Entrepreneurial Intention: Past, Present, and Future Research 2019951885, 9781788975230, 9781788975223

Citation preview

Entrepreneurial Intention

NEW HORIZONS IN ENTREPRENEURSHIP Books in the New Horizons in Entrepreneurship series make a significant contribution to the development of Entrepreneurship Studies. As this field has expanded dramatically in recent years, the series will provide an invaluable forum for the publication of high-quality works of scholarship and show the diversity of issues and practices around the world. Global in its approach, it includes some of the best theoretical and empirical work with contributions to fundamental principles, rigorous evaluations of existing concepts and competing theories, historical surveys and future visions. Titles in the series include: The Role and Impact of Entrepreneurship Education Methods, Teachers and Innovative Programmes Edited by Alain Fayolle, Dafna Kariv and Harry Matlay Entrepreneurial Intention Past, Present, and Future Research Evan J. Douglas

Entrepreneurial Intention

Past, Present, and Future Research

Evan J. Douglas Adjunct Professor of Entrepreneurship, Australian Centre for Entrepreneurship Research, Queensland University of Technology, Brisbane, Australia

NEW HORIZONS IN ENTREPRENEURSHIP

Cheltenham, UK • Northampton, MA, USA

© Evan J. Douglas 2020

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA

A catalogue record for this book is available from the British Library Library of Congress Control Number: 2019951885 This book is available electronically in the Business subject collection DOI 10.4337/9781788975230

02

ISBN 978 1 78897 522 3 (cased) ISBN 978 1 78897 523 0 (eBook)

Contents List of figuresxiii List of tablesxiv Forewordxvi Norris Krueger Prefacexix 1 Introduction

1

2

Taxonomies of entrepreneurial behaviour

18

3

Why do individuals want to be entrepreneurs?

34

4

Issues for entrepreneurial intention research

54

5

Commercial entrepreneurial intention

69

6

Intrapreneurial intention

86

7

Social entrepreneurial intention

104

8

An inclusive model of entrepreneurial intention

122

9

The holistic approach to entrepreneurship intentions

147

10

The future of entrepreneurial intention research

169

Bibliography188 Index  208

v

Extended contents List of figuresxiii List of tablesxiv Forewordxvi Norris Krueger Prefacexix 1 Introduction 1 1.1 Introduction1 1.2 What do we mean by entrepreneurship? 2 1.2.1 Proactivity 2 1.2.2 Risk taking 3 1.2.3 Innovation 4 New products 5 1.2.4 New wealth 8 1.3 Commercial versus social entrepreneurship 8 1.3.1 Commercial entrepreneurship 8 1.3.2 Social entrepreneurship 9 1.4 The entrepreneurial mindset 10 10 1.4.1 Entrepreneurial attitudes 1.4.2 Entrepreneurial abilities 12 1.5 Entrepreneurship as a planned behaviour 14 1.5.1 The perceived desirability of entrepreneurship 15 1.5.2 The perceived feasibility of entrepreneurial behaviour 16 1.6 Summary16 2

Taxonomies of entrepreneurial behaviour 18 2.1 Introduction18 2.1.1 Necessity versus opportunity entrepreneurs 18 2.1.2 Individual versus organisational entrepreneurship 20 2.2 Profit-maximising entrepreneurs 20 vii

viii

Entrepreneurial intention

2.2.1 Growth entrepreneurs 21 2.2.2 Gazelles and unicorns 22 2.3 Profit-satisficing entrepreneurs 23 2.3.1 Independence-oriented entrepreneurs 23 2.3.2 Lifestyle entrepreneurs 24 2.3.3 Salary-substitute entrepreneurs 24 2.3.4 Leisure-seeking entrepreneurs 25 2.4 Social-purpose entrepreneurs 25 2.4.1 Not-for-profit enterprises 26 2.5 Hybrid entrepreneurs 26 2.6 Intrapreneurship27 2.7 Spinoffs and spinouts 28 2.8 Corporate entrepreneurship 28 2.9 Past, present, and future research into types of entrepreneurship 29 2.10 Summary31 3

Why do individuals want to be entrepreneurs? 34 3.1 Introduction34 3.1.1 Utility theory of entrepreneurial behaviour 35 3.2 The perceived desirability of entrepreneurship 37 3.2.1 Need for autonomy 37 3.2.2 Need for competence 38 3.2.3 Need for relatedness 39 3.2.4 Situational needs (for income, risk avoidance, 39 and effort avoidance) 3.2.5 Expectancy-valence theory 40 3.2.6 The non-necessity of anything for entrepreneurial intention41 3.3 The perceived feasibility of entrepreneurship 42 3.3.1 Entrepreneurial tasks and required abilities 43 Opportunity recognition 43 Viability screening 44 Problem solving 46 Management skills 46 3.3.2 Entrepreneurial self-efficacy 47 Self-regulation theory 47

Extended contents

ix

Overconfidence and optimism 49 Solo versus team entrepreneurial intention 51 3.4 Summary52 4

Issues for entrepreneurial intention research 54 4.1 Introduction54 4.2 Construct issues 54 4.2.1 Broad versus narrow entrepreneurial intention constructs54 4.2.2 Latent variable versus indicator variable 56 4.3 Measurement issues 57 4.3.1 Construct clarity 57 4.3.2 Common-method bias 57 4.3.3 Espoused versus revealed data 57 4.3.4 Sex versus gender 58 4.4 Method issues 61 4.4.1 Non-representative samples 61 4.4.2 Cross-sectional versus longitudinal studies 63 4.4.3 Symmetrical correlational analysis 63 4.4.4 Asymmetrical qualitative comparative analysis 66 4.5 Summary67

5

Commercial entrepreneurial intention 69 5.1 Introduction69 5.2 Commercial entrepreneurial intention studies 70 71 5.2.1 Antecedents of commercial motivation 5.3 Profit-maximising versus profit-satisficing entrepreneurial intention73 5.3.1 Profit maximisation 74 5.3.2 Profit satisficing 74 5.4 Growth-oriented versus independence-oriented entrepreneurial intentions76 5.4.1 The hypotheses 77 5.4.2 The empirical study 79 5.4.3 The results 82 5.4.4 Subtypes of profit-satisficing commercial entrepreneurs 82 5.5 Summary84

x

Entrepreneurial intention

6

Intrapreneurial intention 86 6.1 Introduction86 6.2 Development of hypotheses regarding intrapreneurial intention88 6.2.1 Is intrapreneurial intention a distinct construct? 90 6.2.2 Attitudes to the outcomes of intrapreneurship 92 6.2.3 The impact of entrepreneurial self-efficacy on intrapreneurial intention 95 6.3 Sample and method 96 6.4 Analysis and results 97 6.4.1 Separate and distinct constructs? 97 6.4.2 Regression analysis 97 6.5 Discussion99 6.6 Summary102

7

Social entrepreneurial intention 104 7.1 Introduction104 7.2 The “three pillars” of social entrepreneurship 105 7.2.1 The pro-social motivation of social entrepreneurs 107 7.2.2 The profit motivation of social entrepreneurs 107 7.2.3 The innovation motivation of social entrepreneurs 108 7.3 An empirical study of social entrepreneurial intention 109 7.3.1 The model 110 7.3.2 The sample and method 113 115 7.3.3 Regression analysis 7.3.4 Contrast between the different types of entrepreneurial intention118 7.4 Summary120

