked to comment and provide feedback on the Guide Book based on their local experience and knowledge. Consequently, the construction of the RD Guide Book needs to be understood in the context in which it was created. Carlsberg Group has been growing rapidly over the last ten years, and the idea of reaching some level of integration across subunits had become a major priority for the company. Thus, the involvement of regional communication managers and subunits in the development of the Guide Book was thought of as a way to establish,

224 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

maintain, and substantiate the relationship between headquarters and subunits. In this regard, existing local practices were gathered and ‘‘objectified’’ (Czarniawska & Joerges, 1996) so that they could travel within and across the group and materialize in new practices. Thus, consideration and inclusion of local practices served two purposes: To sustain the linkage between headquarters and subunits as well as to serve as an argument and an illustration of ‘‘who we are’’ by virtue of ‘‘what we do’’. The idea of co-shaping is closely linked to the aforementioned ‘‘combination of resources for new purposes’’; still they differ because coshaping is not just a question of combining different resources. What defines coshaping is that actors with different interests are involved in the actual development process, which means that the outcome may be more legitimate within the different represented logics (Garud & Karnøe, 2003). In a comparative study of the creation of wind turbines in Denmark and the United States, Garud and Karnøe (2003) deduce that bricolage is a form of entrepreneurship that is emergent and allows for a mutual coshaping of, in their case, technology. Bricolage is therefore understood as a ‘‘y process of moving ahead on the basis of inputs of actors who possess local knowledge, but through their interactions, are able to gradually transform emerging paths to higher degrees of functionality’’ (2003, p. 296). Their point is that the shaping of the process and the outcome occurred at several interaction points between different interest groups (e.g., floor workers, designers, and policymakers). Garud and Karnøe (2003) note that entrepreneurship through a bricolage process may be particularly relevant in situations where there is complex and nonlinear dynamics between the different actors, artifacts, and rules. Their study thus suggests that the concept of bricolage may be particularly helpful in the study of institutional complexity. While meeting and negotiating, organizational actors were able to find some common ground in the perceived threat to the legitimacy of the organization related to the theme of alcohol abuse and in the social responsibility logic as an important historical aspect of organizational identity. However, they brought different understandings of the RD problem to the negotiations. The marketing employees saw ‘‘responsible drinking’’ as a problem of risk avoidance. This problem was evident in a discussion during a meeting: ‘‘what mindset do we enter with? do we see this as de-sales, or new opportunities to gain a competitive advantage?’’ (Observation notes, 2010). The sales and marketing representative at the meeting was very skeptical at the beginning of the meeting and said: ‘‘I don’t think that it (RD) can be sectioned out like that, it cannot be used as marketing, everybody is doing it (y). There are some consequences,

225

Strange Brew

though: How significant are the negative consequences?’’ (Observation notes, 2010). At this point it became clear that to the marketing representative, RD is not a thing that represents an opportunity; instead, it should be framed as ‘‘risk avoidance.’’ As the meeting continued and appealing examples and cases of a coupling of social responsibility and market logics were presented, this person started to open up and wanted to help operationalize RD ideas. She even offered to help frame the cases and shape the RD Guide Book with a market logic vocabulary. Following this meeting, the Guide Book team at headquarters engaged in a process of institutional bricolage to align RD efforts (the social responsibility logic) with the dominant market logic by creating what is commonly referred to as the ‘‘business case for responsible drinking.’’ The business case would make RD a new concept that was no longer considered to be ‘‘fluffy’’ and ‘‘vague,’’ but something that made sense within the dominant market logic. The key actors within the RD project expect that the development of ‘‘the business case’’ would make it possible to implement RD in organizational practice as emphasized in this statement: We (in Group HQ) want to move away from ‘‘responsible drinking’’ as being a question of belief – to being a business question! One way to get people on board is to make a shift in the way we talk about it. All other topics in our company are driven by business and financial targets, primarily financial targets, so that is what we are brought up with. Therefore we must link ‘‘responsible drinking’’ to that agenda, then it will be much easier to get people on board. (Manager in Group Sales, Marketing, and Innovation, 2011)

Hence, linking RD to the market logic was really a question of making the RD concept something that was immediately relevant and understandable for the organizational members who have the power and the resources to carry out RD initiatives. Concurrently, the effort involved making the RD agenda come alive and become part of the organization’s interaction with varied audiences, and as we have argued, a core aspect of their reconstituted identity.

DISCUSSION AND CONCLUSION In this paper, we have sought to contribute to the institutional logics perspective (Thornton et al., 2012) by exploring the intraorganizational processes by which institutional complexity is resolved. Our understanding of the mechanisms and processes by which organizational actors deal with multiple institutional logics remains limited (Greenwood et al., 2011;

226 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

Pache & Santos, 2010). We drew on the bricolage literature (Boxenbaum & Rouleau, 2011; Duymedjian & Ru¨ling, 2010) to focalize theoretical and empirical research attention on the importance of artifact construction as a key aspect by which institutional complexity gets worked out inside organizations. To distinguish our effort from bricolage research more generally, we label the processes we analyze as institutional bricolage. Our main argument is that institutional bricolage – resolving problems related to institutional complexity inside an organization – fundamentally involves the reconstitution of an organization’s identity. While organizational identity has been identified as a critical aspect of the institutional logics perspective, little attention has been paid to the processes by which extant identities get challenged and reconstructed. As our case of Carlsberg suggests, such situations are moments of acute uncertainty and opportunities for reflection. A group of organizational members were consciously and deliberately trying to change (e.g., via ‘‘making do’’ and ‘‘co-shaping’’) organizational practice as well as audience perceptions of their product. We highlighted how the marginalized social responsibility logic offered a slightly different path that turned out to be symbolically and materially powerful because it was anchored in core foundational aspects of the organization and its heritage (Schultz & Hernes, 2012). While we do not know at this stage whether efforts to reconstitute identity will be successful, it appears that the ability to connect different spaces and the future to the past in this way facilitated the ability to substantively engage in a process of institutional bricolage that combined the social responsibility and market logics (see also Borum & Westenholz, 1995; Garud & Karnøe, 2003). The recombination of existing resources was salient throughout the process; existing models were brought in and modified to fit with other existing models to craft a solution to a problem that the organization faced. This provides an intraorganizational correlate to the idea that established institutional paths provide resources and possibilities for change and the creation of new institutional configurations (Schneiberg, 2007). Pache and Santos (2010) make the claim that differential internal representations may lead to organizational paralyses or breakup. Although this may be the case in some instances, our case suggests a qualification that, under certain conditions, organizational actors may alternatively engage in negotiation and institutional bricolage where they seek to reconstitute their identity to address institutional complexity. When organizational actors start to negotiate across multiple logics, the logics may become more dynamic and take a form that facilitates integration; as we argued, this may

Strange Brew

227

depend on a somewhat profound trigger that makes organizational legitimacy problematic, providing a collective opportunity to renegotiate the relationship among logics. As Selznick (1957, p. 21) proposed, when the organization is understood as an institution, self-maintenance becomes more than just an issue of organizational survival; it becomes a struggle to protect the uniqueness of the organization as it faces changed circumstances and new problems. Therefore, when logic bridging is framed as a means to somehow preserve at least part of the uniqueness of the organization’s identity, it gives diverse organizational members a common goal to work toward. We also highlighted the importance of social mobilization inside the organization – how marginalized organizational actors, namely the communication and CSR people, deliberately made the conversations and negotiations around the Guide Book, an issue of organizational identity. The tinkering and blending of techniques and models infused with different logics was consciously intended to erode the boundaries between the once separated ideas and practices, and strike a new balance between the different logics. Thus, our case observations indicate that a key condition for institutional bricolage of logics within the organization is that internal activists have to be able to skillfully frame key issues as centrally involving identity – ‘‘who we are’’ and ‘‘what we do’’ – and then try to convince others that a marginalized (and to some extend compartmentalized) logic is a core and foundational organizational identity trait. In addition, our case suggests how institutional bricolage may entail an active search for resources and connections across varied organizational units. For instance, examples of subsidiaries’ local RD practices were used to reaffirm that the concept of RD and the social responsibility logic were not (only) a new headquarter’s idea coercively implemented top down. Instead, it gave the Guide Book development process a more democratic and bottom-up flavor of inclusion. This is an aspect that is especially prominent in the case selections from around the Group and in notes like this: ‘‘Contributions to the catalogue were kindly made by your colleagues across the Carlsberg Group.’’ Messages that serve the same purpose mentioned above, to denote that the social responsibility logic is already an ingrained aspect of the organizational identity, and not just in headquarters where the philanthropic heritage is materially present but also in ‘‘your colleagues’’ practices. While we have argued that the bridging of logics requires a reconstitution of organizational identity, future research should focus on the detailed and varied microprocesses and cross-level mechanisms through which

228 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

institutional bricolage happens. We suggest that future research address questions related to how organizational actors might use different identity claims in their response to complexity; how and at what levels – subunit, organizational, or industry, might we consider identity to be relevant in relation to institutional complexity? And how organizational and industry (collective) identities interact and influence organizational responses to complexity? Finally, scholars of the management of multinational enterprises (MNEs) may find this study useful in so far as the management at Carlsberg headquarters chose to engage local subsidiaries and draw from their practice in conceptualizing their ‘‘responsible drinking’’ initiative. In that regard, they sought out more of a consensus building approach to managing the issue. They signaled that they were aware that the issue may be somewhat different across different locations, but that a unified approach was necessary. It would be especially interesting to have more empirical studies of MNEs to uncover how multiple logics are managed across different societal contexts by the same organization. We know little about how MNE headquarters influence subsidiaries’ management of complexity. How do complexity issues travel within and across organizational units? Are logics that influence the different subsidiaries comparable across different societal contexts? For instance, is the social responsibility logic the same in Russia and the United Kingdom? Do efforts of identity reconstitution initiated at headquarters resonate with various local subsidiaries? While our study focused more on detailing institutional bricolage and identity reconstitution efforts at an MNE headquarters, it is important to understand how these processes play out across the entire enterprise. Research of this nature would help expand the scope of the institutional logics perspective while also enhancing our understanding of MNEs and challenges related to globalization.

NOTES 1. The take-over was made together with Heineken and later the activities were split between the two. 2. In April 2012 the Carlsberg City was sold for 2.5 billion Danish kroner, around 400 million U.S. dollar, to a consortium (including Carlsberg who remain a 25 percent ownership). A new neighborhood in Copenhagen is to be developed in the area with around 3000 new accommodations in different price ranges. A number of the historical buildings on the site are not part of the disposal and will remain the property of Carlsberg Group.

229

Strange Brew

3. The Botanical Garden is the largest collection of living plants in Denmark, a collection that is used for research, teaching and public information. 4. ‘‘Panimoliitto’’ is the Finish Federation of the Brewing and Soft Drinks Industry. 5. Dark markets are considered to have high regulations or total prohibition leaving little to no room for marketing communication, for example, no outdoor promotion or TV promotion. In gray markets there is some legislation and restrictions, for example. TV adds after 9 p.m. or no humans in commercial promotion of alcohol. 6. The data collection is still ongoing. Data collected in local subsidiaries has not been included directly in this paper. 7. These meetings also included discussions about the data collection and the findings from subsidiaries. 8. The RD Guide Book consists of the documents: ‘‘The RD Guide Book,’’ ‘‘Why to do RD,’’ ‘‘How to do PR,’’ as well as four schemas to be used before and after planning an RD initiative: ‘‘PR Campaign Mandate,’’ ‘‘PR Campaign Media Outreach Checklist,’’ ‘‘PR Campaign Servicing Media Checklist,’’ and ‘‘PR evaluation Scheme.’’

ACKNOWLEDGMENTS We thank Eva Boxenbaum, Renate Meyer, Majken Schultz, Frans Bevort, and participants at the ABC seminar in Boston in 2011 for their helpful comments on this paper and an earlier version of it. We thank Carlsberg Group, the department of ‘‘Group CSR and Public Affairs’’ and the people from Carlsberg Group who have made this study possible. This research was supported, in part, by the Tuborg Foundation.

REFERENCES Baker, T., & Nelson, R. E. (2005). Creating something from nothing: Resource construction through entrepreneurial bricolage. Administrative Science Quarterly, 50(3), 329–366. Battilana, J., & Dorado, S. (2010). Building sustainable hybrid organizations: The case of commercial microfinance organizations. Academy of Management. 53(6), 1419–1440. Bechky, B. A. (2003). Sharing meaning across occupational communities: The transformation of understanding on a production floor. Organization Science, 14(3), 312–330. Binder, A. (2007). For love and money: Organizations’ creative responses to multiple environmental logics. Theory & Society, 36(6), 547–571. doi: 10.1007/s11186-007-9045-x Borum, F., & Westenholz, A. (1995). The incorporation of multiple institutional models: Organizational field multiplicity and the role of actors. In W. R. Scott & S. Christensen (Eds.), The institutional construction of organizations, international and longitudinal studies (pp. 113–131). Thousand Oaks, CA: Sage.

230 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

Boxenbaum, E., & Rouleau, L. (2011). New knowledge products as bricolage: Metaphors and scripts in organizational theory. Academy of Management Review, 36(2), 272–296. doi: 10.5465/AMR.2011.59330898 Brewers of Europe. (2010). New study shows beer tax increases would jeopardise economic recovery. In news releases. Retrieved from http://www.brewersofeurope.org/asp/news room/l1.asp?doc_id¼217. Accessed on October 29, 2012. Campbell, J. L. (2004). Institutional change and globalization. Princeton, NJ:. Princeton University Press. Carlsberg Group (2010). Annual report Carlsberg A/S. Copenhagen. Christensen, S., & Molin, J. (1995). Origin and transformation of organizations: Institutional analysis of the Danish red cross. In W. R. Scott & S. Christensen (Eds.), The institutional construction of organizations, international and longitudinal studies (pp. 67–90). Thousand Oaks, CA: Sage. Creed, W. E. D., & Scully, M. A. (2000). Songs of ourselves. Journal of Management Inquiry, 9(4), 391–412. Czarniawska, B., & Joerges, B. (1996). Travels of ideas. In B. Czarniawska & G. Sevo´n (Eds.), Translating organizational change (pp. 13–48). Berlin: Walter de Gruyter. DiMaggio, P., & Powell, W. (1983/1991). The iron cage revisited: Institutional isomorphism and collective rationality. In P. DiMaggio & W. Powell (Eds.), The new institutionalism in organizational analysis (pp. 63–82). The University of Chicago Press. Douglas, M. (1986). How institutions think. Syracuse, NY: SyracuseUniversityPress. Dutton, Jane E., & Dukerich, Janet M. (1991). Keeping an eye on the mirror: Image and identity in organizational adaptation. The Academy of Management Journal, 34(3), 517–554. Duymedjian, R., & Ru¨ling, C. (2010). Towards a foundation of bricolage in organization and management theory. Organization Studies, 31(2), 133–151. doi: 10.1177/0170840609347051 Elsbach, K. D., & Kramer, R. M. (1996). Members’ responses to organizational identity threats: Encountering and countering the business week rankings. Administrative Science Quarterly, 41(3), 442–476. Friedland, R., & Alford, R. (1991). Bringing society back in Symbols, practices, and institutional contradictions. In P. DiMaggio & W. Powell (Eds.), The new institutionalism in organizational analysis (pp. 232–263). Chicago, IL: The University of Chicago Press. Garud, R., & Karnøe, P. (2003). Bricolage versus breakthrough: Distributed and embedded agency in technology entrepreneurship. Research Policy, 32(2), 277. Gioia, D. A., & Thomas, J. B. (1996). Identity, image, and issue interpretation: Sensemaking during strategic change in academia. Administrative Science Quarterly, 41(3), 370–403. Glynn, M. A. (2008). Beyond constraint: How institutions enable identities. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The SAGE handbook of organizational institutionalism (pp. 413–430). London: Sage. Glynn, M. A., & Abzug, R. (2002). Institutionalizing identity: Symbolic isomorphism and organizational names. The Academy of Management Journal, 45(1), 267–280. Glynn, M. A., & Lounsbury, M. (2005). From the critics corner: Logic blending, discursive change and authenticity in a cultural production system. Journal of Management Studies, 42(5), 1031–1055. doi: 10.1111/j.1467-6486.2005.00531.x

Strange Brew

231

Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E. R., & Lounsbury, M. (2011). Institutional complexity and organizational responses. The Academy of Management Annals, 5(1), 317–371. doi: 10.1080/19416520.2011.590299 Jarzabkowski, P. (2005). Strategy as practice: An activity-based approach. London: Sage. Kraatz, M., & Block, E. (2008). Organizational implication of institutional pluralism. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The SAGE handbook of organizational institutionalism (pp. 243–274). London: Sage. Lounsbury, M. (2001). Institutional sources of practice variation: Staffing college and university recycling programs. Administrative Science Quarterly, 46, 29–56. Lounsbury, M. (2007). A tale of two cities: Competing logics and practice variation in the professionalizing of mutual funds. Academy of Management Journal, 50(2), 289–307. doi: 10.5465/AMJ.2007.24634436 Lounsbury, M., & Crumley, E. T. (2007). New practice creation: An institutional perspective on innovation. Organization Studies, 28(7), 993–1012. Lounsbury, M., & Glynn, M. A. (2001). Cultural entrepreneurship: Stories, legitimacy, and the acquisition of resources. Strategic Management Journal, 22(6), 545. doi: 10.1002/smj.188 Le´vi-Strauss, C. (1962/1966). The savage mind [La Pense´e savage] (Weidenfield, George and Nicholson Ltd., Trans.). Chicago, IL: University of Chicago Press. Meyerson, D. E., & Scully, M. A. (1995). Tempered radicalism and the politics of ambivalence and change. Organization Science, 6(5), 585–600. Navis, C., & Glynn, M. A. (2010). How new market categories emerge: Temporal dynamics of legitimacy, identity, and entrepreneurship in satellite radio, 1990–2005. Administrative Science Quarterly, 55(3), 439–471. Ocasio, W. (2011). Attention to attention. Organization Science, 22(5), 1286–1296. doi: 10.1287/ orsc.1100.0602 Oliver, C. (1991). Strategic responses to institutional processes. The Academy of Management Review, 16(1), 145–179. Pache, A., & Santos, F. (2010). When worlds collide: The internal dynamics of organizational responses to conflicting institutional demands. Academy of Management Review, 35(3), 455–476. Panimoliitto (2010). Beer taxation. Retrieved from http://www.panimoliitto.fi/panimoliitto/en/ beer_taxation. Accessed on May 17, 2012. Pedersen, J. S., & Dobbin, F. (2006). In search of identity and legitimation: Bridging organizational culture and neoinstitutionalism. American Behavioral Scientist, 49(7), 897–907. doi: 10.1177/0002764205284798 Pratt, M. G., & Foreman, P. O. (2000). Classifying managerial responses to multiple organizational identities. The Academy of Management Review, 25(1), 18–42. Rao, H., Monin, P., & Durand, R. (2005). Border crossing: Bricolage and the erosion of categorical boundaries in French gastronomy. American Sociological Review, 70(6), 968–991. Ravasi, D., & Schultz, M. (2006). Responding to organizational identity threats: Exploring the role of organizational culture. Academy of Management Journal, 49(3), 433–458. Reay, T., & Hinings, C. R. (2005). The recomposition of an organizational field: Health care in Alberta. Organization Studies, 26(3), 351–384. doi: 10.1177/0170840605050872 Reay, T., & Hinings, C. R. (2009). Managing the rivalry of competing institutional logics. Organization Studies, 30(6), 629–652. doi: 10.1177/0170840609104803 Rindova, V., Dalpiaz, E., & Ravasi, D. (2011). A cultural quest: A study of organizational use of new cultural resources in strategy formation. Organization Science, 22, 413–431.

232 LÆRKE HØJGAARD CHRISTIANSEN AND MICHAEL LOUNSBURY

Schneiberg, M. (2007). What’s on the path? Path dependence, organizational diversity and the problem of institutional change in the US economy, 1900–1950. Socio-Economic Review, 5(1), 47–80. doi: 10.1093/ser/mwl006 Schultz, M., & Hernes, T. (2012). A temporal perspective on organizational identity. Organization Science. doi: 10.1287/orsc.1110.0731 Scott, W. R. (2008). Institutions and organizations: Ideas and interests (3rd ed.). London: Sage. Selznick, P. (1957). Leadership in administration: A sociological interpretation. Berkeley, CA: University of California Press. Smets, M., & Jarzabkowski, P. (forthcoming). (Re)-constructing institutional complexity in practice: A relational model of institutional work and complexity. Human Relations. Smets, M., Morris, T., & Greenwood, R. (2012). From practice to field: A multi-level model of practice-driven institutional change. Academy of Management Journal, 55, 877–904. Star, S. L., & Griesemer, J. R. (1989). Institutional ecology, ‘‘Translations’’ and boundary objects: Amateurs and professionals in Berkeley’s museum of vertebrate zoology, 1907– 39. Social Studies of Science, 19(3), 387–420. doi: 10.1177/030631289019003001 Swidler, A. (1986). Culture in action: Symbols and strategies. American Sociological Review, 51(2), 273–286. Thornton, P., & Ocasio, W. (2008). Institutional logics. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The SAGE handbook of organizational institutionalism (pp. 99– 129). London: Sage. Thornton, P. H., Jones, C., & Kury, K. (2005). Institutional logics and institutional change in organizations: Transformation in accounting, architecture, and publishing. Research in the Sociology of Organizations, 23, 125. Thornton, P. H., & Ocasio, W. (1999). Institutional logics and the historical contingency of power in organizations: Executive succession in the higher education publishing industry, 1958-1990. American Journal of Sociology, 105(3), 801–843. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective – A new approach to culture, structure, and process. New York, NY: Oxford University Press. Vergne, J. (2012). Stigmatized categories and public disapproval of organizations: A mixedmethods study of the global arms industry, 1996–2007. Academy of Management Journal, 55(5), 1027–1052. doi: 10.5465/amj.2010.0599 Weick, K. E. (1998). Improvisation as a mindset for organizational analysis. Organization Science, 9(5), 543–555. Westenholz, A., Strandgaard Pedersen, J., & Dobbin, F. (2006). Introduction institutions in the making identity, power, and the emergence of new organizational forms. American Behavioral Scientist, 49(7), 889–896.

INSTANTIATION OF INSTITUTIONAL LOGICS: THE ‘‘BUSINESS CASE’’ FOR DIVERSITY AND THE PREVALENCE OF DIVERSITY MENTORING PRACTICES Shawna Vican and Kim Pernell-Gallagher ABSTRACT Building on recent theoretical insights from the institutional logics perspective, we examine organizational dynamics in the loosely coupled field of corporate diversity management to develop a theory of the process of logic instantiation. We consider a case in which firms subscribed to the same institutional logic, the business performance logic for diversity management, but varied in adoption of diversity mentoring practices. Employing an inductive and iterative approach to analyze over 50 interviews with diversity managers at large U.S. corporations, we explain how four organizational factors mediated the process of logic instantiation in these firms: (1) the diversity manager’s interpretation and framing of the business performance logic, (2) the formal diversity goals of the

Institutional Logics in Action, Part B Research in the Sociology of Organizations, Volume 39B, 233–273 Copyright r 2013 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0733-558X/doi:10.1108/S0733-558X(2013)0039A&B021

233

234

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

firm, (3) the relative organizational power of the diversity manager, and (4) the accepted definition of ‘‘diversity.’’ We discuss implications for theories of social action and diversity management. Keywords: Institutional logics; institutional complexity; practice; institution

Variation in practices among firms that subscribe to a common institutional logic presents a puzzle for the literature on institutions and organizations. The earliest research in the neoinstitutional paradigm sought to explain institutional isomorphism, or convergence in organizational structures, practices, and forms within an organizational field (DiMaggio & Powell 1983; Meyer & Rowan, 1977). Neoinstitutional theory provided a novel explanation for the diffusion of practices: firms seek to adopt practices that embody the taken-for-granted understandings of rational or appropriate behavior advanced by institutions like the state, industry, or the professions (DiMaggio & Powell, 1983). However, critics of the neoinstitutional approach have argued this framing has ‘‘black boxed’’ organizational practices, framing them as ‘‘institutionalized, static elements,’’ and fully derivative of the larger institutional logics that govern the field (Thornton, Ocasio, & Lounsbury, 2012, p. 132). Subsequent research on ‘‘institutional entrepreneurs’’ emerged as a corrective, highlighting the role of powerful actors as change agents in creating and transforming institutions (see Hardy & Maguire, 2008, for a review). The institutional logics approach has reconciled these competing perspectives. This approach rejects both the idea that practices are completely derivative of institutional logics and the idea that practices are the brainchildren of entrepreneurs who operate outside of larger institutional constraints. Instead, the institutional logics perspective frames logics and practices as ‘‘loosely coupled,’’ arguing that logics shape, but do not determine organizational practices (Orton & Weick, 1990). The particular practices selected by an organization should not be seen as purely local, but also reflective of the available logics in an institutional field (Thornton et al., 2012). The duality of the relationship between logics and practices is further evident in the argument that practices are both material instantiations of logics, but also a means of shifting or challenging logics. Socially embedded actors may create or transform practices, using these ‘‘as exemplars in creating, reproducing, and transforming institutional logics’’ (Thornton et al., 2012, p. 127).

Instantiation of Institutional Logics

235

Theoretical advancement in the institutional logics literature has outpaced empirical investigation of these propositions. There is a particularly strong call for research that analyzes how institutional logics ‘‘relate to the intraorganizational dynamics of practices and identities’’ (Thornton et al., 2012, p. 134). Practice and identity are core theoretical concepts in the institutional logics approach, but ‘‘most research to date has not effectively analyzed how institutional logics shape and are shaped by the material instantiations of logics – the practices and identities of concrete actors’’ (Thornton et al., 2012, p. 128). Recent empirical research has explored how multiple institutional logics, and their competing demands, may promote varied organizational responses (Reay & Hinings, 2009). However, we know very little about how internal organizational dynamics shape the instantiation of an institutional logic (Greenwood, Raynard, Kodeih, Micoletta, & Lounsbury, 2011; Pache & Santos, 2010; Thornton et al., 2012, p. 147). We gain analytic leverage on this question by analyzing a case where over 50 large U.S. corporations subscribed to the same institutional logic, but implemented different diversity practices. We draw from this inductive analysis to advance a more nuanced and refined theory for how organizational dynamics shape the process of logic instantiation, or the translation of institutional logics into material practices. We develop a model of the process of logic instantiation, highlighting the importance of organizational dynamics for shaping material manifestations of an institutional logic. First, we argue for the importance of actor interpretations of an institutional logic. Logics are composed of multiple elements or ‘‘building blocks,’’ which together make up the organizing principles of the logic (Thornton, 2004; Thornton et al., 2012). Existing research has yet to explore when, whether, and how actors draw from different elements of an institutional logic – and whether these different logic-element combinations matter for the instantiation of the logic. We argue that organizational decision-makers draw from different foci or combinations of logic elements, and that these different combinations influence logic instantiation by directing organizational attention toward particular practices and away from others. Second, we argue that an organization’s formal goals play an important role in moving institutional logics from the ideational realm to the material realm. The way that a firm interprets an institutional logic is closely linked to its selection of formal goals; however, these factors are not perfectly correlated. Different logic-element combinations make a firm more or less likely to embrace a particular practice, but we suggest that a firm’s formal goals shape how it implements the practice. Third, we argue

236

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

that the relative power of key organizational actors influences the selection and the implementation of practices. We suggest that low-power managers are forced to turn to alternative strategies to achieve their goals, which often results in the use of different practices or different methods of implementing a practice. Fourth, we argue that the relative salience of a competing field frame can influence logic instantiation by leading otherwise likely users of a practice to reject the practice and pursue alternatives.

LOGICS, ORGANIZATIONAL DYNAMICS, AND PRACTICES1 The Institutional Logic Perspective We use Thornton and Ocasio’s (1999) definition of institutional logics as ‘‘the socially constructed, historical pattern of material practices, assumptions, values, beliefs, and rules by which individuals produce and reproduce their material subsistence, organize time and space, and provide meaning to their social reality’’ (Thornton & Ocasio, 1999, p. 804). Institutional logics consist of both the shared, field-level, taken-for-granted understandings about rational organizational behavior or appropriate organizational action and the symbols and material practices that serve as the core manifestations of these understandings. Institutional logics shape organizational behavior by providing the ‘‘rules of the game’’ that influence decision making (Powell & DiMaggio, 1991). Logics highlight or downplay the saliency of particular problems, issues, or sources of power (Ocasio, 1997) and inform solutions to these problems: the ‘‘assumptions, values, beliefs, and rules that comprise institutional logics determine what answers and solutions are available and appropriate in controlling economic and political activity in organizations’’ (Thornton & Ocasio, 1999, p. 806). In short, institutional logics can influence the adoption of an organizational practice indirectly, by legitimating certain bases of organizational power or by defining particular issues as problematic in the eyes of decision-makers, or directly, by providing material that actors may use to theorize the practice as an effective solution (Strang & Meyer, 1993). In this paper, we focus on an institutional logic that spans industry boundaries to shape the activities of many large U.S. corporations: the business performance logic.

Instantiation of Institutional Logics

237

Institutional Logics and Organizational Dynamics The institutional logics perspective highlights three explanations for why institutional logics may manifest in different material practices: ‘‘exogenous shifts in logics, collective identities or practices in an institutional field,’’ performativity, or internal organizational political dynamics (Thornton et al., 2012, p. 137). Our model of the logic instantiation process focuses on the latter factor, organizational dynamics, for two reasons. First, more empirical research on practice variation has investigated the influence of field-level shifts in logics, and the simultaneous existence of multiple logics, than the influence of internal organizational dynamics. At the national level, Fligstein (1990) finds that shifts in logics of corporate control produced corresponding shifts in firm structure among American corporations. At the industry level, Thornton and Ocasio (1999) find that changes in the institutional logic of the higher education publishing industry led to changes in determinants of executive power and succession, and Lounsbury (2007) demonstrates that different performance logics in the Boston and New York mutual fund industries inspired Boston and New York funds to adopt different investment and money management strategies. In contrast to studies of how field-level change catalyzes shifts in organizational structure and practice, the question of how organizational dynamics shape logic instantiation has received less empirical attention (Thornton et al., 2012). Recent studies of institutional complexity, or how firms respond to multiple and competing logics, demonstrate that there is indeed variation in organizational response (Greenwood et al., 2011). Our rich, firm-level interview data is well suited to answer questions about internal organizational processes, and as such the process through which organizational dynamics shape the instantiation of logics. Second, to explore the process of logic instantiation, we chose an empirical setting in which organizational dynamics are more likely to matter. Greenwood and Hinings (1996) point to the extent of tight coupling in the organizational field as a characteristic that may lead to greater variation in organizational practices. Tight coupling ‘‘refers to the existence of mechanisms for dissemination and the monitoring of compliance combined with a focused and consistent set of expectations’’ (Greenwood & Hinings, 1996, p. 1029). Fields are loosely coupled when they contain numerous competing requirements or prescriptions, and when organizations face ambiguity (cf. Meyer & Scott, 1983). This perspective suggests that when firms face unclear or numerous choices in a less tightly coupled organizational field, we should expect to find greater variety in

238

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

organizational practices. The loosely coupled field of corporate diversity management provides an ideal site to examine how organizational factors mediate the process of logic instantiation. In the following section, we provide a history of the field of diversity management to explain why we characterize this field as loosely coupled.

Diversity Management: A Loosely Coupled Field We define the field of diversity management as loosely coupled for two reasons. First, the history of the field makes it clear that there are shifting and often ambiguous prescriptions of appropriate activity. Second, the wide range of practices advanced by diversity management professionals leave firms with ample ways to demonstrate compliance. There is limited evidence of how diversity impacts bottom line, and this ambiguity leaves ample room for interpretation how diversity impacts firm performance. The following brief history of diversity management further develops our argument as to the extent of loose coupling in the field. The shifting meaning and scope of corporate diversity management becomes evident in tracing the history of the field. While corporate attention to ‘‘managing diversity’’ began in the 1980s, the roots of these efforts can be traced to Title VII of the Civil Rights Act of 1964 and Equal Employment Opportunity (EEO) efforts that gained traction in the 1970s (Edelman, Fuller, & Mara-Drita, 2001). In the 1970s, the combination of a legally broadened definition of employment discrimination, new efforts to mandate affirmative action plans, and the increasing power of the Equal Employment Opportunity Commission (EEOC) led to increased attention to corporate antidiscrimination efforts. Firms added equal opportunity departments, often housed within human resources, filled with HR professionals as newly minted equal employment experts. These employment professionals used two main strategies to sell their equal opportunity practices: these practices were argued to reduce a firm’s legal liability through demonstrating that the firm was compliant with Affirmative Action law and to enhance the firm’s efficiency (Edelman, 1992). In the 1980s, diversity management rhetoric shifted away from the compliance rationale to privilege efficiency concerns. When the Reagan administration challenged Affirmative Action, ‘‘equal opportunity’’ programs were rebranded as ‘‘diversity management’’ (Dobbin, 2009; Dobbin & Sutton, 1998). By the early 1990s, most ‘‘equal opportunity’’ departments within firms had been recast as ‘‘diversity management’’ departments. The 1990s were characterized by a decisive

Instantiation of Institutional Logics

239

turn away from compliance-based or social justice arguments toward a focus on the ‘‘business case’’ for diversity (Dobbin, 2009). Are organizations with a more diverse workforce actually more productive or successful? The evidence is mixed (see Richard, Kochan, & Mcmillan-Capehart, 2002 for a review of this literature). Wright, Ferris, Hiller, and Kroll (1995) found some performance differences between diverse and less-diverse firms. However, this study compared firms with the best diversity programs to those who had settled discrimination lawsuits. Research on small groups and teams has found both positive and negative effects of diversity. For instance, a pair of studies about top-management diversity in banking found that more diverse firms were more innovative (Bantel & Jackson, 1989), but also higher rates of turnover among top management (Jackson et al., 1991). Recent work by Herring (2009) suggests that racial diversity positively impacted the bottom line for firms in their sample, but it is difficult to parse this causal relationship: more successful firms may devote more time and energy to diversity issues. The one point of consensus in this literature is that the relationship between corporate diversity and business success is not as simple as lay advocates for a ‘‘business case’’ for diversity imply. However, this has not stopped firms from embracing this rhetoric (Kochan et al., 2003).

Mentoring: An Organizational Practice In this paper, we explore how organizational dynamics shape the embrace of a particular organizational practice, the diversity mentoring program, across a sample of firms that drew from a ‘‘business performance’’ logic to justify diversity management efforts. ‘‘Mentoring’’ refers to a one-on-one relationship between a less experienced member of an organization and a more experienced member of the organization, that is ‘‘prototypically intended to advance the personal and professional growth of the less experienced individual’’ (Mullen 1994, cited in Wanberg, Welsh, & Hezlett, 2003). Firms may develop formal mentoring programs for a variety of reasons, like improving employee morale, increasing employee retention, or generating support for organizational change efforts (Blake-Beard, 2001; Catalyst, 1998; Douglas, 1997; Lawlor, 1997). Bragg (1989) estimates that one third of major corporations in the United States have some kind of formal mentoring program. Diversity mentoring is a particular type of formal mentoring program that is specifically designed to meet the needs of marginalized employees. These programs originated in the 1970s to

240

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

ameliorate the exclusion of women and minorities from the resources embedded in informal social networks Diversity mentoring programs experienced a revival in the 1990s (Wernick, 1994; Winterle, 1992), but they are neither uncommon nor highly prevalent today: Kalev, Dobbin, and Kelly (2006) find that 11 percent of a random sample of U.S. firms required to file an EEO-1 report had a formal diversity mentoring program by 2002. Mentoring relationships are thought to promote employee advancement, satisfaction, and productivity in a variety of ways. Mentors advocate for prote´ge´s within the organization, and provide support, informal coaching, and insight into implicit rules of advancement at the firm (Baron & Pfeffer, 1994; Castilla, 2005; Kanter, 1977). However, for female and minority employees, a tendency toward homophily in organizational social networks and the concentration of white men in upper management leads to few informal relationships with those at the top of the organizational hierarchy (Burt, 1998; Ibarra, 1993; Reskin & McBrier, 2000; Sims, 2002). Previous research demonstrates that female and minority employees are less likely to receive informal mentoring than white males (Cox & Nkomo, 1991; Ragins, 1989; Viator, 2001), but minority and female employees have better career outcomes when they do have access to these relationships (Allen, Eby, Poteet, Lentz, & Lima, 2004; Burke & McKeen, 1997; but see Thomas, 2001). Proponents of diversity mentoring programs argue that these programs provide marginalized employees with access to the valuable coaching, sponsorship, and information that they would otherwise struggle to obtain informally. The verdict is still out on whether formal mentoring programs targeted at diverse employees actually improve career outcomes for these employees. Existing research on this topic suffers from methodological flaws: studies do not distinguish between formal or informal mentoring relationships (Wanberg et al., 2003) or fail to control for important alternative explanations. The work of Kalev et al. (2006) provides an exception to this trend: the authors examine the effects of formal diversity-targeted mentoring programs across firms and across time, and find that the adoption of a diversity-targeted mentoring program was associated with a significant increase in the relative representation of black female managers. Regardless of whether these mentoring programs actually work, the variation in their prevalence across firms who subscribe to the same institutional logic presents an interesting theoretical puzzle. Only 22 of the 53 firms in our sample adopted a formal, diversity-targeted mentoring program, although respondents from all of these firms drew from the business performance logic to justify diversity management initiatives. In-depth

241

Instantiation of Institutional Logics

analysis of this theoretically puzzling case should yield new insights into the process of logic instantiation.

ORGANIZATIONAL DYNAMICS AND PRACTICE VARIATION While an institutional logics perspective suggests that organizational dynamics shape the process of logic instantiation, it is not entirely clear which organizational dynamics matter, or how these dynamics operate to shape logic instantiation. In the following section we draw from literature in organizational sociology to highlight particular organizational factors that may mediate how firms translate institutional logics to the local organizational context.

Capacity for Action To explain organizational change, Greenwood and Hinings (1996) develop a theoretical account for how organizational characteristics can mediate responses to similar institutional pressures, arguing that ‘‘radical change cannot occur without the organization having sufficient understanding of the new conceptual destination, its having the skills and competencies required to function in that new destination, and its having the ability to manage how to get to that destination’’ (p. 1040). In short, the extent to which decision-makers understand available institutional logics, and the extent to which they have access to material and symbolic resources that enable them to translate institutional prescriptions into specific goals and practices, may also shape the process of logic instantiation. Theories of institutional entrepreneurship, drawing in part from models of change in the management literature, also highlight the importance of ‘‘developing a vision,’’ mobilizing others, and sustaining the effort in order to institutionalize change (Battilana, Leca, & Boxenbaum, 2009, p. 78). While we view capacity for action as a key element in the logic instantiation process, we argue that more empirical attention is necessary to refine this concept.