8

An inclusive model of entrepreneurial intention 122 8.1 Introduction122 8.2 Elements of the conceptual model 126 8.2.1 Theoretical lens 126 8.2.2 Profit as a goal of entrepreneurial behaviour 127 8.2.3 Social benefit as a goal of entrepreneurial behaviour 129 8.2.4 Innovation as a goal of entrepreneurial behaviour 130 8.2.5 Job satisfaction as a goal of entrepreneurial behaviour 131

Extended contents

xi

8.3 Development of the general utility model 132 8.3.1 Graphical depiction of the model 133 8.4 A typology of commercial versus social entrepreneurial intention137 8.4.1 The value of typologies for theory building 137 8.4.2 Empirical evidence of eight subtypes 140 8.5 Extending the model to the fourth dimension 141 8.6 Conclusion143 9

The holistic approach to entrepreneurship intentions 147 9.1 Introduction147 9.2 The limitations of symmetrical correlational methods 148 9.3 Case-based analysis of entrepreneurship intention 150 9.3.1 Fuzzy-set qualitative comparative analysis 151 9.3.2 Using fuzzy-set qualitative comparative analysis 152 Constructs and measures 153 Calibration154 Minimum frequency, coverage, and consistency 155 Necessity and sufficiency analyses 156 9.4 Fuzzy-set analysis of entrepreneurial intention data 157 9.4.1 Configurations for social entrepreneurship intention 162 9.4.2 Configurations for independence-oriented entrepreneurial intention164 9.4.3 Configurations for growth-oriented entrepreneurial intentions165 9.4.4 What extra information is revealed by fuzzy-set qualitative comparative analysis? 166 9.5 Summary and conclusions 167

10 The future of entrepreneurial intention research 169 10.1 Introduction169 10.1.1 Summary of preceding chapters 169 10.2 The main problem is the method 171 10.2.1 Fuzzy-set qualitative comparative analysis as a solution 173 10.3 Guidelines for future entrepreneurial intention research 174

xii

Entrepreneurial intention

10.3.1 The holistic decision-making process and case-based analysis175 10.3.2 Multiple pathways to the same objective 176 10.3.3 Asymmetric data relationships 176 10.3.4 Multiple interdependent goals of entrepreneurial behaviour176 10.4 Implications for future research 177 10.4.1 The innovation motivation 177 10.4.2 New primary types of intending entrepreneur 178 10.4.3 Multiple pathways 178 10.4.4 Interdependencies 179 10.4.5 Asymmetries 179 10.4.6 Indicator variables for the higher-level constructs 180 10.4.7 Entrepreneurial self-efficacy 181 10.5 Implications for policy, practice, and entrepreneurial education 183 10.5.1 Implications for policy 183 10.5.2 Implications for practice 183 10.5.3 Implications for entrepreneurship education 184 10.6 Conclusion185 Bibliography188 Index208

Figures 8.1

The utility-maximising combination of social benefits and profit

134

8.2

Indicative paths to utility-maximising choice of new venture characteristics

135

8.3

A typology of intending entrepreneurs (and their firms)

139

8.4

A typology of entrepreneurs in social-benefit, profit, and innovation space

139

8.5

Schematic of primary-entrepreneurial-intention typology

142

xiii

Tables 3.1

Evaluating the perceived desirability of an entrepreneurial opportunity41

3.2

Item loadings for the McGee et al. (2009) entrepreneurial self-efficacy scale

48

3.3

Natural, accidental, and inevitable entrepreneurs

52

5.1

Scales for growth-oriented and independence-oriented entrepreneurial intention

80

5.2

Regression coefficients for the antecedents of GOI and IOI

83

6.1

Factor analysis results: entrepreneurial versus intrapreneurial intention98

6.2

Descriptive statistics and intercorrelation matrix

6.3

Regression results for entrepreneurial and intrapreneurship intention100

7.1

Factor analysis to identify the subcomponents of entrepreneurial self-efficacy

115

7.2

SEI, GOI, and IOI data correlation matrix 

116

7.3

Regression models for three types of entrepreneurial intention

117

8.1

Typology of entrepreneurs based on social-benefits, profit, and innovation attitudes

138

8.2

Empirical frequency of the eight types of entrepreneur

141

9.1

Regression results for growth, independence, and social intentions158

9.2

Reduced regression models for entrepreneurial intention

160

9.3

Sample statistics and calibration parameters for fsQCA

161

xiv

99

Tables

xv

9.4

FsQCA results for social entrepreneurship intention

163

9.5

FsQCA results for independence-oriented entrepreneurial intention165

Foreword Norris Krueger It might be hard to imagine today but circa 1990, the field of entrepreneurship was mired in a research tradition that essentially chose to believe entrepreneurs were born, not made. No matter what the broader fields of psychology and economics might say, we tended to believe (or wanted to believe) that entrepreneurs were this magical creature. As one wag put it, not only did entrepreneurship embrace the dying subfield of personality psychology, we made up for it for executing poorly. But in the early 1990s, we began to see a shift – a tectonic shift we did not immediately see but nonetheless revolutionary. Suddenly we saw tools and theories from social psychology started appearing (and getting past reviewers). Instead of focusing of person variables (P) and situation variables (S) we could embrace P×S variables. For example, in grad school I became enamoured with self-efficacy. One such tool was behavioural intentions – whether Al Shapero’s model of the entrepreneurial event or Ajzen-Fishbein’s models, we suddenly had a very useful tool. As we know now, intentions to be entrepreneurial are much trickier than intentions to buy breakfast cereal but it was a start. Consider the number of citations for “entrepreneurial intentions” versus citations of “entrepreneurship” in Google Scholar.

Entrepreneurial

Before 1993

After 1993

Growth

112

~17,000

150×

intentions

At the same time, entrepreneurship research overall only grew 10×. That is, intentions research has grown more than FIFTEEN times as fast as entrepreneurship research in general. Google Scholar reports over 8200 intentions publications in 2018 alone. That success at least coincided with a rise in other tools from social psychology and cognitive psychology. We also see a 9× increase in entrepreneurial cognition research relative to entrepreneurship in general and it accelerated considerably after 1993, suggesting emotional

xvi

Foreword

xvii

intelligence research was at least a leading indicator of research there (Kaffka & Krueger, 2018).

BUT WHY SO SUCCESSFUL? Certainly intentions models yielded large, shiny R-squareds (if the R-squares is high, it must be right, eh?) and certainly face validity. We had compelling arguments like Shaver and Scott (1991) that the so-called “entrepreneurial personality” was a wee bit more complicated. Still, it does not take a neuroscientist to understand that the rise of the intentions model now opened the door to deeper inquiry into “what lies beneath” intent (Krueger, 2007). We soon had great contributions from Robert Baron (1998) and Ron Mitchell (Mitchell, Smith, Seawright, & Morse, 2000) to pry the door even farther to great theory and methods from cognitive psychology.