Power The institutional logics perspective has long recognized that institutional logics can shape power and political dynamics in firms (Friedland & Alford,

242

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

1991). Institutional logics legitimate particular sources of power and shape which problems or issues are seen as centrally important within organizations. In this way, they influence political struggles (Ocasio, 1997; Thornton & Ocasio, 1999). We build upon this line of research to argue that the relative power of a key organizational actor will influence the process of logic instantiation. A resource dependency view of power implies that actors derive power from control over valued organizational resources (Pfeffer & Salancik, 1978). Borrowing from Scott (2003, p. 310) we take a more relational view of power, defining power as ‘‘the potential for influence.’’ We focus our attention on the social position of a key organizational actor, as this has been shown to enable change efforts (Maguire, Hardy, & Lawrence, 2004). Social position can contribute to power in two ways: through the actor’s formal position in the organizational hierarchy, which affords access to resources, control over decision making, and legitimacy, and through the actor’s informal position in organizational networks, which allows for social ties to powerful actors (Battilana, 2007). For our sample of respondents, there was little variation in formal organizational position. Therefore, we focus on the respondent’s informal position in organizational networks – in particular, the respondent’s relationships with members of upper management. Top-management support had been shown to predict the successful implementation of a number of practices, including information systems (Ragu-Nathan, Apigian, Ragu-Nathan, & Tu, 2004; Young & Jordan, 2008); research and development projects (Green, 1995); and innovations more generally (Bantel & Jackson, 1989).

Collective Identity and Industry Research demonstrates that a firm’s industry affiliation may influence the practices that the firm adopts (Dobbin, Kim, & Kalev, 2011). Firms from different industries may find different problems to be more or less salient. As firms look for solutions, others in their industry – particularly visible or powerful others – may serve as referents (Fligstein, 1990; Haveman, 1993). We consider whether the prevalence of diversity mentoring programs varies across industries, and whether firms from different industries vary in their interpretations of the business performance logic. Our sample includes firms from five industries: food manufacturing, business services, insurance, electronics manufacturing, and healthcare. We do not have a preexisting

Instantiation of Institutional Logics

243

reason to believe that diversity mentoring practices would be more prevalent in some of these industries than others.

METHODS Data Collection We conducted interviews with respondents at 87 establishments that operate in one of five industries: food manufacturing, electronics manufacturing, healthcare, insurance, or business services. These companies were located in one of four metro areas: Atlanta, Boston, Chicago, or San Francisco. Establishments were sampled from a list of firms who completed EEO-1 reports, which means that these firms are public or private companies with 100 or more employees (50 or more employees if the firm is a federal contractor). To be included in our sample, each firm had to have at least one of the following diversity programs: an affirmative action plan, a diversity manager, or a diversity committee. One third of firms in our sample had only an affirmative action plan, while almost 60 percent had two or more diversity practices. For the purposes of this paper, we limit our analysis to those firms that drew from the business performance logic to justify diversity management initiatives: this sample consists of 53 firms. Two of the 53 firms have two distinct diversity mentoring programs, which address separate goals and are thus counted twice in tabular analysis of the data. Therefore, we have 55 firm/mentoring program cases. While we broadly characterize our respondents as ‘‘diversity managers,’’ their formal positions varied by organization. Thirty one of our respondents were identified as full-time diversity managers, with the most common job title being ‘‘Director of Diversity.’’ For firms without a full-time diversity manager, we interviewed the individual who oversaw diversity initiatives. This yielded 22 respondents, mainly HR managers, who reported only part-time involvement with diversity management efforts. We conducted the interviews both in person and over the phone. More of the early interviews were conducted in person, while later interviews were more likely to be conducted over the phone. Interviews lasted anywhere from 30 to 90 minutes and sometimes included a follow-up call if we were unable to ask all questions during the allotted time. Interviewers used a structured interview protocol, but asked unstructured follow-up questions. The interview covered a wide range of topics related to diversity management at the firm, including the respondent’s career history as a diversity

244

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

specialist, firm performance, changes in staffing or leadership team, types of diversity programs, experiences with these programs, and the goals of diversity management. Respondents spoke at length about top-management support for diversity, employee perceptions of diversity initiatives, diversity practices and their perceived efficacy, challenges to diversity management, and future plans for diversity management at the firm.

Data Analysis We used an inductive, grounded theory approach to analyze the data. We alternated between data and theory – examining the data, identifying patterns, turning to theory to help clarify these patterns, and returning to the data. We discuss the measurement and coding of these inductive theoretical constructs in the ‘‘Results’’ section. General Coding Strategy and Process Data was coded using the qualitative analysis software Atlas.ti. Inter-coder reliability was over 90 percent for a sample of 20 coded interviews. We began with a strategy of open coding, but moved to more finite code categories in subsequent read-throughs of the data. The first round of coding highlighted the firm’s diversity programs and policies and discussions of perceived management support. Subsequent rounds of coding focused on how respondents defined diversity and their understandings of the perceived purpose of diversity management. We also coded for specific types of mentoring programs, the antecedents of these programs, and challenges to these programs. We collapsed codes to refine our model as our theoretical perspective emerged. We also wrote memos and created spreadsheets to examine patterns across cases in a more systematic way. Diversity Mentoring Practices We identified three distinct types of diversity mentoring practices: affinity mentoring, high potential mentoring, and general diversity mentoring. ‘‘Affinity mentoring’’ refers to programs where existing employee resource groups (‘‘affinity groups’’) design and oversee the mentoring initiatives. ‘‘High-potential mentoring’’ refers to programs that are designed to develop the ‘‘talent’’ or human capital of diverse employees who have been identified as ‘‘high potential’’ or ‘‘leaders.’’ ‘‘General diversity mentoring’’ refers to programs where mentoring is offered to all diverse employees through a formalized structure. These general programs are ‘‘top-down’’: HR or

245

Instantiation of Institutional Logics

diversity personnel usually oversee these programs, unlike the ‘‘bottom-up’’ affinity mentoring programs. We also identified two subtypes of mentoring practices: ‘‘substitute’’ mentoring practices and ‘‘secret’’ mentoring practices. Substitute mentoring practices are practices that advance mentoring-related goals (to be explained in the next section), without being explicitly defined as ‘‘mentoring.’’ As one example, a hospital implemented a formal ‘‘diverse talent development’’ program, where diverse employees were exposed to career coaching or skilldevelopment via a series of workshops. We coded this kind of program as a ‘‘substitute mentoring practice.’’ Secret mentoring practices are mentoring practices that are not explicitly focused on diversity, but are implicitly used to achieve diversity goals. In other words, a practice becomes a secret mentoring practice when a manager that oversees a universal mentoring program (i.e., a program that is open to all employees, including white men) engages in behind-the-scenes targeting. As one example, a manager at food manufacturing firm took special care to include female and minorities employees to a popular ‘‘high-potential’’ mentoring program, even when this meant selecting a less senior employee.

RESULTS Industry We began the analysis by examining the prevalence of any type of diversity mentoring practice across firms from the five industries included in our sample. We find no clear industry pattern in the prevalence of diversity mentoring programs. In three of the five industries (business services, electronics manufacturing, and food manufacturing), roughly half of the firms in each industry reported a diversity mentoring practice. Approximately one-third of insurance firms had diversity mentoring, while only one eighth of healthcare firms had such a practice. It seems that cross-industry differences are not sufficient to explain why certain firms embrace diversity mentoring practices, while others do not. As developed in the next section, how respondents interpret the business performance logic is a key factor in logic instantiation, but we also find no clear pattern of industry variation in logic interpretation. While there was less variation in how diversity managers interpreted of the business performance logic in the healthcare and insurance industries, even within these two industries we still observed

246

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

diversity managers that drew from a range of different elements of the business performance logic. The lack of a strong industry pattern in the prevalence of diversity mentoring programs or in managerial interpretations of the business performance logic is somewhat surprising. This may be a function of how the field of diversity management is structured. Most of the diversity managers in our sample moved across industries throughout their professional careers, and we speculate that respondents may bring frames and ideas from one firm to another. In this way, frames, ideas, and practices may span industry boundaries.

Elements of the Business Performance Logic When respondents drew from the business performance logic to justify diversity management initiatives, they revealed implicit assumptions about the specific mechanism through which workforce diversity promotes strong business performance. These assumptions were not uniform across firms; instead, we find that diversity managers interpreted the same logic, the logic of business performance, in different ways. We identified four distinct variants or elements of the business performance logic: (1) talent management, (2) employee performance, (3) team performance, and (4) pleasing the customer.2 The element of ‘‘talent management’’ implies that the cultivation of talented employees, especially talented diverse employees, will enhance business performance. Firms that subscribe to the logic element of talent management believe that business success is predicated on attracting and retaining highly skilled workers, for example, ‘‘we recruit highly technical, highly educated talent that is not as readily available to us as we would like to it be’’ (PD 76). Talent management firms may also seek to boost firm performance by developing the human capital of current ‘‘high-performing’’ employees, providing ‘‘development opportunities’’ that are designed to increase an employee’s skills or knowledge. Over a quarter of firms referred to the talent management element of the business performance logic. The element of ‘‘employee performance’’ implies that firms succeed to the extent that they foster employee engagement and satisfaction. Firms that subscribe to the employee performance logic element believe that workers who feel comfortable and engaged at work will be more committed to the firm, more productive, and willing to work harder and longer. Rather than explicitly focusing on highly skilled or top-performing employees, these firms seek to engage and retain their workforce to foster productivity gains.

Instantiation of Institutional Logics

247

The following quote from an ‘‘employee performance’’ respondent exemplifies the theorized causal link: ‘‘[the purpose of diversity management is] basically just [employee] engagement and increasing productivity and getting, making people happy at work so they are more productive and are happy here’’ (PD 70). Nearly two thirds of firms referred to the employee performance element of the business performance logic. The element of ‘‘team performance’’ implies that firms succeed when conditions are conducive to positive team interaction, including conditions that allow for better problem solving within work teams or better communication across team boundaries. Better ideas, and therefore better business results, are thought to follow when employees are able to work together, understand each other, and communicate easily: ‘‘we come up with a better idea, better solutions when we all talk together and work together’’ (PD 71). Furthermore, teams that incorporate a range of diverse viewpoints are thought to be more creative and innovative, and therefore outperform more homogenous teams: ‘‘[as far as] bottom-lines business results and our ability to react and serve our customers better, the more diversity you have in your workplace the more innovation you have, the more viewpoints you have’’ (PD 12). Over a quarter of firms referred to the team performance element of the business performance logic. The element of ‘‘pleasing the customer’’ implies that firms succeed when they understand their customers and appeal to market dynamics, which include the demographics of their customer base. Greater customer satisfaction, and therefore better business results, are thought to follow when employees connect with the target population, for example, ‘‘we recognized that we have an opportunity to really create sustainable growth for this organization by making sure that we are in fact being relevant with our products and our services to this amazingly diverse consumer landscape’’ (PD 6). Diverse employees are valuable because they are expected to connect with or better understand diverse consumers: ‘‘we have recognized that there is an opportunity to engage with more multicultural markets and really to target different market segments than we have been able to historicallyy it’s essential [that more of] the people in our company look like a potential customer’’ (PD 35). Nearly three quarters of firms referred to the pleasing the customer element of the business performance logic. Logic Elements and the Theorization of Diversity Mentoring Friedland and Alford (1991) argue that institutional logics are composed of ‘‘elemental categories or building blocks, which represent the cultural symbols and material practices particular to that [institutional] order’’

248

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

(Thornton et al., 2012, p. 54). Thornton and colleagues (Thornton, 2004; Thornton et al., 2012) elaborate this model to argue that ‘‘[v]arious bases for structure, norms, and symbols are integral parts of any institutional order; they are variable attributes on the Y-axis of different institutional orders’’ (Thornton et al., 2012, p. 51). The authors develop a more detailed typology of the elemental categories or variable attributes that compose institutional logics: institutional orders may differ in root metaphors, sources of legitimacy, control mechanisms, bases of organization or strategy, among other respects (Thornton et al., 2012, p. 73). Thornton et al. (2012) argue that these ‘‘elements of institutional logics, such as practice and identity, are partially autonomous from institutional logics and orders on the X-axis, opening up possibilities for agency in the reconfiguration or recombining of elements to alter extant or create new institutional logics’’ (p. 171). We build on these insights to argue for greater analytical attention to elemental variation within the same institutional logic, not only variation in elemental categories across different institutional logics. As we outlined above, we inductively identify four distinct interpretations or ‘‘elements’’ of a single institutional logic, the business performance logic. Our respondents varied in the elements of the business performance logic they highlighted, and the elements they neglected or downplayed. We argue that how managers frame or interpret a given institutional logic – specifically, how they reference and combine different logic elements – is a key component of the logic instantiation process. Logic elements are not mutually exclusive: respondents often blended multiple elements to make the business case for diversity management. However, we maintain that adherence to different combinations of logic elements influenced the decision to embrace a diversity mentoring program. We expected that certain interpretations of the business performance logic would be more likely to yield diversity mentoring practices. Strang and Meyer (1993) argue organizational practices are often theorized in terms of abstract categories or cause-and-effect relationships, and these ‘‘theoretical accounts of practices y specify and explain the outcomes they produce’’ (Strang & Meyer, 1993, p. 497). If the theorized outcomes of a practice correspond with field-level constitutive understandings about the drivers of firm success, the practice should diffuse more quickly and to a greater extent (Davis & Greve, 1997; Strang & Meyer, 1993). Discussions of mentoring in the management literature imply that this practice is theorized to result in two positive outcomes. First, mentoring is expected to develop the ‘‘talent’’ or human capital of existing employees. Wiser, advanced members of the firm share valuable skills and knowledge

Instantiation of Institutional Logics

249

with their prote´ge´s, increasing the likelihood that these prote´ge´s will advance in the organization (Ragins & Cotton, 1999). Second, mentoring is expected to increase employee engagement and satisfaction by cultivating social ties in the workplace. Employees with strong ties to other workers are thought to be more committed to the organization and more likely to stay put (Payne & Huffman, 2005). The talent management element of the business performance logic is most closely linked to the theorized outcomes of mentoring practices, as it implies that firm success is predicated on cultivating and retaining a highly skilled, talented workforce. The employee performance element also seems related, as it implies that firm success depends on productivity gains made by keeping employees engaged and satisfied with their work. However, the other two elements of the business performance logic (team performance and pleasing the customer) suggest drivers of firm success that do not correspond as readily to the theorized outcomes of mentoring practices. Combinations of Logic Elements Respondents often cited more than one element of the business performance logic, and they varied in the extent to which they combined or blended these different logic elements. We argue that both the type of elements referenced, as well as how these elements were blended with other elements, are key to the process of logic instantiation. We separate firms into four groups, based on four different combinations of logic elements. We define ‘‘Group 1’’ firms as those that referenced the element of talent management. Seventeen firms subscribed to this element, either alone or in combination with other elements of the business performance logic. ‘‘Group 2’’ firms include those firms that subscribed to the employee performance logic element but not the talent management logic. Twenty firms subscribed to this element, either alone or in combination with the team performance or pleasing the customer elements. ‘‘Group 3’’ firms include firms that subscribe to the team performance element but not the talent management or employee performance elements. Five firms subscribed to this element, either alone or in combination with the pleasing the customer element. Finally, we categorize firms in ‘‘Group 4’’ when respondents cited the ‘‘pleasing the customer’’ element as the only logic element. Thirteen firms fall into this category. We expected to find a greater prevalence of diversity mentoring practices at Group 1 and 2 firms than Group 3 or 4 firms, given the stronger correspondence between the theorized outcomes of diversity mentoring and the particular interpretation of the business performance logic at the Group 1 and 2 firms. Our results support these expectations (see Table 1). Just over

250

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

half of Group 1 firms had a diversity mentoring practice, and when substitute or secret mentoring practices are included, 15 of 17 of these firms had a mentoring-related practice. Similarly, just over half of Group 2 firms had a diversity mentoring practice. When substitute and secret practices are included, 15 of 20 of these firms had a mentoring-related practice. By contrast, Group 3 and Group 4 firms were much less likely to have diversity mentoring practice or a mentoring-related practice. Accordingly, we collapse these last two groups of firms in subsequent analyses.

Goals A key insight from the institutional logics perspective is that while logics contain both symbolic and material components, these symbolic and material elements are often loosely coupled. We find that diversity managers drew from different elements of the business performance logic to justify diversity management, and that the prevalence of diversity mentoring programs varied across firms who subscribed to different logic-element combinations. Diversity mentoring programs were relatively common for Group 1 and 2 firms, and relatively rare for Group 3 and 4 firms. However, we find that the specific type of diversity mentoring program did not always map onto to the different logic-element combinations. For example, we Table 1.

Any diversity mentoring Affinity mentoring High potential mentoring General diversity mentoring Substitute practice Secret diversity mentoring No diversity mentoring Firm count

Business Performance Logic and Mentoring Practice. 1: Talent Management

2: Employee Performance

3: Team Performance

4: Pleasing the Customer

9

11

0

2

4 4

5 3

0 0

2 0

1

3

0

0

2 4

3 1

0 1

1 0

2

5

4

10

17

20

5

13

Instantiation of Institutional Logics

251

expected Group 1 firms to implement high-potential mentoring programs, because this type of program corresponds with an emphasis on talent development. However, we found that Group 1 firms were just as likely to implement an affinity-mentoring program as a high-potential mentoring program. With this puzzle in mind, revisiting the data uncovered another key factor in the logic instantiation process: an organization’s formal diversity goals. Group 1 firms, while highlighting the importance of talent management, were equally as likely to have formal diversity goals of talent development as they were goals of employee retention. These divergent goals explain why some firms selected high-potential mentoring while others selected affinity-mentoring. We find loose coupling between logic elements and organizational practices leads to a firm’s formal diversity goals mediating practice selection. Diversity managers had different interpretations as to why diversity was a driver of firm success, but also held different understandings about the specific diversity goals or outcomes that firms should strive for, as well as the ‘‘best practices’’ to meet these goals. For example, while diversity managers across a number of firms argued that team performance was a key driver of firm success, these managers held different ideas about the types of diversity initiatives that will improve team performance. Similarly, while many diversity managers argued that more diverse work groups lead to increased innovation, we find variation in managers’ understandings of ‘‘a diverse team,’’ as well as variation in understandings of ‘‘best practices’’ for inculcating an effective diverse team. We argue that a complete model of logic instantiation requires attention to the firm’s formal diversity goals. To analyze the relationship between these goals and diversity mentoring practices, we coded for all explicit mentions of a firm’s formal diversity management goals, and grouped these goals into five categories. The first four categories include diversity goals that mentoring is theorized to address, while the final category consists of ‘‘all other’’ (non-mentoringrelated) diversity goals. The first goal category includes firms that sought to increase the ‘‘talent’’ or human capital of diverse employees. The second goal category includes firms that sought to increase the representation of diverse employees in leadership positions. The third goal category includes firms that sought to increase employee engagement or satisfaction through maintaining a supportive, enlightened, or respectful workplace. The fourth goal category includes firms that sought to increase employee retention. The final goal category, ‘‘all other goals,’’ includes firms with respondents who exclusively mentioned diversity goals with no clear link to the theorized outcomes of mentoring, like ‘‘engaging with the local community’’ or

252

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

‘‘attracting diverse consumers.’’ These goal categories largely map onto the prevalence of diversity mentoring programs. As shown in Table 2, none of the firms that fell into the ‘‘all other’’ goal category had a diversity mentoring program, while about half of firms with leadership representation or employee engagement goals had a diversity mentoring program, and nearly all firms with talent development or retention goals had a diversity mentoring program. Logic-element combinations and diversity goals were closely linked. Unsurprisingly, the most common diversity goals for Group 1 firms were talent development and employee retention, while the most common goal for Group 2 firms was employee engagement. Group 3 and 4 firms tended to have other, unrelated-to-mentoring, goals. However, it is important to note that a respondent’s interpretation of the business performance logic was not entirely predictive of the firm’s formal diversity goals (see Table 3). Rather, respondent often constructed a broad interpretation of the performance logic, drawing from multiple logic elements, but only reported formal goals related to a subset of these interpretations of the performance logic. Additionally, as discussed in detail in the next section, a number of firms across all logic-element combinations cited a common goal of increasing leadership representation. Table 2.

Diversity Goals and Mentoring Practices.

Goal A1: Leadership Diversity

Goal A2: Talent Goal B1: Goal B2: Goal C: All Development Engagement Retention Other

Any diversity mentoring Affinity mentoring High potential mentoring

7

4

7

6

0

2

0

4

5

0

3

4

1

1

0

General diversity mentoring Substitute practice Secret diversity mentoring No diversity mentoring Firm count

2

0

2

0

0

3

4

0

1

0

3

2

1

1

0

3

0

5

0

1

16

5

13

8

13

253

Instantiation of Institutional Logics

Table 3. Business Performance Logic and Diversity Goals.

Goal A1: Leadership diversity Goal A2: Talent development Goal B1: Engagement Goal B2: Retention Goal C: All other Firm count

1: Talent Management

2: Employee Performance

3: Team Performance

4: Pleasing the Customer

4

6

3

3

5

0

0

0

2 5 1 17

8 3 3 20

0 0 2 5

3 0 7 13

Table 2 outlines the connection between a firm’s diversity goals and the type of diversity mentoring practice the firm implements. Consider the firms categorized as Group 1 (i.e., firms where respondents drew from the talent management logic element): five Group 1 firms reported a talent development diversity goal. Two of the five had high-potential mentoring, two had a substitute practice, and one had secret mentoring. Five different Group 1 firms reported a diversity goal of retention: four of the five had affinity group mentoring and one had secret mentoring. While respondents from all 10 of these firms drew from the logic elements of talent management and employee performance, the particular type of diversity mentoring program these firms implemented was shaped by the formal diversity goals of the firm. Goals Shared Across Logic-Element Groups As described above, we find diversity goals are in large part consistent with different interpretations of the business performance logic. However, we paid close attention to a case in which a variety of firms shared a common diversity goal: increasing the representation of diverse employees in leadership positions. This leadership representation goal was prevalent across all four groups of firms, indicating it was not tied to a particular logic element. We speculate that the ubiquity of this particular goal may be either (1) a remnant of the previous logic of compliance, in which diversity management was viewed as an important tool to discourage lawsuits or (2) somehow related to current Affirmative Action efforts at the firms. We also find that even when firms share the same goal, different logicelement combinations may still shape their implementation of a practice. Group 1 firms that sought to increase the leadership representation of

254

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

diverse employees tended to implement diversity mentoring programs that were targeted toward high-potential employees, or secret diversity programs with the same focus. Group 2 firms with the same goal tended to implement general diversity mentoring programs or substitute practices to achieve this goal. Group 3 or 4 firms with this goal tended to enact affinity mentoring or no mentoring practice at all. Firms that subscribed to different logic-element combinations used different strategies to accomplish the same goal.

Power Different interpretations of the business performance logic and different diversity goals explain much of the observed pattern in prevalence and types of diversity mentoring programs. However, after analyzing cases that fell outside of expected patterns, we identified two additional factors that shaped the logic instantiation process: the relative organizational power of the diversity manager and the definition of ‘‘diversity’’ itself. We coded a diversity manager as ‘‘low power’’ if diversity initiatives received little support from top management. In our interviews, we asked respondents to describe any challenges they faced in achieving their vision for diversity management. One clear roadblock was an absence of support from upper management: I believe we are pushing to have someone at the senior level have as equal of a commitment [to diversity management] as I do y I think it would be less of a challenge [if there were active support for diversity] at a higher level within organization. (PD 55) The organization was a little apprehensive about getting into formalized diversity activities because of some less and positive experiences that some of our leadership had had in other organizations or personally with diversity. (PD 11)

This concern about management support was not limited to firms with newly created diversity initiatives, firms with part-time diversity management roles, or other indicators of minimal organizational investment in diversity management. Rather, the critical nature of top-management support cut across a wide swath of interviews. Even diversity managers who did not currently face this challenge often reflected on past struggles to gain support: I would say initially, back when we were laying out kind of our strategy for diversity within the company, that very first meeting with our senior leadership and you know communicating the importance of diversity as a part of our culture and how that was going to affect our recruitment and our retention, unless you turn that into dollars, you

255

Instantiation of Institutional Logics

know they kind of look at you like yeah so long. And I think that was the original challenge that was the most daunting y And we needed their partnership, we couldn’t succeed without them. (PD 12)

To separate low-power diversity managers from more powerful respondents, we coded for discussion of top-management support for diversity initiatives at the firm, or the lack thereof. In the interviews, we questioned respondents directly about top-management support for diversity management. We also asked whether high-ranking members of the organization participated in diversity management programs, or whether diversity managers had frequent discussions with upper management about their activities. We noted instances where diversity managers cited the lack of topmanagement support as a barrier to their abilities to achieve diversity goals. We find that the power of the diversity manager, as measured by managerial support and resource availability, shaped the kinds of strategies managers used to accomplish their objectives (see Table 4). Among firms with leadership and development diversity goals, two of eight respondents defined as ‘‘low power’’ reported a diversity-mentoring program, compared to seven of thirteen respondents who did not face similar constraints. Among firms with employee satisfaction and retention goals, 1 of 6 lowpower respondents reported a diversity mentoring program, compared to 12 of 15 respondents who did not face similar constraints. For a healthcare firm with the goal of increasing diversity in leadership positions, a lack of top-management support for diversity initiatives meant a Table 4.

Diversity Manager Power and Mentoring Practices. Goal A: Leadership Goal B: Engagement Representation or Talent or Retention Development

Goal C: All Other Diversity Goals

Low power All other Low power All other Low power All other (8) (13) (6) (15) (4) (9) Any diversity mentoring Affinity High potential General

25%

54%

17%

80%

0%

0%

0% 13% 13%

15% 31% 8%

17% 0% 0%

60% 7% 13%

0% 0% 0%

0% 0% 0%

Substitute practice Secret div mentor None

25% 25% 25%

23% 15% 8%

17% 33% 33%

0% 0% 20%

0% 0% 100%

0% 0% 100%

256

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

mentoring program that targeted diverse high-potential employees was not feasible. Prior efforts to secure top-management support for diversity efforts had failed, which led the manager to focus on informal initiatives, with few formal programs and a focus on employee education and awareness: I think there are a variety of reasons [that diversity management is a low priority] y from a business perspective, because the hospital itself is in such a challenging environment, most of the attention especially [in] the last two years from the senior management has been on just keeping us afloat, and so it is kind of like, when you are in a crisis mode, you know, buying a pretty dress isn’t as important. They perceive diversity as, yeah, it’s nice, but you guys go do that, because we don’t have time to really focus on it y So what we have done that’s very different y [is that] all of our efforts have been focused at the employee level. (PD 13)

Unsurprisingly, firms with a low-power diversity manager were less likely to have a formal diversity mentoring practice. However, a subset of low-power respondents managed to achieve their diversity goals through the use of ‘‘secret’’ mentoring practices. The majority of secret mentoring practices occurred in firms with a low-power diversity manager. An example from one such firm illustrates how respondents used these programs to achieve their goals. In this firm, the low-power respondent asked if they could go ‘‘off the record’’ (i.e., stop the voice recorder) to speak about the lack of leadership support for diversity management. This respondent cited a diversity goal of increasing the representation of diverse employees in leadership positions, and described how a recent internal survey of employees had revealed that minority and female employees had no ‘‘clear path for moving into management.’’ This framing of the problem is consistent with the framing among firms that have formal diversity mentoring programs, but this lowpower respondent chose to implement a universal, rather than diversityfocused, program instead. However, the respondent worked behind the scenes to use the program to address the lack of development opportunities for diverse employees, with our respondent reporting that gender, race, and culture ‘‘came into play’’ when she made assignments (PD 42).

Defining Diversity Although all of our respondents agreed that the purpose of diversity management was to improve the bottom-line performance of their firm, respondents held different understandings of what ‘‘diversity’’ entailed. We coded for instances where the respondent discussed the type of diversity the firm sought to manage. These definitions of diversity could be implicit or

Instantiation of Institutional Logics

257

explicit: respondents typically explicitly defined ‘‘diversity’’ when discussing the firm’s diversity goals, but they also implicitly defined diversity in discussions of particular practices or employee groups. We also noted mentions of ‘‘diversity and inclusion’’ initiatives (‘‘D and I’’). ‘‘Inclusion,’’ in the context of diversity management, is defined in opposition to the traditional U.S. definition of diversity (i.e., diversity as variation in the race or gender composition of the workforce). Respondents who define diversity as ‘‘inclusion’’ argue that firms should downplay a focus on race and gender and highlight a wide range of other potential differences instead. Nine respondents fully endorsed the inclusion definition of diversity. The following quotes illustrate the types of responses that are consistent with an inclusion framing: [Our focus is] beyond diversity y it’s more about the differences, more about inclusion, differences in thoughts, in lifestyles, in education, so it’s gone beyond [diversity]. (PD 48) We have a very holistic view of what diversity is [and] it is not the traditional limited view of ethnicity or gender, for example. For us it’s about, it sounds like a sort of a ridiculous term, but we talk about diversity of thought in the company. And for your reference, [we use an approach developed by prominent diversity consultants] y of a diversity [as a] concentric circle. And the central circle is personality and then you know internal influences and the next one is external influences y it’s education and gender and age and you know schooling and religion and so forth. And it’s all these components that make us unique in our views, in our reactions and so forth. (PD 63)

Half of respondents subscribed to a traditional U.S. ‘‘diversity’’ definition of diversity, for example, diversity as race or gender differences. The remaining third of respondents employed a ‘‘blended’’ definition of diversity, which incorporated elements from both the inclusion and diversity definitions. The use of inclusion language has diffused broadly across the firms in our sample: nearly every diversity manager mentioned the word ‘‘inclusion’’ at least once. However, what separated the three ‘‘definition of diversity’’ groups was the frequency or intensity of the use of ‘‘inclusion’’ versus ‘‘diversity’’ language. Two thirds of respondents could be easily classified at either end of the spectrum (i.e., diversity vs. inclusion), but the remaining third fell somewhere in the middle. For example, a blended diversity respondent might initially define diversity as ‘‘all differences,’’ and explicitly reject a race-or-gender definition, but then later outline goals or programs designed to increase the representation of minorities or women in the firm. A blended respondent might also draw predominantly from a traditional definition of diversity, but specify an explicit inclusion focus in their job title or in their formal diversity goals.

258

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

Redefining Diversity: Power and Competing Field Frames We anticipated some mention of diversity-as-inclusion in our interviews, because this rhetoric has become popular in the HR and diversity practitioner literature, but we did not expect to find such a stark difference between firms that we categorized as ‘‘inclusion’’ focused versus those we categorized as ‘‘diversity’’ focused. The introduction of the inclusion frame was more substantial than a simple shift in rhetoric. Instead, this was a shift in the underlying logic about what defines diversity and as such influenced the support and implementation of mentoring practices. It appears as though diversity management, as a profession, may now have a competing field frame with regard to exactly what diversity entails. While a field frame is similar to an institutional logic, ‘‘akin to strategic framing, it is endogenous to a field of actors and is subject to challenge and modification’’ (Lounsbury, Ventresca, & Hirsch, 2003, p. 72). While research has documented the shift to inclusion, there has been only limited attention to whether this represents a shift in rhetoric or a change in material actions and practices (Roberson, 2006). Firms that subscribed to an inclusion definition of diversity were overrepresented among firms with low-power diversity managers: 5 of 9 inclusion firms had low-power diversity managers, compared to 4 of 18 blended definition firms and 8 of 28 diversity definition firms. Respondents at all five low-power respondent/inclusion firms reported difficulty communicating a business case for diversity to others at the firm and a general lack of management support. In-depth analysis of these five cases revealed that these respondents, who struggled with difficulty securing buy-in actively chose to reframe diversity around inclusion in a bid for greater support. For example, one respondent explained that they struggled to make the ‘‘straight business case’’ for diversity to others at the firm, and that they lacked support from top management. This respondent chose to take a ‘‘less divisive’’ inclusion approach in light of these constraints: Unlike a lot of companies that usually approach gender or race as the foray into the topic [of diversity management], I made a decision to make it intercultural diversity because I felt that was a topic that most employees could relate to and it would be least divisive. (PD 61)

Another respondent provided a similar reason for why they had abandoned a diversity framing in favor of an ‘‘inclusion’’ framing: I think over time, diversity has become kind of a word that we have moved away from simply because I think when you start talking about diversity, it kind of feels like a force fit for people who are working in the industry where they don’t think it’s a

Instantiation of Institutional Logics

259

problem. And I think our management wasn’t aware of it. They just never thought about [diversity] in that way before because when you are working in an environment of like and like, and all of your clients are just like you, that’s just how you operate. And so there was the thought that we are doing well, we are making money, what’s the problem. (PD 49)

Another diversity manager explained how their firm actively sought to move from a diversity focus to an inclusion focus, in part to obtain white male employee buy-in: The diversity committee was something that had been around for a while on and off up until two years ago when they changed their mission and name to Employee Interaction Committee. You know when people hear diversity they think race and ethnicity. Then, you know, a white male manager may not want to take part in such a committee. By changing the name and mission we changed people’s opinions, we opened their eyes, took the blinders off. It’s about including everyone, and everyone should have a say. (PD 67)

This strategy of shifting the definition of diversity was also apparent at a firm where financial restructuring had ended the tenure of a full-time diversity manager, where there were currently several part-time diversity managers spread across various HR functions. One of the part-time managers explained why the diversity committee was now called ‘‘Valuing Differences’’: Well [the diversity committee is] actually called Valuing Differences, so we sort of went for the differences versus the diversity and you know just more about people being different, what does that meany they are just sort of a group that is consulting for the organization. So for example if marketing is running a new ad, they may run it by the group to just see if there is anybody that sees anything that might necessarily be offensive or be, having a reaction to, whether that reaction’s positive or negative. And there is a centralized place where people get together and try to reflect a respect for difference. (PD 31)

At another firm, resistance from upper management led to this shift in the definition of diversity: The hospital itself doesn’t have a diversity goal. If you look at our strategic plan there is nothing in there that says you know we have any goal around diversity. But I will quote you our mission and goal of our diversity council we say to cultivate an environment where we recognize and respect each other’s diverse ideas and qualities. The reason for that we want to focus on diversity not as age, race, and gender and so forth, but more about the diversity of thought. (PD 13)

Even some of the ‘‘diversity-as-inclusion’’ respondents who were not defined as ‘‘low power’’ indicated that they had embraced inclusion to address the

260

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

lack of support for diversity efforts. A relatively powerful diversity manager explained: Well [the mentoring program has] been in place in some way, shape or form forever, but it doesn’t do well because it is a diverse mentoring program and it loses steam. It has got to be shifted [away from diversity into] the way that we develop our employees or our own leadership or our performance management process y If you have [mentoring] as a separate subject people just won’t do it. (PD 89)

The inclusion field frame had implications for organizational practices. Diversity managers that subscribed to an inclusion definition of diversity viewed most targeted diversity programs as inappropriate means to achieving the firm’s diversity goals. However, these respondents were often unsure about which practices should replace these targeted programs, as the newer inclusion definition of diversity does not yet imply an obvious set of material organizational practices

DISCUSSION AND CONCLUSION The key premise of the business performance logic is that diversity management is valuable because it improves bottom-line organizational performance. We consider a case in which firms that subscribed to the business performance logic implemented different diversity mentoring programs, or sometimes no mentoring program at all. Through the empirical analysis of this case, we refine existing theory of the logic instantiation process. We focus on the organizational factors that shape the translation of logics into organizational practices and identify four organizational factors that mediated the process of logic instantiation: (1) the interpretation and framing of the business performance logic; (2) the selection of formal diversity goals; (3) the organizational power of the manager responsible for diversity initiatives; and (4) the prevailing definition of diversity at the firm. We demonstrate how each factor operates to shape the logic instantiation process. See Fig. 1 for a visual representation of our theoretical model.

Interpreting and Framing an Institutional Logic Institutional logics are composed of multiple elements or components (Thornton, 2004; Thornton et al., 2012). We find that when actors interpret

Instantiation of Institutional Logics

Fig. 1.

261

Process of Logic Instantiation.

and frame an institutional logic, they often privilege certain logic elements and neglect others – and these different logic-element combinations yielded different practices. Previous research from the theorization perspective finds that a practice is more likely to diffuse when its theorized outcomes correspond with dominant constitutive understandings about the sources of organizational success. Our findings are consistent with this approach. Diversity mentoring programs, theorized to increase employee human capital or employee engagement, were an easier sell at firms that championed the ‘‘talent management’’ or ‘‘employee performance’’ elements of the business performance logic, when compared to firms that focused on the ‘‘team performance’’ or ‘‘pleasing the customer’’ elements. Variation in Framing Why did the diversity managers in our sample frame the business case for diversity in different ways? We suggest three potential explanations. First,

262

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

the variation in the framing of the business performance logic may be a product of the wide range of framing options open to diversity managers. The theorization of diversity management in the management literature implies that diversity initiatives can offer competitive advantages across a wide range of performance dimensions, including ‘‘cost, resource acquisition, marketing, creativity, problem-solving, [and] organizational flexibility’’ (Cox & Blake, 1991, p. 45). Given the many ways that diversity management is theorized to help firms, combined with a lack of concrete empirical evidence regarding the effects of diversity on firm performance, it is not surprising that diversity managers employ different strategies to connect diversity practices with measurable business results. Second, firms may frame the business performance logic in different ways because of the ongoing legacy of prior institutional logics. While the business performance logic of diversity management is currently the dominant logic, artifacts from the former compliance-based logic may persist within firms. Many diversity managers were careful to mention that their work was not about compliance or EEO efforts, and affirmative-action activities were often functionally separate from diversity efforts and overseen by compliance or legal departments. However, elements of the compliance logic continued to inform diversity goals in these firms. Our sample of firms, which subscribed to the business performance logic, shared goals with firms that subscribed to the compliance logic (e.g., increase the number of diverse employees or increase diverse representation in upper management). Many of the firms in our sample also had Affirmative Action plans and diversity managers. Respondents from firms that retain goals or structures from the compliance era may have more difficulty selling the business performance logic. This tension between business performance and compliance logics may provide additional incentive for diversity managers to actively construct and redefine a business case for diversity, a surprising outcome given the acknowledged dominance of the business case logic in this field. Our finding supports the institutional logics view that ‘‘practice and identity variation or change can create ambiguity that garners the attention of actors and catalyzes social interactions aiming to resolve the ambiguity generated’’ (Thornton et al., 2012, p. 133). Third, we find some evidence that the professional background of the diversity manager may influence the interpretation of the business case logic. Diversity managers in our sample were split between full-time diversity management professionals and human resource managers with part-time diversity roles. We find full-time diversity managers were more likely to cite the talent management and employee performance elements of the business

263

Instantiation of Institutional Logics

performance logic, while HR managers with part-time diversity roles often relied on the pleasing the customer logic element. Full-time diversity managers were also more likely to enact diversity mentoring practices than were their HR counterparts, even given the same diversity goals. For instance, when considering just those firms with the diversity goals of employee engagement or retention, 12 full-time diversity managers report diversity mentoring as compared to zero HR managers. While HR managers cite a number of constraints to enacting practices, there was no difference in the proportion of HR versus diversity managers citing framing barriers and diversity managers citing these types of problems manage to create mentoring programs through strategies such as using existing employee networks to run affinity-mentoring groups. While a full analysis of the difference between HR managers and diversity managers with respect to framing and practice creation is outside the scope of this paper, our results suggest that perhaps HR managers with part-time diversity roles have a less sophisticated view of diversity management, or less professional knowledge than their full-time diversity manager counterparts.