A HIDDEN LEVER? I got into intentions because my dissertation supervisor was in marketing and intentions were widespread there, even boring. But I was not the only person to take advantage of this experience base and, more important, the broader tool kit from social psychology.1 If you are going to talk about perceived feasibility and perceived desirability, maybe it will help us to understand... perceptions? While Evan Douglas had done some clever and impactful work already (Douglas & Shepherd, 2002a) he had begun a delightful deep dive into the various perceptual lenses that entrepreneurs might deploy. (Remember that Busenitz and Barney had forced us to engage with Kahneman and Tversky’s marvellous findings.) Evan’s Babson paper on entrepreneurial perceptions is pretty much my favourite Babson presentation. It showed that entrepreneurship was firmly on a course to using the best concepts and tools from mainstream psychology. An early piece I did with Alan Carsrud on using social psychology tools in entrepreneurship didn’t get many citations until papers like Evan’s started appearing and now people read it on Researchgate and message me “how passé”. A humorous sign that we have moved far? 1 And there were other good scholars doing good work. Lars Kolvereid was another early adopter of the theory of planned behaviour and contributed a long stream of solid, useful work. Jill Kickul made contributions that tested the limits of the intentions model and Malin Brännback brought new tools like Vallerand’s passion scales and Rick Bagozzi’s theory of trying. Francisco Liñán is another intentions scholar whose work I highly recommend. And Evan’s early co-author, Dean Shepherd, has been one of the rock stars of our field.

xviii

Entrepreneurial intention

For the Carsrud–Brännback book, The Entrepreneurial Mind (2017), I asked “is the field of entrepreneurship growing... or just growing bigger?” Evan’s new book is terrific evidence that, yes, we are growing! Entrepreneurially yours, Norris Krueger, June 2019, Boise, Idaho

Preface The purpose of this book is to serve as a primer for the researcher who is interested in studying the process by which individuals form entrepreneurial intention – i.e. their introspective process that leads them to the conditional decision that they want to behave entrepreneurially. The process of entrepreneurial intention formation occurs over time, and might start with the individual’s recognition of a specific entrepreneurial opportunity, or conversely might start with the realisation that entrepreneurship is the preferred career path (and thus prompt a search for a desirable entrepreneurial opportunity). The process of entrepreneurial intention formation then continues through a series of exploratory phases including viability screening, risk recognition and mitigation, new venture modelling, and estimation of the valuation of the new venture, eventually culminating in the decision (or not) to exploit the new venture opportunity and to actually launch the new business venture. The decision to actually behave entrepreneurially, i.e. to launch a new venture, is conditional on the outcome of this intention-formation process, such that entrepreneurial intention is solidified (or weakened) as time passes by the discovery of confirming (or disconfirming) information about the financial feasibility and the social desirability of the evolving new venture concept, not to mention the confirmation by the individual that the new venture is (or is not) sufficiently congruent with their attitudes (i.e. perceived desirability) and abilities (i.e. perceived feasibility) toward entrepreneurship. At some point, the conditional decision to become an entrepreneur transforms into the decision to actually start a new venture, and the individual transitions from being an intending entrepreneur to a practising entrepreneur (Douglas, 2013). Thus, the target readers for this book include research degree candidates and more seasoned entrepreneurship researchers who wish to conduct research into the entrepreneurial process, particularly the underlying drivers of entrepreneurial intention. This book brings you up to date with the past and present research on entrepreneurial intentions, and then offers new theoretical perspectives by developing an integrated model of entrepreneurial intention, and argues for a novel analytical method, that together will reinvigorate entrepreneurial intention research and allow a more fine-grained understanding of what causes the formation of entrepreneurial intention for the various types of entrepreneurship that we observe in practice. xix

xx

Entrepreneurial intention

There are five main themes running through this book. First, we are concerned with the holistic decision-making process of the individual, whereby the individual takes into account their personal cognitions, preferences, capabilities, and situational factors before making the decision that best serves the objective(s) of that individual. Sample averages may provide general information about the drivers of entrepreneurial intention, but reflect the commonalities (rather than the differences) among the respondents, and thus suppress fine-grained detail about the behaviour of individual entrepreneurs. Second, there is vast heterogeneity among entrepreneurial individuals, and also among the entrepreneurial opportunities that they undertake. It is a folly to expect there to be a simple formula to explain why individuals choose to become an entrepreneur, even for a particular type of entrepreneurship (e.g. social entrepreneurship), because the preferences, cognitions, capabilities, and situational factors impinging on their decisions are both many and various, meaning that there are likely to be several alternative combinations of causes for the decisions they make (e.g. to become a social entrepreneur). Third, individuals undertake entrepreneurship in pursuit of differing goals. Three main goals are currently discussed in the literature. Historically the entrepreneur’s goal was to “do well” by earning profits that allow the satisfaction of the material and emotional needs of themselves and their families. A second goal, and a primary goal for some, is to “do good” by providing benefits to others who are socially or economically disadvantaged, or by avoiding destruction of the natural environment and other life forms. A third goal is to achieve “job satisfaction” by gaining psychic income from the non-monetary benefits associated with being one’s own boss, choice of location and working conditions, and congenial interaction with customers, suppliers, and fellow employees. A fourth and conceptually separate goal, introduced in this book, is to “make a difference” in the world, by pioneering a new technology or new product/service category, with resultant self-satisfaction and accolades from the local or global audience. It is highly likely that individuals might choose to pursue more than one of these goals, and perhaps all four, simultaneously, to a greater or lesser extent. A study of entrepreneurial intention that considers pursuit of only one or two of these goals neglects the nuances (and explanatory power) that might be discovered if all four goals were included in the analysis. Fourth, the motivation to pursue these four goals (or any subset of them) is provided, in part, by the individual’s attitudes to each of the outcomes related to these goals, where these outcomes are the magnitudes of profit, social impact, job satisfaction, and innovations that are expected to be achieved in the pursuit of a given entrepreneurial opportunity. Expectancy-valence theory posits that the “perceived desirability” of the focal entrepreneurial opportunity is the sum of the utility part-worths of the outcomes, where each utility part-worth is the product of the valence of (i.e. attitude towards) the outcome,

Preface

xxi

and the expectancy (i.e. the magnitude) of the outcome. Motivation to pursue the opportunity is created if the perceived desirability is higher than for any other use of the individual’s resources, and if the “perceived feasibility” (i.e. the individual’s self-confidence that they can successfully achieve the desired outcomes) is sufficient to cause the decision to proceed with the entrepreneurial behaviour. Fifth, given the heterogeneity of individuals, the diversity of their goals, and the multiplicity of circumstances in which they operate, it is impossible to imagine that a one-size-fits-all explanation for their decision to start a business venture would provide a prescription that would apply to many, if any, of the individuals in the sample. Yet this is what we effectively do when we utilise symmetric correlational methods (like multiple regression analysis and structural equation modelling) to find a single dominant “net-effects” model of the relationship between the entrepreneurial decision and its antecedent independent variables. Such sample-level findings are not likely to represent closely any individual in the sample. Rather than a single prescription to explain entrepreneurial intention, we expect that there will be multiple, equally valid, pathways to the same entrepreneurial decision exhibited by identifiable subgroups of individuals. Moreover, asymmetric data relationships, whereby an antecedent variable may be positively related to the dependent variable for some respondents and negatively related for others, are suppressed by symmetrical correlational methods, despite being potentially important for theory building to better understand entrepreneurial behaviour. Thus we are advocating the case-level analysis of entrepreneurial decision making, which can be facilitated by fuzzy-set Qualitative Comparative Analysis (fsQCA), which is still a relatively novel analytical tool in the entrepreneurship literature. The chapters of this book are organised as follows. Chapter 1 provides an introductory foundation and discussion of the main constructs involved in the study of entrepreneurial intentions. Chapter 2 considers a taxonomy of main entrepreneurship types that have been identified by past research, which has focused on two main types, namely commercial and social entrepreneurship and their various subtypes. Chapter 3 outlines the main theoretical lens used in this book, this being the theory of planned behaviour, and considers the perceived desirability and the perceived feasibility of entrepreneurship as the main drivers of entrepreneurial intention. Chapter 4 considers a range of issues that must be considered when undertaking empirical analysis of entrepreneurial intention, before we turn to empirical studies in the next three chapters. Chapters 5, 6, and 7 investigate the statistical antecedents of entrepreneurial intention, where the dependent variable in the regression equation relates to commercial entrepreneurship, intrapreneurship, and social entrepreneurship, respectively. In each case, that main type of entrepreneurial intention is investigated in conjunction with one or more other types of entrepreneurial

xxii

Entrepreneurial intention

intention. In Chapter 5, profit-seeking entrepreneurship is contrasted with psychic income-seeking entrepreneurship. In Chapter 6, self-employed entrepreneurship is contrasted with corporately employed intrapreneurship. In Chapter 7, pro-social motivation is contrasted with both profit motivation and innovation motivation. With this background the reader is ready for the integrative conceptual model presented in Chapter 8, where all four motivations (profit, social impact, psychic income, and innovation) are incorporated into a general model of entrepreneurial intention. Building on the foundation laid by the preceding chapters, the integrative conceptual model includes four main motivations, reflecting the primary goal of the intending entrepreneur to either (i) earn profit; (ii) gain psychic satisfaction; (iii) provide social impact; or (iv) be the pioneer in the development of a new technology and its implementation in related market categories. In line with the holistic theory of human decision making, these motivations are postulated to interdependently underlie the formation of entrepreneurial intention. The conceptual model suggests a four-way hybridity of entrepreneurship, and generates a taxonomy of four main types of entrepreneurial intention where the primary purpose is to be either profit seeking, social impact seeking, psychic income seeking, or pioneership seeking. This four-goal model is consistent with the holistic theory of human decision making (Magnusson & Torestad, 1993) and suggests a wide range of new research opportunities for entrepreneurship researchers. Chapter 9 presents fsQCA, a relatively new methodology for the study of entrepreneurial intention and other entrepreneurship phenomena. This chapter argues that traditional quantitative methods, such as multiple regression analysis and structural equation modelling, which rely on assumptions of discrete independent variables, data symmetry, and data-relationship symmetry, necessarily produce a single dominant “net-effects” prescription for the antecedents of entrepreneurial intention. This, despite our observation that heterogeneous individuals seem to pursue multiple and different pathways to entrepreneurship. Moreover, traditional methods suppress the asymmetric data relationships that are observed in entrepreneurial phenomena. As a result these methods are unable to reveal information that is of vital interest to the researcher seeking to build a new theory to explain the observed heterogeneity of entrepreneurial behaviour. By considering the conjunctural interdependence of antecedent conditions on the focal outcome, fsQCA is able to reveal further finer-grained information that exists in the data. It finds multiple pathways to the focal decision (e.g. entrepreneurial intention). It also identifies causal asymmetry in data relationships (i.e. where an antecedent condition may be positively related to the outcome for some individuals and negatively related to the outcome for other individuals), when it exists. Accordingly fsQCA is complementary to symmetric correlational methods in that it provides

Preface

xxiii

finer-grained empirical evidence about entrepreneurial behaviour that the traditional methods leave hidden in the sample data, and which can form the basis for new theory development. Finally, Chapter 10 offers a summary and suggestions for further research using the four-goals model and fuzzy-set thinking, hopefully having built the reader’s motivation to build theory by testing propositions embodied in the general model, and to use fsQCA to identify new information presented by the interdependencies and the asymmetries that lie hidden in any data set. Indeed this new technique provides the legitimate opportunity to re-examine past studies, both to replicate them (or not) and to provide an empirical basis for new theory development.

1. Introduction 1.1 INTRODUCTION This book is about what motivates individuals to form the intention to become an entrepreneur. This is an important question, because the practice of entrepreneurship contributes to the economic and social welfare of any nation, and by extension, to global prosperity and human wellbeing. Entrepreneurs practise innovation, and innovation is the source of new profit and social opportunities. New commercial ventures introduce new technologies and provide new employment opportunities and company tax revenues, making entrepreneurship a key objective of politicians globally. Similarly, new social ventures introduce new ways to solve social problems and to relieve the suffering of the needy, complementing the work of governments to help solve the “market failure” problem (whereby profit-seeking firms will not undertake unprofitable ventures). Accordingly, entrepreneurial behaviour can be regarded as a renewable resource that provides economic and social welfare. The supply of new entrepreneurs needs to be replenished continuously as prior entrepreneurs fail, tire, or grow complacent and conservative. Given the high failure rate of new ventures, an ongoing supply of newly minted entrepreneurs, launching new business and social ventures or taking over the reins of existing companies and social organisations, is important for the economic and social welfare of nations. If firms exiting the marketplace are not replaced by the next crop of new and repeat entrepreneurs, the consequences for economic and social wellbeing are potentially disastrous. The formation of entrepreneurial intention (and subsequent entrepreneurial behaviour) is just as important for the managers of existing business firms and organisations as it is for individuals contemplating starting a new business or social venture. Within individual business firms, relentless product innovation is required as their previously new products are driven out of the market by newer products that render the earlier products obsolete or unprofitable to produce. For social entrepreneurship, the ongoing persistence of social and economic inequity calls for new social entrepreneurs to arise and add their energies and empathy to the solution of those wicked problems. 1

Entrepreneurial intention

2

Moreover, the human cost of new venture failure (Shepherd, 2003; Ucsbasaran, Shepherd, Lockett, & Lyon, 2013) behoves us to better understand entrepreneurial intentions, because some significant proportion of new ventures are started without the would-be entrepreneur possessing the necessary entrepreneurial attitudes and abilities that would allow them to survive and thrive as “babes in the woods” amongst more experienced incumbents in the marketplace. If entrepreneurship educators can better enunciate and develop the attitudes and abilities that foster successful entrepreneurship, then hopefully the supply of successful entrepreneurs will be increased, or at least the failure rate of new ventures will be reduced as new ventures that are doomed to fail will not be started. Instead, intending entrepreneurs may be encouraged to wait until they are suitably equipped, with a sufficient armoury of entrepreneurial attitudes and abilities, and subsequently start new ventures with a higher probability of success. There are many differing views of entrepreneurship, and of entrepreneurs, such that communication about that behaviour and those individuals can easily be at cross purposes unless we start by defining terms to ensure that we are all talking about the same phenomena. Accordingly, in this introductory chapter we will lay a foundation for understanding entrepreneurial attitudes, abilities, and intentions, by introducing and clarifying the meaning of important terms and concepts necessary for the discussion of this subject.

1.2

WHAT DO WE MEAN BY ENTREPRENEURSHIP?