Goals We argue that a firm’s diversity goals serve as an important bridge between a diversity manager’s interpretation of an institutional logic and the materialization of this logic in an organizational practice. Different interpretations of an institutional logic provide loose prescriptions for diversity management initiatives. However, the loose coupling of logic elements and practices means that even those actors who interpret an institutional logic similarly may have different ideas about specific strategies that will lead to success. An organization’s diversity goals influence the instantiation of the business performance logic by calling attention to particular variants of diversity mentoring programs. Thus we find more variation in practice implementation than a traditional neoinstitutional account of diffusion might predict. For example, some firms in our sample used affinity groups to oversee mentoring, while others included mentoring as part of an employee talent and leadership development strategy. Still others had general, HRsponsored mentoring programs that were open to all diverse employees. We also find that some firms developed ‘‘secret’’ diversity mentoring practices. While additional data would be needed to fully disentangle the relationship between interpretations of the business performance logic and formal diversity goals, we provide two tentative explanations for why goals link

264

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

particular logic elements to related practices. First, the active interpretation of the business performance logic by diversity managers leads to a wide range of appropriate diversity goals and practices. Diversity managers report continual efforts to persuade other organizational actors as to the value of diversity management efforts, suggesting that broader constructions of the performance logic may be a strategy to secure support from a range of other actors. As such, it is not surprising that most firms do not have formal diversity goals that map one-to-one onto each logic element. Additionally, while logic-element combinations represent the diversity manager’s interpretation of the logic – and are likely influenced by both professional identity and firm specific or contextual factors – formal diversity goals move beyond the cognitive or ideational. Formal organizational diversity goals are endorsed by organizational leaders, frequently incorporated into annual strategic planning, and communicated to a range of internal and external actors. Finally, we find evidence that some diversity goals, such as increasing leadership representation, may be artifacts of prior logics. After finding instances of the leadership representation goal across firms in all logicelement combinations of the business performance logic, we revisited our full sample, including firms categorized as fitting the ‘‘compliance’’ logic. The increasing leadership representation goal was cited by six firms whose diversity management efforts fit the compliance logic, suggesting that some goals may be remnants of previous logics and once formalized may survive shifts in logics.

Power and Field Frames We find that the relative power of diversity managers, measured in terms of top-management support, influenced the instantiation of logics. Previous research from the institutional logics perspective suggests that power is institutionally contingent (Friedland & Alford, 1991; Thornton & Ocasio, 1999; Thornton et al., 2012, p. 65). We argue that the process of logic instantiation is also contingent on the relative power of key actors within the firm. Top-management support appeared to be an important component of diversity managers’ power for a number of reasons. Executive endorsement of diversity management efforts lent legitimacy, encouraged the participation of middle-managers whose support was necessary in the day-to-day enactment of diversity efforts, and was a key element in acquiring resources.

Instantiation of Institutional Logics

265

While we lacked systematic data about diversity management budgets, and as such did not use this as a separate component of power, we did code for instances where respondents reported that a lack of resources, including time, money, or staff, seriously hindered their ability to achieve goals. We also paid particular attention to discussion of relative resources and how this affected diversity managers’ ability to implement programs. While nearly all respondents mentioned needing more time or manpower to fully achieve goals, a subset of these respondents saw this as a serious constraint. Diversity managers may lack resources for a number of reasons, but one common reason was the lack of a true commitment to diversity initiatives at the top of the organization. While top managers often gave lip service to the importance of diversity management, this verbal commitment was not always accompanied by a resource commitment: [Executives] are very supportive of [diversity management]. I know that when I talk with them about [new strategies] they have said what is it that we can do to help support that y They support it but when it comes time to actually taking the action that’s where I fall flat. I can’t seem to get them to make that one extra step in terms of commitment and I can use some help there. (PD 22)

A resource-dependency view of power might suggest that low-power managers would fail to get things done, because they lack the resources that are necessary to achieve their goals. To be sure, some low-power respondents never found effective strategies for achieving their visions, and remained frustrated by their impotence. Diversity mentoring programs were generally less prevalent at firms with a low-power diversity manager. However, we also find numerous examples where low-power managers achieved their goals through the use of creative, alternative strategies. One notable strategy was the use of a secret mentoring practice – a practice that was ostensibly open to all employees, but implicitly targeted to benefit diverse employees. A low-power diversity manager could obtain crucial buy-in by implementing a universal mentoring practice, and then surreptitiously use the practice to address racial or gender disparities in the firm. Another important strategy was the use of the competing ‘‘inclusion’’ field frame, which allowed managers to redefine the firm’s diversity goals altogether. Existing theory from the institutional logics perspective suggests that actors are more likely to turn to competing logics in unstable fields: When activation is not driven by automatic attention y the opportunity for individual agency increases. But social actors do not construct new identities, goals, or schemas de novo. In this context, alternative institutional logics may serve as toolkits for action (Swidler, 1986). Rather than relying on those logics that are most accessible, social actors

266

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

may invoke and activate other available logics drawn from other situational contexts. (Thornton et al., 2012, pp. 100–101)

We build upon this insight to advance a slightly narrower argument: the relative power of key organizational actors shapes how and when firms embrace competing field frames. A number of low-power diversity managers had switched to the inclusion framing because they thought that this strategy would give them greater organizational support for their initiatives. We argue that explicit attention to the relative power of key organizational actors, as well as how this power interacts with other important elements of the logic instantiation process, can help to explain organizational outcomes that might otherwise seem puzzling.

CONCLUSION Our empirical findings suggest a number of broader theoretical implications. In what follows, we highlight the two most important theoretical insights from this study. We conclude with a discussion of the limitations and promising directions for future research.

Logic Elements We argue that understanding the process of logic instantiation requires paying greater attention to how actors draw from different element of the same institutional logic. We find that different combinations of logic elements directed actors toward particular practices and away from others. Research on the translation and transposition of managerial practices has already recognized the importance of interpretative processes (Boxenbaum & Battilana, 2005). Our findings suggest future research in this area should pay more attention to interpretative processes at the logic-element level, given the demonstrated importance of logic-element combinations for the materialization of an institutional logic.

Unreflective and Reflective Action The relative importance of agency versus structure is a topic that continues to motivate research throughout the social sciences. By bridging insights from early neoinstitutional theory and the literature on institutional work,

Instantiation of Institutional Logics

267

the institutional logics perspective provided researchers an important set of theoretical tools to address this question. Powell and Colyvas (2008) write that ‘‘institutional logics are instantiated in and carried by individuals through their actions, tools, and technologies y [and in some situations] afford considerable latitude for human agency and interpretation’’ (p. 277). Our data are particularly suited to generate new insights into the complex interplay between agency and structure. We argue that the variation we find in combinations of logic elements reflects the interpretative work of embedded managers. While our findings highlight internal organizational dynamics, this should not imply that we see field-level factors as unimportant for the process of logic instantiation. However, what we found most striking was how much these institutionally embedded diversity managers actively worked to frame the business case for diversity. Our data imply that the interpretation and framing of the business performance logic was often a deliberate, strategic act on the part of socially embedded diversity managers. As one diversity manager explained, many of these managers are fully conscious of the myriad framing choices they face when they join a new organization: But when you are new to a company y you can run real, real, real, real fast and redefine what it is you are trying to do. And in that period of time once you have been there for 18 months you cannot change [how diversity is framed] because now they already know either, its marketplace, workplace or workforce or its affirmative action. You cannot rebrand it as a person. It’s really, really challenging to re-brand diversity from the inside if you are from the inside. (PD 56)

In short, the vast majority of analyses that consider the relationship between institutional logics and organizational practices focus on the taken-for-grantedness of institutional logics or the habitual and unreflective actions of organizational decision-makers. We demonstrate how and when actors also use logics reflectively, and highlight how different strategic frames influence the materialization of institutional logics into practices. The diversity managers in our sample were certainly institutionally embedded: their different interpretations of the business performance logic fell into clearly defined categories (e.g., talent development, employee engagement, team performance, pleasing the customer) and were not idiosyncratic to each firm. Furthermore, these interpretation-categories varied across industry, at least to some extent. Our argument is simply that, at least in contested, loosely coupled organizational fields, institutionally embedded actors use logics much more purposively and

268

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

intentionally than a conventional account of institutions and organizations might predict.

Limitations and Directions for Future Research Generalizability We explain how a single institutional logic, within a single organizational field, materializes in a particular organizational practice: the diversity mentoring program. It is possible that the results we present here do not generalize to other kinds of fields or practices. Our selection of a loosely coupled field likely overstates, to some degree, the relative importance of organizational dynamics. However, the primary purpose of this paper was to develop a more refined model of the process of logic instantiation, through the inductive analysis of a theoretically puzzling empirical case. We believe that our model contributes to a superior theoretical understanding of the process of logic instantiation, but defining the limits of the applicability of this model remains a task for future research. Additional Actors Expansion of the theoretical model might consider the influence of a broader array of actors. We focus on the interpretations and characteristics of diversity managers, who are embedded in particular types of firms and industries. However, these managers are also professionals who are embedded in multiple institutional orders and as such influenced by a range of other actors, organizations, and institutions. Our respondents report they gather ideas about diversity management strategies and practices from a range of sources, including professional associations, consultants, and colleagues. Previous research demonstrates that professional associations may play an important role in theorizing change and spreading innovation in an institutional field (Greenwood, Suddaby, & Hinings, 2002). Future research might explore the implications of these extraorganizational, extra-industry factors for the process of logic instantiation. Reciprocal Influence Variation in organizational practice may have a reciprocal influence on an institutional logic – that is, ‘‘practice variation in and across organizations is a key endogenous mechanism that can trigger profound organizational change’’ (Thornton et al., 2012, p. 172). In this chapter, we explore only one side of this feedback loop. However, we expect that the tensions we observed

269

Instantiation of Institutional Logics

at the organizational level (e.g., competing field frames, loose coupling between logics and practices, defining current efforts in opposition to prior logics) are likely to refract back to the field level. The field of diversity management is a volatile one: it has undergone major changes since the 1970s, and continues to evolve (Dobbin, 2009). Future research might provide the other half of our account by exploring how the observed variation in organizational diversity practices shapes the evolution of an institutional logic.

NOTES 1. Although we present the theoretical overview before our data analysis, we employed an inductive method to our research and theoretical concepts were developed as they emerged from the data (Strauss & Corbin, 1990). As patterns and themes emerged we iterated between the data and the relevant literature to refine our theory. 2. To create the analytical construct of the business performance logic and identify its component elements, we coded for explicit references to a ‘‘business case,’’ as well as references to specific performance indicators, including ‘‘ROI,’’ ‘‘the bottom-line,’’ ‘‘value proposition,’’ ‘‘revenue,’’ and ‘‘productivity.’’ When respondents mentioned desired performance goals or particular performance challenges for the broader organization, we coded for the content of these goals or challenges. Discussions of company culture and values, with respect to making the firm competitive, also informed this construct.

ACKNOWLEDGMENTS We would like to thank Michael Lounsbury and Eva Boxenbaum for valuable feedback on previous drafts of this paper. We are also grateful to Frank Dobbin, Alexandra Kalev, and the entire ‘‘Equal Opportunity Innovations at Work’’ research team. This research was supported by the National Science Foundation (grant SES-0620534).

REFERENCES Allen, T. D., Eby, L. T., Poteet, M. L., Lentz, E., & Lima, L. (2004). Career benefits associated with mentoring for proteges: A meta-analysis. Journal of Applied Psychology, 89(1), 127. Bantel, K. A., & Jackson, S. E. (1989). Top management and innovations in banking: Does the composition of the top team make a difference? Strategic Management Journal, 10(S1), 107–124.

270

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

Baron, J. N., & Pfeffer, J. (1994). The social psychology of organizations and inequality. Social Psychology Quarterly, 57, 190–209. Battilana, J. (2007). Initiating divergent organizational change: The enabling role of actors’ social position. Academy of Management Proceedings, (1), 1–6. Battilana, J., Leca, B., & Boxenbaum, E. (2009). How actors change institutions: Towards a theory of institutional entrepreneurship. The Academy of Management Annals, 3, 65–107. Blake-Beard, S. D. (2001). Taking a hard look at formal mentoring programs: A consideration of potential challenges facing women. Journal of Management Development, 20(4), 331–345. Boxenbaum, E., & Battilana, J. (2005). Importation as innovation: Transposing managerial practices across fields. Strategic Organization, 3(4), 355–383. Bragg, A. (1989). Is a mentor program in your future? Sales & Marketing Management, 141(11), 54–63. Burke, R. J., & McKeen, C. A. (1997). Not every managerial woman who makes it has a mentor. Women in Management Review, 12(4), 136–139. Burt, R. S. (1998). The gender of social capital. Rationality and Society, 10(1), 5–46. Castilla, E. J. (2005). Social networks and employee performance in a call center. American Journal of Sociology, 110(5), 1243–1283. Catalyst. (1998). ‘‘Introduction’’, advancing women in business – The catalyst guide. San Francisco, CA: Josey-Bass Publishers. Cox, T. H., & Blake, S. (1991). Managing cultural diversity: Implications for organizational competitiveness. The Executive, 5(3), 45–56. Cox, T., & Nkomo, S. (1991). A race and gender-group analysis of the early career experience of MBAs. Work and Occupations, 18(4), 431–446. Davis, G. F., & Greve, H. R. (1997). Corporate elite networks and governance changes in the 1980s. American Journal of Sociology, 103(1), 1–37. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48(2), 147–160. Dobbin, F. (2009). Inventing equal opportunity. Princeton, NJ: Princeton University Press. Dobbin, F., Kim, S., & Kalev, A. (2011). You can’t always get what you need: Organizational determinants of diversity programs. American Sociological Review, 76, 386–411. Dobbin, F., & Sutton, J. (1998). The strength of a weak state: The employment rights revolution and the rise of human resources management divisions. American Journal of Sociology, 104, 441–476. Douglas, C. A. (1997). Formal mentoring programs in organizations: An annotated bibliography. Greensboro NC: Center for Creative Leadership. Edelman, L. B. (1992). Legal ambiguity and symbolic structures: Organizational mediation of civil rights law. American Journal of Sociology, 97(6), 1531–1576. Edelman, L. B., Fuller, S. R., & Mara-Drita, I. (2001). Diversity rhetoric and the managerialization of law. American Journal of Sociology, 106, 1589–1641. Fligstein, N. (1990). The transformation of corporate control. Cambridge, MA: Harvard University Press. Friedland, R., & Alford, R. R. (1991). Bringing society back in: Symbols, practices, and institutional contradictions. In W. W. Powell & P. J. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 232–266). Chicago, IL: University of Chicago Press. Green, S. G. (1995). Top management support of R&D projects: A strategic leadership perspective. IEEE Transactions on Engineering Management, 42, 223–232.

Instantiation of Institutional Logics

271

Greenwood, R., & Hinings, C. R. (1996). Understanding radical organizational change: Bringing together the old and the new institutionalism. Academy of Management Review, 21, 1022–1054. Greenwood, R., Raynard, M., Kodeih, F., Micoletta, E., & Lounsbury, M. (2011). Institutional complexity and organizational response. The Academy of Management Annals, 5, 317–371. Greenwood, R., Suddaby, R., & Hinings, C. R. (2002). Theorizing change: The role of professional associations in the transformation of institutionalized fields. The Academy of Management Journal, 45(1), 58–80. Hardy, C., & Maguire, S. (2008). Institutional entrepreneurship. In R. Greenwood, C. Oliver, R. Suddaby & K. Shalin-Anderson (Eds.), Thousand Oaks, CA: Sage. Haveman, H. A. (1993). Follow the leader: Mimetic isomorphism and entry into new markets. Administrative Science Quarterly, 38(4), 593–627. Herring, C. (2009). Does diversity pay?: Race, gender, and the business case for diversity. American Sociological Review, 74(2), 208–224. Ibarra, H. (1993). Personal networks of women and minorities in management: A conceptual framework. The Academy of Management Review, 18, 56–87. Jackson, S. E., Brett, J. F., Sessa, V. I., Cooper, D. M., Julin, J. A., & Peyronnin, K. (1991). Some differences make a difference: Individual dissimilarity and group heterogeneity as correlates of recruitment, promotions, and turnover. Journal of Applied Psychology, 76, 675–689. Kalev, A., Dobbin, F., & Kelly, E. (2006). Best practices or best guesses? Assessing the efficacy of corporate affirmative action and diversity policies. American Sociological Review, 71, 589–617. Kanter, R. M. (1977). Men and women of the corporation. New York, NY: Basic Books. Kochan, T., et al. (2003). The effects of diversity on business performance: Report of the diversity research network. Human Resource Management, 42, 3–21. Lawlor, J. (1997). Mentoring meets networking in formal programs. The New York Times, 30(November), 215–229. Lounsbury, M. (2007). A tale of two cities: Competing logics and practice variation in the professionalizing of mutual funds. Academy of Management Journal, 50, 289–307. Lounsbury, M., Ventresca, M., & Hirsch, P. (2003). Social movements, field frames and industry emergence: A cultural-political perspective on U.S. recycling. Socio-Economic Review, 1, 71–104. Maguire, S., Hardy, C., & Lawrence, T. B. (2004). Institutional entrepreneurship in emerging fields: HIV/AIDS treatment advocacy in Canada. The Academy of Management Journal, 47(5), 657–679. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83, 340–363. Meyer, J. W., & Scott, W. R. (1983). Organizational environments: Ritual and rationality. CA: Sage. Ocasio, W. (1997). Towards an attention-based view of the firm. Strategic Management Journal, 18, 187–206. Orton, J. D., & Weick, K. E. (1990). Loosely coupled systems: A reconceptualization. Academy of Management Review, 15, 203–223. Pache, A.-C., & Santos, F. (2010). When worlds collide: The internal dynamics of organizational responses to conflicting institutional demands. Academy of Management Review, 35, 455–476.

272

SHAWNA VICAN AND KIM PERNELL-GALLAGHER

Payne, S. C., & Huffman, A. H. (2005). A longitudinal examination of the influence of mentoring on organizational commitment and turnover. The Academy of Management Journal, 48(1), 158–168. Pfeffer, J., & Salancik, G. (1978). The external control of organizations: A resource dependence perspective. Palo Alto, CA: Stanford University Press. Powell, W. W., & Colyvas, J. A. (2008). Microfoundations of institutional theory. In R. Greenwood, C. Oliver, R. Suddaby & K. Sahlin (Eds.), Handbook of organizational institutionalism (pp. 276–298). London: Sage. Powell, W. W., & DiMaggio, P. J. (1991). The new institutionalism in organizational analysis. Chicago, IL: University of Chicago Press. Ragins, B. R. (1989). Barriers to mentoring: The female manager’s dilemma. Human Relations, 42, 1–22. Ragins, B. R., & Cotton, J. L. (1999). Mentor functions and outcomes: A comparison of men and women in formal and informal mentoring relationships. Journal of Applied Psychology, 84(4), 529–550. Ragu-Nathan, B. S., Apigian, C. H., Ragu-Nathan, T. S., & Tu, Q. (2004). A path analytic study of the effect of top management support for information systems performance. Omega, 32, 459–471. Reay, T., & Hinings, C. R. (2009). Managing the rivalry of competing institutional logics. Organization Studies, 30(6), 629–652. Reskin, B. F., & McBrier, D. B. (2000). Why not ascription? Organizations’ employment of male and female managers. American Sociological Review, 65, 210–233. Richard, O. C., Kochan, T. A., & Mcmillan-Capehart, A. (2002). The impact of visible diversity on organizational effectiveness: Disclosing the contents in Pandora’s black box. Journal of Business and Management, 8, 1–26. Roberson, Q. M. (2006). Disentangling the meanings of diversity and inclusion in organizations. Group & Organization Management, 31(2), 212–236. Scott, W. R. (2003). Institutional carriers: Reviewing modes of transporting ideas over time and space and considering their consequences. Industrial and Corporate Change, 12(4), 879–894. Sims, A. D. (2002). An inter-group examination of African-American executives’ mentoring relationships: Traversing the invisible hurdles of corporate America. Unpublished doctoral dissertation. Rutgers, The State University of New Jersey. Strang, D., & Meyer, J. W. (1993). Institutional conditions for diffusion. Theory and Society, 22, 487–511. Swidler, A. (1986). Culture in action: Symbols and strategies. American Sociological Review, 51(2), 273–286. Thomas, D. (2001). The truth about mentoring minorities: Race matters. Harvard Business Review, 79, 98–107. Thornton, P. (2004). Markets from culture: Institutional logics and organizational decisions in higher education publishing. Stanford, CA: Stanford University Press. Thornton, P. H., & Ocasio, W. (1999). Institutional logics and the historical contingency of power in organizations: Executive succession in the higher education publishing industry, 1958–1990. American Journal of Sociology, 105, 801–843. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure, and process. Oxford: Oxford University Press. Viator, R. (2001). The association of formal and informal public accounting mentoring with role stress and related job outcomes. Accounting, Organizations and Society, 26, 73–93.

Instantiation of Institutional Logics

273

Wanberg, C., Welsh, E., & Hezlett, S. (2003). Mentoring research: A review and dynamic process model. Research in Personnel and Human Resources Management, 22, 39–124. Wernick, E. D. (1994). Preparedness, career advancement, and the glass ceiling. Washington, DC: U.S. Department of Labor Glass Ceiling Commission. Winterle, M. (1992). Work force diversity: Corporate challenges, corporate responses. New York, NY: Conference Board. Wright, P., Ferris, S. P., Hiller, J. S., & Kroll, M. (1995). Competitiveness through management of diversity: Effects on stock price valuation. Academy of Management Journal, 38(1), 272–287. Young, R., & Jordan, E. (2008). Top management support: Mantra or necessity? International Journal of Project Management, 26, 713–725.

THE INTERNAL COMPLEXITY OF MARKET LOGICS: FINANCIAL SOPHISTICATION AND PRICE DETERMINATION Vince Feng ABSTRACT Economic theory posits a universal sociocultural orientation toward pricing complicated only by systematic cognitive biases. While institutional and organizational theorists have challenged the purported homogeneity of market logics, they have not linked market heterogeneity to price outcomes. If market logics are internally complex with multiple orientations toward pricing, skilled actors should be able to influence prices through market logics. This study utilizes qualitative analysis of interview data with a stratified random sample (75 percent response rate) of key participants to examine how investment banks (underwriters) instantiate a hybrid market logic in the Initial Public Offering (IPO) market. Underwriters exploit their status position to promulgate IPO pricing methods contradicting neoclassical rationality, behavioral models of pricing, and the underwriters’ own calculative mode of behavior. They successfully create this hybrid logic for issuers while hiding the nature of

Institutional Logics in Action, Part B Research in the Sociology of Organizations, Volume 39B, 275–317 Copyright r 2013 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0733-558X/doi:10.1108/S0733-558X(2013)0039A&B022

275

276

VINCE FENG

their market power through deceptive use of vocabulary from the market logic itself. Hence, the internal complexity of market logics directly impacts financial prices, with skilled actors achieving superior outcomes. This study concludes with an assessment of the implications for price theory, developing propositions to guide future research on market logics and pricing. Keywords: Hybrid logics; market vocabulary; price theory; initial public offering; sociology of markets

Modern sociology has largely abdicated price theory to neoclassical and behavioral economics, and in so doing failed to investigate the universal sociocultural orientation postulated in such price models. While institutional and organizational theorists have challenged the homogeneity of market logics (Dobbin, 1994; Fligstein, 1990; Lounsbury, 2007; Marquis & Lounsbury, 2007; Zajac & Westphal, 2004), they have not linked these insights to the market price mechanism. Without articulating a link between market heterogeneity and prices, analysts implicitly accept the economic assertion of a culturally homogeneous price mechanism even in markets exhibiting institutional pluralism. This study challenges the assertion of cultural homogeneity in market pricing by investigating the micro-level processes driving price formation within a segment of US equity capital markets. The core hypothesis of the institutional logics paradigm is that rationality and values vary by institutional orders (Thornton, Ocasio, & Lounsbury, 2012, pp. 2–4), or as Weber termed them ‘‘value spheres’’ (1958 [1916]). Alongside material forces, vocabularies and symbolic action associated with institutional logics express and enact these value spheres. If markets truly exhibit varying rationality and values by institutional orders, skilled or powerful actors should be able to utilize these differing logics to affect the price outcome. I apply the institutional logics framework, in particular the work on how vocabularies generate meaning in instantiating logics (Burke, 1984 [1935]; Mills, 1939; Loewenstein, Ocasio, & Jones, 2012), to the microlevel process of pricing Initial Public Offerings (IPOs) in US capital markets. This study examines how powerful investment banks (underwriters) successfully create a hybrid logic for issuers while hiding the nature of their market power through deceptive use of vocabulary from the market logic. This hybrid logic contradicts the homogeneous market logic of neoclassical

The Internal Complexity of Market Logics

277

rationality, behavioral models of pricing, and the underwriters’ own calculative mode of action. IPOs are the initial listing of a private company’s shares on a public equity market. When a company wants to list their shares for the first time in the United States, they must engage an underwriter to manage the offering. Underwriters have a vested interest in pricing IPOs attractively for buyers and attempt to indoctrinate neophyte issuers in a ‘‘sophisticated’’ orientation that such price outcomes are proper. Due to this active proselytization by underwriters, IPO prices exhibit high first-day returns contradicting neoclassical theory that economists term ‘‘underpricing.’’ Despite over 30 years of research in financial economics, IPO underpricing remains an unresolved price anomaly (Loughran & Ritter, 2002; Ritter & Welch, 2002). Underwriters promote action oriented toward a standard of ‘‘sophistication’’ grounded in the vocabulary of financial economics, but actually at odds with economic price models. In other words, the IPO market exhibits institutional pluralism, with underwriters skillfully utilizing vocabularies to instantiate a hybrid market logic for issuers. I investigate this market hybridization by studying the vocabularies of practice (Loewenstein et al., 2012, pp. 45, 63; Thornton et al., 2012, p. 159) that shape reasoning and perceived rationality, conducting in-depth interviews with the primary participants in IPOs: underwriters, issuers, and large institutional buyers of IPOs. The interview respondents are drawn from a stratified random sample with an overall response rate of 75 percent. I first explore the sociocultural homogeneity underlying economic theories of market price before examining how institutional logics theoretically, and IPO markets empirically, complicate this assertion. This study concludes with a discussion of the implications for prominent behavioral theories of IPO pricing and propositions to guide future research on market logics and pricing.

PRICE THEORY AND MARKET HOMOGENEITY1 Neoclassical price theory formalizes calculative rationality among atomized actors: market participants with rational preferences and expectations process information to maximize utility against resource constraints (Dybvig & Ross, 2003; Manski, 2000). Rational preferences involve making normatively ‘‘acceptable’’ choices consistent with Savage’s subjective expected utility (SEU), and rational expectations entail observational learning based on ‘‘appropriately’’ updating beliefs with new information

278

VINCE FENG

in accordance with Bayes’ theorem (Barberis & Thaler, 2003).2 The collective action of individual rational actors through the market mechanism of arbitrage determines prices, which Coleman denotes as the micro– macro link in neoclassical theory (1990). Applied to stock prices, actors should perform mean-variance optimization: the mean excess return for each asset should be proportional to the marginal contribution of volatility in the actor’s optimal portfolio. The resulting asset-price model – Capital Asset Pricing Model (CAPM) – equates a stock’s excess return over risk-free returns to its exposure to relevant risks (Fama & French, 1992).3 Efficient Market Hypothesis (EMH) further asserts that a functioning market does not require all actors being rational. As long as irrational reactions are random and follow a normal distribution so that the net price impact cannot be exploited to make excess returns, then market prices remain the best indicator of intrinsic value. Hence, EMH predicts that stock prices equal intrinsic value, defined as the discounted present value (PV)4 of future dividends (Fama, 1965, 1976, 1990). Equivalently, returns from purchasing stock at prevailing market prices should only reflect compensation for exposure to systematic risks, with no investor being able to earn excess returns in the long run. Neoclassical theory thus entails an extremely strong assertion of the cultural homogeneity of market logics. In particular, systematic action must be rational, with irrationality limited to random occurrences. To the extent market heterogeneity exists, it must be peripheral to the logic of rationality in pricing, not affecting SEU preferences, Bayesian updating, or the ability of rational actors to arbitrage away irrational price behavior. Hence, divergence from the sociocultural logic of neoclassical price determination cannot involve socially shared perceptions, as these entail systematic rather than random variation. While ostensibly at odds with neoclassical theory, behavioral finance also asserts a culturally homogenous model of action within the market logic. Behavioral theory relaxes either the assumption of Bayesian updating or rational SEU preferences, claiming that actors are systematically biased or nonrational (Barberis & Thaler, 2003; Hirshleifer, 2001). Nonrational investor models within behavioral finance posit the presence of nonBayesian investors, generally referred to as ‘‘noise traders’’ or ‘‘sentiment investors.’’ Baker and Stein (2004) demonstrate that sentiment investors can dictate prices by driving rational investors out of the market due to shortsale constraints. These sentiment investors are prone to nonrational optimism and underweighting information relevant to calculating intrinsic values, violating Bayes’ theorem (Kaustia & Knu¨pfer, 2008; Odean, 1998).

The Internal Complexity of Market Logics

279

When sentiment levels are high, these noise traders increase both trading volume and prices beyond rational levels, as rational investors cannot short sell the overpriced shares.5 Similarly, nonrational issuer models within behavioral finance generally relax the assumption of SEU preferences. Prospect theory remains the dominant non-SEU theory of preferences within behavioral finance (Barberis & Huang, 2008; Barberis, Huang, & Santos, 2001; Barberis & Thaler, 2003; Benartzi & Thaler, 1995; Kahneman & Tversky, 1979; Tversky & Kahneman, 1992). Instead of normative preferences, prospect theory utilizes cognitive psychology to argue that utility is defined relative to an arbitrary reference level, with actors exhibiting loss-aversion.6 Given these empirical preferences, framing and mental accounting effects matter since the utility of actors are reference-level dependent and nonlinear (Kahneman, Knetsch, & Thaler, 1991; Thaler, 1999). For neoclassical theory, actors are rational and observe Bayesian updating and normative preferences; for behavioral theory, systematic variation in cognitive biases exists, as actors are nonrational and operate under these nonrational models of behavior (systematically non-Bayesian updating and nonrational preferences). Both theory camps thus reject sociocultural variation within the market logic for pricing. Market struggles among actors espousing preferences (whether normative or empirical) and processing information in the same way (whether Bayesian or non-Bayesian) determine pricing. I argue that even within the market logic of high finance, social power influences rationality as ‘‘rationality’’ is in fact a set of shared meanings that actors enact to establish order (Dobbin, 1994; Dobbin & Dowd, 2000). Differing market logics could thus entail different forms of rationality that alter how actors price securities.

INSTITUTIONAL LOGICS AND MARKET HETEROGENEITY Institutional logics are cultural assumptions, values, and beliefs that inform how actors perceive and interpret the environment (Friedland & Alford, 1991; Thornton et al., 2012). As Friedland (forthcoming) explains in his contribution to this volume, the concept of an interinstitutional system of oftentimes contradictory cultural orders resonates with Weber’s work on ‘‘social life as a polytheism of values in combat with one another’’ (Gerth & Mills, 1958 [1946], p. 70). Importantly, the logics emerging from these

280

VINCE FENG

conflicting value spheres entail distinctly different cultural systems for measuring value (Friedland & Alford, 1991, pp. 232–235). While Friedland and Alford discuss heterogeneity at the macro-societal level, research has increasingly pushed the concept of institutional heterogeneity down to the organizational level (Greenwood, Raynard, Kodeih, Micelotta & Lounsbury 2011; Thornton et al., 2012). Building on the core insight that cultural systems for generating and measuring values can vary systematically across social groups, the institutional logics literature has directly challenged the cultural homogeneity of markets. Numerous studies delineate how markets and organizations espouse coexisting or hybrid logics incorporating elements from multiple cultural systems of meaning (e.g., Battilana & Dorado, 2010; Dunn & Jones, 2010; Greenwood & Hinings, 1996; Pache & Santos, 2010; Reay & Hinings, 2009; Tracey, Phillips, & Jarvis, 2011). Scholars have already established that heterogeneity exists in financial markets as well. This heterogeneity could result from shifts in logics, such as those experienced in financial advisory (Lounsbury, 2002) and financial markets (Zajac & Westphal, 2004), or from political contestation and coexisting logics, such as with the commercial banking (Marquis & Lounsbury, 2007) and mutual fund industries (Lounsbury, 2007; Lounsbury & Rao, 2004). While analysts have identified institutional pluralism in financial markets, they have not linked such sociocultural heterogeneity with price outcomes. This is troubling for several reasons. First, if markets truly exhibit heterogeneity in logics, powerful or skilled actors should be able to utilize these differing cultural systems of measuring values to affect price outcomes. Second, even in the absence of active manipulation by powerful actors, institutional groups espousing different cultural systems of measuring values should generate different price outcomes. If market heterogeneity cannot be linked to actual price outcomes, the price mechanism remains culturally homogeneous. In short, markets would still enforce a universal cultural orientation for price determination. Price theorists could then safely disregard institutional logics as either superfluous or epiphenomenal. This study contributes to the literatures on institutional logics and price determination by explicitly examining how sociocultural heterogeneity impacts financial prices. In particular, I delineate how powerful actors can skillfully employ vocabularies associated with the market logic to undermine such logic with regards to financial pricing.

Vocabularies Instantiating Logics Vocabularies of practice are ‘‘systems of labeled categories used by members of a social collective to make sense of and construct organizing practices’’

The Internal Complexity of Market Logics

281

(Thornton et al., 2012, p. 159). Understanding such vocabularies is critical as logics necessarily entail a symbolic component (Thornton et al., 2012, p. 150), with vocabulary structure helping identify implicit categories that comprise institutional logics (Loewenstein et al., 2012, p. 72). The institutional logics literature has experienced a renewed focus on vocabularies (e.g., Jones & Livne-Tarandach, 2008; Nigam & Ocasio, 2010; Suddaby & Greenwood, 2005). Furthermore, studies have repeatedly shown that vocabularies affect financial market outcomes. For example, Hirsch (1986) analyzed how the linguistic framing of hostile takeovers legitimated what was initially a ‘‘deviant innovation,’’ enabling its subsequent diffusion in financial markets; Babb (1996) identified and traced signifying words in the mid-19th century that helped align interests in the social movement of debtors against creditors; Suddaby and Greenwood (2005) examined how actors strategically developed vocabularies in the contestation over the legitimacy of the multidisciplinary organizational form; Ferraro, Pfeffer, and Sutton (2005) explored how the vocabulary of economic theory reinforces the theory’s assertions on human motivation; and Zhong, Loewenstein, and Murnighan (2007) demonstrated that the labeling of the choices available in a prisoner’s dilemma game significantly impacted player actions as labels such as ‘‘cooperate’’ and ‘‘defect’’ imposed meaning on otherwise numeric alternatives. While a prisoner’s dilemma game can be broadly construed as price negotiation, none of these studies directly linked either vocabularies or logics to real-world market price outcomes. As evidenced by the above examples, ‘‘vocabularies not only convey but also generate meanings constitutive of practice y [they are] structured systems of cultural categories that generate meaning and enable and constrain social practices’’ (Loewenstein et al., 2012, p. 63). Vocabularies enable both persistence and change. While market actors learn how to perceive the environment based on the relevant vocabularies of practice, they also select vocabularies and words to shape action since words are cultural resources generating meaning (Burke, 1984 [1935]; Jones & LivneTarandach, 2008; Mills, 1939; Suddaby & Greenwood, 2005). An investigation of market vocabulary thus allows for the examination of not only market heterogeneity, but also how skilled actors can affect the price outcome through such heterogeneity. Just as the institutional logics paradigm resonates with Weber’s view of value polytheism, analyzing vocabularies to explain economic outcomes harkens back to Weberian methodology. Weber’s detailed exegesis of the Protestant term ‘‘calling’’ (Beruf) forms a central part of his argument that the Protestant logic arising from the Reformation pharisaically imbued the chrematistic orientation with ethical duty, fundamentally altering economic

282

VINCE FENG

action. As Weber states, ‘‘the religiously lively epoch of the seventeenth century bequeathed to its utilitarian heirs was above all a startlingly clear conscience – we can say without hesitation, pharisaically good – as concerns the acquisition of money’’ (Weber, 2009 [1905], p. 154). For Weber, this effect was pharisaical in part because the effect defied the logic of Lutheran dogma that originated the term.7 Similarly, I will show how underwriters use the vocabulary of the market logic to achieve pharisaical outcomes: strategic use of market vocabulary instantiates a hybrid logic that actually contradicts the price mechanism of the original market logic, linking market heterogeneity directly to price determination.

Initial Public Offerings and Heterogeneity in Pricing The IPO market is an excellent research site for investigating market heterogeneity in pricing, as financial markets should represent the high capitalism of utmost calculative rationality. When a company wants to access equity capital markets for the first time in the United States, they must engage an investment bank to underwrite the new issuance of stock. In the United States, underwriters price IPOs in a two-stage process. In the first stage, only certain institutional investors – primarily hedge, mutual, pension, and private equity funds – negotiate with issuers to determine offer prices through the underwriter-mediated order collection and allocation (bookbuilding) process. The underwriter recommends an indicative price range to approach institutional investors with, and issuers either acquiesce or negotiate the price range with the lead underwriter. Once the price range is agreed upon, meetings with institutional investors (roadshow) commence and bookbuilding demand helps determine the final offer price to institutional investors.8 After underwriters complete this first stage of allocating and selling the offering to institutional investors, the stock begins trading on the general market, and retail investors can purchase shares from these institutional investors or from the lead underwriter in the second stage of IPO pricing. The first-day return is the increase in share price from the offer to the closing price on the first day of trading (i.e., the difference between the two stages of pricing). IPO pricing remains a significant area of study due to its theoretical import as a test of EMH. Since the second-stage outcome generally occurs only one day after the first-stage outcome, IPO first-day returns are purged of the explanatory factors underlying neoclassical theory. In other words, each company remains unchanged in its exposure to systematic risks to

283

The Internal Complexity of Market Logics

market returns, size, and value (CAPM), and no new information on future dividends could cause rational investors to update their estimates of intrinsic value (EMH). Hence, economists describe first-day returns as ‘‘underpricing’’ since no new information or changed circumstances could account for such immediate increases in share price from a rational actor perspective. IPO first-day returns averaged 19.0 percent from 1980 to 2011 on a proceeds-weighted basis, costing issuers a cumulative total of $132 billion, or on average $4.1 billion a year (Fig. 1). The persistence of positive first-day returns and the inability of EMH to accommodate such underpricing are findings that remain robust across econometric methods and time periods, having been replicated in numerous studies over the past three decades (Ritter & Welch, 2002). Economists have attempted to weaken EMH to account for first-day returns by noting agency

60

40

40

30

30 20 20

10

Average First-day Returns (percentage)

Loss due to Underpricing (US$ billions)

50

10

0

0 1980

1985

1990

1995

2000

2005

2010

Year

Fig. 1. US Initial Public Offering First-day Returns (1980–2011). Note: Average first-day returns calculated on a proceeds-weighted basis. Data accessed from Jay R. Ritter’s IPO database: http://bear.warrington.ufl.edu/ritter/ipodata.htm

284

VINCE FENG

costs, substitution costs, information asymmetries, or employing behavioral theories (e.g., Beatty & Ritter, 1986; Habib & Ljungqvist, 2001; Krigman, Shaw, & Womack, 1999; Loughran & Ritter, 2002, 2004; Ljungqvist & Wilhelm, 2003; Ritter, 1984, 1991; Ritter & Welch, 2002). Agency theory, while espousing EMH, predicts that underpricing may be necessary to align the interests of underwriters with issuers (Fama & Jensen, 1983; Jensen & Meckling, 1976). Others have suggested that underpricing is a substitute for marketing and legal expenses, or that issuers purchase analyst coverage with such underpricing. Information asymmetry predicts that the reputation of knowledgeable backers – underwriters, venture capital, and private equity investors – certifies the quality of the issuer to the broader market, reducing first-day returns (Carter, Dark, & Singh 1998; Carter & Manaster, 1990). Nonrational investor models from behavioral finance hypothesize that noise traders and limits to arbitrage underlie first-day returns. For instance, the changing risk competition hypothesis predicts that riskier IPOs will exhibit higher first-day returns (Ritter, 1984; Loughran & Ritter, 2004), or information revelation and sentiment models that predict higher underpricing during periods of higher investment sentiment (Derrien, 2005). Nonrational issuer models from behavioral finance hypothesize that the issuer either accepts or seeks underpricing. For instance, the realignment of incentives hypothesis predicts that professional managers with lower share ownership versus owner-managers may seek higher first-day returns (Ljungqvist & Wilhelm, 2003), or prospect theory loss-aversion and mental accounting that predict increased secondary selling lowers underpricing. However, these explanations do not adequately explain the degree, period effects, or international differences in first-day returns. I argue that these theories have produced mixed results because they overlook market heterogeneity in logics. In particular, underwriters promulgate a ‘‘sophisticated’’ orientation for issuers complicating the assertion of a homogeneous logic for pricing.

METHODOLOGY I investigate the IPO negotiation through in-depth interviews with key actors drawn from the IPO pricing process. I chose one-on-one interviews as the primary method of data collection to better explore how actors conceive of themselves and understand vocabularies of practice. Interviewing in a mixed environment where different parties are all present would have hindered such an exploration given the deception involved in IPO

The Internal Complexity of Market Logics

285

interactions described in the findings. While an ethnographic study of the IPO process from start to finish with different issuers and investment banks could also address the research question, those meetings are strictly confidential. However, by exploring how each individual actor views the process, I can accurately capture their interactions while better highlighting their differing individual viewpoints. Although interviews are the primary data collection method, I was in one instance able to gather ethnographic data at a respondent’s workplace while his team was coordinating several IPOs over the course of an eight-hour period. Also, in my former capacity working in investment banking and private equity, I was involved as either an underwriter or investor in over 30 IPOs.

Stratified Random Sampling I applied scientific standards of sampling to increase representativeness and minimize selection bias, focusing on interviewing senior executives at leading underwriters working in the divisions most directly responsible for IPO pricing. Internally, underwriters enforce a strict professional hierarchy, with managing directors (MDs) at the top of the organizational pyramid making executive decisions and representing the firm to outside actors. Externally, underwriters conform to a status hierarchy best documented by the stringent rule-based name placement ceremony involved in advertising completed transactions.9 Underwriters perform numerous roles and serve different clients during the IPO process: the Investment Banking Division (IBD) serves the issuers and prepares the documentation; Equity Sales serves the institutional investors and contacts them about upcoming roadshow meetings; and Equity Capital Markets (ECM) coordinates and manages the bookbuilding process, centralizing investor orders into an order book and allocating shares among institutional investors. ECM also performs the crucial functions of finalizing the price range indicated to investors during the roadshow and recommending a final offer price based upon bookbuilding demand. I thus focused on interviewing MDs in the IBD and ECM divisions of leading investment banks. I set the maximum number of interviews at 20 and stratified the respondents according to the three categories of actors (underwriters, institutional investors, and issuers). With underwriters as the focus of the study, I reserved half (10) of the interview slots for investment banks, and divided the remainder between issuers and investors (5 each). Among the investment banks, I selected the eight leading underwriters as the population

286

VINCE FENG

for sampling.10 I further stratified the investment banks along the two divisions most heavily involved in the IPO process, IBD and ECM, resulting in 16 possible combinations of banks and divisions. Then, I randomly selected a bank and a division using a random number generator, removing selected combinations from the pool, and contacted the highest ranking MD at the departments selected, usually the head or co-head of the department. I was generally directed to other MDs for interviews, but directly interviewed the department head in two instances. Given the far greater number and diversity of institutional investors, I asked bankers whom I interviewed which investors they recently sold IPO shares to, cross-referencing their nominations with publicly available rankings of the top mutual and hedge funds in the United States. I then compiled a population universe of the 20 most mentioned funds, all of which were among the top 50 funds in the United States in terms of assets under management, and randomly selected five investors. As the population of issuers over a representative timeframe is overwhelming, I decided to only focus on the issuers with the strongest negotiating position vis-a`-vis underwriters and institutional investors. Hypothesizing that issuers backed by private equity firms would be better able to negotiate the IPO process, I randomly selected 5 of the top 20 US private equity firms in terms of funds under management and contacted them for an introduction to their portfolio company that had most recently gone public. Of the 20 slots, I secured interviews with eight underwriters (three from IBD and five from ECM), four institutional investors, and three issuers for a combined response rate of 75 percent. I had not met face-to-face with any of the interviewees prior to this study, although six mentioned that they had worked on transactions involving my former firm. Most interviews lasted at least an hour, and I was able to partially audio record 10 of the interviews. I transcribed all recordings and summarized hand-written field notes within 24 hours of interviews, and analyzed all transcriptions and interview notes in Atlas.ti (version 5). All interviews took place between September 2009 and August 2010. Please refer to Appendix B for a detailed interview roster and description of the data analysis procedures.

STUDY FINDINGS Emerging from the interviews is the conflict of interests and contentious negotiation typifying the IPO process. As respondents repeatedly emphasized without prompting, negotiation is at the very heart of the IPO pricing

The Internal Complexity of Market Logics

287

process: ‘‘At the end of the day, everything is a negotiation between the company [issuer] and the underwriter, and between the underwriter and the institutional investor base that prices the offering’’ (IBD2). While issuers, underwriters, and institutional investors all contest the outcome, underwriters occupy a privileged position in the negotiations and wield significant influence over its outcome: The IPO process is nothing but a negotiation between the interests of the issuer and the investors, and the bankers [underwriters] can push the issuers back or they can push the investors back. Right? So if they push the investors back they get a higher valuation, if they push the issuers down the investors get a lower valuation and more upside. (ECM1)

None of the respondents contradicted the finding that the negotiation process is the pricing process and that underwriters significantly influenced the negotiation outcome. The following sections explore the role of underwriters in shaping the perceived self-interest of issuers.

Conflicting Interests The conflict of interests inherent in the price negotiation drives the division of labor within the investment bank: When the roadshow starts you have two sides of the bank working simultaneously on the deal who have very opposite client interest. The bankers [IBD] obviously work with their client, the issuer, and are trying to get the best outcome for the issuer, whereas the sales force [Equity Sales division] covers the investors, trying to get the institutional investors the best outcome, either through a better deal or better allocation. So there is a tension exerted by both sides of the investment bank. Sitting in the middle is ECM, which deals with both sides. (IBD1)

Clearly, the underwriter serves clients with opposing interests and can align themselves with either group or remain neutral, as the following ECM banker articulates: ‘‘when we price a deal, again we try to not overprice it and have a bad aftermarket performance, so our investor clients lose money. But we also don’t want to really underprice it, because one you don’t want to have a reputation of always consistently underpricing deals’’ (ECM1). Issuers are often aware of this conflict of interests, as evidenced by the following comment from a private-equity backed company that listed several years ago: The role of the investment bank is to sell shares at the highest possible price at a given point in time, just like any vendor looking to sell products. Because the selling process is essentially a one-off event from the fee-making perspective, their interests are never

288

VINCE FENG

aligned with their clients’ [issuers] long-term best interests and are very often conflicted. (PEIS1)

Whose interests the underwriter serves depends on the relative importance of the clients to the investment bank, as well as the degree to which the client’s interests align with those of the underwriter’s. While investment banks earn less on the underwriting fee (gross spread) with increased firstday returns as fees are always quoted as a percentage of the offering size, they more than make up for the reduced fees in increased stabilization trading profits, rent-seeking behavior from investors during the allocation process (spinning), and quid pro quo with institutional buyers of the IPO. In order to facilitate the creation of an efficient and orderly market in the newly listed stock, the lead underwriter serves as the ‘‘stabilization agent’’ for a period of several weeks by buying and selling the issuer’s stock with the goal of stabilizing the price.11 Economic studies have shown that high firstday returns lead to increased volatility, translating into significantly increased trading profits for the lead underwriter approximating 2 percent of the offering during the first three months of the stabilization period (Aggarwal, 2000; Ellis, Michaely, & O’Hara, 2000, 2002; Ritter & Welch, 2002). A senior trader at one of the largest global hedge funds describes the quid pro quo aspect: The way it really works is like this. If ECM or the investment bank keeps on pricing things that make no money, they fuck their clients [the institutional investors], their clients won’t trade with them, it has much bigger repercussions to other parts of the business. Right? So if they price it cheap, their clients make money, like put it this way. If [a leading investment bank] comes out with an IPO, I make money because day one it goes up 16 percent, what am I gonna do? I’m gonna give them a couple more trades and say thank you for the IPO, thanks for the allocation, right? (HF2)

Hence, investment banks and institutional investors are generally aligned against the interests of the issuer from a purely pecuniary perspective. Deception as Art Most of the respondents unabashedly discussed times where they deceived others in the negotiation process. Indeed, they almost condoned deception (without calling it deception) as a normal form of discourse in the IPO process, and in that way reconciled it with their self-conceptions as good people who valued relationships and trust. The following banker discusses why they use discounted cash flow (DCF) analysis to estimate the price of an unlisted company’s stock. While he never uses the term deception, it is fairly

The Internal Complexity of Market Logics

289

apparent this method is chosen simply because it inflates the value of the company’s stock, which appeals to issuers at the very beginning of the IPO process: The reason we use DCF is because it actually gives a very high valuation, because obviously the issuers want to hear that they are worth much higher, as much as possible y at the pitchbook stage you want to show a very, very strong valuation, and DCF usually reflects a very strong valuation. Many variables you can change. DCF really looking far into the future. (ECM1)

Importantly, underwriters do not utilize DCF to estimate PV intrinsic values as hypothesized by neoclassical theory because they claim DCF calculations are necessarily inaccurate. Underwriters describe predicting the future period cash flows and discount rates (FV and i in equations from footnote 4) required for DCF analysis as ‘‘pure guesswork,’’ or in less flattering terms, ‘‘I pull those numbers out my ass most of the time’’ (IBD2). Instead, the emphasis is squarely on manipulating a pliable method (‘‘many variables you can change’’) to produce an exaggerated value, making the underwriter more attractive to issuers. But the inflated attractiveness of the underwriter’s early promises is usually ephemeral. As another banker noted, ‘‘the period just before we actually drop our price range on the cover [of the draft prospectus] y is the only time that really matters y any discussion of price before that is all around the beauty contest’’ (ECM2). The same banker goes on to state quite candidly ‘‘bankers and others talk about many, many, many different methods of valuation, all that’s baloney too, OK?’’ Despite the prevalence of deception, most market participants emphasized the importance of trust, relationship, and reputation: ‘‘the most important thing is relationship y general reputation is also very important’’ (IBD1), and ‘‘[if] I don’t trust these guys, that’s it. You don’t go in, it doesn’t matter what they say’’ (HF2). These same respondents also describe instances where they utilized deception. Many seem to reconcile deception with trust by indicating how difficult it is to price IPOs without utilizing deception. Again, respondents never called it deception, instead describing the process as ‘‘art not science,’’ ‘‘gut feel,’’ or ‘‘rule of thumb.’’ The art of deception is needed, respondents claim, because scientific methods of pricing such as DCF are inadequate: ‘‘for a DCF, 90 percent of the time usually DCF will give you an inflated valuation and is highly theoretical, the problem with DCF is that it relies on swings in forecasts’’ (IBD3). Because of this fundamental problem with scientific pricing methods, underwriters

290

VINCE FENG

explain that they must use other less scientific methods, encapsulated in the ‘‘art not science’’ refrain: I think anyone can come up with a high number [for the IPO price] if they want, but to make it credible you must have the right positioning y very much art not science y dollar amount is irrelevant, more relevant is positioning and the relevant comps. (IBD1)

Almost universally, underwriters claim that the ‘‘right’’ methodology is to look at already listed companies similar to the issuer (known as ‘‘comparables,’’ ‘‘comps,’’ or ‘‘peer’’ group). Underwriters and investors agree that you cannot get a deal done today without talking about the ‘‘comps.’’ The price to sales, assets, or earnings ratios (trading multiples) of comparable companies are multiplied to the sales, assets, or earnings of the issuer to calculate a reasonable valuation for the issuer once it becomes a public company. This valuation is also known as the ‘‘fully distributed’’ value of the firm. But even with comparable company analysis, underwriters must resort to ‘‘experience’’ to price an offering and not simple-mindedly accept ‘‘fully distributed’’ values, as I will discuss in the next section. The capriciousness of markets and inability of pricing methods to specify valuations with certainty condone deceptive industry practices in the minds of underwriters and investors. As one banker strongly emphasized, the markets are not predictable and this is referred to within the industry as ‘‘shit happens.’’ Because neophyte issuers are new to the market and do not yet comprehend that ‘‘shit happens,’’ underwriters justify deceiving them not only at the start of the process, but throughout, valorizing the ‘‘gut feel’’ approach to pricing as art. Issuers are also aware of the deception at times, and often utilize deception in their own right. As a banker notes, ‘‘the more sophisticated issuer, they would be looking for more real valuation rather than like a bullshit valuation that is inflated’’ (IBD3). Issuers admit to playing underwriters off one another, or feigning ignorance at critical junctures when it suits their purposes. But underwriters generally approve of the ‘‘sophisticated’’ issuer that understands and utilizes deception, perhaps since such issuer behavior justifies their own use of deception and further dignifies deception as a ‘‘sophisticated’’ art underappreciated by neophyte issuers.

Sophistication Has Its Costs The typology of ‘‘sophisticated’’ versus ‘‘unsophisticated’’ is extremely important to underwriters. Underwriters universally used and understood

The Internal Complexity of Market Logics

291

the meaning of this distinction as all eight of the investment bankers interviewed mentioned sophistication without prompting: There may be the very unsophisticated [issuer] that looks for highest [price] number outcome y although the company can push for another 50 cents or another dollar on the table, but I think the more sophisticated clients understand that the pricing outcome is a combination of their business, capital markets environment, how well their peer group is trading, and how the marketing process is coming along. (IBD2)

Sophistication entails not only an interpretation of the proper price outcome, but also an understanding of the art of deception involved in such price outcomes: ‘‘relatively unsophisticated clients y don’t know how an IPO works, how bankers can promise one thing at the very beginning and modify it at IPO time’’ (ECM1). All of the investment banking respondents agreed and used this definition of sophistication. The term itself denigrates issuers who do not accept a lower price outcome, making the moral judgment that they are somehow inadequate or inexperienced. As such, underwriters view it as their duty to educate new issuers to appreciate the nuances and aesthetics of the price negotiation: ‘‘in fairness they’re [issuers] not in the IPO market everyday, most of them are doing this their one and only time, so yeah it definitely requires some education about the whole process about how investors look at it’’ (ECM2). At stake for the underwriters is how issuers interpret the price negotiation and assess its outcome. For underwriters, sophisticated issuers would be able to look beyond such rudimentary measures as ‘‘fully distributed’’ or DCF values and understand what the ‘‘proper’’ outcome should be: ‘‘There’s got to be some discount, right? The so-called IPO discount. IPO discount is a function of a few things, I think of it as sort of the price of admission, OK?’’ (ECM2). Or in the words of a hedge fund manager, ‘‘has to be a discount. If there is no discount, then you know, hell, why’d you buy it? y you think you can make a few percent, or 5 to 10 percent because they are pricing it properly, then you go in’’ (HF2). A ‘‘properly’’ priced offering needs to be underpriced relative to comparable companies. IBD senior MDs (who represent the issuer’s interest within investment banks) to a person defined a successful IPO from the issuer’s perspective as one that is not discounted too much, but discounted ‘‘just right’’: I think a successful deal y trades up not too high, maybe between 15 to 30 percent good first day performance. (IBD1) I think a very good outcome is one where y the stock probably goes up at least in the high single digit percentage, anywhere between call it 8 to 20 percent people would say that is the sweet spot. (IBD2)

292

VINCE FENG

I want to see a stock price increase on the first day y pop 10 to 20 percent, that would be very nice. At least all the investors are happy, company won’t feel that we priced the IPO too low. That is a good, 10 to 20 percent, that is a good number. (IBD3)

Given the uniform conviction of all IBD MDs on this topic, I asked why the perfect outcome is a 10 percent discount rather than parity to ‘‘fully distributed’’ value. The question stumped them, one responding with a ‘‘I don’t know’’ reaction: hands outstretched with palms up, shrugging his shoulders while looking askance, a wry smile on his face. Another replied with laughter: ‘‘yeah, I don’t know why its 10 to 15 percent, it just sounds about right. Over time about 10 to 15 percent, because that’s probably the right amount to compensate people for the risk of taking on the IPO [laughter]’’ (ECM1). After the initial surprise, some respondents explicitly linked the IPO discount to ‘‘classic’’ economic theory (see ‘‘Upholding the Market Logic’’ section below). Others replied that pricing is ‘‘art not science,’’ explaining that experience gained from working on IPOs for the better part of two decades informed their judgment and conviction that IPOs must be underpriced by that amount for the proper outcome and the best interests of issuers. Underwriters utilize the art of deception to simultaneously reject scientific pricing methods while referencing economic theory to justify discounting to comparable company valuations, defining a ‘‘proper’’ outcome that ‘‘sophisticated’’ issuers should readily appreciate. And to some extent, they have been successful at inculcating this orientation in the minds of issuers. As ‘‘sophisticated’’ issuers remarked: ‘‘yes it [IPO discount] makes sense, but I certainly don’t think it should be double digits’’ (PEIS2) and ‘‘think the [IPO] discount could be a bit smaller, like 10 percent’’ (PEIS1). In other words, the underwriter has successfully moved the debate to the degree of underpricing, rather than the underpricing itself.

Control Empowers Proselytization Control of the IPO process is in the hands of the underwriter, who effectively manages the entire process on behalf of the issuer’s senior management team: We are like the project manager, we help submanage the IPO project, as you can imagine the CEO and CFO they have a lot of daily work, dangerous to get the CEO away from his daily job to focus on the IPO, we are more like the project manager, we help the company deal with lawyers, accountants, everyone together to get everything done. We only get management to review the disclosures to make sure its correct. We are more like the project manager role, and then at the same time we are also the marketers, we help

The Internal Complexity of Market Logics

293

market the story to investors, that is the research analyst job. We are also the sales team, our sales team will help the issuer to sell the stock to investors. Given that time, they almost outsource this IPO project to us, the underwriter, to help them to get everything done. That is basically the underwriter’s role. (IBD3)

As depicted in Fig. 2, underwriters occupy a critical position in the social system bridging IPO issuers to the capital markets.12 As network theorists have noted, ‘‘operative in a system of exchange with access restrictions might be ‘middleman’ behavior [involving] y of course a ‘commission’ or ‘fee’ in the process’’ (Marsden, 1983, p. 714). This privileged network position augments both the efficiency and status position of investment banks, enabling them to better distribute new issuer stock as well as enhancing their power of influence. Furthermore, decision-making and negotiations are often influenced by idiosyncratic personal differences, with each MD possessing his own tendencies and biases: Comes down to a very human level [laughter], so which banker is pushing which, which ECM guys are, if the ECM guy is very pro sales and trading, and he happens to be a very

Fig. 2. Exchange Network for US Initial Public Offerings. Note: Solid lines indicate frequent direct interactions; dotted lines indicate infrequent or impersonal interactions.

294

VINCE FENG

strong-headed ECM guy, a very established ECM guy, then the IPO is usually priced at a lower value. (ECM1)

For this very reason, issuers often choose investment banks based on personal ties or connections. Trust becomes ever more important in a situation where the broker can influence outcomes. The ability of the investment bank to push IPO prices up or down is due to the inherent uncertainty of financial market prices, where sometimes ‘‘perception is reality’’: An IPO that is perceived to be getting very good demand, of course the perception is coming from the communication between ECM and the investors, from ECM through to sales and trading to the investors. So IPOs that are perceived to be getting very strong demand will get more demand. And then usually more demand means that you can price the IPO at closer to the top end of the [price] range. (ECM1)

The vagaries of market outcomes paradoxically underpin both trust and deception in the IPO process. Because underwriters can influence market outcomes through the art of deception by convincing investors that a transaction has very strong demand from other investors, investors must also rely upon deception, often withholding their true demand information from the underwriter until the very last day of the roadshow. Importantly, this contradicts the assumptions of information revelation explanations (e.g., Benveniste & Spindt, 1989) where underwriters supposedly need to pay investors with lower offer prices for their demand information. Instead, this study suggests that underwriters and investors are simply engaged in mutual deception. Because underwriters can influence the price outcome by deceiving investors, investors reciprocate with deception as well. Since deception is so prevalent and market outcomes malleable, trust becomes highly valued: The other key thing is really the relationship between the banker and the company. For any IPO really, if one of the top investment banks wants to pull their full resources behind it, you can have a very successful IPO, in terms of generating demand, getting the best clients to subscribe to the deal, in terms of research coverage, getting the best coverage to come out to market frequently with research, and even have more upbeat research reports. So with any company any bank can do a good job, have resources to do a good job. But of course the banks have other IPO deals to work with. So essentially from an issuer’s perspective, if you have a very strong relationship with a banker, who is influential, who is a mover and shaker who can best mobilize his bank’s resources, then you feel more comfortable giving him the IPO mandate to that bank. (ECM1)

The underwriter’s control of the process facilitates their art of deception in an environment of extreme uncertainty, engendering a strong regard for trust by issuers and investors who are in less advantaged positions for the

The Internal Complexity of Market Logics

295

negotiation process. In other words, issuers and investors simultaneously recognize the rarity and value of trust and accept the prevalence of deception.

Status and Power Asymmetries In addition to their control of the situation, underwriters also command a distinct advantage in power and social position vis-a`-vis issuers. Issuers relinquish control to underwriters partly because they are new to the financial markets, as one investor describes: You’ve got two out of the three components of this deal that know the system very well, which is the buyers, a bunch of institutional investors, and the investment banks. And then when it comes to the issuer, they’re kind of like the new guy, that does this once in a corporate lifetime, and there are some investors who know that better y but for the most part, you’ve got a pretty big asymmetry and so, its easy to go in and say, gee we need an IPO discount y (HF1)

Underwriters utilize their privileged network position intermediating issuers and the capital markets to control the IPO process and educate issuers on the normative role behavior expected of them by the markets: the cultural frames of interpretation encapsulated in the term ‘‘sophistication.’’ Another investor notes how the inexperience of new issuers allows underwriters to accomplish this indoctrination to the markets: The issuers, they have no idea, right? Most of these issuers have never done it before, maybe a half of them or a third of them have done it before, the other two thirds have not, they’ll just believe what ECM tells them is a fair price when there needs to be a discount, right? (HF2)

Recognizing the conflicting interests inherent in the IPO negotiation, underwriters developed a standard practice of IPO pricing that involves: rejection of scientific methods of valuation such as DCF, upholding the use of alternative valuation methodologies such as ‘‘comps,’’ construction of the ‘‘IPO discount’’ as an inviolable aspect of pricing, and maintaining that the IPO discount is congruent with the market logic. Through their control of the IPO process due in large measure to their advantageous social position, they justify these practices to issuers by glorifying the importance of experience and gut feel, emphasizing that IPOs are ‘‘art not science,’’ and dutifully educating issuers on how to appreciate this art by increasing their ‘‘sophistication.’’ Importantly, issuers espousing this hybrid logic generate IPO first-day returns, as underwriters cannot force issuers to sell their shares

296

VINCE FENG

at a discount. Issuers must freely assent to such pricing, in the process violating the calculative rationality of neoclassical theory.

Upholding the Market Logic Underwriters repeatedly use word-to-word connotation to link this hybrid logic to the market logic of neoclassical theory, with ‘‘sophistication’’ in reference to ‘‘financial sophistication.’’ Thus, underwriters simultaneously valorize neoclassical theory while they promulgate the ‘‘sophisticated’’ hybrid logic that violates calculative rationality. Underwriters explain the seeming paradox of why IPOs must be priced at a discount to already listed comparable companies (if they are comparable why is the issuer’s stock discounted?) by asserting that ‘‘classic theory’’ demands there be a discount: The classic theory about IPOs is that you have to give an IPO discount. So what you usually do is you look at a company and there are 5 to 10 comps and you look at their average valuation [multiple], average PE valuation or average EBITDA valuation, and then you say give a discount of 10 to 20 percent. (ECM1)

Other underwriters explain that unsophisticated issuers do not understand financial markets and economic theory. ‘‘The sophisticated issuers know how to react to the price negotiation y the problem is when we are dealing with unsophisticated companies that don’t understand how classic theory values companies. So they question why they need to leave any money on the table [price at a discount]’’ (ECM5). Many underwriters proclaim that economic theory demands there be an IPO discount, and they need to educate issuers on how to understand this sophisticated financial pricing model: ‘‘when we deal with the very unsophisticated issuer, we need to spend a lot of time educating them on financial pricing y you know, many of these guys don’t know much finance, that’s partly where we earn our gross spread, educating them’’ (ECM4). ‘‘Classic theory,’’ presumably from economics or finance, can only be construed as a figure of speech here. Not only does financial economics not theorize that an IPO discount should exist, but the persistent underpricing of IPOs relative to their already listed comparable companies contradicts neoclassical theory. When questioned further, underwriters also offered specific rational actor explanations for why an IPO discount is necessary beyond simply asserting that economic theory demands so. These explanations usually revolve around costs for the institutional investors of buying into an IPO, such as

The Internal Complexity of Market Logics

297

the costs of switching, uncertainty, or volatility. The following quote succinctly summarizes most of these points made by other respondents: The theory behind the IPO discount is that you are trying to entice institutional investors to buy into the IPO y if you price the IPO same valuation as existing publicly trading comps, mainly institutional investors will ask why do I want to switch? That is the first, enticing them to switch from existing holdings to this IPO. And also there is the existing IPOs have longer track records, proven y they are just more experienced, and investors are more familiar with their management teams, and with the companies because they have been listed longer. For these reasons you need to offer a discount to entice them to buy into the IPO. Of course, when you IPO, you usually look at the trading volatilities of the IPO, they are usually very high, especially at the beginning very first few months, so again to compensate them for the added volatility that they take on, you have to give them [institutional investors] a discount. (ECM1)

Again, these explanations are phrased as part of mainstream economic theory. However, these reasons contradict both econometric analysis and common sense. If institutional investors are unwilling to bear the putative costs of purchasing IPO shares, issuers should just sell directly to retail investors who obviously are willing to bear these supposed costs given firstday returns. The actual extent, or even existence, of these costs is also questionable. For instance, econometric research cited earlier shows that volatility in IPO shares is directly related to first-day returns (Ellis et al., 2000); in other words, the less institutional investors pay, the more ‘‘volatility cost’’ they need to bear. Finally, another oft-cited rationale for the IPO discount is the purported trade-off between maximizing IPO proceeds and long-term performance of the stock as explained by the same respondent: ‘‘unsophisticated issuers y are more focused on the IPO valuation, trying to maximize proceeds during the IPO y obviously there is a trade off between short term and longer term [IPO returns to the issuer]’’ (ECM1). Again, econometric studies provide a different picture of the longer-term returns to IPOs relative to first-day returns (Krigman et al., 1999; Ritter, 1991; Ritter & Welch, 2002). In short, the specific reasons given to support the contention that economic theories demand an IPO discount are largely unfounded. Despite the inconsistencies in their reasoning, underwriters remain adamant in upholding the market logic as demanding an IPO discount. When told that neoclassical asset-price models actually consider IPO firstday returns an inexplicable anomaly, one underwriter remained silent for over 10 seconds whereas a hedge fund senior executive unleashed a profanity-laden protestation. The reactions of other respondents, while less visceral, still suggest real angst and consternation. Perhaps this

298

VINCE FENG

consternation is due to the underwriters having linked ‘‘sophistication’’ to an amorphous financial economic price theory justifying the IPO discount, a theory that is beyond the comprehension of unsophisticated neophyte issuers and purportedly in agreement with mainstream economic theory. This espousal of both ‘‘sophistication’’ and economic price theory is unsurprising given the strategic interest of underwriters in promoting this orientation; it is only surprising given the presence of numerous economists within investment banks (albeit primarily not working in IBD and ECM on IPOs, perhaps reflecting compartmentalization and the existence of coexisting logics within underwriters). Interestingly, investment bankers in this study did not repeat explanations previously offered to economists. Economic studies have cited investment bankers professing that potential investors anchor to the midpoint of the indicative price range, so underwriters cannot increase IPO prices as market conditions improve (Ritter & Welch, 2002). This predicts a negative demand curve (Welch, 1992), but has been refuted since IPOs actually exhibit a strong positive demand curve with investors buying more when prices increase (Loughran & Ritter, 2002). Other economists have noted the ‘‘leaning against the wind’’ hypothesis: investment bankers claim that they are looking out for the issuer’s best interests by ensuring lower first-day returns when markets are irrationally exuberant. This hypothesis predicts a negative correlation in revisions in offer price with long-term returns but has also been refuted, as revisions in offer price are empirically unrelated to three-year returns (Logue, Rogalski, Seward & Foster-Johnson, 2002; Loughran & Ritter, 2002). One IBD MD even scoffed at the ‘‘leaning against the wind’’ hypothesis remarking, ‘‘everyone knows that’s just a lot of hot air.’’ Apologists and former underwriters argue ‘‘that the IPO process is not as mysterious, nor are the participants as nefarious as some would have us believe,’’ and that underwriters are only looking out for the interests of the common investor: ‘‘in fact one of the primary reasons that most major banks do not embrace the pure auction form of IPO is they do not want ‘widows and orphans’ to be the marginal bidders for such a risky asset class. They would prefer to let most of that risk be assumed by professional money managers like mutual funds and hedge funds’’ (Topper, 2012). As with the other subsequently refuted explanations offered by investment bankers, the ‘‘protecting widows and orphans’’ argument simply makes no sense given the quid pro quo dynamic between hedge funds and underwriters discussed earlier. Underwriters have enshrined the IPO discount as an institutional norm for IPO pricing by overtly linking it to financial sophistication in the capital

The Internal Complexity of Market Logics

299

markets. ‘‘Sophistication’’ is a code word for a hybrid market logic: a specific sociocultural orientation toward measuring and understanding values in the context of financial market pricing. As previously discussed, the homogeneous market logic envisioned by neoclassical theory recommends rational self-interest in pricing; issuers should demand offer prices at parity to listed comparable companies. Yet the logic of ‘‘sophistication’’ clearly subverts the market logic of financial economics, recommending the acceptance of an IPO discount as appropriate. Ironically, underwriters contravene financial economics in part by linking the hybrid logic to financial economics through repeated reference to financial sophistication. But underwriters are only able to promulgate this hybrid logic because of their status position and control of the IPO situation.

DISCUSSION AND FUTURE RESEARCH Culturally homogeneous market logics cannot explain IPO pricing. EMH avers that capital markets price assets efficiently and accurately, reflecting intrinsic value. From the perspective of EMH, however, market participants inexplicably ‘‘underprice’’ IPOs by approximately 20 percent based on firstday returns. Persistent IPO first-day returns costing issuers billions of dollars every year contradicts EMH as well as its fundamental assumptions: rationality and sociocultural homogeneity in pricing. For IPO prices, the latter assertion proves more problematic. In-depth interviews with leading market participants suggest that IPO pricing is a negotiation of interests and power around the normative definition of IPO price; that is, the sociocultural orientation by which one measures the proper price for IPOs. In fact, IPO first-day returns only represent ‘‘underpricing’’ to a neoclassical orientation of atomistic calculative rationality. Underwriters influence firstday returns by utilizing their status position to promulgate a hybrid market logic – financial ‘‘sophistication’’ – endorsing the IPO discount norm. Thus, IPO first-day returns reflect the social power asymmetry between issuers and underwriters as adjudicated through logics and status. Market heterogeneity impacts pricing because actors perceive the environment and understand their self-interests differently based on their understanding of the situation. At the micro-level, actors utilize vocabularies to communicate their understanding of the situation to each other, as well as to generate meaning. Here, actors utilize market vocabulary to describe and understand the IPO pricing process. However, skilled actors can deceptively utilize market vocabulary to shape the actions of others in

300

VINCE FENG

price determination. Leveraging their status position and intermediary role to educate neophyte issuers on the proper understanding of IPO pricing, underwriters borrow words from financial economics to construct a hybrid market logic that actually undermines the price predictions of neoclassical and behavioral theory. In short, vocabularies are the pathways linking logics to tangible price outcomes. More generally, skillful use of vocabularies could underlie hybridity and institutional change. IPO pricing clearly shows that powerful actors can achieve their strategic interests through localized hybridization without overturning the dominant logic. This case study would suggest a wider prevalence of hybrid logics in other markets where powerful actors wish to influence outcomes without overtly renouncing the dominant logic. The strategic use of vocabularies clearly demonstrates the compatibility of logics with calculative rationality. Underwriters are calculative agents deceptively using market vocabulary to advance their own interests. In broad agreement with a strategic action field perspective on other market phenomena such as the mortgage securitization crisis (e.g., Fligstein & Goldstein, 2010), this study shows how a small set of powerful actors can influence market-wide phenomena. However, logics could also underpin forms of behavior incongruent with a strict strategic approach. Interests and strategic action do not explain everything in isolation from logics, as issuer acquiescence to the hybrid logic advanced by underwriters cannot be understood without considering the independent influence of vocabularies and logics on actors. This is a critical point of divergence from purely strategic perspectives. Without market homogeneity, accurate price models require detailed contextual knowledge of markets. Abstract models built on a foundation of sociocultural homogeneity inadequately explain IPO pricing. For instance, this case study complicates one of the main nonrational investor theories of IPO pricing: information revelation and investor sentiment. Derrien (2005) builds on the work of Miller (1977), Benveniste and Spindt (1989), and De Long, Shleifer, Summers, and Waldmann (1990) to model how underwriters price IPOs in the presence of sentiment investors in France. Derrien hypothesizes that underwriters rationally maximize profits, consisting of the underwriting fee as a percentage of the offer price less the cost of price support post-IPO. In the model, underwriters buy the offering from the issuer for resale to two groups of investors with different price predictions, rational and sentiment. Underwriters themselves do not know the intrinsic value of the issuer, so they conduct a two-stage IPO process, soliciting price predictions from the group of rational investors first. Rational investors

The Internal Complexity of Market Logics

301

must be enticed to reveal private information on their estimation of the intrinsic value of the issuer, so underwriters price the offering below the irrationally inflated price predictions of sentiment investors. Given such a model, the offer price to rational investors increases with investor sentiment, but usually does not reach the price predictions of sentiment investors, thus explaining the persistence of first-day returns.13 Furthermore, issuers are not upset with this underpricing, since the offer price is priced above the intrinsic value of the firm, or the predicted price by rational investors (Purnanandam & Swaminathan, 2004). Finally, rational investors are happy to sell the shares to sentiment investors on the first day of trading for a quick profit. This model represents one of the most complete sentiment investor accounts of IPO first-day returns and has been extended to explain long-run IPO underperformance (Ljungvist, Nanda, & Singh, 2006) and why underwriters induce sentiment investors into the market (Cook, Kieschnick, & Van Ness, 2006). Non-Bayesian investor models such as the one offered by Derrien generally assume that underwriters buy and resell (hard underwrite) the offering to investors, the underwriting fee is their primary source of income, and post-IPO price support is costly. However, in-depth contextual knowledge of IPO markets clearly invalidates these core assumptions. US underwriters do not hard underwrite IPOs, but instead act as agents helping issuers sell shares on a best-efforts basis. Furthermore, the alternative sources of income from an offering (spinning, quid pro quo with investors, and stabilization profits) could rival or exceed that from the underwriting fee. Far from being costly, post-IPO stabilization activity generates significant trading volume and profit for the lead underwriter if the IPO experiences high first-day returns (Aggarwal, 2000; Booth & Chua, 1996; Boehmer & Fishe, 2004; Ellis et al., 2000, 2002). These empirical practices fundamentally alter the abstracted profit-maximization constraints facing underwriters.14 As discussed earlier, underwriters are not lowering offer prices to buy private information from institutional investors. Instead, they are doing so because it benefits them. Hence, IPO first-day returns should vary with issuer resistance to the hybrid logic of ‘‘sophistication,’’ rather than the profit-maximization equations presented by information revelation and sentiment investor accounts of IPO pricing. I conclude this study with propositions for future research on logics and pricing. If the qualitative findings of this study are veracious, we should be able to observe several quantitative regularities with IPO pricing. First, contradicting both information asymmetry and status signal theory, increasing underwriter status should increase rather than reduce IPO first-day

302

VINCE FENG

returns. While underwriter status may exert the efficiency effect predicted by status signal theory and a reputation certification effect mitigating information asymmetries, underwriters also utilize their status position to inculcate issuers with the hybrid logic of sophistication increasing first-day returns. This is a critical test of my qualitative findings as previous studies have identified the first two effects but not the third for debt and equity underwriting. A quantitative analysis of status that can identify these three distinct effects of status (efficiency, reputation certification, and power) would advance not only the conclusions of this study but also the status and information asymmetry literatures. Second, issuers espousing institutional logics exhibiting differing levels of resistance to the hybrid logic of sophistication should experience systematically different IPO first-day returns. In other words, heterogeneity at the level of institutional logics would predict varying levels of IPO pricing. I believe that quantitative analysis of private-equity backed issuers would highlight pricing differences between issuers espousing different institutional logics. Third, macrocultural differences between societies (e.g., socialist vs. capitalist, East Asian vs. Western, etc.) should interact with organizational institutional logics in determining IPO first-day returns due to the macro-cultural acceptance or resistance to Western financial ‘‘sophistication.’’ I anticipate future research will more fully explore these propositions, highlighting the importance of sociocultural logics for price behavior. The investment banking, private equity, and hedge fund industries are all highly sensitive to confidentiality. I was able to gain access due to my previous professional ties, having worked in investment banking and private equity for 15 years.