Over 30 years ago, as research into the practice of entrepreneurship blossomed and scholars debated what is and what is not entrepreneurship, Gartner (1988) circuitously stated that “Entrepreneurship is what entrepreneurs do” to emphasise that entrepreneurship is a behaviour. It is not an attitude, and not an ability, but rather is the consequence of possessing entrepreneurial attitudes and abilities. In this book I shall adopt what I believe to be the mainstream definition of entrepreneurship, namely that it is proactive, risk-taking, innovative behaviour, intended to create new wealth, involving the introduction of new products, new processes, and/or new business models (see, e.g. Covin & Slevin, 1991; Lumpkin & Dess, 1996; Shane & Venkataraman, 2000). That definition contains several jargon words that deserve further examination. 1.2.1 Proactivity Proactivity implies that entrepreneurs tend to act first, before others do, in the pursuit of opportunities to create new wealth. To act first, one must first perceive the opportunity. Opportunity recognition is thus a foundational element of the entrepreneurial process. Some entrepreneurial opportunities

Introduction

3

are discovered, while others are created (Davidsson, 2015). An opportunity might be sitting there waiting to be discovered by a person who recognises two things, first, the existence of a known but unserved need (i.e. market demand), and second, the existence of a way (i.e. the technology) to serve that need. Alternatively, the need may be unknown at first, but an entrepreneur might create the opportunity by developing a new product or service (such as WiFi) that serves a useful purpose that, once revealed to potential customers, is demanded by them because it saves them time, money, or pain, or because it brings them pleasure, health, or other personal benefits. This “demand– technology nexus” lies at the heart of the entrepreneurial opportunity. The capability to recognise an opportunity is known as “entrepreneurial alertness” (Kirzner, 1999). This alertness can be understood as knowledge and creativity on both sides of the demand–technology nexus. Knowing and understanding what potential customers need (or might want) allows the individual to begin searching for a solution for that need or want. On the other side of this nexus, knowing the possibilities and capabilities of various technologies allows the entrepreneur to envision the technological solution that will solve a customer’s problem. Thus, entrepreneurial alertness underlies opportunity recognition, which in turn underlies proactive entrepreneurial behaviour. 1.2.2

Risk taking

A risk is the possibility that an event could happen which would have adverse consequences for the individual (or other entity). The offending event for the entrepreneur might include a lack of demand for the new product; the failure of the technology to solve the customer’s problem; the advent of rival producers who vigorously contest the marketplace; and regulations and legal restrictions imposed by governments. The adverse consequence might involve monetary losses and/or psychic costs including stress, upset, and the loss of physical and/or mental health. In the proactive pursuit of entrepreneurial opportunities the entrepreneur must necessarily bear the risks of such events, because these events cannot be perfectly predicted nor completely mitigated or insured against. Entrepreneurship takes place in an environment of risk and uncertainty. Strictly, risk implies a known probability of the adverse event and known consequences of the adverse event. So if you and I agree to flip a coin to decide who pays for our $40 lunch, the probability of the adverse event (i.e. not having a free lunch) is 50 per cent and the expected value of the event is a loss of $40 × 50% = $20. But most entrepreneurship does not involve known probabilities and known losses. Instead, entrepreneurship is mostly concerned with “uncertainty” (Knight, 1921) where the adverse events include both predicted and unpredicted events, and neither the probabilities of these events occurring,

4

Entrepreneurial intention

nor the magnitudes of the consequences, can be reliably estimated. Estimating the probabilities and magnitudes of adverse events requires a history of similar “trials” in which these events either did or did not happen, but since entrepreneurship involves proactive pursuit of innovative new products there is often no relevant prior experience from which to draw reliable estimates of probabilities or magnitudes of the adverse outcomes. The more radical the innovation, the less relevant prior information exists about its costs and demand situations. As a result, entrepreneurs necessarily face both risk and uncertainty when seeking to create new wealth. For convenience of exposition we tend to use “risk” to include both risk and uncertainty, but keeping the distinction in mind is important for clarity in many discussions. Being exposed to risk (and uncertainty) is unpleasant for most people, meaning that they incur psychic disutility (or dissatisfaction) over the prospect of monetary and/or psychic loss. The prospect of these losses tends to cause people to avoid exposure to risk, to a greater or lesser extent according to their “risk tolerance” (Busenitz & Barney, 1997; Forlani & Mullins, 2000). Risk tolerance means one might choose to tolerate risk if that risk is necessary to achieve a reward that would be large enough (in terms of the psychic utility it generates) to more than compensate for the psychic disutility of the risk exposure. The flip side of risk tolerance is risk aversion – that is, a low degree of risk tolerance implies a high degree of risk aversion. We expect most entrepreneurs to be risk averse but risk tolerant (to a degree), but it is an oversimplification to assume that all entrepreneurs exhibit low risk aversion (Palich & Bagby, 1995; Simon, Houghton, & Aquino, 2000). Exposure to risk is just one of the issues underlying the formation of entrepreneurial intentions, such that even a highly risk-averse individual may want to become an entrepreneur if the monetary and psychic rewards of entrepreneurship are sufficiently high (Douglas, 2013). 1.2.3 Innovation The proactive, risk-taking, innovative behaviour that entrepreneurs “do” culminates in new products, new processes, and/or new business models. First, let us note that innovation differs in concept from invention. Invention is the creation or ideation of a new technology (e.g. Bluetooth near-field communication) whereas innovation is the embodiment of that technology in a new product, process, or business model (e.g. cashless payment using your smartphone). Inventors may become entrepreneurs of course, and subsequently innovate in that capacity, but more often entrepreneurs create innovative new business (or social) venture opportunities using the technologies invented by others.

Introduction

5

New products New products mean innovative new goods or services that are produced by individuals or firms. They are the end result of a production process, and for simplicity of communication we shall henceforth use the term “new products” to mean whatever new goods and services the firm produces. It may mean a new physical good, or a new intangible service, or a new combination of both goods and services supplied as a package. The question immediately arises “how new (i.e. how different) must a new product be, to be called a new product?” A new product could be entirely different from anything that went before it, like the Segway personal transporter was back in 2001.1 Such “radical” or “really new” products (Song & Montoya-Weiss, 1998) create a new product category in which the pioneer firm has an initial monopoly position until rivals arise to compete with imitative products in the new product category. These imitative products might be identical if the rivals can access all the necessary resources2 to perfectly imitate the pioneer’s product, but more likely will be unable to copy the pioneer’s patented technology, registered trademarks, designs, and/or copyrighted materials, and will consequently be differentiated relative to the pioneer’s product. This raises the question whether the entry of a differentiated product into an existing product category constitutes a “new product” if it is essentially similar to other products in the product category (e.g. tennis racquets) except for relatively minor differences. The answer is in the affirmative if these differences incorporate at least one innovation. Even a copycat product, similar in all technical aspects, can be regarded as a new product if it is differentiated by its brand (e.g. a new brand of milk introduced to food markets). So, new products include a new version of an existing product, or a new brand entering an existing product category, if that product incorporates innovations. Drucker (2014) called these “creative imitations” which are consistent with the general description of the product category but contain innovations that allow the product to be differentiated in the marketplace. Thus, new products include a new entry to an existing product category with a differentiated and innovative version of the generic product already sold in that market by existing firms. Such new products typically embody product or process innovations that are new to the market. There is a substantial literature on the “innovativeness” of new products, with ongoing attempts to measure the degree of innovativeness of new