DATA GATHERING AND INTERVIEW EFFECTS The investment banking, private equity, and hedge fund industries are all highly sensitive to confidentiality. I was able to gain access due to my previous professional ties, having worked in investment banking and private equity for 15 years. I interviewed most of the respondents by telephone as travel schedules complicated face-to-face meetings. However, I was able to meet five of the respondents in person, including three at their offices during work hours. Most interviews lasted over an hour with several lasting over two hours. I was able to partially audio record 10 of the interviews, although many of those recordings are truncated at the request of the respondents at the time of the interview, and nearly half of the respondents called back

The Internal Complexity of Market Logics

303

requesting that certain portions of their responses be removed from the record. I do not believe that any of the deleted materials compromise or contradict the findings reported in this study. Many of the respondents remarked that they were surprised at how candidly they had responded. I transcribed all recordings and summarized hand-written field notes within 24 hours of interviews. I was also able to conduct an impromptu on-site ethnography on the ECM floor of one of the interviews over the course of an eight-hour period. Wishing to allow the respondents to speak at length without prompting, I developed a simplistic interview guide that encouraged them to educate me on how the IPO process works. In all the interviews, I tried to the extent possible to dispel their notions of what I already knew and asked them to explain and interpret common terminology and processes. Some of the respondents seemed to enjoy teaching someone about their business, and talked at length about their own personal viewpoints. While introductions from high up abet gaining access, they could potentially impede the actual interview by making the respondents self-conscious. Always cognizant of this potential effect, I consciously downplayed those introductions and believe that in general I was successful in soliciting honest responses.

DATA ANALYSIS I began analysis of the transcripts on Atlas.ti (version 5) shortly after each interview, usually within two days. I did not code line-by-line as I did not want to force a code arbitrarily onto every line, but began by coding carefully all interpretive observations, normative statements, or descriptions of processes made by respondents, including overt emotional reactions to questions or topics. Hence, some areas of the transcripts are densely coded while others sparser. On average, I reviewed the transcripts twice at this level of initial coding before moving on to a higher more theoretical level of coding. Throughout the process, I jotted down memos as ideas emerged from the codes. By the third round of review, I began to move toward major categories emerging from the data, chosen based on conceptual density. Sometimes, a phrase that was repeated consistently with the same connotation or forcefulness identified categories. Other times, the same described pattern of behavior emerged as a category. I tested these categories with the respondents who called back (as mentioned earlier, about half the group) to confirm the categories’ representativeness. Later in the process, I formed

304

VINCE FENG

hypotheses on conceptual relations between categories. I actively looked for both consistencies and discrepancies between the responses and the extant theories on pricing. Overall, the collection and formulation of categories was a gradual process, but not overly difficult as the categories appeared frequently in the data.

NOTES 1. I have altered the original notations in the economic models to present a consistent set of notation throughout this paper. P refers to price; superscript  refers to predictions; non-superscripted P are observed prices; subscript t refers to time period; subscript or superscript R and S refer to rational and sentiment investors, respectively; p or P() refers to probability; R refers to returns; Q refers to quantity; d refers to sentiment; and e to uncorrelated disturbance terms. 2. Subjective expected utility (SEU) describes how rational actors P choose between decision alternatives; formally, actors maximize SEU value V ¼ ni¼1 Uðxi Þpi where U(.) is the individual’s utility function, xi is the vector of goods in the ith state of the world, and pi is the probability of the ith state of the world occurring. Savage (1972 [1954]) demonstrated that preferences should adhere to seven axioms for SEU maximization to occur. Bayes’ theorem relates the conditional probabilities of events A and B: PðBjAÞ ¼ ððPðAjBÞPðBÞÞ=ðPðAÞÞÞ. Rational actors should update their probability of B occurring when receiving new information A based on Bayes’ theorem. 3. Formally: ðRit Rf Þ ¼ ai þ bi ðK m Rf Þ þ si SMBt þ hi HMLt þ it where Rf is the risk-free rate, Km is the market rate of return, SMB (small minus big) and HML (high minus low) are the differences in returns between portfolios by market capitalization and book-to-market (value) ratios, and the relevant coefficients are firm-specific exposures to such risks (market, size, and value risk). 4. Discounted cashflow (DCF) analysis derives stock valuations by estimating the stream of dividends accruing to shareholders over the entire future life-course of the company and discounting that stream of payments back to the present. Intrinsic value is equal to the discounted of future dividends. Formally for P present value (PV) t discrete PV ¼ N t¼0 ðFVt =ð1 þ iÞ Þ; or for continuous cash flows: R T cash flows: PV ¼ 0 FVðtÞelt dt, where l ¼ log(1+i). In actual practice, underwriters usually calculate the enterprise value of the entire firm, replacing FV with operating free cash flow and i with the weighted average cost of capital for the firm. 5. Formally, rational investors initially value a company based on the PV of its future dividend stream: PnR1 ¼ PV1 while sentiment investors value the same company: PnS1 ¼ PV1 þ d; where d represents investor sentiment and PV1 is the correct discounted present value of future dividends at time=1. At time=2, new information is revealed and rational investors correctly update their price expectations: PnR2 ¼ PV2 ¼ PV1 þ xn while sentiment investors incorrectly update their price expectations: PnS2 ¼ PV1 þ d þ y xn where x is the expected change to future dividends revealed and ½oyo1 represents the underweighting of such information by sentiment investors. Demand for shares is: QitD ¼ maxfci ðPnit Pt Þ; 0g where i 2 fR; Sg and c is the risk-tolerance factor for that type of investor (rational

The Internal Complexity of Market Logics

305

or sentiment). At low sentiment levels ½doz0 o0 such that PnSt oPnRt ðQ=cR Þ, sentiment investors withdraw from the market and rational investors determine prices: Pt ¼ PnRt ðQ=cR Þ. At high sentiment levels ½d4z1 40 such that PnSt 4PnRt þ ðQ=cR Þ, rational investors are driven from the market due to shortsale constraints and sentiment investors determine pricing: Pt ¼ PnSt ðQ=cS Þ. At intermediate sentiment levels ½z0 odoz1 such that PnRt ðQ=cR Þ  PnSt  PnRt þ ðQ=cR Þ, rational and sentiment investors jointly determine price, which approximates a risk-tolerance weighted average of price expectations: Pt ¼ ½cR =ðcR þ cS ÞPnRt þ ½cS =ðcR þ cS ÞPnSt ½Q=ðcR þ cS Þ. 6. Formally, cumulative prospect theoryP (Tversky & Kahneman, 1992) hypothesizes that people assign gambles the value i pi vðxi Þ, where v ¼ xa if xZ0 and v ¼ lðxÞa if xo0; pi ¼ wðpiÞ  wðpni Þ; wðpÞ ¼ ðpg =ðpg þ ð1pÞg Þ1=g Þ; and p is the probability that the gamble will yield outcomes at least as good as x. A wide range of studies supports lE2 (coefficient of loss-aversion, a measure of relative sensitivity to gains and losses), violating SEU preference axioms since the sensitivity to gains and losses should be uniform. 7. The effect was also pharisaical for at least three additional reasons: it was unintended, the dogma itself was self-righteously condemnatory and censorious, and once established, the effect flourished without the need for the original dogma or the social carriers of that original creed. 8. One could justifiably ask why an investment bank must underwrite IPOs and price them in a two-stage bookbuilding process. In fact, IPOs have been priced by auction methods or sold directly to public investors in other countries and even here in the United States, with a specialized investment bank, WR Hambrecht + Co, championing Dutch auction IPOs since the late 1990s. However, such non-bookbuilt IPOs remain a rarity. An exposition on the institutionalization of current US IPO practices following the aftermath of the Great Depression is beyond the scope of this paper. Please refer to Appendix A for a summary of current US IPO practices and terminology. 9. Please refer to the entry for ‘‘Tombstone’’ in Appendix A for more details. 10. Bank of America (Merrill Lynch), Citigroup, Credit Suisse First Boston, Deutsche Bank, Goldman Sachs, J.P. Morgan, Morgan Stanley, and Union Bank of Switzerland (UBS) based on the total dollar value of deals underwritten from January 1, 2006 to September 30, 2009 and Carter-Manaster rankings. I adjusted the rankings to account for Bank of America’s purchase of Merrill Lynch. 11. This is the only legal exception to US anti-manipulation trading laws. The stabilization agent can short sell the stock if necessary, and cover the shorts by calling on more shares from the issuer through a ‘‘greenshoe’’ mechanism. See Appendix A for more details. 12. In the language of exchange theory, underwriters are situated on the point vulnerability of a restrictive positive network. The network is positive in that underwriters facilitate transactions between issuers and investors, whereas transactions in negative networks do not facilitate other transactions; it is restrictive because transactions between issuers and investors can take place only through the pivotal actor. 13. Formally, the first-day close price for the issuer’s shares fully incorporating sentiment investor’s expectations is: P1 ¼ Pnt þ zd1 where Pnt is the rational predicted

306

VINCE FENG

price and zd1 is the intensity of noise trader sentiment at t=1 and is a random variable uniformly distributed on ½0; 2zd0  so E½zd1 jzd0  ¼ zd0 (integrating Baker and Stein’s formulation in footnote 5 to Derrien’s model, if doz0 then sentiment investors withdraw from the market, zd1 ¼ 0, and P1 ¼ Pnt ; if dWz0, P1 is strictly increasing with increasing d). The underwriters set the offer price POPR by maximizing fees earned against the cost of price support activities post-IPO. Total fees are the gross spread times the offer fPOPR , where f is the gross spread. Cost R P price: Pn of price support is: EðCostjPOPR Þ ¼ 0 OPR t ð1=ð2zd0 ÞÞ ðPOPR Pnt xÞdx ¼ ð1=ð4zd0 ÞÞ 2 ðPOPR Pnt Þ ; underwriters maximize earnings by pricing the IPO offer price (obtained by solving @=@POPR ½fPOPR ð1=ð4zd0 ÞÞ at: PnOPR ¼ Pnt þ 2f zd0 2 ðPOPR Pnt Þ  ¼ 0). The numerator for first-day returns is: Pfdr ¼ P1 PnOPR ¼ Pnt þ zd0 ðPnt þ 2f zd0 Þ ¼ ð12f Þzd0 ; as f is generally fixed at 0.07, investor sentiment should predict first-day returns (if dWz0, Pfdr is strictly increasing with increasing d). 14. If we take Derrien’s (2005) model of Runderwriter profit-maximization, the lack P Pn of a costly price support EðCostjPOPR Þ ¼ 0 OPR t ð1=ð2zd0 ÞÞðPOPR Pnt xÞdx means underwriters should increase POPR to the maximum price given d sentiment levels, or PnOPR ¼ P1 ¼ Pnt þ zd1 . In such a formulation, d would predict price level but not returns, as P1=POPR and first-day returns should again equal zero, with underwriters effectively selling to sentiment investors when dWz0. Of course, it is possible to rewrite the cost constraint as a function of angering institutional buyers of IPOs, but that is not the premise of these models.

ACKNOWLEDGMENTS I would like to thank Frank Dobbin, Christopher Marquis, Peter Marsden, Orlando Patterson, and participants at the Stanford Economic Sociology workshop, Harvard-MIT Economic Sociology seminar, and ABC Network ‘‘Organizing Institutions’’ conference for their comments and suggestions.

REFERENCES Aggarwal, R. (2000). Stabilization activities by underwriters after initial public offerings. Journal of Finance, 55, 1075–1104. Babb, S. (1996). A true American system of finance: Frame resonance in the U.S. labor movement, 1866–1896. American Sociological Review, 61, 1033–1052. Baker, M., & Stein, J. C. (2004). Market liquidity as a sentiment indicator. Journal of Financial Markets, 7, 271–299. Barberis, N., & Huang, M. (2008). Stocks as lotteries: The implications of probability weighting for security prices. American Economic Review, 98, 2066–2100. Barberis, N., Huang, M., & Santos, T. (2001). Prospect theory and asset prices. Quarterly Journal of Economics, 107, 291–311.

The Internal Complexity of Market Logics

307

Barberis, N., & Thaler, R. (2003). A survey of behavioral finance. In G. M. Constantinides, M. Harris & R. M. Stulz (Eds.), Handbook of the economics of finance (Vol. 1B, pp. 1053– 1124). Financial Markets and Asset Pricing. Oxford, UK: Elsevier North-Holland. Battilana, J., & Dorado, S. (2010). Building sustainable hybrid organizations: The case of commercial microfinance organizations. The Academy of Management Journal, 53(6), 1419–1440. Beatty, R. P., & Ritter, J. R. (1986). Investment banking, reputation, and the underpricing of initial public offerings. Journal of Financial Economics, 15, 213–232. Benartzi, S., & Thaler, R. (1995). Myopic loss aversion and the equity premium puzzle. Quarterly Journal of Economics, 100, 75–92. Benveniste, L. M., & Spindt, P. A. (1989). How investment bankers determine the offer price and allocation of new shares. Journal of Financial Economics, 24, 343–361. Boehmer, E., & Fishe, R. P. H. (2004). Underwriter short covering in the IPO aftermarket: A clinical study. Journal of Corporate Finance, 10(4), 575–594. Booth, J. R., & Chua, L. (1996). Ownership dispersion, costly information, and IPO underpricing. Journal of Financial Economics, 41, 291–310. Burke, K. (1984 [1935]). Permanence and change: An anatomy of purpose. Berkeley, CA: University of California Press. Carter, R. B., Dark, F. H., & Singh, A. K. (1998). Underwriter reputation, initial returns, and long-run performance of IPO stocks. Journal of Finance, 53(1), 285–311. Carter, R. B., & Manaster, S. (1990). Initial public offerings and underwriter reputation. Journal of Finance, 45, 1045–1067. Coleman, J. S. (1990). Foundations of social theory. Cambridge, MA: Harvard University Press. Cook, D. O., Kieschnick, R., & Van Ness, R. A. (2006). On the marketing of IPOs. Journal of Financial Economics, 82, 35–61. De Long, J. B., Shleifer, A., Summers, L. H., & Waldmann, R. (1990). Noise trader risk in financial markets. Journal of Political Economy, 98, 703–738. Derrien, F. (2005). IPO pricing in ‘Hot’ market conditions: Who leaves money on the table? Journal of Finance, 60, 487–521. Dobbin, F. (1994). Forging industrial policy: The United States, Britain, and France in the Railway Age. New York, NY: Cambridge University Press. Dobbin, F., & Dowd, T. J. (2000). The market that antitrust built: Public policy, private coercion, and railroad acquisitions, 1825 to 1922. American Sociological Review, 65(5), 631–657. Dunn, M. B., & Jones, C. (2010). Institutional logics and institutional pluralism: The contestation of care and science logics in medical education, 1967–2005. Administrative Science Quarterly, 55, 114–149. Dybvig, P. H., & Ross, S. A. (2003). Arbitrage, state prices and portfolio theory. In G. M. Constantinides, M. Harris & R. M. Stulz (Eds.), Handbook of the economics of finance (Vol. 1B, pp. 606–637). Financial Markets and Asset Pricing. Oxford, UK: Elsevier North-Holland. Ellis, K, Michaely, R., & O’Hara, M. (2000). When the underwriter is the market maker: An examination of trading in the IPO aftermarket. Journal of Finance, 55, 1039–1074. Ellis, K., Michaely, R., & O’Hara, M. (2002). The making of a dealer market: From entry to equilibrium in the trading of Nasdaq stocks. Journal of Finance, 57, 2289–2316. Fama, E. F. (1965). The behavior of stock-market prices. Journal of Business, 38, 34–105.

308

VINCE FENG

Fama, E. F. (1976). Foundations of finance: Portfolio decisions and securities prices. New York, NY: Basic Books. Fama, E. F. (1990). Stock returns, expected returns, and real activity. Journal of Finance, 45, 1089–1108. Fama, E. F., & French, K. R. (1992). The cross-section of expected stock returns. Journal of Finance, 47, 427–465. Fama, E. F., & Jensen, M. C. (1983). Separation of ownership and control. Journal of Law and Economics, 26(2), 301–325. Ferraro, F., Pfeffer, J., & Sutton, R. I. (2005). Economics language and assumptions: How theories can become self-fulfilling. Academy of Management Review, 30(1), 8–24. Fligstein, N. (1990). The transformation of corporate control. Cambridge, MA: Harvard University Press. Fligstein, N., & Goldstein, A. (2010). The anatomy of the mortgage securitization crisis. Research in the Sociology of Organizations, 30, 29–70. Friedland, R., & Alford, R. R. (1991). Bringing society back in: Symbols, practices, and institutional contradictions. In W. W. Powell & P. J. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 232–263). Chicago, IL: University of Chicago Press. Gerth, H. H., & Mills, C. W. (1958 [1946]). Introduction. In H. H. Gerth & C. Wright Mills (Eds. & Trans.), From Max Weber: Essays in sociology (pp. 3–74). New York, NY: Oxford University Press. Greenwood, R., & Hinings, C. R. (1996). Understanding radical organizational change: Bringing together the old and the new institutionalism. Academy of Management Review, 21, 1022–1054. Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E. R., & Lounsbury, M. (2011). Institutional complexity and organizational responses. The Academy of Management Annals, 5(1), 317–371. Habib, M. A., & Ljungqvist, A. P. (2001). Underpricing and entrepreneurial wealth losses in IPOs: Theory and evidence. The Review of Financial Studies, 14, 433–458. Hirsch, P. M. (1986). From ambushes to golden parachutes: Corporate takeovers as an instance of cultural framing and institutional integration. American Journal of Sociology, 91, 800– 837. Hirshleifer, D. (2001). Investor psychology and asset pricing. Journal of Finance, 56, 1533–1597. Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs, and ownership structure. Journal of Financial Economics, 3, 305–360. Jones, C., & Livne-Tarandach, R. (2008). Designing a frame: Rhetorical strategies of architects. Journal of Organizational Behavior, 29, 1075–1099. Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1991). Anomalies: The endowment effect, loss aversion, and the coase theorem. Journal of Economic Perspectives, 5, 193–206. Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47, 263–291. Kaustia, M., & Knu¨pfer, S. (2008). Do investors overweight personal experience? Evidence from IPO subscriptions. Journal of Finance, 63, 2679–2702. Krigman, L., Shaw, W. H., & Womack, K. L. (1999). The persistence of IPO mispricing and the predictive power of flipping. The Journal of Finance, 54, 1015–1044. Ljungvist, A., Nanda, V., & Singh, R. (2006). Hot markets, investor sentiment, and IPO pricing. Journal of Business, 79, 1667–1702.

The Internal Complexity of Market Logics

309

Ljungqvist, A., & Wilhelm, W. J., Jr. (2003). IPO pricing in the dot-com bubble. Journal of Finance, 58, 723–752. Loewenstein, J., Ocasio, W., & Jones, C. (2012). Vocabularies and vocabulary structure: A new approach linking categories, practices, and institutions. The Academy of Management Annals, 6(1), 41–86. Logue, D. E., Rogalski, R. J., Seward, J. K., & Foster-Johnson, L. (2002). What is special about the roles of underwriter reputation and market activities in initial public offerings? The Journal of Business, 75(2), 213–243. Loughran, T., & Ritter, J. R. (2002). Why don’t issuers get upset about leaving money on the table in IPOs? Review of Financial Studies, 15, 413–443. Loughran, T., & Ritter, J. R. (2004). Why has IPO underpricing changed over time? Financial Management, 33, 5–37. Lounsbury, M. (2002). Institutional transformation and status mobility: The professionalization of the field of finance. Academy of Management Journal, 45, 255–266. Lounsbury, M. (2007). A tale of two cities: Competing logics and practice variation in the professionalizing of mutual funds. Academy of Management Journal, 50, 289–307. Lounsbury, M., & Rao, H. (2004). Sources of durability and change in market classifications: A study of the reconstitution of product categories in the American mutual fund industry, 1944–1985. Social Forces, 82(3), 969–999. Manski, C. F. (2000). Economic analysis of social interactions. Journal of Economic Perspectives, 14, 115–156. Marquis, C., & Lounsbury, M. (2007). Vive la Re´sistance: Competing logics in the consolidation of community banking. Academy of Management Journal, 50, 799–820. Marsden, P. V. (1983). Restricted access in networks and models of power. American Journal of Sociology, 88(4), 686–717. Miller, E. M. (1977). Risk, uncertainty, and divergence of opinion. Journal of Finance, 31, 1151– 1168. Mills, C. W. (1939). Language, logic and culture. American Sociological Review, 5, 670–680. Nigam, A., & Ocasio, W. (2010). Event attention, environmental sensemaking, and change in institutional logics: An inductive analysis of the effects of public attention to Clinton’s health care reform initiative. Organization Science, 21(4), 823–841. Odean, T. (1998). Volume, volatility, price, and profit when all traders are above average. Journal of Finance, 53, 1887–1934. Pache, A.-C., & Santos, F. M. (2010). When worlds collide: The internal dynamics of organizational responses to conflicting institutional demands. Academy of Management Review, 35(3), 455–476. Purnanandam, A. K., & Swaminathan, B. (2004). Are IPOs really underpriced? Review of Financial Studies, 17, 811–848. Reay, T., & Hinings, C. R. (2009). Managing the rivalry of competing institutional logics. Organization Studies, 30(6), 629–652. Ritter, J. R. (1984). Signaling and the valuation of unseasoned new issues: A comment. The Journal of Finance, 39(4), 1231–1237. Ritter, J. R. (1991). The long-run performance of initial public offerings. The Journal of Finance, 46, 3–27. Ritter, J. R., & Welch, I. (2002). A review of IPO activity, pricing, and allocations. The Journal of Finance, 57, 1795–1828. Savage, L. J. (1972 [1954]). The foundations of statistics. New York, NY: Dover Publications.

310

VINCE FENG

Suddaby, R., & Greenwood, R. (2005). Rhetorical strategies of legitimacy. Administrative Science Quarterly, 50, 35–67. Thaler, R. H. (1999). Mental accounting matters. Journal of Behavioral Decision Making, 12, 183–206. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure, and process. New York, NY: Oxford University Press. Topper, D. (2012). The IPO process: Separating fact from fiction. Retrieved from http:// www.generalatlantic.com/en/news/article/1843. Accessed on July 10, 2012. Tracey, P., Phillips, N., & Jarvis, O. (2011). Bridging institutional entrepreneurship and the creation of new organizational forms: A multilevel model. Organization Science, 22(1), 60–80. Tversky, A., & Kahneman, D. (1992). Advances in prospect theory: Cumulative representation of uncertainty. Journal of Risk and Uncertainty, 5(4), 297–323. Weber, M. (1958 [1916]). Religious rejections of the world and their directions. In H. H. Gerth & C. Wright Mills (Eds. & Trans.), From Max Weber: Essays in sociology (pp. 323–359). New York, NY: Oxford University Press. Weber, M. (2009 [1905]). In S. Kalberg (Trans.). The protestant ethic and the spirit of capitalism: With other writings on the rise of the west. New York, NY: Oxford University Press. Welch, I. (1992). Sequential sales, learning, and cascades. Journal of Finance, 47(2), 695–732. Zajac, E. J., & Westphal, J. D. (2004). The social construction of market value: Institutionalization and learning perspectives on stock market reactions. American Sociological Review, 69, 433–457. Zhong, C.-B., Loewenstein, J., & Murnighan, J. K. (2007). Speaking the same language: The cooperative effects of labeling in the prisoner’s dilemma. Journal of Conflict Resolution, 51(3), 431–456.

The Internal Complexity of Market Logics

311

APPENDIX A: US INITIAL PUBLIC OFFERING PRACTICES AND TERMINOLOGY Beauty contest

Bookbuilding

Comparables

ECM

Fully distributed value

The bidding process for an issuer’s IPO mandate by underwriters. Investment banks primarily present pitchbooks detailing how many deals they have underwritten recently, the strength of their research coverage of comparable stocks, and their estimation of the issuer’s market value. The cost of underwriting services is generally not negotiable (see ‘‘Gross Spread’’) The underwriter mediated order collection and allocation process with institutional investors during the roadshow. The lead underwriter’s ECM division handles the order book and allocation process; hence, the lead underwriter is also known as the ‘‘bookrunner.’’ Large IPO mandates may entail multiple co-lead bookrunners. Already-listed companies who should be comparable to the issuer in terms of trading multiple valuation criteria (such as price-earnings growth or price-EBITDA ratios) because they are in similar industry sectors or have similar operations. Equity Capital Markets. The division within the investment bank coordinating the conflicting client interests of the underwriter during the IPO process. ECM coordinates and manages the order-taking (bookbuilding) process, centralizing orders from institutional investors into an order book and allocating shares after determination of the final offer price. ECM also performs the crucial functions of recommending the indicative price range for the roadshow and the final offer price to institutional investors based upon bookbuilding demand following the roadshow. The hypothetical value of the issuer once it has listed its shares and begins trading on a public exchange. Underwriters calculate the fully distributed value by

312

VINCE FENG

Appendix A. (Continued )

Greenshoe

Gross spread

IBD

Institutional investor

IPO

multiplying the average trading multiples of comparable companies to the relevant financial metrics of the issuer. Named after the first company to grant an over-allotment option to its underwriters (Green Shoe Manufacturing Company, since renamed Stride Rite Corporation), the greenshoe is basically a free call option for an additional 15 percent of the original number of IPO shares at the final offer price to institutional investors granted to the lead underwriter by the issuer (see ‘‘Stabilization Agent’’) The underwriting fee (cost of underwriting services for an IPO), quoted as a percentage of the IPO proceeds raised based on the final offer price to institutional investors. In the United States, the gross spread is fixed at 7 percent for the vast majority of IPOs. Of all US IPOs between 2001 and 2009, 71.6 percent had underwriting fees of exactly 7 percent. This trend toward a fixed industry-wide gross spread of 7 percent has been increasing over time and is dominant for moderate size IPOs: 96.9 percent of all US IPOs with offering sizes of $25–$100 million in inflation-adjusted 2007 dollars had underwriting fees of exactly 7 percent over the same time period. For detailed summary information, please refer to http://bear.warrington.ufl.edu/ritter/ipodata. htm Investment Banking Division, sometimes referred to as Corporate Finance. The division within the investment bank representing the issuer’s interests in the IPO process. IBD works directly with the issuer in preparing the prospectus. Professionally managed institutional pools of capital that underwriters actively solicit to invest in IPOs during the roadshow: primarily hedge, mutual, pension, and private equity funds. Initial Public Offering. The initial listing of a private company’s shares on a public equity stock exchange.

The Internal Complexity of Market Logics

313

Appendix A. (Continued )

IPO discount

Issuer

Offer price

Price Range Prospectus

Retail investor

Roadshow

In the United States, almost all IPOs are underwritten by investment banks. The discount applied to the fully distributed value of the issuer when determining the optimal offer price to institutional investors. Underwriters sometimes refer to this as the ‘‘illiquidity’’ or ‘‘liquidity’’ discount. The company undertaking to list its shares in an IPO. The offering shares could be new shares issued by the company (primary shares) or existing shares sold by pre-IPO shareholders of the company (secondary shares). The final price at which the IPO shares are allocated and sold to institutional investors following the roadshow based upon bookbuilding demand. The offer price can be priced below, within, or above the price range. The indicative price range quoted to institutional investors at the beginning of the roadshow. SEC-mandated report to IPO investors detailing the issuer’s financials, business outlook, shareholding structure, legal disclosures, and other relevant information. Noninstitutionally affiliated public market investors. These investors can purchase shares on the first-day of trading from institutional investors allocated shares during the bookbuilding process or from the lead underwriter in its capacity as the ‘‘stabilization agent.’’ Meetings with global institutional investors generally lasting two weeks prior to the first day of trading. The underwriter recommends an indicative price range incorporating the IPO discount to approach institutional investors with, and issuers either acquiesce or negotiate the price range with the lead underwriter. Once the price range is agreed upon, the roadshow commences and bookbuilding demand helps determine the final offer price to institutional investors. The stock begins trading and is available for retail investors to purchase generally on the day following allocation of

314

VINCE FENG

Appendix A. (Continued )

Secondary selling

Stabilization agent

Tombstone

Underwriters

IPO shares to institutional investors at the end of the roadshow. Issuers can offer either newly issued shares (primary) or already issued shares held by pre-IPO shareholders (secondary) for sale in the IPO. The issuer only receives proceeds raised from the sale of primary shares; secondary selling does not raise any money for the company, but represents an (partial) exit by pre-IPO shareholders. The lead underwriter’s role for the first several weeks following the IPO as a primary trader for the newly listed stock with the goal of stabilizing the price. This is the only legal exception to anti-manipulation trading laws allowed by the SEC. The stabilization agent can short-sell the stock if necessary, and cover the shorts by calling on more shares from the issuer through a greenshoe mechanism (see ‘‘Greenshoe’’) Advertisements of completed IPO transactions. These advertisements are known as ‘‘tombstones’’ because they were originally published in the Wall Street Journal facing the obituaries section. Underwriters conform to a strict status hierarchy involving a stringent rule-based name placement ceremony for all tombstones. The top-tier underwriters form a distinct group and are referred to as the ‘‘bulge-bracket’’ due to their type font and placement on the tombstones. We can derive status measures from these tombstones by analyzing the order placements, with the most common measures being the Carter-Manaster rank in economics and eigenvector centrality measure in sociology. Investment banks retained by issuers to conduct the IPO. Technically, underwriting is the purchase of shares for resale, but IPO transactions are no longer pre-bought by investment banks today. Despite the fact that all equity underwriting is on a best-efforts basis, investment banks are still referred to as underwriters. The lead underwriter assists the issuer in preparing the

The Internal Complexity of Market Logics

315

Appendix A. (Continued )

Underwriting syndicate

story they will tell investors (with the aid of research analysts from the underwriting syndicate), drafts the filings and documentation required by the SEC, coordinates the auditors and lawyers providing the financial and legal disclosures, and assembles a broader group of investment banks (underwriting syndicate) to assist with marketing and distributing the offering shares. The group of investment banks marketing and distributing the IPO shares. The lead underwriter will usually organize a group of other investment banks (comanagers) to help market and distribute the IPO. These co-managers are usually lower status investment banks with specialized research capabilities. An external status hierarchy protects high and low-status firms from competing with each other, establishes a clear delineation of roles (‘‘bulge-bracket’’ bookrunners vs. ‘‘mid-market’’ co-managers), and disciplines rogue firms that break rank and attempt to undercut fees. High-status firms discipline rogue firms by freezing them out of future underwriting syndicates. IPO issuances in the United States usually involve one to four high-status co-lead bookrunners who jointly manage the bookbuilding process and a syndication of numerous low-status co-managers who mainly provide additional research coverage and some incremental distribution capabilities. Depending on the size of the co-manager syndicate, these low-status firms as a group may receive upwards of 40 percent of the underwriting fee. Due to the strict status hierarchy, low-status firms cannot replace high-status firms simply by underbidding.

316

VINCE FENG

APPENDIX B: INTERVIEW ROSTER Code

Position

Interview Length

IBD1

Managing Director, Investment Banking Division Managing Director, Investment Banking Division Managing Director, Investment Banking Division Managing Director, Equity Capital Markets

1 hour; portion audio-recorded; transcription word count 2,033 1.5 hours; portion audiorecorded; transcription word count 2,543 0.5 hours; audio-recorded; transcription word count 2,940 1 hour; audio-recorded; transcription word count 4,352 1.5 hours; portion audiorecorded; transcription word count 3,490 1.5 hours; portion audiorecorded; transcription word count 1,900 2 hours

IBD2

IBD3

ECM1

ECM2

Co-head of Global Equity Capital Markets

ECM3

Managing Director, Equity Capital Markets

ECM4

Managing Director, Equity Capital Markets Co-head of Equity Capital Markets Managing Partner, Leading Global Hedge Fund

ECM5 HF1

HF2

Senior Trader, Leading Global Hedge Fund

HF3

Senior Partner, Leading Global Hedge Fund

1.5 hours 1.5 hours; portion audiorecorded; transcription word count 4,176 1 hour; portion audio-recorded; transcription word count 2,350 1 hour; portion audio-recorded; transcription word count 1,608

317

The Internal Complexity of Market Logics

Appendix B. (Continued ) Code

Position

Interview Length

PE4

Managing Director, Leading Global Private Equity Fund CFO, NYSE listed privateequity backed issuer CEO, Nasdaq listed privateequity backed issuer CFO, Nasdaq listed privateequity backed issuer

0.5 hours; portion audiorecorded; transcription word count 860 1 hour

PEIS1 PEIS2 PEIS3

1.5 hours 0.5 hours

TAKING STOCK OF INSTITUTIONAL COMPLEXITY: ANCHORING A POOL OF INSTITUTIONAL LOGICS INTO THE INTERINSTITUTIONAL SYSTEM WITH A DESCENDENT HIERARCHICAL ANALYSIS Thibault Daudigeos, Ame´lie Boutinot and Ste´phane Jaumier ABSTRACT Institutional pluralism is an intriguing phenomenon for institutional scholars. How the balance among logics evolves within a field and what kind of trajectories a set of logics may experience over a long-term period remain unclear. In particular, extant literature tends too often to downplay institutional complexity by focusing on two dominant logics, and to ignore modes of interaction among logics other than competition. In order to address these issues, we offer a novel methodology for measuring institutional complexity – multiple institutional logics and their Institutional Logics in Action, Part B Research in the Sociology of Organizations, Volume 39B, 319–350 Copyright r 2013 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0733-558X/doi:10.1108/S0733-558X(2013)0039A&B023

319

320

THIBAULT DAUDIGEOS ET AL.

change. In particular, we highlight the utility of descendent hierarchical classification models, and demonstrate their relevance by analyzing articles published in a leading French trade journal over more than 100 years to study logics related to workplace in the construction industry. We identify a pool of six field-structuring logics over a period of one century; they reveal the composite nature of such logics, which we characterize as combining several higher institutional orders. Additionally, our results bring to light new mechanisms that can explain the composition of institutional logics. Keywords: Institutional complexity; pool of institutional logics; workplace safety; construction industry; descendent hierarchical classification model December 3rd, 1902. A rainy day, on a construction site in the North of France. Mr Dupont drives the new mechanical digger his company has just acquired. While moving a large stone, a mudslide occurs; unfortunately, the device slips and overturns onto its left side. Mr Dupont hurts his head, and breaks his left arm and hip. Following the 1898 occupational security law, Mr Dupont gets paid a fixed compensatory amount by a private insurer; the employer’s penal responsibility is disengaged by contributing to such an insurance system. December 3rd, a hundred years later. Same situation, same occupational injury. But this time, while driving the device, Mr Dupond was wearing a safety helmet and some gloves offered by his company, and he was trained to drive it safely. The whole company is hurt by such an accident. It will damage the company’s name, its statistics, and the yearly bonus of the team in charge of the building site.

INTRODUCTION Institutional pluralism is an intriguing phenomenon for institutional scholars (Greenwood, Raynard, Kodeih, Micelotta, & Lounsbury, 2011; Kraatz & Block, 2008). How the balance among logics evolves within a field and what kind of trajectories a set of logics may experience over a long-term period remain unclear (Dunn & Jones, 2010). To address these questions, we need a fuller understanding of the way logics interact. The literature that has so far dealt with these questions suffers from one major limitation, namely its tendency to downplay the complexity of the field under study. The first reason for this lies in the literature’s usual focus on two dominant or particularly salient logics. Unfortunately, such an approach overlooks the

Taking Stock of Institutional Complexity

321

potential impacts of less-salient but nonetheless influential logics (Greenwood et al., 2011). It also orients the perception of interactions between logics toward the ideas of competition and conflict, where a more pluralistic approach would allow identifying forms of compatibility among logics (Greenwood, Diaz, Li, & Lorente, 2010). The second reason lies in a methodological bias toward analyzing data based on a fixed set of preconceived logics. Positing ex ante the logics to be studied precludes the possibility not only of capturing less-salient logics but also of identifying dominant logics in their original formulation. In this paper, we aim at remedying these two issues. We introduce a multistep methodology that, first, encourages multiple logics to emerge from data and, second, provides a way to properly characterize the logics identified and to relate them to the broader interinstitutional system in order to illuminate their complex composition (Thornton, Ocasio, & Lounsbury, 2012). Empirically, we look at the organizational field of workplace safety in the French construction industry. As Hoffman has already noted, fields may aggregate around social issues (Hoffman, 1999, p. 352), and this particular field of workplace safety is of interest by virtue of having been a longstanding issue since the birth of capitalism, continuously marked by strong debates about the various possible representations of the issues at stake (Aldrich, 1997). We track the evolution of a pool of logics by measuring over time the dominant meaning systems within the field (DiMaggio & Mullen, 2000; Mohr, 1998). To do this, we postulate that these meaning systems are reflected in the content of the main weekly trade journal in the industry. We thus collect a sample set of the articles dealing with safety in the construction industry that have been published in this journal from its creation in 1902 to the present day. In the first step of our data analysis, we apply to our corpus a relational quantitative textual analysis, based on a descendent hierarchical classification model (Leenen, Van Mechelen, De Boeck, & Rosenberg, 1999; Leenen, Van Mechelen, Gelman, & De Knop, 2008; Ventresca & Mohr, 2002). This allows us to identify six institutional logics at work over the century. In a second step, we characterize these six institutional logics (DiMaggio & Mullen, 2000; Thornton et al., 2012), and make their composition explicit by relating them to the broader interinstitutional system (Thornton et al., 2012). In doing so, our contribution to the perspective of institutional logics is twofold. First, thanks to our novel methodology, our results fully recognize the existence of institutional complexity. In addition to identifying six field-structuring logics over a one-century period, our study reveals the composite nature of institutional logics. Indeed, we show that most institutional logics are the

322

THIBAULT DAUDIGEOS ET AL.

result of combining several institutional orders. Second, our study throws some light on new mechanisms that can be mobilized to explain the composition of institutional logics. Our argument is organized as follows. The first section draws on extant literature in order to introduce the main theoretical concepts and document the research gaps. The second section presents our field of study and methodology. The third section details the results of our study. The fourth section discusses the contributions of the paper and presents concluding remarks.

THEORETICAL FRAMEWORK In this section, we draw on extant literature in order to theoretically inform our research work. The first subsection situates the institutional logics perspective and summarizes extant knowledge on interactions among institutional logics. The following subsections challenge this knowledge by building on the notion of institutional complexity. They introduce the notion of a pool of logics (second subsection) and the relationship between institutional logics and the broader interinstitutional system (third subsection) as a means to adequately capture institutional complexity.