See www​.economist​.com/​science​-and​-technology/​2001/​12/​06/​is​-that​-it. For the “resource-based view” of competitive advantage, see Barney (1991, 2001). 1 2

6

Entrepreneurial intention

products, since it is widely expected that innovation drives firm profitability and that more innovative products should mean more profitable outcomes (Ganbaatar & Douglas, 2018). The literature distinguishes between the “technical novelty” and the “market novelty” of new products. Technical novelty involves a difference in the style, the performance of, and/or the needs served by the new product. Style differences (such as new colours, shapes, and/or textures of a physical product, or new ways to present a given service, e.g. more flamboyant hairdressers) are purely cosmetic in nature, simply changing the appearance of the product or service, and not changing the performance or functionality of the product (e.g. a new container for milk, or a new process for the same haircut). Greater technical novelty (and hence innovativeness) is evident in innovations that provide improved capability for the product in particular product functions (such as more pixels in a smartphone screen allowing greater clarity, or greater battery capacity for electric cars allowing longer trips). Still greater technical novelty (and innovativeness) is exhibited by the addition of new functions embodied in the product (such as when a camera was first incorporated into the smartphone, or when automated parallel parking was incorporated into cars). Market novelty is concerned with the newness of the above technical novelties to the customer. The least degree of market newness (and innovativeness) is when a technical change embodied in the focal firm’s product is only new to that firm, because one or more other firms have previously introduced that innovation to customers in the firm’s marketplace. Such “new to the firm” innovation is indicative of catch-up, or copycat, behaviour by the firm involved. A greater degree of market newness (and innovativeness) occurs when a technical novelty is “new to the market” – that is, when the focal firm pioneers the introduction of the technical novelty to customers in a particular geographic market. The first-mover advantages associated with this greater degree of innovativeness should be expected to garner greater profits from that market, at least initially (Lieberman & Montgomery, 1988; Suarez & Lanzolla, 2007). An even greater degree of market novelty (and innovativeness) occurs when the technical novelty is “new to the world” – meaning it has not been introduced to any market (globally) previous to that point in time. Note that any new product might have technical novelties of all three types (style, functional improvements, and new functions), each of which might be either new to the firm, new to the market, or new to the world, such that a combined measure of a new product’s innovativeness requires a complex calculation (see Ganbaatar & Douglas, 2018). The greatest degree of product innovativeness is exhibited by a “really new” product (Song & Montoya-Weiss, 1998) which effectively creates a new product category because it comprises a majority of technical novelties that are new to the world

Introduction

7

and is thus quite unlike any pre-existing product (e.g. the Segway personal transporter; Facebook; YouTube, and other examples). A new product might also be a “disruptive” product, in the sense that it has severe negative impacts on firms in another industry, potentially putting some or all of these disrupted firms out of business. For example, the printing industry, bricks-and-mortar travel agents, video-rental stores, and recorded-music stores have been disrupted by new products on the back of the IT revolution. But note that a radical new product might not be disruptive, while a creative imitation might actually be disruptive, in their respective markets. The Segway personal transporter was a radical new product but did not have a significant disruptive impact on any particular market category, since its sales grew relatively slowly initially (Day, 2007) and its impact was spread over a broad range of other people-moving markets, including bicycles, motorcycles, automobiles, taxis, buses, and walking shoes. On the other hand, a creative imitation product employing an innovative new technology might achieve a similar result for the customer at a substantially lower cost, or a consistently higher quality at a similar price, and thus disrupt the demand for other suppliers in the same market using older less-efficient technologies, as customers shift their purchases to the innovative new product. An example of a disruptive creative imitation is Uber, who used an innovative business model (and ordinary cars and drivers) to disrupt the taxi industry. A disruptive radical product is one that creates a new product category that attracts demand away from an existing product category, such that the latter product category withers and dies as customers transfer their allegiance to the new product category – perhaps an example of this will be delivery of goods by drones (quadcopters) that may disrupt the parcel delivery industry. These are innovative new ways of production, marketing, and distribution of the firms’ products. These new technologies add value to either or both the producer and the consumer. IKEA disrupted the traditional furniture industry by mass producing items that are sold “fully knocked down” and enlisting the customer’s support for both the delivery and assembly of the items. Automated new processes allow IKEA to save production and shipping costs, and/or to produce a higher-quality product, and provide the consumer with a better-value proposition as the result of a higher-quality product and/or a lower price to purchase the product. These are innovative new ways of organising and managing the business firm, such that the firm reduces its costs and thereby increases its profit, and/ or allows the customers to gain greater value in terms of increased product quality, faster delivery, better service, and so on. Historic examples of new business models that deliver value to both producers and consumers include department stores disrupting independent butchers, bakers, and candlestick

Entrepreneurial intention

8

makers; Uber people moving; Airbnb accommodation; Expedia travel services; and many others. 1.2.4

New wealth

The “new wealth” that entrepreneurs intend to create (but they may fail) can include personal monetary and non-monetary rewards for stakeholders of the firm, and/or societal benefits for others external to the firm. This dichotomy of new wealth gives rise to two main types of entrepreneurial behaviour observed, namely “commercial” entrepreneurship and “social” entrepreneurship.

1.3

COMMERCIAL VERSUS SOCIAL ENTREPRENEURSHIP

It has become conventional to dichotomise entrepreneurs as either commercial or social entrepreneurs, on the basis of whether their primary purpose is to make money or to create social impact. This recognises that an entrepreneur may have both commercial and social motivations, notwithstanding that some treatises portray the commercial entrepreneur as totally committed to profit making, while others portray social entrepreneurs as totally selfless individuals who eschew profit making in favour of maximising the social impact of their ventures. This derives from the classical (economic) theory of entrepreneurship that assumed a profit-seeking individual with little or no pro-social attitude, while early research on social entrepreneurship assumed only pro-social attitude (with little or no preference for profit). Hybrid models of entrepreneurship now recognise that both of these attitudes are likely to precede the intention to become either a commercial or a social entrepreneur. 1.3.1

Commercial entrepreneurship

Commercial entrepreneurship is self-oriented, where individuals undertake entrepreneurship primarily for their personal gratification, which is received from both monetary and non-monetary rewards, the latter also being known as “psychic income” (Gimeno, Folta, Cooper, & Woo, 1997). Psychic income in this context refers to personal satisfaction derived from the practice of entrepreneurship, from sources such as decision-making autonomy (being one’s own boss); flexibility of when to work, where to work, and what products to produce; and work enjoyment derived from achievements gained and working with a hand-picked team of partners and employees. Monetary income (from salary and profit) also generates psychic satisfaction when spent on goods and services or saved for future spending.