Interaction among Institutional Logics Institutional logics designate: the socially constructed, historical pattern of material practices, assumptions, values, beliefs, and rules by which individuals produce and reproduce their material subsistence, organize time and space, and provide meaning to their social reality. (Thornton & Ocasio, 1999, p. 804)

They thus provide individuals and organizations with a broad set of symbolic and material resources for building their identities and goals. The main contribution of the institutional logics perspective indeed lies in the way it reaffirms the duality of structure and action (Thornton et al., 2012). Embedded within multiple logics, individual actors benefit from partial autonomy. Rather than being solely constrained by social structures, they can now be seen to be at the same time as enabled by them. In other words, the institutional logics perspective allows us to understand how actors, while conditioned by institutions, also change them. By doing so, this perspective

Taking Stock of Institutional Complexity

323

has so far made a strong contribution to answering DiMaggio’s (1988) call to reintroduce further consideration of the role of agency within the institutional research stream. To date, the need to understand the way various logics may interact over time within a field has resulted in the formulation of several models, mostly drawn from natural sciences (Thornton, Jones, & Kury, 2005). The most commonly described model – the evolutionary or sequential one – presents the coexistence of multiple logics as a transitory phenomenon. Indeed, it shows how a formerly dominant logic is progressively displaced by a newly dominant one, and has been observed in fields as diverse as publishing (Thornton & Ocasio, 1999), finance (Lounsbury, 2002) and French cuisine (Rao, Monin, & Durand, 2003). The punctuated-equilibrium model is characterized by discrete jumps that mirror regulatory changes (Thornton et al., 2005). The cyclical model comprises two logics that take turns to occupy the foreground, as is the case in the field of architecture with the esthetic and efficiency logics (Thornton et al., 2005), or in that of medical education, where the logics of science and care alternate in taking precedence over the other (Dunn & Jones, 2010).

Toward A More Refined Understanding Of Institutional Complexity What is common to these studies, which address the question of the models of evolution of institutional logics within a field, is their focus on two logics, which they typically analyze in terms of conflict. Although the reliance on this dualistic scheme can be praised for its relative simplicity and vivid explanatory record, it may entail a great deal of reductionism (Greenwood et al., 2011). We therefore see a major opportunity to extend these contributions by going one step further in the degree of complexity under study. Restricting the scope of inquiry to two logics may indeed have undesired implications (Greenwood et al., 2011). By missing the interactions that take place among ruled-out logics, and the possible compatibility between and reinforcement of these, the complexity of the field may be overlooked, thus leading to severe misinterpretations of its dynamics. Complexity can in fact be understood along two complementary dimensions (Greenwood et al., 2011). First, complexity stems from the sheer number of logics involved in a given field: the more numerous the logics at work within a field, the higher the complexity to be managed by social actors participating in the field. Whereas most studies consider the coexistence of multiple logics as transitory, or as a problem to be solved

324

THIBAULT DAUDIGEOS ET AL.

(Friedland, this volume), it becomes more and more clear that a field may actually host several logics over rather long periods of time (Goodrick & Reay, 2011). These findings suggest that, rather than a few narrowly identified logics, the attention of scholars should focus on a pool of logics, that is, a comprehensive set of logics that are made available to actors participating in a given field. The present article aims at doing so, in a way that resonates with Goodrick and Reay’s (2011) idea of constellation of logics. Second, complexity depends on the degree of compatibility that is exhibited by the different logics that coexist within a pool. The degree to which the incompatibility among logics that compose the pool is more salient, the higher is the complexity to be dealt with by actors involved in the field. Again, whilst the studies that focus on two dominant logics put forward the idea of incompatibility through the notions of competition and conflict (Lounsbury, 2007; Marquis & Lounsbury, 2007), the interest in a more comprehensive pool of logics allows the identification of other forms of coexistence. For instance, several less-salient logics may tend to reinforce or complement each other when confronting a more dominant one (Greenwood et al., 2010).

Connecting Institutional Logics And Orders To Understand Complexity Some scholars have introduced the concept of the interinstitutional system to describe a set of predefined societal orders. In their view, institutional logics at the field level translate combinations of higher institutional orders into actors’ practices (Lounsbury & Crumley, 2007). Looking more deeply, how logics are supported by institutional orders may certainly open the way to a better understanding of institutional complexity and therefore of interactions between logics. Indeed, if institutional logics are to be understood as field-level specific compositions of broader societal orders, the analysis of their composite nature should prove to be key in grasping when some logics may complement or reinforce – rather than oppose – each other (Thornton et al., 2012). Additionally, when we draw the link between institutional logics and orders, it makes even more sense that the institutional logics perspective as an organizational research stream originally built on the foundational formulation by Friedland and Alford (1991) of the interinstitutional system that governs Western societies (Friedland, this volume). In its most recent update (Thornton et al., 2012), the interinstitutional system is composed of seven institutional orders – namely market, corporation, profession, state,

Taking Stock of Institutional Complexity

325

family (Friedland & Alford, 1991), religion (Thornton, 2004), and community (Greenwood et al., 2010). While these institutional orders are constitutive of the societal level, the institutional-logics perspective has developed (Thornton & Ocasio, 1999, 2008), in part, because neoinstitutionalist scholars have recognized the importance of accounting for multiple symbolic and material elements at more microlevels (be they an industry, organizational field, or organization). Returning more systematically from institutional logics to institutional orders – from the field to the society – would therefore contribute to closing the institutional loop (Thornton et al., 2012). The present paper aims, therefore, to contribute to a better understanding of the interactions among logics by focusing more systematically the question of complexity. To that purpose, we intend, first, to introduce a further level of complexity in the analysis of interactions between logics within a field through the notion of a pool of logics and, second, to systematically relate institutional logics to higher institutional orders so as to benefit from the explanatory power of the former’s composition.

DATA AND METHODS: TRACKING THE EVOLUTION OF A POOL OF LOGICS OVER A LONG PERIOD OF TIME Research Design And Setting How can we capture and characterize the evolving pool of logics at work in a field over a long time period? Tracking this evolution supposes that we delineate at least one organizational field to study. This is all the more challenging when considering, as Friedland & Alford (1991) contend, that the boundaries of an organizational field can hardly be defined a priori. On the contrary, boundaries coevolve with the institutional logics that inhabit the field. In line with previous research (Franzosi, 1987; Hoffman, 1999), we expect that a form of media that occupies a central position within an observed field may reflect this coevolution between the field and the logics that are enacted there. Workplace Safety in The Construction Industry as an Organizational Field Our empirical setting is the organizational field of workplace safety in the French construction industry. We selected this setting for three reasons.

326

THIBAULT DAUDIGEOS ET AL.

First, we understand it as an organizational field, composed of several social actors whose frequent interactions are structured by common issues, shared understanding, rules, professional norms, and social policies. When a stonemason on a building site gets new protection gear for the ears, a whole set of actors may have been connected to this action – for example, occupational health and safety professionals, managers, civil servants, the safety inspectorate, and public research institutes. As Hoffman has already noted, fields may aggregate around social issues, such as the protection of the natural environment in the chemical industry, or workplace safety in the construction industry (1999, p. 352). Second, we understand workplace safety in the French construction industry as a plural field, where several logics are involved over a long period of time. The concern for workplace safety has grown with the development of capitalism in the Western world (Aldrich, 1997), and has always been preeminent in the construction industry. As an illustration of the early existence of workplace safety controversies in this industry, the first massive strike ever reported in France was when Parisian stonemasons campaigned for better working conditions as early as 1848. Because of the historical layers of organizations and professions involved in the issue of safety in the construction industry, we can expect to observe the structuring effects of several logics, be they market or state, as in microfinance in the bank industry (Battilana & Dorado, 2010), or access to employment in social enterprises (Pache & Santos, working paper). Even now, construction is the most dangerous industry to work for in France, although it has received the attention of public authorities for decades. In recent years, many public controversies arose in French media about workplace safety on building sites, and this may reveal the presence of frictions between logics within the field. As Mohr and Duquenne contend, social policies are one of the main arenas of institutional change in modern societies. Therefore, we expect workplace safety in the construction industry to be an excellent place to observe institutional life over a long period of time. Third, construction is a rather mature industry, with limited change in its main outcomes and foundational techniques over time. The most recent radical innovation, which upset the industry in France, was the invention of concrete at the end of the nineteenth century (Boxenbaum & Daudigeos, 2008). Because of this, the multiple logics we expect to observe cannot be attributed to institutional change that occurred as a consequence of drastic shifts in the nature of the tasks and products delivered by this industry; this therefore allows us to control for competing explanations of institutional change.

Taking Stock of Institutional Complexity

327

Capturing Meaning Systems Through the Lens of a Form of Media How can we measure institutional logics? We share the opinion that capturing meaning systems through the study of discourses is a way to track the structuring cognitive effect of institutional logics (Mohr, 1998; Phillips & Malhotra, 2008; Schneiberg & Clemens, 2006; Scott, 2008 [1995]). According to these authors, institutions are cognitive constructs that are building blocks of meaning systems. In this sense, our research design is consistent with their call for a discursive approach to institutional change. Alvesson and Ka¨rreman (2000) describe two irreconcilable views on the nature of discourse: discourses are either loosely coupled with, or even decoupled from, practices and are political in essence (Boxenbaum & Jonsson, 2008), or they reflect the inner subjectivity of social actors and the meaning they give to practices. In the present study, we lean toward the second option and consider that discourses are at the same time visible marks of cognitive processes in the field and provide resource with which a field is socially constructed. As Lounsbury, Ventresca, and Hirsch explain: Field frames are forged, maintained and eroded through discourse in policy forums such as congressional hearings as well as in industry media (y) (2003, p. 77). Indeed discourses are produced and reproduced to forge and enact dominant social structures, but they may be helpful as well in constructing an arena in which to fight this dominant order. Discourses are thus a privileged object for studying institutional change (Phillips & Hardy, 2002; Suddaby & Greenwood, 2005). Discourses may be captured through direct contact with the observed sample or through secondary data considered as marks of past discourses. In our case, we use trade-journal articles as a proxy for the discourses of key professionals in the field we study. The form of media we have selected to reflect the evolution of the configuration of logics in the field of workplace safety in the construction industry is a weekly trade journal called Le Moniteur des Travaux Publics et du Baˆtiment1 (The Monitor of Public Works and Construction). Founded in 1902 and dedicated to some of the main professions in the construction industry – contractors, architects, industrialists, promoters, and engineers (Fig. 1) – it has since occupied a central and unique position in the construction industry in France.2 In the 1960s, it became the most distributed trade journal in France (Bellanger, Godechot, Guiral, & Terrou, 1975).

Data Collection We had full access to all the issues of the targeted trade journal, from the first issue in August 1902 to the present time, except some rare missing

328

THIBAULT DAUDIGEOS ET AL. Readership (%)

12%

30%

5%

Contractors Architects

11% Private promoters Public promoters 5% 16%

21%

Materials providers Services providers Institutions and education

Fig. 1.

Readership of Le Moniteur des Travaux Publics et du Baˆtiment in 2011 (%). Source: Le Moniteur des Travaux Publics et du Baˆtiment.

issues. To build our sample, we decided to select the journal articles relating to safety that were published in one year out of every 10 years from 1902 to 2012. For issues published in 1902 to 1996 inclusive, we analyzed print copies, and from 1997 onward we used online scanned archives of the journal. We thus applied two distinct methodologies to collect the data according to the printed or digital nature of the data. – From 1902 to 1996 inclusive, we manually selected the articles relating to safety. To do this, we defined a list of keywords that should be present at least once in the selected articles: ‘‘accident,’’ ‘‘safety,’’ ‘‘occupational injury,’’ ‘‘prevention,’’ ‘‘precaution,’’ ‘‘disability,’’ ‘‘unfitness for work,’’ ‘‘death,’’ ‘‘victim,’’ ‘‘injury,’’ ‘‘wound,’’ ‘‘casualty,’’ ‘‘protection,’’ ‘‘work inspectorate,’’ and ‘‘compensation.’’ We also opened the list to words based on the same lexical roots, such as ‘‘dead’’ for ‘‘death’’ and ‘‘injured’’ for ‘‘injury.’’ Using this selection criterion, we obtained a first set of articles. Articles varied from 100 to several thousand words. We then carefully read all these articles to exclude those that contain these keywords but did not in fact relate to workplace safety in the construction industry. For instance, we systematically removed the articles that related to workplace health but were not linked to any accident. As an illustration of this, the asbestos tragedy in the construction industry was not considered a safety issue, because it did not imply any accident. – From 1997 and after, we used a search engine to select the targeted articles using the same keywords. From this sample, we then manually selected the articles that really dealt with our area of interest, in the same way as with the print copies of the earlier period.

329

Taking Stock of Institutional Complexity

We thus collected a total of 187 articles covering the whole period of analysis on a discrete basis (see Table 1). We then built a corpus gathering all collected articles. A signal variable was additionally created to link each article with its year of publication. This signal variable would later be used to relate the results of our analysis to time periods.

Data Analysis: A Descendent Hierarchical Classification Model Different analytical methodologies may be applied to infer meaning systems from textual data (Berelson, 1952; Mohr, 1998; Ventresca & Mohr, 2002). Archival research itself has given rise to a diverse set of analytical methodologies. One of them, content analysis, has already been widely used to study institutional processes (Krippendorff, 2004). Content analysis may be divided into two distinct streams: thematic analysis and textual analysis. To choose between these, we refer to the categorization of methodologies by Ventresca and Mohr (2002, p. 819), who propose basing this on the research goal of the archival researcher (see Table 2). Our own empirical goal in this research is to track meaning systems over a long time period. As this is a mixed objective in the typology of Ventresca and Mohr, we adopted a mixed strategy. Our main object of investigation was a set of professional discourses in a trade journal as it is a good data source for studying meaning systems. But from that, we decided to analyze this source with a descendent hierarchical classification model since this methodology fits with our need to obtain a longitudinal analysis of the data. How Do We Infer Meaning Systems from Discourses Using a Descendent Hierarchical Classification Model? A hierarchical classification model may be used to conduct a structural analysis of meaning as coined by Mohr (1998). Following Ferdinand de Table 1. Number of Articles Relating to Safety from 1906 to 2007. Year

1906 1920 1927 1937 1947 1957 1967 1977 1987 1997 2007 Total

Number of 11 articles relating to workplace safety

0

7

5

7

24

15

27

17

25

49

187

330

THIBAULT DAUDIGEOS ET AL.

Table 2. Objects of Investigation, Data Sources and Analytical Methodologies in Archival Research (from Ventresca & Mohr, 2002, p. 819). Objects of Investigation

Data Sources

Analytical Methodologies

Meaning systems

Professional discourse (journal articles, trade publications) Procedural talk (emails) Organizational identity statements (directories, initial public offerings, annual reports)

Content analysis Semantic grammars Semiotics Multidimensional scaling

Institutional logics

Classification statements (directories, industry reports, organizational narratives)

Galois Lattice Correspondence analysis Hierarchical classification model

Saussure’s tradition, a structural interpretation of meaning contends that the meaning of isolated words is fundamentally arbitrary and that words derive their meaning from their placement within larger systems. As a consequence, the pattern of relations among words is more important than the words themselves (Krippendorff, 2004, p. 290). Hence, this research stance is often referred to as a relational view on measuring meaning (Carley, 1994; Franzosi, 1995; Kennedy, 2005). The second core element of the structural tradition is the dialectical relationship between cultural meaning and social structures; social and cultural structures should then be captured together. Diverse analytical tools compete to reduce and simplify the data source to find the substantial structure of the text. Hierarchical classification models are still one of the best instruments to account for the duality that inheres between cultural and social structures (Mohr, 1998). They support a two-mode data-analysis strategy because they order both columns and rows of a matrix data. In our case, we contend that the hierarchical classification model is especially relevant for the longitudinal analysis of data over a long period of time. Even if some algorithms of hierarchical classification have already been discussed in medical, psychological and sociological studies (Breiger, 2000; Leenen et al., 1999; Mohr, 1998), these models remain largely obscure.

Taking Stock of Institutional Complexity

331

Focus on the Algorithm of the Descendent Hierarchical Classification Model Following the structural stance on meaning, the algorithm of a descendent hierarchical classification model3 looks at the distribution of lexical forms within a corpus through the analysis of repeated segments and lexical association. More precisely, it tracks co-occurrences of lexical forms. Their exact format, their position within a sentence and their number are not significant. What matters is their absence from or presence in some text segments, which could lead to a deviation from the uniform distribution of words within the entire corpus being analyzed. A deviation from the uniform distribution may signal the intention of the author of the discourse to associate words in a special manner in order to create meaning. The first objective of the algorithm is to create a simplified view of the textual data source while not losing the main characteristics of its initial structure. The algorithm produces a binary matrix (objects x attributes) that helps the ordering of the textual data. To do so, clusters are organized in both columns and rows of the matrix. As mentioned by Mohr (1998, p. 363), ‘‘the classes of objects are hierarchically ordered, the classes of attributes are hierarchically ordered and the two hierarchical orders are related to one another.’’ The hierarchical classification is thus a process of categorization and reduction. It is called a descendent classification because the starting point is the full corpus, which is then divided into classes that simplify it while maximizing the structure of the text. The second objective is then to build these classes of words that maximize the deviation from uniform distribution. Chi2 (Karl Pearson’s w2) is used to measure the actual distance from the uniform distribution and thus the strength of the link between forms and classes. Classes are composed not only of words extracted from the articles but also of signal variables, that is, the years significantly associated with the class in our case. For example, the fact that the year 1927 is part of Class 1 means that the meaning system identified by Class 1 is significantly associated with the 1920s. It also means that this class of words is relatively dominant in the 1920s in comparison with all the other periods. The words that comprise Class1 and their associated meanings may be present in other places in the corpus, however, and so may appear at different periods. While a thematic analysis would remain largely interpretative, lexical cooccurrence provides an objective track of the meaning of a discourse. Also, in the case of a descendent hierarchical classification, the interpretation of the researcher and the naming of the classes that have been constructed by the algorithm come at the end. By employing such a model, we intend to contribute to the intense methodological conversation that has taken place

332

THIBAULT DAUDIGEOS ET AL.

in recent years about the influence of the measure itself on the way we investigate the questions of institutional complexity (Hoffman, 1999; Marquis & Lounsbury, 2007; Thornton & Ocasio, 2008).

Data Analysis: From Meaning Systems To Characterized Institutional Logics Building on the several classes of lexemes and corresponding meaning structures identified through the descendent hierarchical classification of the analyzed corpus, we then followed a three-step process to characterize the corresponding institutional logics. First Step: Classifying Lexemes Every class is composed of several tens of words that relate significantly. In our first step, we sorted these lexemes and allocated them to the categories that DiMaggio and Mullen (2000) proposed for characterizing the logics they study in the historic context of progressive America. These categories include the social issue at stake, the actors, the repertoire of actions, and the objects of action mobilized by a given logic. Table 3 summarizes these various categories and provides their definitions. In applying this model, we decided to further break down the ‘‘actors’’ category into three subcategories that are pertinent to the field of workplace safety. We therefore made the distinction between primary and secondary actors, and, within primary actors, between responsible and victim actors. Second Step: Naming and Characterizing Institutional Logics In our second step, we further characterized the institutional logics. First, we used the network analysis feature offered by our software to identify the most Table 3.

Framework of Analysis of Institutional Logics.

Dimensions of Institutional Logics Social issue at stake Actors Repertoire of legitimate actions Objects of actions

Definitions

Framing of the central issue around which the field is constituted Organizational roles that transcend individuals’ roles and local contexts Content and ways of doing organizational tasks performed by actors Tools that support and shape actors’ repertoire of legitimate actions

Taking Stock of Institutional Complexity

333

central lexemes within each class. Second, we compared each institutional logic with its corresponding historical context through the study of secondary data, namely historical literature relating to the field of workplace safety. By doing so, we were able to name each logic and then to infer its main features in line with the model of analysis proposed by Thornton et al. (2012, p. 73) for describing institutional orders. Following their recommendation to select the features that are most salient to the research questions and context (2012, p. 59), we used eight of their nine elemental categories – that is, root metaphor, source of identity, source of legitimacy, source of authority, basis of norms, basis of attention, basis of strategy, and economic system. Table 4 provides our own definitions of the eight features we chose to focalize. Third Step: Relating Institutional Logics to Institutional Orders Finally, in our third step, we made the link between our field-level institutional logics and the broader interinstitutional system at the societal level. To do so, we tried to systematically recognize, in the description of our logics, elements that pertain to the higher institutional orders, as described by Thornton et al. (2012, p. 73). This systematic work of comparison was facilitated by the fact that we had used the same model to characterize our logics as these authors used to characterize the higher institutional orders. For each of the identified logics, this allowed us to recognize the presence and influence of at least one institutional order, and more often of two of them.

RESULTS In this section, we describe the evolution of workplace safety logics in the French construction industry throughout the 20th century. For each logic, we give the classification of lexemes associated with it, our interpretation of the elemental subcategories that define it, and the institutional order(s) that support it. Using primary and secondary data, we thus meet our initial research objective: tracking the evolution of a pool of institutional logics and relating this evolution to the interinstitutional system.

The Evolution Of Workplace Safety Logics In France Over The 20th Century Over the period 1906–2007, the descendent hierarchical classification model was able to identify six stable distinct categories of lexemes. For each of the

Committee, funds, intercompany, subcontractors, ministers, inspection, social insurance, master owner

Recommendation, measure, intervention, adopt, establish, communicate, conform, coordinate, require, approve Decree, bylaw, probate, plans, notebooks

Court, judge, representative, inspection

Decision, compensation, court order, dispute, payment, submit a case, legal redress, impose, sentence

Act, law, judgment, bill, regulation, fine, evidence

Secondary actors

Repertoire of legitimate actions

Objects of actions

X

X

1947, 1977 Social insurance, prevention

Victim, employee, workman

1907, 1927, 1937 Offense, misdemeanor, violation of the rights of an employee, liability, responsibility, safety Owner

State

Colloquiums, conferences, symposiums, studies, science, propaganda, events, reports, treaties

Federations, doctors, associations, International Association of Social Insurance, institute, National Safety Institute, OPPBTP, offices, specialists Chair, cooperate, document, underline, communicate, organize, influence, gather, insist, outline

X

X

1957, 1967 Prevention, technological, safety

Professions

Tools, cranes, vehicles, trials, equipment

Lift, prop up, rescue, fit out, commercialize, filter out, insulate, excavate, install, use, protect

Firemen, squad, laboratory

Miners, individual, man

1967 Protection, rescue, precaution, danger, explosion, failure, fall down X

Technical

X

Rate, AT (occupational injury rate), figures, numbers, table, ratings, allowance, statistics, ratemaking

Calculate, indemnify, count, contribute, pay, markup, make a census, account, increase, decrease

Death, temporary disability, permanent disability, employees, serious, seriousness, fatal, disease CRAM (local health insurance fund), CPAM (state health insurance fund)

1977, 2007 Injury

Accounting

Lexemes Significantly Related to the Categories Identified by Alceste.

Primary actor: victim

Primary actor: responsible

Year Social issue at stake

Judicial

Table 4.

Training period, information, programs, zero accident, quality, charter, objectives, video

Coordinator, Centre Pierre Caloni (training center), ASE (corporate safety officers’ professional association), OPPBTP, journeymen, contracting owner, partners, master owner Training, explain, organization, integrate, hand out, improve, organize, make, realize, reward, control

Corporation, staff, construction site, SPIE (company), director, foreman, Colas (company), Bouygues (company), subsidiary, function Trainees

1997, 2007 Safety, prevention, risk, behavior

Managerial

Taking Stock of Institutional Complexity

335

6 categories identified, Fig. 2 lists the 24 lexemes that are the most significantly related to the category, as measured by the w2 distance. Some years (identified by asterisks) appear to be significantly related to the categories identified by Alceste. These relationships between categories and years of publication allow us to recount the constitution over time of the discourses corresponding to a category and to build representations of the underlying institutional logics. For each of the six categories identified by Alceste, we then categorized the significantly associated lexemes according to the framework of analysis of institutional logics developed in our theoretical section. Table 4 presents the lexemes that are significantly related to the six identified logics. Using primary and secondary data, we define elemental subcategories that compose the logics, and propose an interpretation of the institutional orders that support them.

Fig. 2.

The Six Stable Categories of Lexemes Identified by Alceste.

336

THIBAULT DAUDIGEOS ET AL.

Category 1: The Judicial Logic (1900s, 1920s, And 1930s) The first category of lexemes is significantly associated with the 1900–1940 period (1906, 1927, and 1937, with respective w2 of 152, 296, and 61). Strikingly, this category is characterized exclusively by judicial vocabulary, even in the lexemes that relate to the repertoire or objects of action: ‘‘court,’’ ‘‘judge,’’ ‘‘act,’’ ‘‘law,’’ ‘‘judgment,’’ ‘‘fine,’’ ‘‘evidence,’’ ‘‘court order,’’ ‘‘submit a case,’’ and ‘‘legal redress.’’ This first result means that the judicial logic is dominant in the first half of the twentieth century relative to the rest of the period studied. Workplace safety is mainly understood as an issue dealing with ‘‘misdemeanors’’ or ‘‘offenses’’ that engaged the ‘‘liability’’ of individuals. The corporation is represented by its ‘‘owner’’ and the injured worker is portrayed as the ‘‘victim.’’ Our framework of analysis evidences a system of relationships between individuals where the organizational dimension is totally absent (Table 4). From this classification of the lexemes significantly associated with the judicial logic, we then define the elemental categories that compose the logic (Table 5). Under the judicial logic, workplace safety is primarily considered a matter of justice. The study of our secondary data allows us to relate this judicial logic to the issuance of the 1898 law, which is the foundational instrument of French modern public policies applying to occupational injuries (Ewald, 1986; Le Goff, 2004 [1985]). The rule of law is the general model of society supporting the logic. Every citizen is protected by the law equally. The 1898 law distinguishes between two different types of cases in workplace safety. First, it sets the principle of the systematic civil responsibility of the employer for all injuries taking place within the company. The victim employee can thus ask for financial compensation, without having to prove any employer’s misdemeanor. The compensatory amount is fixed and the employer benefits in exchange for the payment, from penal immunity. However, if a specific case occurs when a violation of the rights of the employee is suspected, the employer can be put before a criminal court, and sentenced if the violation is evidenced. As a consequence of the new law, there is intense activity in the production of laws and in jurisprudence at the beginning of the twentieth century in order to define a violation of the rights of the employee. The production and enforcement of laws thus confer authority, while laws and jurisprudence define what is legitimate. The judicial power targets a certain level of legal conformity by qualifying and punishing offenses. Our results show that a judicial view on workplace safety characterizes the first half of the century, in contrast to the rest of the period studied. Behind the judicial logic, we can infer a unique combination of two institutional

Contributor to the social system

Citizenship membership

Basis of norms (=criterion according to which one’s behavior is driven by the logic)

Involvement in the operational work on the building sites

Engineering skills, standardization bodies

State Public bodies

Production and enforcement of laws (Parliament, court)

Source of authority (=what confers authority)

Professional associations, production and communication of scientific knowledge Membership of communities of practice and professional associations

Technical specifications Technical innovation

Expertise

Source of legitimacy (=what confers legitimacy)

Submission to or mastering of a physical event

Recognition as an expert

Economic classification (owner, subcontractor) Values of the Republic (equality, fraternity)

Technological change

Technical

Division between work and capital (owner versus workman) Laws, jurisprudence Proof Trust, liability

Source of identity (=what makes the actors visible)

Professions Science

State

Social welfare

Justice

Root metaphor (=safety as a matter ofy)

Judicial

Accounting scope

Public production and mastering of statistics

Figures Transparency

Accounting Classification of victims

Valuation and measurement

Accounting

Table 5. The Elemental Categories of the Six Institutional Logics.

Within the frontier of the firm (which has implemented a safety policy)

Hierarchy (supported by public bodies and professional associations)

Strategy, process, key performance indicators Management knowledge

Productivity Continuous improvement process Position in the corporation and contractual status

Managerial

Organization of mutualization

Control of conformity/ deviance

Welfare capitalism

Legal conformity

Qualify and punish offenses

Rule of law

Basis of strategy (=main action to address the logic base of attention) Economic system (=model of society the logic promotes)

State

Basis of attention (=what the logic focuses on)

Judicial

Technocracy

Formalization of knowledge

Prevention

Professions

Technical

Technocracy

Rationalization of workplace tasks and technical conformity Optimize technical solutions

Table 5. (Continued )

Market

Reliability and availability of information

Measures, ranking, normality curves

Accounting

Managerial capitalism

Training, action plans

Quality and productivity objectives

Managerial

Taking Stock of Institutional Complexity

339

orders: state and family, as defined in the interinstitutional system (Thornton et al., 2012). As in the institutional order of state, citizenship membership delineates the basis of norms. People are ruled by this safety logic as soon as they are under the protection of the French law. Besides, the logic recognizes social classes, that is, division between work and capital, as a source of identity – as in the institutional order of state. Indeed, laws and jurisprudence set rights and duties differently for owners and for workers, as defined in the 1898 law and the following jurisprudence. However, we do not have here a pure manifestation of the state order based on welfare principles and redistribution mechanisms; the state order is hybridized with the family order, which confers very specific modalities. In the judicial logic, many principles of the family order are translated on a national scale. This way of exerting power via the centralized production and enforcement of laws is close to a patriarchal model of society, whereas there is a striking analogy between the liability of the owner of the firm in the judicial logic and the liability of the chief of the family in the family order.

Category 2: The State Logic (1940s and 1970s) Another category of lexemes is significantly associated with the period following World War II (WWII), especially the decades 1940s and 1970s (1947 and 1977, with respective w2 of 73 and 182). This category is characterized by social welfare vocabulary, such as ‘‘social insurance,’’ and ‘‘prevention.’’ Our analysis of secondary data relates the appearance of this new logic to the birth in 1946 of the current regime of public administration of occupational injuries (Ewald, 1986; Le Goff, 2004 [1985]). Industrial medicine becomes compulsory on October 11, 1946, and the law of 30 October the same year gives birth to the social security system and its related obligations. State intervention is strongly emphasized through a list of action-based words (‘‘measure,’’ ‘‘recommendation,’’ ‘‘intervention,’’ ‘‘adopt,’’ ‘‘establish,’’ ‘‘conform,’’ ‘‘require,’’ and ‘‘approve’’) and instruments of public policies (‘‘decree’’ and ‘‘bylaw’’). Occupational injuries are no longer perceived as consequences of individual deeds, but rather as random events as part of a complex work environment. Consequently, the figures of responsible and victim actors disappear from the associated lexemes, and safety issues are mutualized and managed by large state-owned bodies. These bodies use to ‘‘coordinate’’ the organization of the social contributions system, and make sure that large firms that deviate too much pay more and thus get incentive to improve their safety performance. ‘‘Conformity’’ is a key issue in the social system, and public authorities

340

THIBAULT DAUDIGEOS ET AL.

provide ‘‘advice,’’ ‘‘communicate,’’ and give ‘‘recommendations’’ to follow. Such terms are presented in Table 4. From these lexemes, we built the elemental categories that compose the logic (Table 5). Under the state logic, workplace safety is primarily considered a matter of social welfare, where every employee is protected and the richest contribute the most. In 1946, the insurance regime becomes public and universal in France. Corporations no longer pay the compensatory amount direct to the victim but instead contribute to dedicated funds through the payment of taxes. Large companies pay according to their actual accident rate, whereas smaller firms contribute a fixed, mutualized amount. This contributions system thus recognizes the random nature of workplace accidents and introduces a protection for the small economic actors, who can be drastically affected by sudden serious workplace accidents. Welfare capitalism is the general model of society supporting the logic, that is, resources are mutualized and equally redistributed, providing protection from and prevention of injuries for those who financially contribute to the social system. The public authorities and public bodies confer authority to social actors, policies and processes in the field, while the values of the French Republic (equality and fraternity) define the basis of legitimacy. The organization of mutualization and redistribution mechanisms are made possible by controlling conformity and deviance through the governance of mechanisms for social contributions and the work of the safety inspectorate. This change does not, however, totally overrule the past regime. The 1898 law remains in force and becomes part of the new social security system. The employers’ responsibility remains limited except in the case of violation of the rights of the employee. The only changes are, first, that the state now organize the compensatory scheme and, second, that a rationale of mutualization is introduced. Behind the state logic, we can infer a unique institutional order: the state as it is defined in the interinstitutional system. The state logic of workplace safety just after WWII completely echoes the state order on several points complementary to those cited above: contributing or not to the social system delineates the basis of norms, while the logic recognizes economic classification as a source of identity. Category 3: The Professional Logic (1950s and 1960s) The third category of lexemes is significantly associated with the 1950s and 1960s (1957 and 1967, with respective w2 of 45 and 215). This category is mostly characterized by scientific vocabulary, highlighting knowledge, information and lobbying activities, which are reflected in their repertoire

Taking Stock of Institutional Complexity

341

of actions (‘‘document,’’ ‘‘communicate,’’ ‘‘influence,’’ ‘‘cooperate,’’ and ‘‘chair’’) as well as objects of action (‘‘colloquiums,’’ ‘‘conferences,’’ ‘‘symposiums,’’ ‘‘studies,’’ and ‘‘reports’’). A new range of stakeholders is engaged in such activities: ‘‘specialists,’’ ‘‘federations,’’ ‘‘doctors,’’ ‘‘associations,’’ ‘‘institutes,’’ ‘‘offices,’’ and so on. Responsible and victim actors are still absent from the framework of analysis. Referring to Scott (2008 [1995]), we have named the logic influencing this period ‘‘professional,’’ meaning that professional associations and experts seek to theorize and stabilize meanings (Table 4). From this classification of the lexemes significantly associated with the professional logic, we then define the elemental categories that compose the logic (Table 5). Under the professional logic, workplace safety is primarily considered a matter of science. The technocracy is the general model of society supporting the logic: new stakeholders, such as professionals and the community of experts, cannot be ignored and give directions on workplace safety. As a consequence, the professional associations – and the production and communication of scientific knowledge – confer authority, while expertise defines what is legitimate. The state still plays a role in the implementation of this professional logic, as it hosts experts and produces scientific knowledge in public bodies such as the OPPBTP (Organisme Professionnel de Pre´vention du Baˆtiment et des Travaux Publics = Professional Office for Prevention in the Construction Industry), whose mission is the development of prevention knowledge. Behind this logic, we can infer a combination of the professional and the state institutional orders, as defined in the interinstitutional system. The professional logic completely echoes the professional order on several points: the professional associations target a certain level of prevention by formalizing knowledge about workplace safety; the membership in communities of practices and professional associations delineates the basis of norms; and the logic recognizes the status of experts as a source of identity. But the state order is also implicitly present, in that it can be considered an instrument of formalization and enforcement of the knowledge created with professional associations. Moreover, important actors of this logic have a hybrid status between a public authority and a private professional association. As such, the sources of authority often rely on public bodies. For instance, the OPPBTP’s main mission is to advise and train corporations on anticipating the risks arising from their activities. Similarly, the state bodies organizing the French public-insurance system start to connect with similar foreign counterparts and to play an important role in supporting and diffusing the creation of knowledge at the international level.

342

THIBAULT DAUDIGEOS ET AL.

Category 4: The Technical Logic (1960s) The technical logic is significantly associated with the 1960s (1967 with a w2 of 298). This category is mostly characterized by risk vocabulary, concretely expressed in the notions of ‘‘danger,’’ ‘‘explosion,’’ ‘‘failure,’’ and ‘‘fall.’’ The victim reappears as a primary actor, referred to as an ‘‘individual’’ or ‘‘man.’’ The risk of accident is tackled on the building site through technical remediation. It is necessary to ‘‘lift,’’ ‘‘prop up,’’ ‘‘insulate,’’ and ‘‘excavate’’ in order to ‘‘filter out’’ potential risks; this implies the use of ‘‘cranes,’’ ‘‘vehicles,’’ ‘‘tools,’’ etc. (Table 4). Table 5 introduces the elemental categories associated with this logic, under which workplace safety is primarily considered a matter of technical change. The logic relates to the period of technical progress associated with the post-WWII economic boom (Ewald, 1986; Le Goff, 2004 [1985]). In the same way that the rationalization and mechanization of a growing number of tasks results at the time in strong productivity gains, it is considered within this logic that the prevention of safety issues can be significantly improved through the issuance of ad hoc or standardized technical solutions. The logic gives salience to two types of actors according to whether they have the capacity to master a physical event or they are affected by it. The issue of safety therefore appears as an operational issue, integrated into technocracy as a general model of society. Behind the technical logic, we can infer a combination of two institutional orders, as defined in the interinstitutional system: the profession and the corporation orders. The technical logic echoes the professional order: the engineering skills and standardization bodies, as sources of authority, aim to optimize technical solutions, which relate to experts’ knowledge and to professional associations mentioned in the professional order. Technical specifications and solutions define what is legitimate, and these can also be considered an extension of the expertise of the professional order. Nevertheless, the corporation order seems to influence the technical logic too. The rationalization induced by technical progress mainly targets workplace tasks on building sites and echoes productivity enhancement programs, which constitute the basis of attention in the corporation order. Likewise, in the technical logic, the involvement in operational work delineates the basis of norms. There is here a clear analogy with some elements of the managerial logic as defined in category 6, with technocracy invading corporate boundaries. Category 5: The Accounting Logic (1970s and 2000s) The 1970s (1977 with w2 of 73) are characterized by a new category of lexemes that focus on measurement and evaluation. These lexemes highlight that actors

Taking Stock of Institutional Complexity

343

are now interested in the measuring and costing of occupational injuries. They want to ‘‘count,’’ ‘‘make a census’’ of or ‘‘account’’ for accidents, the ‘‘rates’’ of which ‘‘decrease’’ or ‘‘increase.’’ The financial aspect is ever present, since corporations ‘‘calculate,’’ ‘‘indemnify,’’ ‘‘contribute,’’ or ‘‘pay’’ based on the ‘‘ratemaking’’ provided by the social-insurance administration. In summary, the accounting logic deals with ‘‘figures,’’ ‘‘numbers,’’ and ‘‘statistics’’ organized in ‘‘tables.’’ The representation of the victim takes various forms, as defined by the administration: ‘‘death,’’ ‘‘temporary disability,’’ or ‘‘permanent disability’’ (Table 4). From this classification of the lexemes, and thanks to the analysis of our secondary data, we can now define the elemental categories that compose the logic (Table 5). Under the accounting logic, workplace safety is primarily considered a matter of valuation and measurement. Such disclosure based on quantitative indicators starts in the 1940s, with the implementation of the social security system, and progressively develops to become salient in the 1970s (Ewald, 1986; Le Goff, 2004 [1985]), embracing the development of market capitalism as a model of society in the 1970s. Actors are provided with a dispassionate overview of the risk evolution, and public bodies regularly disclose key statistics – such as the Health Insurance Fund’s frequency index of occupational injuries, quarterly report on occupational injuries, and yearly financial and technical report on occupational injuries; and the Ministry of Labor’s annual outline summary. In this way, the public production and mastering of statistics confer authority, while figures and transparency define what is legitimate. The basis of norms focuses on the accounting scope, and the latter provides a framework of accounting classifications of victims, which sets the basis of identity. Behind this logic, we can infer a combination of the market and state institutional orders. The market order is explicitly related to the accounting logic, especially through their common basis of attention (measures, rankings, numbers, and normality curves) and strategy (the need for transparency and reliability). The state order is also implicitly present in the accounting logic, but more as a support for the market side: the public bodies produce figures and tables and organize the transparency of the field at that time; these then serve as a basis for the regulation of safety by a market logic. Category 6: The Managerial Logic (1990s and 2000s) The category of lexemes significantly associated with the 1990s and 2000s (1997 and 2007, with respective w2 of 435 and 150) focuses on continuous improvement processes, especially within the corporation. The repertoire

344

THIBAULT DAUDIGEOS ET AL.

and objects of action are typical of management: ‘‘train,’’ ‘‘explain,’’ ‘‘organize,’’ ‘‘reward,’’ based on ‘‘training periods,’’ ‘‘information,’’ ‘‘zero accident programs,’’ and well-defined ‘‘objectives.’’ A parallel is traced with ‘‘quality’’ programs and their principle of continuous ‘‘improvement.’’ The management board is put upfront as a responsible actor in this logic, but interestingly in a completely different way from in the judicial logic: unlike the latter, corporations are no longer personified by their owners. Now, management appears in its organizational dimension (‘‘corporation,’’ ‘‘subsidiaries,’’ and ‘‘construction sites’’), through the whole hierarchical chain (‘‘staff,’’ ‘‘director,’’ and ‘‘foreman’’) and staff professionals (‘‘functions’’ and ‘‘safety officers’’). Large French companies are often named, for example, ‘‘SPIE,’’ ‘‘Colas,’’ and ‘‘Bouygues.’’ Construction companies also relate to other actors in the value chain, ‘‘contracting owners’’ or ‘‘master owners.’’ The administration appears only through its advisory and training bodies: ‘‘Centre Pierre Caloni’’ (=Pierre Caloni Center, which used to be a state-owned training center for safety officers, work inspectorsy), and ‘‘OPPBTP’’ (Table 4). Regarding the elemental categories that comprise the managerial logic (Table 5), workplace safety is primarily considered a matter of productivity and continuous improvement. The managerial logic is actually the only one that places the corporation at the center of the field of workplace safety with a structuring role. Through its strategy and processes, the setting of performance indicators, and the development of specific management knowledge about safety, the corporation defines legitimate issues in the field. The frontier of the firm and the position in the firm hierarchy are sources of identity and define who is impacted by the logic. As a consequence, the managerial logics in the 1990s and 2000s echoes the managerial capitalism that can be considered the model of society in this period. These elemental categories are evidence that the corporation order – as defined in the interinstitutional system – supports the managerial logic. The sources of identity and authority, the basis of norms, and the economic system are clearly related to the elemental categories of the corporation order. Nevertheless, our analysis also suggests other influences. The associated lexemes and the elemental categories emphasize the development of a dedicated safety-management knowledge that is clearly related to the rise of new professionals: corporate safety officers. As secondary actors, these officers reinforce the authority of the hierarchy through their practice and the knowledge they create and diffuse. Lastly, the managerial logic is also supported by the market order. As shown in Tables 4 and 5, productivity and economic performance are central to the logic and define

Taking Stock of Institutional Complexity

345

not only the basis of attention in the field but even the root metaphor, which considers safety as an issue of productivity. Besides, the identity of the social actors rests not only on their role and position in the corporation, but also on contractual statuses, which are structural elements in a market order.

DISCUSSION AND CONCLUSION Our results confirm the fact that the usual focus on two dominant logics (Dunn & Jones, 2010; Lounsbury, 2002; Rao et al., 2003; Thornton & Ocasio, 1999; Thornton et al., 2005) runs the risk of seriously downplaying the richness and plurality of institutional life. Our study identified six logics of workplace safety during the twentieth century in France. The precise number of logics we identify is certainly not as important as the order of magnitude revealed. Compared to previous studies, we chose a methodological approach that aimed to more fully capture institutional complexity, as it is traced inductively from the data. This has the advantage of surfacing less-salient logics that contribute to the complexity of the field. In addition, our data collection covers a whole century, in contrast to previous studies which mostly focused on a few decades. And with the help of the descendent hierarchical analysis, the range and quantity of the collected data enabled us to identify a wider range of logics in action than might be otherwise identified. For all these reasons, we believe our methodological approach can enhance our understanding of institutional complexity, contributing to a more complete approach to the longitudinal dynamics of logic plurality (Dunn & Jones, 2010; Goodrick & Reay, 2011; Thornton et al., 2005). Our second contribution is to develop a better understanding of the nature of logics: through the connections we have made between logics and orders of the interinstitutional system, as described by Thornton et al. (2012), we can now better describe the ‘‘composite nature’’ of logics. Previous studies elaborated on the fact that logics are mere translations of societal-level understandings into actors’ practices (Lounsbury & Crumley, 2007; Thornton et al., 2012), suggesting that institutional logics can be conceived as combinations of higher institutional orders (Thornton et al., 2005). Following this idea, and by breaking down each logic into underlying orders, we show that logics are actually based on a combination of several institutional orders described by Thornton et al., (2012). For instance, our case reveals that the judicial logic can be linked to both family and state orders. Likewise, the managerial logic can be considered a combination of

346

THIBAULT DAUDIGEOS ET AL.

the corporation, profession and market orders. Moreover, orders that may potentially be antagonist in their interest or duties, such as the state order (which goes with an interest in social welfare and equal redistribution of resources) and the market order (which goes with a profit interest), can actually be connected under the banner of one single logic (in this case, an accounting logic). Contrary to previous studies that highlighted conflicts between orders, our study suggests that logics, through being composed of a plurality of orders, can actually reveal how these may complement each other. Thus, we highlight and further develop the complexity that lies in the very composite nature of logics. Our third contribution concerns a more dynamic approach to a pool of logics, by unveiling complementarities among different logics. Indeed, the trajectory of our pool of logics seems based more on these than on conflicts, contrary to what previous studies have focalized (Lounsbury, 2007; Marquis & Lounsbury, 2007). Our results suggest a coexistence of logics within a pool, with one logic nourishing another. For instance, the professional logic in the 1950s and 1960s created some knowledge that was applied in solutions, processes and tools on building sites during the 1960s, when technical logic was relatively strong. As such, they represent two sides of an interest in technical approaches (the professional logic for knowledge, and the technical one for its application), and echo the history of the constitution of the field. Likewise, the managerial logic finds inspiration in the accounting logic, in that the productivity and continuous improvement processes are possible because the accounting logic previously built valuation mechanisms and measurement systems (and because all the formalization of knowledge and technical solutions was previously elaborated in the professional and technical logics). Such complementarities between logics can be explained by the presence of one order supporting, to different degrees, several distinct logics. Indeed, the state order is present in four of the six logics revealed in our study, and at various degrees: if it perfectly fits the state logic, it is combined with the professional order in the professional logic, and with the market order in the accounting logic. In these two cases, the state order looks much more supportive than constitutive of those logics. The state order can actually create a link on which several logics of a pool are together constructed. In the same vein, we show that the professional order is supporting not only the professional logic, but also the technical and corporation ones. The scientific knowledge created and diffused under the professional logic in the 1950s and 1960s is then applied to technical improvement on the building sites under the technical logic, later entering corporate management

Taking Stock of Institutional Complexity

347

programs in the 1990s and 2000s. Therefore, we respond to the call from Thornton and colleagues (2012) for more empirical studies into the evolution of the interinstitutional system, and provide empirical evidence of a reconfiguration of orders of the interinstitutional system. As such, we suggest that, instead of following a sequential evolution, as mentioned in Thornton et al. (2012), the orders of the interinstitutional system can evolve and be reconfigured over time. A fourth and final contribution deals with a deeper look at the combination mechanisms between orders, in that our results suggest two ways of connecting orders to support logics. The first mechanism echoes the translation of some characteristics of an order into another, changing the scale of the application of the categorical elements of an order. For instance, in the judicial logic, the family order is combined with the state one, in that the family values, basis of identity, and other categories of this order are translated at the state level, that is, at a broader scale than the close circle of the family. Likewise, in the technical logic, the values, attention and other categorical elements of the professional order are translated at the scale of the corporation. The second mechanism of combination between orders is based on one order being an instrument or a support for another. For instance, the professional and state orders coexist in the professional logic, but our case shows that the state authorities create rules and norms only thanks to the professional knowledge. We can therefore consider the state authorities as being instrumental in or supporting the process of professional expertise and knowledge finding its place among the field actors. In the same vein, within the accounting logic, the market and state orders coexist, where the market supports the state values and basis of attention, among others. Our study does have limitations, which we think can open up avenues for future research. First, so as to correct any sample bias, our data collection could be extended. Indeed, although we chose to select one year per decade and analyzed articles from each, we may make our methodology more robust by tracing more years within each decade. Second, our methodology highlights the most visible logics in each decade. Even if we were able to trace both dominant and less visible logics, our methodology may miss more implicit ones. So to compensate for this issue, we used secondary data to get a better understanding of the context of the logics we measured. But more research into alternative solutions could be developed to improve our methodology. Third, future research can improve our point about pools of logics by studying other organizational fields and seeing how other pools of logics may be constructed and evolve in time. This may address a potential

348

THIBAULT DAUDIGEOS ET AL.

limitation of our study, namely that it is focused on a field with its own specificities and rules. One way to tackle this would be to study other fields that are composed around an issue, following Hoffman’s (1999) view, on which this paper is based. This would improve our understanding of such fields, contrary to many previous studies that empirically investigated fields constituted around an industry. Finally, this study calls for future research about mechanisms of combination between orders. This study suggests two mechanisms of combination, but they do not represent an exhaustive list.

NOTES 1. Biweekly from 1902 to 1940, and weekly since then. 2. In 1936, it acquired its main competitor, Le journal des travaux publics, du baˆtiment et des fournitures administratives. 3. For the data analysis in the present study, we used a French software program called Alceste.

REFERENCES Aldrich, M. (1997). Safety first: Technology, labor and business in the building of work safety, 1870–1939. Baltimore, MD: Johns Hopkins University Press. Alvesson, M., & Ka¨rreman, D. (2000). Varieties of discourse. On the study of organizations through discourse analysis. Human Relations, 53, 1125–1149. Battilana, J., & Dorado, S. (2010). Building sustainable hybrid organizations: The case of commercial microfinance organizations. Academy of Management Journal, 53, 1419– 1440. Bellanger, C., Godechot, J., Guiral, P., & Terrou, F. (1975). Histoire ge´ne´rale de la presse francaise, tome 4: De 1940 a` 1958. France: Presses Universitaires de France. Berelson, B. (1952). Content analysis in communication research. New York, NY: Free Press. Boxenbaum, E., & Daudigeos, T. (2008). Institutional factors in market creation: Concrete theorization of a new construction technology. Academy of Management, best papers proceedings, Anaheim, CA. Boxenbaum, E., & Jonsson, S. (2008). Isomorphism, diffusion and decoupling. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 78–98). London: Sage. Breiger, R. L. (2000). A tool kit for practice theory. Poetics, 27, 91–116. Carley, K. M. (1994). Extracting culture through textual analysis. Poetics, 22, 291–312. DiMaggio, P. (1988). Interest and agency in institutional theory. In L. G. Zucker (Ed.), Institutional patterns and organizations: Culture and environment (pp. 3–21). Cambridge, MA: Ballinger. DiMaggio, P., & Mullen, A. L. (2000). Enacting community in progressive America: Civic rituals in national music week, 1924. Poetics, 27, 135–162.

Taking Stock of Institutional Complexity

349

Dunn, M. B., & Jones, C. (2010). Institutional logics and institutional pluralism: The contestation of care and science logics in medical education, 1967–2005. Administrative Science Quarterly, 55, 114–149. Ewald, F. (1986). L’Etat providence. Paris: Grasset. Franzosi, R. (1987). The press as a source of socio-historical data: Issues in the methodology of data collection from newspaper. Historical Methods, 20, 5–16. Franzosi, R. (1995). Computer-assisted content analysis of newspapers. Quality and Quantity, 29, 157–172. Friedland, R., & Alford, R. R. (1991). Bringing society back in: Symbols, practices and institutional contradictions. In W. W. Powell & P. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 232–263). Chicago, IL: University Chicago Press. Goodrick, E., & Reay, T. (2011). Constellations of institutional logics: Changes in the professional work of pharmacists. Work and Occupations, 38, 372–416. Greenwood, R., Diaz, A. M., Li, S. X., & Lorente, J. C. (2010). The multiplicity of institutional logics and the heterogeneity of organizational responses. Organization Science, 21, 521–539. Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E., & Lounsbury, M. (2011). Institutional complexity and organizational responses. Academy of Management Annals, 5, 1–55. Hoffman, A. J. (1999). Institutional evolution and change: Environmentalism and the U.S. chemical industry. Academy of Management Journal, 42, 351–371. Kennedy, M. T. (2005). Behind the one-way mirror: Refraction in the construction of product market categories. Poetics, 33, 201–226. Kraatz, M. S., & Block, E. S. (2008). Organizational implications of institutional pluralism. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 243–275). London: Sage. Krippendorff, K. (2004). Content analysis: An introduction to its methodology (2nd ed.). Beverly Hills, CA: Sage. Le Goff, J. (2004 [1985]). Du silence a` la parole : Une histoire du droit du travail des anne´es 1830 a` nos jours. Rennes: Presses Universitaires de Rennes. Leenen, I., Van Mechelen, I., De Boeck, P., & Rosenberg, S. (1999). INDCLAS: A three-way hierarchical classes model. Psychometrika, 64, 9–24. Leenen, I., Van Mechelen, I., Gelman, A., & De Knop, S. (2008). Bayesian hierarchical classes analysis. Psychometrika, 73, 39–64. Lounsbury, M. (2002). Institutional transformation and status mobility: The professionalization of the field of finance. Academy of Management Journal, 45, 255–266. Lounsbury, M. (2007). A tale of two cities: Competing logics and practice variation in the profession of mutual funds. Academy of Management Journal, 50, 289–307. Lounsbury, M., & Crumley, E. T. (2007). New practice creation: An institutional perspective on innovation. Organization Studies, 28, 993–1012. Lounsbury, M., Ventresca, M., & Hirsch, P. M. (2003). Social movements, field frames and industry emergence: A cultural-political perspective on US recycling. Socio-Economic Review, 1, 71–104. Marquis, C., & Lounsbury, M. (2007). Vive la resistance: Competing logics and the consolidation of U.S. community banking. Academy of Management Journal, 50, 799–820. Mohr, J. W. (1998). Measuring meaning structures. Annual review of sociology, 24, 345–370. Phillips, N., & Hardy, C. (2002). Discourse analysis: Investigating processes of social construction. London: Sage.

350

THIBAULT DAUDIGEOS ET AL.

Phillips, N., & Malhotra, N. (2008). Taking social construction seriously: Extending the discursive approach in institutional theory. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 702–736). London: Sage. Rao, H., Monin, P., & Durand, R. (2003). Institutional change in Toque Ville: Nouvelle cuisine as an identity movement in French gastronomy. American Journal of Sociology, 108, 795–843. Schneiberg, M., & Clemens, E. S. (2006). The typical tools for the job: Research strategies in institutional analysis. Sociological Theory, 24, 195–227. Scott, W. R. (2008[1995]). Institutions and Organizations: Ideas and Interests. Thousand Oaks, CA: Sage. Suddaby, R., & Greenwood, R. (2005). Rhetorical strategies of legitimacy. Administrative Science Quarterly, 50, 35–67. Thornton, P. H. (2004). Markets from culture: institutional logics and organizational decisions in higher education publishing. Stanford, CA: Stanford University Press. Thornton, P. H., Jones, C., & Kury, K. (2005). Institutional logics and institutional change in organizations: Transformation in accounting, architecture, and publishing. Research in the Sociology of Organizations, 23, 125–170. Thornton, P. H., & Ocasio, W. (1999). Institutional logics and the historical contingency of power in organizations: Executive succession in the higher education publishing industry, 1958–1990. American Journal of Sociology, 105, 801–843. Thornton, P. H., & Ocasio, W. (2008). Institutional logics. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 99–129). London: Sage. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure and process. Oxford: Oxford University Press. Ventresca, M. J., & Mohr, J. W. (2002). Archival research methods. In J. A. C. Baum (Ed.), The Blackwell companion to organizations (pp. 805–828). Malden, MA: Blackwell.

FROM CULTURAL REPERTOIRES TO INSTITUTIONAL LOGICS: A CONTENT-ANALYTIC METHOD Klaus Weber, Hetal Patel and Kathryn L. Heinze ABSTRACT Much of contemporary institutional theory rests on the identification of structured, coherent, and encompassing logics, and from there proceeds to examine multilevel dynamics or the relationship between logics in a field. Less research directly studies the internal properties and dynamics of logics and how they are structured over time. In this paper, we propose a method for understanding the content and organization of logics over time. We advocate for an analysis of logics that is grounded in a repertoire view of culture (Swidler, 1986; Weber, 2005). This approach involves identifying the set of cultural categories that can make up logics, and measuring empirically the dimensions that mark a cultural system as more or less logic-like. We discuss several text analytic approaches suitable for discourse data, and outline a seven-step method for describing the internal organization of a cultural repertoire in term of its ‘‘logicness.’’ We provide empirical illustrations from a historical analysis of the field of alternative livestock agriculture. Our approach provides an

Institutional Logics in Action, Part B Research in the Sociology of Organizations, Volume 39B, 351–382 Copyright r 2013 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0733-558X/doi:10.1108/S0733-558X(2013)0039A&B024

351

352

KLAUS WEBER ET AL.

integrated theoretical and methodological framework for the analysis of logics across a range of settings. Keywords: Institutional logics; institutional complexity; practice; institution

The concept of logics has seen a resurgence in institutional and cultural analysis, often in order to understand issues of diversity, durability, conflict, and change (Sewell, 2005; Thornton, Ocasio, & Lounsbury, 2012). Much of this resurgence has taken place within organizational institutional theory. Institutional logics describe socially constructed, historical patterns of material practices, assumptions, values, beliefs, and rules by which individuals produce and reproduce their material subsistence, organize time and space, and provide meaning to their social reality (Friedland & Alford, 1991, p. 243; Thornton & Ocasio, 2008). Institutional researchers have identified and examined logics at the level of countries (Biggart & Guille´n, 1999), sectors (Scott, Ruef, Mendel, & Caronna, 2000), and industries (Thornton & Ocasio, 1999). These analyses share several aspects in the understanding of logics: first, logics reside at the field level – they organize and regulate collective action and cannot be understood by looking at single actors (Lounsbury, 2007; Scott & Meyer, 1983). Logics also connect signification/meaning, on the one hand, with social structures and practices on the other, for example, by linking actors’ identities and actions to cultural values (Breiger & Mohr, 2004; Rao, Monin, & Durand, 2003; Sewell, 2005; Weber, Heinze, & DeSoucey, 2008); and logics are carried by social groups that participate in a field of activity (Marquis, Glynn, & Davis, 2007; Scott, 2003). The logics perspective is especially important for institutional theory because it provides an avenue for analyzing historically situated pluralities; or the nature of fields as having, at some point, the capacity to carry more than one paradigm (Greenwood, Raynard, Kodeih, Micoletta, & Lounsbury, 2011; Kraatz & Block, 2008). It should be noted that the analytical approaches involved in analyzing institutional logics closely resemble those of analyzing pluralities in other research streams, such as cultural repertoires (Lamont, 1992; Swidler, 1986), social sensemaking vocabularies (Mills, 1940; Weick, 1995), justification orders (Boltanski & The´venot, 2006 [1991]), or varieties of capitalist systems (Hall & Soskice, 2001; Thelen, 1999). The institutional work on logics is, in fact, part of a more general analytic approach in the social sciences that Sewell (1998, pp. 250–251, 2005)

From Cultural Repertoires to Institutional Logics

353

characterizes as semiotic. In this approach, explanation takes the form of identifying codes or paradigms that prompt human agents to perform certain practices; this in contrast to mechanistic forms of explanation that seek to identify cause and effect. Identifying pertinent codes and paradigms must fundamentally be an empirical task for institutional researchers; as logics are meant to represent lived social realities rather than merely presenting analytic ideal types and models that the researcher identifies a priori for analytic convenience (Schu¨tz, 1967 [1932]; Weber, 1995 [1904]). This is despite Thornton et al.’s (2012, pp. 52–57) praise of ideal types as a tool for theory construction. Ideal types may be formed to distill the essence of a logic, but the construction of the ideal type by the researcher must be grounded in empirical observation, not simply abstract formal reasoning. This is where the ideal type approach differs from formal (mathematical) theory – see, for example, Max Weber’s original formulation of ideal types of domination systems from historical observation. A challenge for advancing the work on logics is thus to find reliable methods for empirically establishing logics and for assessing their distinctiveness and coherence of over time. Much of the extant research on logics in the institutional tradition has, however, treated the existence of logics as the starting point of analysis rather than the end point: logics and their distinctiveness are treated as natural and well-defined. This is evident in the early formulation of Friedland and Alford (1991) who distinguish logics anchored in societal domains of action, such as the family or market. Plurality here coincides with well-established societal sectors and domains. A parallel body of research, inspired by practice theorists in the tradition of Boltanski and Thevenot (2006 [1991]), examines logic-like justification regimes that also originally organized different societal spheres (e.g., civic, industrial, or domestic worlds), but that are now linked to these domains metaphorically rather than indexically. Recent empirical research in organization theory has largely moved from these broad sectoral logics to logics at more intermediate levels, especially organizational fields (Thornton et al., 2012). Because these more ‘‘local’’ logics are not predefined by the macro organization of society into spheres and domains, identifying logics and pluralities is empirically more challenging. The approach commonly employed is to claim and perhaps anecdotally illustrate, coherent and encompassing preexisting logics, and then focus on showing their effect on action or the relationship between different logics. For example, Thornton’s (2004) study of the publishing industry traces a shift from an editorial to a market logic; and Lounsbury (2007) documents how different historically formed logics account for practice variation in the financial service industry. While this approach

354

KLAUS WEBER ET AL.

offers indirect pragmatic proof for the conceptual utility of the logics concept, it provides limited understanding of the logics themselves. Far less research directly studies the internal properties and dynamics of logics and how they are structured over time. A few more recent studies have examined rhetorical processes in the rise of new logics (Green, 2009; Jones, Maoret, Massa, & Svejenova, 2012; Ocasio & Joseph, 2005; Suddaby & Greenwood, 2005); but studies of the temporal dimension of logics, and especially their emergence and change, remain rare. This leaves important questions unanswered: When can we speak of a logic that has the structuring power suggested by institutional theorists? How can we represent the ‘‘content’’ of logics to examine their influence on institutional agents? Are there dimensions of logics that allow for a comparison of their coherence, strength, and scope? The purpose of this paper is to propose a general method for directly studying logics, as semiotic systems, over time. We advocate for an analysis of logics that is grounded in a repertoire view of culture (Swidler, 1986; Weber, 2005). This approach involves identifying the set of cultural categories that can make up logics, and measuring empirically the dimensions that mark a cultural system as more or less logic-like. Discursive data has the advantages of being produced by participants themselves, of being an unobtrusive form of observation, and of offering rich unstructured information about meaning making that is increasingly available in electronic format. Although this is not our focus here, we believe that the procedures described below can be analogously applied to non-textual data. Our approach thus provides an integrated theoretical and methodological framework for the analysis of logics across a range of settings. The paper is structured as follows. We briefly review some of the prior approaches to conceptualizing and analyzing logics, then introduce the setting from which the empirical illustrations are drawn: the field of alternative livestock agriculture. The main part of the paper describes a seven-step analysis process for assessing the emergence of a single logic without presuming its existence. For each step, we identify options and methodological choices and use examples from the study as illustrations.

CONCEPTUAL ISSUES IN THE STUDY OF LOGICS As discussed above, much existing institutional research starts with the presence of alternative logics and examines their interplay. The methodological approach that follows is to identify and track key indicators of the

From Cultural Repertoires to Institutional Logics

355

given logics. The rise and fall of an existing logic can then be documented through event history or count model studies that show the diffusion of archetypical practices or forms (Davis, Diekmann, & Tinsley, 1994; Marquis & Lounsbury, 2007; Thornton & Ocasio, 1999), or through interpretive methods, such as content analysis (Nigam & Ocasio, 2010; Scott et al., 2000). This ‘‘indicator approach’’ to logics starts with the assertion of a logic based on a holistic understanding of a field, and then proceeds to trace the logic over time or in comparison to others. While appropriate in many contexts, a major limitation of this approach is that it lacks a disciplined way for establishing the origin, emergence, and possible transformation of a logic. Thornton et al. (2012, pp. 52–58), for example, advocate for identifying ideal-typical logics but do not provide much guidance for how ideal types should be derived. The indicator approach also detracts from examining the internal structure of logics and from comparing different logics on properties that may explain their institutional power. Instead, all logics are implicitly treated as equivalent when in fact, as cultural systems, they may vary in coherence, complexity, scope, and other dimensions (Sewell, 1992; Swidler, 2001b). Understanding these dimensions may offer useful insights into institutional dynamics. Some recent approaches to conceptualizing and measuring logics have taken a more emergent perspective and looked at the processes through which a new logic is articulated in the first place. In particular, several studies look at how rhetorical processes and strategies, such as theorization, lead to the emergence of new logics (Jones et al., 2012; Nigam & Ocasio, 2010; Suddaby & Greenwood, 2005). Using inductive text analysis, these studies focus on whether words and passages from informative publications (identified as key words or coded into themes) form a pattern that reveals distinct logics. Implicit in these approaches is a view of logics as emerging from cultural processes of social construction (Weber & Dacin, 2011). Our goal is to build on these advances and offer a more general methodological framework for empirical studies of the constitution, evolution, and comparison of logics. Prior to being recognizable as a more or less legitimate alternative, a set of innovative practices and symbolic representations has to attain the internal coherence and value infusion that distinguishes a logic from a less structured set of ideas and activities. The study of logic emergence benefits from an analytic shift from a holistic starting point for conceptualizing logics – in terms of basic principles, archetypes, and root metaphors such as market and family – toward a dynamic and detailed view of logics as empirical relationships between basic cultural categories (cf. Thornton et al., 2012).

356

KLAUS WEBER ET AL.

The emergence of a logic is akin to the structuring of relationships between elements of a community’s cultural register (Barley & Tolbert, 1997): A loose toolkit of meanings and practices becomes increasingly ordered, aligned, expansive, and cohesive, in the understanding of participants. This process of expansion and organization is a distributed, but nevertheless an active, process of construction; at its core, this process involves theorization (Greenwood, Suddaby, & Hinings, 2002; Lawrence, Suddaby, & Leca, 2009; Strang & Meyer, 1993), the self-conscious development and specification of abstract categories and the formulation of patterned relationships such as chains of cause and effect (Strang & Meyer, 1993, p. 492). Theorization aims to rationalize innovative practices by aligning different cultural categories, and in the process may structure the meaning system into emerging logic. If and how this structuring of signification coincides with the structuring of practices and social relations can then be studied empirically, allowing for useful distinctions between integrated institutional logics and related but more partial concepts, such as ideologies, collective myths, epistemic cultures, or theories (Foucault, 1970; Hirsch, 1986; Sahlins, 1978; Sewell, 2005). Analyzing the construction of a logic, therefore, requires identifying the cultural categories that in the aggregate may constitute a logic; and then assessing the degree to which these elements are structured as a systematic logic. In somewhat simplistic terms, logics can be seen as a special case of cultural registers or toolkits that are highly structured and impose rigidities on the selection of cultural resources (Swidler, 1986; Weber, 2005). This view raises the questions of what are the cultural building blocks of logics, and how the degree of structuration is conceptualized. Regarding the former, we follow the tradition of social anthropologists and sociologists who have identified cultural categories – conventionally defined concepts that organize experience through classification – as the elementary building blocks of meaning systems (D’Andrade, 1995; Douglas, 1986; Durkheim & Mauss, 1963 [1903]; Levi-Strauss, 1966 [1962]). Practice theory inspired approaches to institutions point to three core cultural meta-categories that typify social reality (Breiger, 2000; Mohr, 2000; Mohr, Bourgeois, & Duquenne, 2004; Swidler, 2001b): actor identities, classes of social practices, and dimensions of value. Each meta-category corresponds to cultural codes that mark categorical distinctions or taxonomic relationships between the included cultural categories. For example, the work of organization ecologists centers on social codes that define identities (Hannan, Polos, & Caroll, 2007), while cultural psychologists have been concerned with practice codes that give rise to concepts that typify behaviors into categories

From Cultural Repertoires to Institutional Logics

357

such as ‘‘cooperation’’ (Keller & Loewenstein, 2011). While the institutional logics perspective also draws on categories at the level of broad societal domains, such as the family or the market, we are in this paper concerned with the ‘‘smaller’’ concept categories that allow for some degree of bricolage and assembly into more complex formations. Through theorization or recurrent sensemaking, practices, identities, and values are aligned into ready-made assemblies, such as scripts of actor-practice combinations (Weber & Glynn, 2006), strategies of action (Swidler, 2001a) – or, complex systematic wholes (logics). Notably, each of the elementary cultural building blocks can, in turn, be embodied in discourse, interactions, social, and other structures. Fig. 1 offers a schematic representation of a hypothetical transition as a result of increased structuration from a flexible repertoire of cultural categories to a coherent institutional logic. In addition to identifying the building blocks of logics, to analyze their construction, we need attributes that mark a set of identities, practices, and logics as more or less structured. Conceptual work on theorization (Greenwood et al., 2002; Strang & Meyer, 1993) alludes to some dimensions associated with a move toward a logic-like system of cultural categories. Here we suggest that a more systematic conceptualization of the ‘‘logic-ness’’ of a cultural system should include at least four dimensions: (1) the expansiveness of the system; (2) the cohesiveness, or the degree to which elements are seen as interrelated and connected to each other; (3) the extent of internal

Fig. 1.

Schematic Transition from Repertoire to Logic.

358

KLAUS WEBER ET AL.

structure and organization; and (4) the distinctiveness of the system within a larger context. In terms of expansiveness, to accommodate the complexity of most institutional fields, logics need to bring a wide range of practices and identities within the scope of a common framework. Logics also need to densely interrelate different elements to create perceptions of coherence (Greenwood et al., 2002). And the organization of meaning in logics often rests on maintaining central core meta-categories as anchors, such as archetypical practices and identities, and on developing an internal structure of associated categories, for example, in terms of connected subclusters of ideas (Swidler, 2001b). Lastly, to be recognizable, logics, at a minimum, must be meaning systems with distinct content; but logics are also normally explicit representations of boundaries in relation to cultural categories that should be considered not compatible with the logic (see Weber et al., 2008 for an example of oppositional semiotic codes that mark such boundaries). A meaning system that is narrowly focused on a small set of element lacks the power to organize larger complex fields. A meaning system that lacks connectivity between elements lacks the power to provide strong guidance and is less likely to be recognized as a whole.

EMPIRICAL CONTEXT FOR ILLUSTRATIONS: ALTERNATIVE LIVESTOCK AGRICULTURE To illustrate our proposed approach, we use data from the context of alternative livestock agriculture, specifically discourse in trade journals of this field. We collected data on the field of livestock agriculture over a period of 27 years, focusing on the possible emergence of a logic around alternative approaches centered on pasture-based farming, whereby animals are kept on pasture rather than brought to a feedlot. This type of farming involves economic and technological innovations to make consistent, high-quality products for geographically and culturally dispersed consumers; however, pasture-based farming is also a social innovation. Practices are not only designed to change the environmental and health impact of food production, but also aim to reconfigure the agricultural supply chain and the social organization of rural economies away from a corporate industrial model (see Weber et al., 2008). Pasture-based practices have led to a successful and growing grass-fed meat and dairy market. What were once considered inferior products are now served in high-end restaurants, sold at a premium by retailers, and

From Cultural Repertoires to Institutional Logics

359

referenced in prime news outlets. The diffusion of pasture-based over the past 30 years is driven by the institutional work of a diverse alliance of participants who sought to articulate and promote alternatives to the incumbent industrial feedlot system. Through central publications, these actors, including those in agriculture as well as journalists and nutrition experts, helped to connect, legitimate, and disseminate the practices, values, and identities around pasture-based production and consumption. From early on, ideas around pasture-based production were communicated through a central outlet – the Stockman Grass Farmer. The Stockman Grass Farmer is a monthly trade publication ‘‘devoted to the art and science of making a profit from grassland agriculture’’ (website). Since the early 1990s, the Stockman has been the main source of information for farmers and ranchers who are focused on pasture-based animal husbandry (or grass-fed production). The magazine includes detailed profiles of successful farmers and ranchers; articles on scientific discoveries relevant to pasture-based; step-by-step ‘‘how-to’’ pieces on the specifics of production, marketing, and business plans; and editorials. In addition, the Stockman Grass Farmer hosts conferences and workshops and publishes books. All together, the Stockman has helped to create a community of farmers and ranchers; and the publication, in particular, has served as a vehicle for discourse around pasture-based agriculture. Although the relationship between discourse and practice is not our primary concern, the practice of pasture-based agriculture has evolved, too. Early innovations focused on the practicalities of the production process, as farmers and ranchers struggled with learning practices that deviated from existing paradigms in their industry. As new ideas became connected with pasture-based agriculture, such as the nutritional and environmental benefits, these helped to further motivate producers and inform choices about technology, processes of production, and how to connect with consumers. Producers adopted rotational grazing systems, new fencing technology, and methods for growing forage suitable to local climate conditions. They identified breeds of cattle best suited to a grass diet, and selected distribution channels and numerous other practices and technologies. Eventually, issues of marketing became more central, as challenges around production eased. As this narrative suggests, a new logic around pasture-based practices may have emerged over time as new participants joined a movement that sought to articulate an alternative to the logic or industrial agriculture. Yet, the resulting ‘‘logic-ness’’ of this theorization is an open empirical question that we will assess empirically.

360

KLAUS WEBER ET AL.

AN ANALYTIC STRATEGY The cultural repertoire approach described above can be applied to study several aspects of logics, including material practices, individual subjectivities, and social relationships. Here, we focus on the communicative dimension of institutions, that is, discourse (Phillips, Lawrence, & Hardy, 2004). To map the emergent discursive meaning system, we use data from the Stockman, the publication described above, to analyze the structuration of the pasture-based logic. Our sample includes all issues of the Stockman for the years 1981–2007, resulting in a total of 3,692 articles. This is not to suggest that institutions or culture can be reduced to discourse, rather we believe that to analyze institutional processes it is useful to analyze their diverse dimensions separately and treat the relationships between, for example, discourse and material practices as objects of empirical analysis in their own right. The relationship between cultural categories (our focal unit of analysis), specific linguistic labels in manifest discourse, and material practice or subjectivities, can be understood as a semiotic relationship between referent, signifier, and signified (Fig. 2). The relationships between the three semiotic components are based on social convention, not necessity, and are hence subject to empirical analysis (see, e.g., Ruef, 1999). The analytic process we describe and illustrate here involves seven steps: first, the conceptualization of the boundaries of an emerging meaning system; second, the identification of data sources to capture collective meaning; third, the identification of the cultural categories at play; fourth, the creation and validation of measurement instruments; fifth, text analysis

Fig. 2.

Semiotic View of Practice, Discourse, and Culture in Logics.

From Cultural Repertoires to Institutional Logics

361

of the categories found in data sources; sixth, the mapping of aggregate changes in the categories; and finally, dynamic modeling of the relationships among categories over time. We begin the discussion of each step with identifying key issues and decision points, followed by illustration from the field of alternative agriculture.

Step 1: Conceptualizing the Boundaries of the Cultural System Our approach to studying logics takes as its starting point not the existence of a logic, but of a system of meaning whose content and properties can be analyzed to assess the presence and strength of one or several logics. The question is then, what are the boundaries of the system to be analyzed? The challenge for the researcher is to identify appropriate boundaries, since the inclusion and exclusion of discourses and institutional agents directly affect the properties of the system. Yet by the very nature of meaning being studied as emergent rather than presupposed, boundaries are not easily defined. This challenge is complicated by the fact that boundary creation is endogenous to the process of structuration that is central to the rise of logics. One solution to this dilemma is to define the boundaries of a meaning system with reference to its carriers, so that the social boundaries of communities may be used to approximate boundaries of meaning. For example, different cities with stable populations may support different cultural repertoires and institutional logics (Lounsbury, 2007; Marquis et al., 2007). A second solution involves a focus on particular issues and activities across society. Discourses and meaning systems can be structured around focal ‘‘keywords’’ (Williams, 1985), practices (Fine, 1996), or ‘‘problems’’ (Hilgartner & Bosk, 1988), with or without giving rise to institutionalized logics that address these ideas. Another common solution to identifying boundaries is to fall back on the more comprehensive concept of the institutional field or arena (Fligstein & McAdam, 2012; Scott, 2001). At least in more settled institutional settings, the boundaries of fields may be relatively transparent and also important for meaning making. The best basis for identifying the boundaries of the meaning system in question likely depends on the scope of the research question and should be driven by the researcher’s contextual knowledge. In our case of the field of alternative agriculture, we used a combination of carriers, practices, and preexisting fields. We started with agriculture as a well-recognized institutional field, but then focused on the social community

362

KLAUS WEBER ET AL.

of alternative farmers and ranchers within this field, since our question concerned the structuredness of meaning in this subfield. Had our interest been in understanding the potential plurality, comparison or blending of logics in Western agriculture, we would have had to remain at the field level. From our contextual understanding of the agricultural community, we also recognized the centrality of grass-fed and pasture-based production practices involving animals, and the market context of production that extended the cultural system to activities related to production, exchange, and consumption.

Step 2: Identifying Data Sources A second major task is to determine and locate data sources that represent the meaning system. The requirements for these data sources are twofold: first, the data must allow us to draw valid conclusions about the meaning system within its particular context; and second, the data must be collected over a period of time that permits the observation of temporal variation and change (Zaheer, Albert, & Zaheer, 1999). Each of the building blocks of potential logics – cultural categories of identities, practices, and values – can be represented in discourse. The data for analyzing the possible construction and existence of a logic should be representative or encompassing of, or influential for the discourses in question. Especially useful are data that offer real time traces of communications produced by a specific community, such as trade journals, mail or email lists, recorded interviews, and company documents. In our illustration, we used a corpus of 3,692 articles published in a leading trade journal in the field. This journal, the Stockman Grass Farmer, is recognized as the grass-fed and pasture-based farming movement’s central publication. We obtained all issues of the Stockman Grass Farmer for the years 1981–2007. The journal was repeatedly identified during interviews as a key source of information on alternative agriculture. In addition, the long time span of this publication and the diversity of its content make it an excellent source for studying the alignment of the three key dimensions of logics over time. By virtue of its format, editorial policy, and circulation, the Stockman has long provided a forum for many participants in the discourse around alternative livestock agriculture. The journal is subscription-based, and consists of articles written by both practitioners and professionals of alternative agriculture. The intended audience of these writers consists of producers or potential producers. In addition, the articles are diverse in

From Cultural Repertoires to Institutional Logics

363

authorship, purpose, and topical focus. Formats range from accounts of concrete experience (e.g., ranchers providing tutorials or accounts of their particular challenges and practices) to general or abstract knowledge (e.g., scientific research reports, regulatory issues, and general news) and editorials. The topical focus is broad enough to include a range of issues across the agricultural value chain, such as grazing, soil, animal handling, business models, marketing and distribution, human health, and food consumption. In this sense, the observable discourse includes the full spectrum of the market of alternative agriculture, from potential production and production, to distribution and consumption; and discourse is sampled directly from those who participate in the community. An additional component of the determination of a textual data source is a solid understanding of the context of that source. Texts only acquire significance in regard to their meaning and interpretations in the specific contexts of their use (Krippendorff, 2012). A proper understanding of the interaction setting that a researcher is interested in studying helps to inform later decisions during the analytic process. In our study of alternative agriculture, we drew on extensive contextual data to supplement our main text corpus. We consulted third party accounts of the field, conducted 61 semi-structured interviews, and drew on extensive archival sources, such as websites, books on alternative agriculture, and articles in national newspapers. In addition, we obtained 23 hours of audio recordings of presentations and panel discussions at conferences from 1986 to 2004, and podcasts of speeches by prominent journalists from 2005 to 2010. We also constructed biographies of eight individuals that were widely credited in the community with having made major contributions during the time period of our study. These data are described in detail in an earlier article (Weber et al., 2008). As a whole, these data give us the background knowledge to perform an informed text analysis.