Introduction

9

There may also be negative psychic income (i.e. psychic costs) associated with the entrepreneurial process, such as risk taking and hard work, and these will detract from the overall psychic satisfaction derived. When the psychic income expected from both the monetary and (net) non-monetary sources are added together, this is either sufficient to cause the individual to decide to become an entrepreneur (i.e. to form entrepreneurial intention), or conversely is not expected to be as rewarding as being an employee within another firm or organisation (Douglas & Shepherd, 2000), such that entrepreneurial intention is not formed in the latter case. 1.3.2

Social entrepreneurship

Social entrepreneurship is undertaken primarily for “other-oriented” reasons, to provide benefits to others who are not involved in the ownership of the firm or organisation, and who may be economically and socially disadvantaged. This “pro-social” motivation of social entrepreneurs is driven by feelings of altruism, empathy, compassion, and guilt (see, e.g. Austin, Stevenson, & Wei-Skillern, 2006; Grant & Berry, 2011; Mair & Marti, 2006; Miller, Grimes, McMullen, & Vogus, 2012; Zahra, Newey, & Li, 2014). Observing the economic or social disadvantages faced by less fortunate people, social entrepreneurs are motivated to act to relieve the hardship and suffering of those people. That is, they are motivated to use innovative methods to produce goods and services needed by disadvantaged persons, and may or may not seek to make personal profit from that endeavour. Social entrepreneurs derive personal satisfaction from helping others in need (Bacq & Alt, 2018; Mair & Marti, 2006), and like commercial entrepreneurs may also be expected to gain at least some psychic satisfaction from the entrepreneurial process and from any profits and non-monetary income earned. Individuals form the intention to become a social entrepreneur if the sum of those satisfactions is expected to exceed the satisfaction to be gained by any other use of their time and energy (Douglas & Shepherd, 2000). Hybrid social entrepreneurs (Battilana & Lee, 2014) seek to earn profits as a secondary purpose, subjugated to their primary purpose of providing social benefits. Similarly, commercial entrepreneurs, whose primary purpose is to make personal monetary and psychic income, may also seek (as a secondary purpose) to provide social benefits (from which good deeds they will derive additional psychic income). This secondary purpose may be driven by “corporate social responsibility” which in turn may reflect the innate preference of the entrepreneur and/or result from external pressure to provide social as well as private benefits, placed on the firm by investors, customers, and potentially other stakeholders.

10

1.4

Entrepreneurial intention

THE ENTREPRENEURIAL MINDSET

A mindset is a way of thinking – it describes the way a person views and interprets the world around them, which in turn influences the decisions they make. Thus a mindset underlies the individual’s behaviour, essentially motivating particular actions that appear to be “the best thing to do” after consideration of the things that matter most to that individual. The entrepreneurial mindset describes how entrepreneurs think, including what issues underlie their formation of entrepreneurial intentions and their decision to become an entrepreneur. Fundamental to the entrepreneurial mindset are the attitudes that the individual holds towards the outcomes of entrepreneurship and their perception of their ability to cope with and succeed at the tasks involved in the entrepreneurial process. 1.4.1

Entrepreneurial attitudes

Simplistically, entrepreneurial attitudes are the attitudes one has toward being an entrepreneur, or at a higher level of analysis, they are the attitudes toward the outcomes that are associated with being an entrepreneur. These outcomes may be categorised as monetary income (salaries and profit), decision-making autonomy, other non-monetary benefits, non-monetary costs (Douglas & Shepherd, 2000), and pro-social outcomes. An individual’s attitude to monetary income may vary from negative (e.g. for a recluse who wants to subsist in the wilderness without recourse to the monetary economy) to strongly positive (e.g. for a person who is highly acquisitive of goods and services). Confining our purview to entrepreneurs, attitudes to income may be relatively low or slightly negative for some social entrepreneurs who provide unpaid (volunteer) services and eschew profit making because that would reduce the social benefits that can be provided. Other social entrepreneurs, operating hybrid social ventures, exhibit a positive attitude to monetary income, but restrict their profit making to being the secondary objective of the firm, again because greater profits would come at the expense of reduced social benefits. The attitudes of commercial entrepreneurs to monetary income can be expected to range from very strong for the classical profit-maximising entrepreneur (Davidsson, Steffens, & Fitzsimmons, 2009) to less strong for those motivated to earn a lesser but satisfactory level of profit supplemented by psychic income derived from non-monetary rewards – such profit-satisficing individuals (Simon, 1972) have been called “independence-oriented” (Douglas, 2013), and “salary-substitute” and “lifestyle” entrepreneurs (Barringer & Ireland, 2006: 45). In Douglas (2013) I found that attitude to income was

Introduction

11

positively related to entrepreneurial intention for profit and growth-oriented intending entrepreneurs, but insignificantly related to independence-oriented intentions, indicating that some independence-oriented individuals did want to make profit while others did not, such that no clear picture emerged, on average sample-wide. Attitudes to autonomy are expected to be positive for entrepreneurs because autonomy is an innate human need, according to the theory of self-determination, which argues that individuals will strive to fulfil their innate (and situational) needs (Deci & Ryan, 1985; Gagné & Deci, 2005; Ryan & Deci, 2000). The heterogeneity of entrepreneurs (Gartner, 1988; Shane & Venkataraman, 2000) ensures that some will have more strongly positive attitudes to autonomy than will others. Some people want to be in charge, whereas others are happy to be part of a team and make decisions collaboratively. Some do not enjoy the responsibility and/or stress that goes with decision-making authority, and may seek a less innovative kind of entrepreneurial new venture involving fewer unknowns and therefore requiring fewer decisions where the knowledge and guidance of others would be needed. Decision-making autonomy is a non-monetary benefit to the entrepreneur, assuming such autonomy is viewed positively, that is, as net-utility providing. It seems unlikely that an individual with a negative attitude to decision-making autonomy would willingly choose to be a solo entrepreneur, given that entrepreneurial behaviour requires a substantial amount of decision-making independence, but that person may wish to start an entrepreneurial new venture within a top management team, where other members of the team take greater responsibility for decision making. Attitudes to other non-monetary benefits provided by the entrepreneurial process are expected to be (net) positive for most intending entrepreneurs, since surely most would enjoy the status of being a self-employed entrepreneur: the flexibility of choosing the place and working hours for their venture and the interpersonal relationships they can enjoy with “fellow travellers” on their entrepreneurial journey, be they customers, employees, suppliers, or fellow management team members. As mentioned earlier there will be non-monetary (psychic) costs associated with entrepreneurship, and these derive from exposure to risk, exposure to hard physical work, and exposure to mental stress. Intending entrepreneurs will most likely have negative attitudes toward these outcomes of entrepreneurial behaviour, unless they enjoy pain and punishment. Earlier we spoke of risk tolerance, which is the willingness to bear risk if it seems worthwhile in terms of the total monetary and non-monetary rewards envisioned. Given that risk and monetary returns are generally expected to be positively related, profit-seeking entrepreneurs might be expected to have relatively low risk aversion, seeing higher risk as a necessary correlate of higher profit possibilities. On the contrary, profit-satisficing entrepreneurs may be