Step 3: Identification of the Cultural Register of Practices, Identities, and Values The next step in our methodology involves the identification of the register of cultural categories available. We use the term register to indicate the comprehensiveness of this set of cultural categories across time and participants, and in contrast to the term repertoire, which represents the set of categories available to individual participants or subgroups (see also Weber, 2005). The analysis of logics from a cultural repertoire perspective

364

KLAUS WEBER ET AL.

imposes two challenges for the identification of categories for subsequent text analysis. First, the conceptual unit of analysis is the cultural category, while the text corpus data is composed of linguistic units (words). The cultural categories are latent and must be extracted from the manifest use of words in the text. As discussed in the context of the semiotic representation in Fig. 2, the relationship between conceptual categories – such as ‘‘livestock,’’ as the signified, and manifest words, such as ‘‘Angus,’’ ‘‘dairy cow,’’ or ‘‘cattle,’’ as signifiers – is by social convention, and requires the researcher to establish a reliable correspondence between words and categories over the temporal and spatial scope of the analysis. (The distinction between signified and signifier is similar to the distinction between type and token in linguistics – the analysis is at the level of types, not tokens.) Second, the analysis of discourse is inherently inter-textual. Different documents and the cultural categories employed in them are located in time and space, and the contextual information about this location offers important information about discursive coherence, change, and other parameters of interest. Hence, the register of categories used in the analysis must facilitate the representation and comparison of the entire set of documents that comprise the text corpus. Fig. 3 combines the two dimensions (within text – context, and concept – signifier) to distinguish our emphasis on the analysis of a cultural register from other units used in textual analysis. A register is at the level of cultural concepts and is

Fig. 3.

Focus of Different Forms of Textual Analysis.

From Cultural Repertoires to Institutional Logics

365

constructed not from a single text, but from a larger discourse that puts individual texts or clusters of text in context. In contrast, repertoires pertain to specific texts or producers of text (Weber, 2005); and lexicons and vocabularies are equivalent terms that describe manifest words or labels. To identify the cultural categories that are part of a meaning system’s register, researchers have at least three fundamentally different options. One choice is to take a deductive approach and derive categories from extant research. This approach is similar to the ‘‘indicator approach’’ described above, in that the categories to be measured are not grounded in the text corpus at hand but decided upon a priori. Arguably, this approach is difficult to justify when the meaning system is assumed to be emergent or evolving. A second broad approach is to identify the register of cultural categories from the data in hand. Two methods hold most promise in this regard. One uses standard qualitative-interpretive procedures, such as thematic coding, or more structural coding approaches found in semiotic and rhetoric theory (Feldman, 1995; Manning, 1987; Weber, 2005). The researcher sifts through the data and draws on interpretive insight to understand the categories used by the producers of discourse. This method is most promising when the researcher has access to the broader culture and social position of the text producer. An alternative method of deduction uses computational-algorithmic procedures to identify concept categories. More commonly used in information science and computational linguistics, this family of methods uses statistical analyses of word frequencies in the text corpus, co-occurrence patterns, and universal linguistic rules to extract latent meaning structures from the text corpus. Most suitable for identifying categories are some applications of latent semantic analysis (LSA, Deerwester, Dumais, Furnas, Landauer, & Harshman, 1990; Dumais, 2005), generative topic models (Blei, 2012; Blei, Ng, Jordan, & Lafferty, 2003), and statistical corpus linguistics (Manning, Raghavan, & Schu¨tze, 2008; McEnery & Hardie, 2011). This methodological approach is most useful for very large corpora and when non-textual information yields little additional information. Regardless of the choice of methods, it is important to utilize documents across the entire corpus to create the categories that make up the cultural register. It is also beneficial to use multiple interpreters (researchers) or computational algorithms to increase the robustness of the categories identified. In our illustrative case, we aimed to construct an exhaustive set of categories that could represent the entire time period and had the potential to go beyond the discourse manifest in the text corpus. Our methodology

366

KLAUS WEBER ET AL.

was qualitative-inductive. We not only sampled documents from different years, but we also included other data sources, such as interviews and audio recordings. We used inductive semiotic techniques to identify the conceptual categories corresponding to the meta-categories of practices, identities, and values. For example, the meta-category of practices included the categories of ‘‘heritage breeding’’ (as a type of production) and ‘‘buy-local’’ (as a type of consumption). The process started with two of the authors initially constructing categories individually from samples of text, as well as contextual understandings of the setting. The authors then refined the emerging category scheme over multiple iterations of discussion, checking for logical consistency and cross-references categories with the sample data. While extensive background knowledge of a particular setting is a key component of determining the substance of cultural categories in a cultural system, qualitative semiotic techniques can help to structure this process and increase a researcher’s confidence that the elements selected are discrete. We used semiotic techniques including paradigmatic clustering, commutation tests, and semiotic squares that allowed us to have more certainty regarding the distinctiveness of each of our concepts (Weber, 2005). The technique of paradigmatic clustering suggests that words can be grouped with other words if they act as signifiers of the same concept, while commutation tests help establish the distinctiveness of concepts by testing whether one concept can be substituted for another without changing the meaning in a statement (Barthes, 1967 [1964]; Feldman, 1995). Using these methods, we achieved saturation with a resulting category scheme consisting of 52 exhaustive categories. Table 1 shows the category scheme.

Step 4: Developing a Measurement Instrument The fourth step involves creating and validating measurement instruments for the categories defined in the previous step. Measurement in text analysis establishes the relationship between a latent concept and manifest signifier in a reliable way (see Figs. 2 and 3). Measurement instruments allow the researcher to represent a cultural register through key words and phrases that reflect each category. The computational-algorithmic methods described above use statistical algorithms to select and weigh words that are most informative for discriminating between categories and documents (e.g., the commonly used term frequency–inverse document frequency (tf/idf) statistic). Note that most computational algorithms jointly and simultaneously optimize

Identities

Practices

Meta-Category

Agroindustry

Animal product

Animals

Producer

Consumption – Food preparation

Consumption – Purchase

Exchange – Retail

Production – Processing Exchange – Marketing

Production – Animal care

Production – Animal feeding

Production – Soil management

Production – Animal breeding

Other alternative Conventional Grass fed Cattle/Bison Other livestock Dairy Meat Agroindustry

Heritage breeding Performance breeding Chemical treatment Nonchemical treatment Grazing Feedlot Conventional animal care Natural animal care Processing Internet marketing Mass marketing (conventional) Word of mouth Selling direct Selling through mass retail Buy chain Buy local Buy online Food preparation

Category

Rodale, Steiner, biodynamic Large+farm, industrial+farm Small+farm, grazier Cow, steer, buffalo Horse, sheep, goat Cheese, butter, cream Pork, steak, vennison Kraft, Dole, Dean

Devon, Angus, Galloway Holstein, cross-bred Pesticide, chemical+soil Manure, compost Grazing, pasture, paddock Corn+pen, CAFO Antibiotics, hormone treatment Natural sheep care, Pat Coleby Butcher, slaughter Online+marketing, internet+advertise Agrimarketing, television+advertise Customer referral, word of mouth Direct+sale, sell+on the farm Middleman, sell+supermarket Buy+chain, mass+distribution Farmers+market, buy+local Buy+internet, mail order Recipe, prepare+food

Example of Key Phrases for Coding

Coding Categories for Practices, Identities, and Values.

Category Grouping

Table 1.

From Cultural Repertoires to Institutional Logics 367

Value dimensions

Meta-Category

Taste quality

Financial

Regulation–freedom

Sickness–health

Humane–inhumane

Sustainable–exploitative

Natural–artificial

Authentic–manipulated

Media Government

Advocates

Consumer Expert

Retailer

Category Grouping

Authentic Manipulated Natural Artificial Sustainable Exploitative Humane Inhumane Sickness Health State regulation Individual freedom Financial (good) Financial (bad) Meat taste (high) Meat taste (low)

Media Government

Distributor Mass retailer Restaurant Consumer Academics Consultants Consumer groups Producer groups

Category

Table 1. (Continued )

Heritage, genuine, raw Processed, modified Pure, organic, natural Pollution, contaminated Stewardship, preserve Depletion, erosion, chemical contamination Humane treatment, animal rights Abuse, poor treatment+animal BSE, cholesterol, sickness Nutritious, omega-3 Tax, law, regulation, industry standard Liberties, antigovernment Profit(able), money making, make a living Shortfall, income loss, debt Tender, juicy, taste+flavorful Gamey, dry+meat, taste+cardboard

Supplier, distributor Retail chain, mass+retailer Cafe´, restaurant, eating place Buyer, spender, client Professor, researcher, scholar Extension service, soil consultant Chefs collaborative, Eatwild American Grassfed Association, producer+association Reporter, journalist, press EPA, agent, government

Example of Key Phrases for Coding

368 KLAUS WEBER ET AL.

From Cultural Repertoires to Institutional Logics

369

categories and indicators (although technically, categories could be deduced from a different sub-corpus). The analogy to this approach in (psychometric) measurement theory is the notion of formative measurement, where the manifestations (words) directly constitute the variable (category). The advantage of this approach is that the algorithm makes use of all data and is highly consistent; the challenge is most often the substantive interpretability of the resulting categories (see, e.g., for topic model, Blei, 2012). Category schemes developed through a qualitative-interpretive approach can be made measurable via classic content analysis (Krippendorff, 2012; Neuendorf, 2001). Coding guides for manual coding and coding dictionaries for computer-automated coding provide rules for assessing the presence of different categories in texts. The phrase lists in coding dictionaries can be understood as analogous to items on a reflective scale in measurement theory. Reflective scales require some validation that the items (coding rules or word phrases) reliably reflect the latent variable (cultural category). This can be established via inter-rater reliabilities between different coders, including between human and computerautomated coding of the same text segments (Krippendorff, 2012). An advantage of this form of measurement over formative measurement in computational approaches is that text-specific ambiguities in language can be addressed more directly and that a closer link to theoretical interested can be maintained. The challenge is that the words used to measure cultural categories may be relatively sparse in the text and that proper validation of a measurement scale for a specific text corpus can be time consuming. To measure the categories developed in step 1, we followed a contentanalysis approach. We created a coding dictionary by identifying the key words and phrases that act as signifiers for the categories we established. We used a variety of methods in combination in order to construct this set of signifying key words and phrases. The first method was through a straightforward search of text samples. Words and acronyms that writers commonly used to refer to specific categories were included as signifiers of those categories. Second, we used KWIC (keyword-in-context) searches of text samples. Many text analysis software packages offer the option to examine the textual ‘‘neighborhood’’ of a word, a word pattern, or a specific category that is already partially determined. This allowed us to identify additional words and phrases that signified the concept categories from the previous step. Additionally, we utilized a thesaurus-based approach to search for words that may similarly signify the same category. These steps in combination not only allowed us to construct a set of reliable search phrases

370

KLAUS WEBER ET AL.

for our overarching categories, but also helped us to disambiguate the meaning of more complex words and phrases. Just as in the previous step, a crucial element of this process is to assure that the measurement instrument is reliable. The reliability of these categories as a measurement instrument was tested in two rounds of coding, by computing inter-coder reliabilities using Krippendorff’s alpha (Krippendorff, 2012). In the first round, two human coders manually coded approximately 1,000 sentences for the presence of each category. The rules and a list of indicator phrases were then refined for any categories with a Krippendorff alpha of o0.7. A third human coder then manually coded 600 sentences for each category and we calculated reliability statistics between the human coder and a computer-automated coding of the same sentences. The average resulting Krippendorff alpha across all categories was 0.78.

Step 5: Textual Analysis Once a measurement instrument is established, the coding can be applied to the full text corpus. Applying the category scheme to the documents produces quantitative measures that form the basis of all subsequent analyses. Two types of text analysis – computational and content – are straightforward and valuable methods for measuring (a) the presence and prevalence of cultural categories and (b) their relationships to each other. We first discuss options for measuring the presence and prevalence of a category, normally represented as a raw or normalized frequency count. Manual coding of a text corpus is time consuming and does not necessarily produce higher reliabilities than computer-automated approaches. However, manual coding has advantages that under some circumstances may justify the cost. Manual coding is especially valuable when concepts and relationships are implicit or expressed in indirect ways in the analyzed documents. This may be the case with certain emotional states, or in text corpuses that make extensive use of humor, metaphorical, or highly idiosyncratic language that is targeted at a unique and narrow audience. Because human coders can take context and tone into account more easily, they tend to perform better than more automated approaches. When manual coding is unavoidable for very large corpuses, sampling text units such as sentences, paragraphs, or documents, can yield satisfactory results while reducing manpower requirements (Krippendorff, 2012). Computer-automated content coding is easily implemented for concepts that are explicitly expressed in language, such as most social identity

From Cultural Repertoires to Institutional Logics

371

categories; and supporting software packages are readily available and relatively easy to use. Computer-automated coding offers substantial time savings over manual coding in the large corpuses that are typical in institutional analysis. A common limitation is that words and phrase expressions often can signify different cultural categories depending on the context in which they are used. A rule-based disambiguation of the intended meaning can be difficult, especially in corpuses that contain heterogeneous genres and styles and hence limit the ability to specify general rules. Fully computational approaches again use algorithms and statistical principles to measure the presence of categories across documents. Advanced computational approaches offer a number of advantages over conventional content analysis. They routinely correct for the base rate expectancy with which a phrase occurs across corpuses, and hence prevent interpretations of differences in frequency counts as informative, when they are, in fact, generated by generic language use. Some computational approaches also use machine learning routines so that coding algorithms can be ‘‘trained’’ to code more complex constellations of phrases in unambiguous ways; others make use of grammatical information to achieve the same task (Corman, Kuhn, McPhee, & Dooley, 2002; Manning & Schutze, 1999; Manning et al., 2008). While these techniques can, in principle, be applied to the traditional content-analysis approach (e.g., taking into account different base rates of categories), they are not commonly implemented in content-analysis software. Because fully algorithmic approaches are mostly ‘‘blind’’ to theoretical considerations that may privilege categories that are rare or ambiguous, they are powerful in purely inductive designs, but of more limited use when the coding involves many such categories. In our analysis of discourse on alternative agriculture, we used automated content analysis to apply the coding dictionary described above to the full text corpus. Automated text analysis was suitable because the text corpus was large and relatively homogeneous, since it was generated within a single setting. We used WordStat 6.1, a software suite that supports standard content analysis as well as relational analyses through word co-occurrences. Using this software, we coded for the binary presence of each identity, practice, and value category at the sentence level – the most natural linguistic unit of text (Krippendorff, 2012). The second type of analysis involves the measurement of observed relationships between categories. One of the key dimensions of a logic is the coherence of categories, and one way to establish this coherence is to look at the relationship between concepts as represented in discourse. Most text

372

KLAUS WEBER ET AL.

analysis approaches use the proximity of two concepts in a text to infer relationships. The most basic, and commonly used, form of proximity is cooccurrence: Two cultural categories that co-occur within a certain number of words, or in the same sentence, paragraph, or document, are related to each other. Note that the statistical probability of two categories cooccurring at random in a text unit should be taken into account in interpreting co-occurrences. This base rate co-occurrence of categories is, for example, influenced by the total frequency of each category in the corpus. Using co-occurrences as a measure of relationships is based on a notion of meaning systems as a web of associations between heterogeneous categories. The idea is that if an author of a text uses two concepts in close proximity, these ideas are seen as related in the writer’s mind. By contrast, if two concepts are used in very different parts of the same document, or in different documents, this implies that the author compartmentalizes these concepts; or, if the documents are produced by different authors, the concepts are not integrated in individuals’ minds. Advances in information retrieval-based approaches to text analyses and the increase in computing power have greatly expanded the family of algorithms available to establish similarity between two text units. More complex relationships, beyond dyadic co-occurrence, can be used to measure proximity. For example, vector space models represent two text units as high-dimensional vectors of many categories (at the level of words or cultural categories); thus, the similarity metric of two pieces of text can take into account the presence and absence of many categories at the same time. A widely used metric is the cosine similarity measure, developed for document classification and indexing, but applicable more generally (Blei et al., 2003; Dumais, 2005; Salton, Wong, & Yang, 1975). Currently, these approaches are limited to computational text analysis, but conceptually, they can easily be extended to and combined with traditional contentanalytic approaches, to establish associations between categories beyond the use of dyadic co-occurrences. A more general downside of utilizing an associational approach to meaning is that this approach subsumes a variety of types of relationships. Two concepts that are associated with one another could be described by the author as in opposition to one other, as complementary to one another, or perhaps even as causally linked to one another. Thus, an alternative method of conceptualizing meaning is to analyze specific relationships directly. One prominent example of this approach is causal map analysis (Axelrod, 1976; Huff, 1990; Nadkarni & Narayanan, 2007). Though not easy to capture with algorithmic and automated approaches, casual relationships can be

From Cultural Repertoires to Institutional Logics

373

measured through the content coding of specific causal verbs and phrases (e.g., ‘‘causes,’’ ‘‘is due to,’’ ‘‘leads to,’’ etc.), followed by coding for cultural categories that occur directly before and after the causal term. This form of relational analysis corresponds closely to the notion in institutional theory that theorization involves the establishment of relationships of cause and effect between cultural categories (Strang & Meyer, 1993). In our analysis of alternative agriculture, we operationalized the relationships between concepts in the form of co-occurrences of two categories within a sentence. We used Jaccard coefficients (the binary presence of two categories) rather than more sophisticated metrics, since the sentence-level coding of cultural categories resulted in a low dimensional and sparse vector of category presences. We found computer-automated coding of causal relationships difficult to perform in a comprehensive way.

Step 6: Mapping Aggregate Properties and Changes in Meaning Systems Although quantitative text analysis offers a way to count basic category occurrences and co-occurrences, it only delivers the raw material for constructing variables that corresponds to the conceptual dimensions of logics identified earlier. The next step of our analysis procedure consists of mapping out descriptive properties of a meaning system to evaluate the extent to which the system amounts to a logic. A preliminary choice in this analysis is the extent to which data is aggregated, for example, from the sentence level to documents, authors, or time periods. Aggregation decisions are consequential, especially for establishing associative networks between cultural categories. The more aggregate the data, the more likely associations can be observed; but the extent of aggregation of discourse data should not exceed the level of aggregation from which actual associations can be inferred. For example, aggregating documents to five-year periods will result in a greater number of concept co-occurrences, but it may be unwarranted to infer strong associations in the meaning system from relationships observed years apart. While in some instances, natural aggregation units exist, such as quarterly and annual reporting periods in corporate disclosure, the extent of aggregation is driven mostly by theoretical considerations. The types of questions to consider are: Does the research seek to understand differences between producers of texts, or identify multiple epistemic communities or logics within a discourse? Is the interest in short-term variability or longterm trends?

374

KLAUS WEBER ET AL.

From these data, we can then calculate descriptive statistics that correspond to the key dimensions of meaning systems outlined at the beginning of the paper: (1) the expansiveness of the system; (2) its cohesiveness, in terms of interrelations of categories; (3) the extent of internal structure and organization; and (4) the distinctiveness compared to other meaning systems. A straightforward variable of expansiveness is simply the number of categories that are in active use at a point in time. This metric is enabled by the repertoire approach to identifying the cultural register as the maximum set of categories that could be in use (see step 3). Expansiveness can then be used comparatively (how broad in scope a meaning system is) or temporally (whether it becomes more or less encompassing over time). Cohesiveness parameters can be constructed from the coding of relationships between concepts, then applying standard network analysis procedures to network graphs in which the nodes are cultural categories and the ties are observed relationships. Associational ties result in undirected graphs while, for example, causal relationships result in directed graphs. Standard whole network parameters for cohesion, such as network cohesion, transitivity or path length-based measures can then be computed and compared (Borgatti, Everett, & Freeman, 2002; Wasserman & Faust, 1994). The degree and type of internal structuring of a meaning system can take different forms, such as the centralization of associations around a few pivotal categories (Corman et al., 2002), or dense clusters of concepts that become ready-made subassemblies, acting as institutionalized shorthands for more complex argument structures (Green, 2009). The calculation of parameters can again draw on standard network metrics like centralization or clustering coefficients. Lastly, from a cultural repertoire perspective, the distinctiveness of a meaning system can be represented simply by the extent to which the same cultural categories are used (Swidler, 2002). In addition, distinctiveness can also arise from the relative prominence (frequency) of categories, and from different patterns of relationships between the same cultural categories. A range of similarity metrics can easily be calculated from the coding of categories and relationships, including the above-mentioned Jaccard, Mahalanobis-family, and cosine similarity metrics (see Weber, 2005 and Manning et al., 2008 for overviews). As discussed above, topic modeling (Blei, 2012) allows for a comprehensive approach to mapping differences and changes over time based on vector space models of texts, and analogous mappings can be produced from content-analysis data if coding categories are substituted for topics. For our illustrative example, we aggregated category frequencies and cooccurrence matrices to the year level, and computed some simple graphical representations and selective metrics that correspond to the dimensions

375

From Cultural Repertoires to Institutional Logics

identified above. Figs. 4 and 5 show a comparative snapshot of the years 1982 and 2005, and two illustrations of graphically mapping changes in expansiveness (proportion of register categories used) and cohesion (network cohesion index). -Concept Network, 1982 -

46 186 0.52 1.87 0.48 16.88 2.32 26.78

# of Nodes # of ties Density Average Distance Cohesion Centralization (EV) Clustering Transitivity (% of triplets)

Fig. 4.

Fig. 5.

-Concept Network, 2005 -

47 349 1.32 1.71 0.63 14.17 6.147 29.85

Concept Networks, 1982 versus 2005.

Changes in Network Parameters, 1980–2007.

376

KLAUS WEBER ET AL.

For example, the graph of repertoire expansiveness over time tells us that additional categories were used by authors in later time periods; and a more detailed analysis showed that the additional categories primarily related to customers and third parties, while the categories related to agricultural production were in use throughout the time period. Another potential descriptive mapping can come from plotting the density of category associations over time. While there are numerous ways that theorists can measure cohesion, such as transitivity or path length-based measures, for our illustration, we used one simple measure of cohesion: the density of the co-occurrence network, based on the proportion of hypothetically possible ties to those ties actually present in the data (Wasserman & Faust, 1994). The graph in Fig. 5 suggests that the cohesiveness of concept associations in discourse on alternative agriculture fluctuated over time; this provides evidence counter to the linear process of increasing cohesion that one might expect in the gradual emergence of a new logic.

Step 7: Dynamic Statistical Models The final step in analyzing the possible emergence of logics from a cultural repertoire perspective looks to uncover the mechanisms of evolution in the concept webs over time. This analysis is, of course, premised on the researcher being able to observe in the previous step a descriptive pattern that suggests an increasing logic-ness of discourse. While the previous steps produced an overview of changes over time, they could not speak to the mechanisms of change. Longitudinal statistical analyses, using the parameters and network matrices from the prior analysis, allow for stronger tests of what drives changes in meaning systems, and hence the emergence or transformation of logics. Conventional statistical analysis offers a path to examine the effect of non-textual factors, such as events, social mobilization, and power struggles, on the meaning system in a field. Another option is the use of a dynamic network to test the effect of endogenous dynamics on discrete network transitions over time and, hence, changes in the overall structure of the associational network represented by the discourse. Stochastic dynamic network models have, for example, been implemented in the SIENA (Simulation Investigation for Empirical Network Analysis) software modules (Snijders, Steglich, Schweinberger, & Huisman, 2009) and can be applied to concept networks. The models specified and tested in this approach can combine network-statistical drivers of evolution as well as covariates of substantive interest. While the application of these models to

From Cultural Repertoires to Institutional Logics

377

cultural networks is in its infancy, it offers exciting future opportunities for institutional researchers.

CONCLUSION In this paper, we sought to expand the conceptual and methodological toolkit for the growing research on institutional logics and change. Despite important theoretical advances, studying empirically the emergence and internal dynamics of logics remains challenging. To move toward a better understanding of how logics are constructed and transformed we propose a bottom-up perspective on institutional logics that is grounded in a repertoire perspective of culture and sees logics as one end of a continuum in the structuration of a meaning system. From this perspective, we derived concepts, dimensions of logics, and an empirical method that allow researchers to empirically assess the ‘‘logic-ness’’ of a meaning system and analyze the micro-mechanisms of change in a given logic. Our focus on quantitative text analysis expands the methodological toolkit of institutional analysis, which has been dominated by qualitative-interpretive and indicator-based methods for understanding logics. We illustrated some of these methodological tools in the empirical example of discourse in alternative agriculture. What conclusions does our proposed approach permit, and how do they compare to those of an indicator approach to studying institutional logics? Putting side by side the changes in the four parameters we tracked – extensiveness, cohesion, organization, and distinctiveness – a more nuanced picture of the changing meaning system emerges. If an institutional logic exists when all four parameters show a rising trend, then the conclusion is that over the time period we studied, we did not see the emergence of a complete logic. In fact, while some parameters show the expected pattern (e.g., a gradual rise in expansiveness), others do not (e.g., fluctuating levels of discursive cohesion). In this sense, our approach toward measuring the emergence of a logic proves advantageous in that it allows for greater precision and moves away from linear and binary conclusion of the form ‘‘a new logic did/did not emerge.’’ Our method allows us the flexibility of observing various processes that may suggest the congealing of a logic but also a range of other changes in a meaning system, such as cycles of topics or partial structuration. The relative merit of this approach compared to indicator-based approaches and the use of analytic ideal types can perhaps best be

378

KLAUS WEBER ET AL.

understood in terms of the inherent tradeoffs between types of explanations first suggested by Thorngate (1976): An explanation cannot at the same time be simple, general, and accurate. Arguably, existing approaches, such as the construction of ideal types and indicator-based approaches to logics aim for simple and general modes of understanding, while the approach described here privileges accuracy. Our hope is thus that the expansion of the empirical and conceptual toolkit we proposed will facilitate theoretical developments in our understanding of logics in action.

REFERENCES Axelrod, R. (1976). Structure of decision. Princeton, NJ: Princeton University Press. Barley, S. R., & Tolbert, P. S. (1997). Institutionalization and structuration: Studying the links between action and institution. Organization Studies, 18, 93–117. Barthes, R. (1967 [1964]). Elements of semiology (A. Lavers & C. Smith, Trans.). London: Jonathan Cape. Biggart, N. W., & Guille´n, M. F. (1999). Developing difference: Social organization and the rise of the auto industries of South Korea, Taiwan, Spain, and Argentina. American Sociological Review, 64, 722–747. Blei, D. M. (2012). Introduction to probabilistic topic models. Communications of the ACM, 55, 77–84. Blei, D. M, Ng, A. Y., Jordan, M. I., & Lafferty, J. (2003). Latent Dirichlet allocation. Journal of Machine Learning Research, 3, 993–1022. Boltanski, L., & The´venot, L. (2006 [1991]). On justification (C. Porter, Trans.). Princeton, NJ: Princeton University Press. Borgatti, S. P., Everett, M. G., & Freeman, L. C. (2002). Ucinet 6 for Windows: Software for social network analysis. Harvard: Analytic Technologies. Breiger, R. L. (2000). A tool kit for practice theory. Poetics, 27, 91–115. Breiger, R. L., & Mohr, J. W. (2004). Institutional logics from the aggregation of organizational networks: Operational procedures for the analysis of counted data. Computational & Mathematical Organization Theory, 10, 17–43. Corman, S. R., Kuhn, T., McPhee, R. D., & Dooley, K. J. (2002). Studying complex discursive systems: Centering resonance analysis of communication. Human Communication Research, 28, 157–206. D’Andrade, R. (1995). The development of cognitive anthropology. New York, NY: Cambridge University Press. Davis, G. F., Diekmann, K. A., & Tinsley, C. H. (1994). The decline and fall of the conglomerate firm in the 1980s: A study in the de-institutionalization of an organizational form. American Sociological Review, 59, 547–570. Deerwester, S., Dumais, S. T., Furnas, G. W., Landauer, T. K., & Harshman, R. (1990). Indexing by latent semantic analysis. Journal of the American Society of Information Science, 41, 391–407. Douglas, M. (1986). How institutions think. Syracuse, NY: Syracuse University Press.

From Cultural Repertoires to Institutional Logics

379

Dumais, S. T. (2005). Latent semantic analysis. Annual Review of Information Science and Technology, 38, 188–230. Durkheim, E., & Mauss, M. (1963 [1903]). Primitive classification (R. Needham, Trans.). Chicago, IL: University of Chicago Press. Feldman, M. S. (1995). Strategies for interpreting qualitative data (Vol. 33). Thousand Oaks, CA: Sage. Fine, G. A. (1996). Justifying work: Occupational rhetorics as resources in restaurant kitchens. Administrative Science Quarterly, 41, 90–115. Fligstein, N., & McAdam, D. (2012). A theory of fields. New York, NY: Oxford University Press. Foucault, M. (1970). The order of things. London: Tavistock. Friedland, R., & Alford, R. R. (1991). Bringing society back in: Symbols, practices, and institutional contradictions. In P. J. DiMaggio & W. W. Powell (Eds.), The new institutionalism in organizational analysis (pp. 232–263). Chicago, IL: University of Chicago Press. Green, S. E. (2009). Suspended in self-spun webs of significance: A rhetorical model of institutionalization and institutionally embedded agency. Academy of Management Journal, 52, 11–36. Greenwood, R., Raynard, M., Kodeih, F., Micoletta, E. R., & Lounsbury, M. (2011). Institutional complexity and organizational responses. Academy of Management Annals, 5, 317–371. Greenwood, R., Suddaby, R., & Hinings, C. R. (2002). Theorizing change: The role of professional associations in the transformation of institutionalized fields. Academy of Management Journal, 45, 58–80. Hall, P., & Soskice, D. (2001). Varieties of capitalism: The institutional foundations of comparative advantage. New York, NY: Oxford University Press. Hannan, M. T., Polos, L., & Caroll, G. R. (2007). The logics of organizational theory: Social codes and ecologies. Princeton, NJ: Princeton University Press. Hilgartner, S., & Bosk, C. L. (1988). The rise and fall of social problems: A public arenas model. American Journal of Sociology, 94, 53–78. Hirsch, P. M. (1986). From ambushes to golden parachutes: Corporate takeovers as an instance of cultural framing and institutional integration. American Journal of Sociology, 91, 800– 837. Huff, A. S. (1990). Mapping strategic thought. In A. S. Huff (Ed.), Mapping strategic thought (pp. 11–49). New York, NY: Wiley. Jones, C., Maoret, M., Massa, F. G., & Svejenova, S. (2012). Rebels with a cause: Formation, contestation, and expansion of the De Novo category ‘‘Modern Architecture,’’ 1870–1975. Organization Science, 23(6), 1523–1545. Keller, J., & Loewenstein, J. (2011). The category of cooperation: A cultural consensus model analysis for China and the US. Organization Science, 22, 299–319. Kraatz, M. S., & Block, E. (2008). Organizational implications of institutional pluralism. In R. Greenwood, C. Oliver, R. Suddaby & K. Sahlin-Andersson (Eds.), Handbook of organizational institutionalism (pp. 243–275). Thousand Oaks, CA: Sage. Krippendorff, K. (2012). Content analysis: An introduction to its methodology. Beverly Hills, CA: Sage. Lamont, M. (1992). Money, morals, and money. Chicago, IL: University of Chicago Press. Levi-Strauss, C. (1966 [1962]). The savage mind. Chicago, IL: University of Chicago Press.

380

KLAUS WEBER ET AL.

Lawrence, T. B., Suddaby, R., & Leca, B. (2009). Institutional work: Actors and agency in institutional studies of organization. Cambridge, UK: Cambridge University Press. Lounsbury, M. (2007). A tale of two cities: Competing logics and practice variation in the professionalizing of mutual funds. Academy of Management Journal, 50, 289–307. Manning, C., & Schutze, H. (1999). Foundations of statistical natural language processing. Cambridge, MA: MIT Press. Manning, C. D., Raghavan, P., & Schu¨tze, H. (2008). Introduction to information retrieval. Cambridge, UK: Cambridge University Press. Manning, P. K. (1987). Semiotics and fieldwork. Newbury Park, CA: Sage. Marquis, C., Glynn, M. A., & Davis, G. F. (2007). Community isomorphism and corporate social action. Academy of Management Review, 32, 925–945. Marquis, C., & Lounsbury, M. (2007). Vive la re´sistance: Competing logics and the consolidation of US community banking. Academy of Management Journal, 50, 799–820. McEnery, T., & Hardie, A. (2011). Corpus linguistics: Method, theory and practice. Cambridge, UK: Cambridge University Press. Mills, C. W. (1940). Situated actions and vocabularies of motive. American Sociological Review, 5, 904–913. Mohr, J. W. (2000). Introduction: Structures, institutions, and cultural analysis. Poetics, 27, 57–68. Mohr, J. W., Bourgeois, M., & Duquenne, V. (2004). The logic of opportunity: A formal analysis of the University of California’s outreach and diversity discourse. CSHE Research & Occasional Paper: CHSE.9.04, University of California, Berkeley, CA. Nadkarni, S., & Narayanan, V. K. (2007). Evolution of collective strategy frames in high and low velocity industries. Organization Science, 18, 688–710. Neuendorf, K. A. (2001). The content analysis guidebook. Beverly Hills, CA: Sage. Nigam, A., & Ocasio, W. (2010). Event attention, environmental sensemaking, and change in institutional logics: An inductive analysis of the effects of public attention to Clinton’s health care reform initiative. Organization Science, 21, 823–841. Ocasio, W., & Joseph, J. (2005). Cultural adaptation and institutional change: The evolution of vocabularies of corporate governance, 1972–2003. Poetics, 33, 163–178. Phillips, N., Lawrence, T. B., & Hardy, C. (2004). Discourse and institutions. Academy of Management Journal, 29, 635–652. Rao, H., Monin, P., & Durand, R. (2003). Institutional change in Toque Ville: Nouvelle cuisine as an identity movement in French gastronomy. American Journal of Sociology, 108, 795–843. Ruef, M. (1999). Social ontology and the dynamics of organizational forms: Creating market actors in the healthcare field, 1966–1994. Social Forces, 77, 1403–1432. Sahlins, M. D. (1978). Culture and practical reason. Chicago, IL: University of Chicago Press. Salton, G., Wong, A., & Yang, C.-S. (1975). A vector space model for automatic indexing. Communications of the ACM, 18, 613–620. Schu¨tz, A. (1967 [1932]). The phenomenology of the social world (G. Walsh & F. Lehnert, Trans.). Evanston, IL: Northwestern University Press. Scott, W. R. (2001). Institutions and organizations. Thousand Oaks, CA: Sage. Scott, W. R. (2003). Institutional carriers: Reviewing modes of transporting ideas over time and space and considering their consequences. Industrial and Corporate Change, 12, 879–894. Scott, W. R., & Meyer, J. W. (1983). The organization of societal actors. In J. W. Meyer & W. R. Scott (Eds.), Organizational environments: Ritual and rationality (pp. 129–153). Beverly Hills, CA: Sage.

From Cultural Repertoires to Institutional Logics

381

Scott, W. R., Ruef, M., Mendel, P., & Caronna, C. (2000). Institutional change and healthcare organizations: From professional dominance to managed care. Chicago, IL: University of Chicago Press. Sewell, W. H. (1992). A theory of structure: Duality, agency and transformation. American Journal of Sociology, 98, 1–29. Sewell, W. H. (1998). Language and practice in cultural history: Backing away from the edge of the cliff. French Historical Studies, 21, 241–254. Sewell, W. H. (2005). Logics of history: Social theory and social transformation. Chicago, IL: University of Chicago Press. Snijders, T. A. B., Steglich, C. E. G., Schweinberger, M., & Huisman, M. (2009). Manual for SIENA version 3.2. Groningen, NL: ICS. Strang, D., & Meyer, J. W. (1993). Institutional conditions for diffusion. Theory and Society, 22, 487–511. Suddaby, R., & Greenwood, R. (2005). Rhetorical strategies of legitimacy. Administrative Science Quarterly, 50, 35–67. Swidler, A. (1986). Culture in action: Symbols and strategies. American Sociological Review, 51, 273–286. Swidler, A. (2001a). Talk of love: How culture matters. Chicago, IL: University of Chicago Press. Swidler, A. (2001b). What anchors cultural practices. In T. R. Schatzki, K. Knorr-Cetina & E. Von Savigny (Eds.), The practice turn in contemporary theory (pp. 74–92). New York, NY: Routledge. Swidler, A. (2002). Cultural repertoires and cultural logics: Can they be reconciled? Comparative and Historical Sociology, 14, 1–6. Thelen, K. (1999). Historical institutionalism in comparative politics. Annual Review of Political Science, 2, 369–404. Thorngate, W. (1976). Possible limits on a science of social behavior. In L. H. Strickland, F. E. Aboud & K. J. Gergen (Eds.), Social psychology in transition (pp. 121–139). New York, NY: Plenum Press. Thornton, P. H. (2004). Markets from culture: Institutional logics and organizational decisions in higher education publishing. Stanford, CA: Stanford University Press. Thornton, P. H., & Ocasio, W. (1999). Institutional logics and the historical contingency of power in organizations: Executive succession in the higher education publishing industry, 1958–1990. American Journal of Sociology, 105, 801–843. Thornton, P. H., & Ocasio, W. (2008). Institutional logics. In R. Greenwood, C. Oliver, K. Sahlin & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 99–129). Thousand Oaks, CA: Sage. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure and process. New York, NY: Oxford University Press. Wasserman, S., & Faust, K. (1994). Social network analysis. New York, NY: Cambridge University Press. Weber, K. (2005). A toolkit for analyzing corporate cultural toolkits. Poetics, 33, 227–252. Weber, K., & Dacin, M. T. (2011). The cultural construction of organizational life. Organization Science, 22, 286–298. Weber, K., & Glynn, M. A. (2006). Making sense with institutions: Thought and action in Karl Weick’s theory. Organization Studies, 27, 1639–1660.

382

KLAUS WEBER ET AL.

Weber, K., Heinze, K., & DeSoucey, M. (2008). Forage for thought: Mobilizing codes in the movement for grass-fed meat and dairy products. Administrative Science Quarterly, 53, 529–567. Weber, M. (1995 [1904]). Die Objektivita¨t sozialwissenschaftlicher und sozialpolitischer Erkenntnis. Schutterwald: Wissenschaftlicher Verlag. Weick, K. E. (1995). Sensemaking in organizations. Thousand Oaks, CA: Sage. Williams, R. (1985). Keywords: A vocabulary of culture and society. New York, NY: Oxford University Press. Zaheer, S., Albert, S., & Zaheer, A. (1999). Time scales and organizational theory. Academy of Management Review, 24, 725–741.