12

Entrepreneurial intention

satisfied with less than maximal profits, because pursuit of those would expose the individual to intolerable levels of risk. But, as I argued in Douglas (2013), attitude to risk is but one of the factors underlying the formation of entrepreneurial intentions, and that decision rests on the overall total of the utility part-worths of the several outcomes (i.e. profit, psychic income, and psychic costs) anticipated for the contemplated entrepreneurial opportunity. Thus the apparently paradoxical situation of a highly risk-averse individual pursuing a high-risk opportunity is nonetheless possible, and equally a person with very low risk aversion might choose an opportunity with very low risk. This will be addressed further in later chapters. A second main non-monetary (psychic) cost of entrepreneurship relates to the fact that entrepreneurs are humans, with limited physical and mental stamina. Entrepreneurship generally requires long and hard hours of work, far exceeding the 40-hour working week of employees (Bird & Jellinek, 1988; Douglas & Shepherd, 2000). Frequently, cash-strapped entrepreneurs cannot afford to pay additional workers to share the workload, and as a consequence need to work evenings and weekends in addition to the standard working week. While physical and mental activity is healthy and desirable up to a point, it is likely that many entrepreneurs go beyond that point and experience psychic disutility from the harder work and the emotional stress required of them in operating a new business or social venture. These net results of the non-monetary psychic benefits and costs that accrue to the individual in the practice of entrepreneurship can be referred to as “job satisfaction” (Lee, Wong, Foo, & Leung, 2011; Locke & Latham, 1990), which is one of the specified four goals of the integrated model of entrepreneurial intention to be developed in Chapter 8. It is expected that intending entrepreneurs would derive utility from job satisfaction associated with participation in the entrepreneurial process. Note that in addition to the above sources of psychic income and costs (job satisfaction), we are also recognising that intending entrepreneurs will have an attitude toward the provision of social benefits to others, and as a result expect to gain psychic incomes and incur psychic costs, and thus net utility, from the social impact of their new venture. For expositional clarity of the four-goals integrated model of entrepreneurial intention we shall keep separate the utility deriving from the “other-oriented” (pro-social) behaviour and the utility derived from the “self-oriented” (commercial) behaviour. 1.4.2

Entrepreneurial abilities

Wanting to be an entrepreneur is one thing, and being a successful entrepreneur is another. Self-determination theory (Deci & Ryan, 1985) argues that humans have an innate need for competence, and thus tend to pursue activities they

Introduction

13

think they will be “good at” while avoiding activities where they do not expect to exhibit competence. This need for competence is underpinned by one’s need for achievement (McClelland, 1961) and, on the flipside, a preference to not expose oneself as incompetent and thereby suffer embarrassment and “loss of face”. Thus the intending entrepreneur is likely to assess whether or not he/she possesses the necessary abilities to succeed as an entrepreneur. These abilities derive from the individual’s accumulated knowledge, prior education, and related experiences. This introspective assessment of one’s capabilities is known as “self-efficacy” (Bandura, 1982), and is task-specific. For example, a person may think they are quite efficacious as a teacher, but not as a singer or golfer – these activities require different sets of abilities to attain success. Entrepreneurial self-efficacy (ESE) therefore relates to the expectation that one can successfully complete the tasks associated with entrepreneurship (Boyd & Vozikis, 1994; Wilson, Kickul, & Marlino, 2007; Zhao, Seibert, & Hills, 2005). Several scholars have proposed specific tasks which are salient for entrepreneurship, and have tested ESE scales (e.g. Chen, Greene, & Crick, 1998; De Noble, Jung, & Ehrlich, 1999; Liñán & Chen, 2009; McGee, Peterson, Mueller, & Sequiera, 2009; Thompson, 2009). McGee, et al., (2009) follow the four main stages of the entrepreneurial process suggested by Stevenson, Roberts, and Grousbeck (1985), these being the searching, planning, marshalling, and implementation stages. The searching stage includes opportunity recognition, viability screening, and business model validation tasks (Blank & Dorf, 2012). The planning stage involves the development of the strategic plan for the new venture, including its marketing, production and operations, human resource, and financial plans. The marshalling stage involves the gathering of the necessary resources for the execution of the business plan, particularly the new venture’s funding sources and its “strategic resources” which are expected to be, or to become, valuable, rare, hard to copy, and non-substitutable (Barney, 1991). The implementation stage involves launching the new business venture and the management tasks to ensure that the firm survives and prospers. In their study McGee et al. (2009) found empirically that the self-efficacy for the implementation stage separated into two discrete factors, representing people-management skills and financial -management skills. As we shall discuss further in Chapter 3, ESE acts as a “reality check” on the individual’s decision to form entrepreneurial intentions. If they are not confident that they can successfully handle the tasks involved in the specific entrepreneurial new venture under consideration, they are not likely to form intention to pursue that entrepreneurial opportunity. Instead they might investigate other self-employment opportunities that better suit their skill set, or alternatively seek to build their skill set (and ESE) through formal education or a period of employment in another firm. In these cases the individual’s

14

Entrepreneurial intention

realisation that they do not have the necessary competences may result in the postponement, rather than the abandonment, of entrepreneurship as a career alternative. Alternatively they may seek to partner with one or more partners to form a top management team that collectively possesses the needed abilities and skills.

1.5

ENTREPRENEURSHIP AS A PLANNED BEHAVIOUR

Although entrepreneurship may seem in some cases to be an impetuous or impulsive reaction to a recognised entrepreneurial opportunity, in general it is a planned behaviour (Ajzen, 1985, 1991, 2002; Ajzen & Fishbein, 1970). It is a reasoned action (Fishbein & Ajzen, 2011) that is preceded by the formation of the intention to act, which in turn is preceded by the attitudes to the act. We have outlined above that in the case of entrepreneurship the “attitude toward the act” can be disaggregated to include the attitudes toward the expected outcomes of the act, namely the individual’s attitudes toward profits, decision-making autonomy, work enjoyment, risk, hard work, and the provision of social benefits. These attitudes tend to be dispositional, residing in the mindset of individuals, essentially awaiting the recognition of a desirable entrepreneurial opportunity to trigger the formation of entrepreneurial intention for that particular opportunity.3 Entrepreneurship takes place at the nexus of the individual and the opportunity (Shane & Venkataraman, 2000; Venkataraman, 1997). By this we mean that the characteristics of the individual (their attitudes and abilities) and the characteristics of the opportunity (the expected profit, autonomy, work enjoyment, risk, hard work, and social benefits) intertwine to drive the individual’s decision to form (or to not form) entrepreneurial intention for that particular opportunity (Douglas, 2013). We all recognise entrepreneurial opportunities that we think someone else ought to undertake, which McMullen and Shepherd (2006) called “third-person” opportunities. They argued that individuals form entrepreneurial intention only for a “first-person” opportunity, where the person recognises an opportunity that best suits their own attitudes and abilities. The theory of planned behaviour argues that formation of the intention to undertake an act is driven by the perceived desirability of the act, the perceived 3 In the four-goals model the outcomes of entrepreneurial behaviour will be restated as profit, social impact, innovation, and job satisfaction, where the latter goal encompasses all the non-monetary sources of psychic income or costs (except social impact), these including decision-making autonomy, risk exposure, work effort, and work enjoyment that have been enumerated previously (Douglas, 2013).

Introduction

15

feasibility of the act, and by the approval or disapproval of referent others, generalised as social norms (Ajzen, 1991; Shapero & Sokol, 1982). In the context of entrepreneurship, Krueger (1993); Krueger and Brazeal (1994); and Krueger, Reilly, and Carsrud (2000) reconciled Shapero and Sokol’s theory of the “entrepreneurial event” with Ajzen’s theory of reasoned action, and effectively reduced the drivers of entrepreneurial intention to the perceived desirability and the perceived feasibility of entrepreneurship. Social norms as a separate driver of intentions has often not found empirical support (e.g. Carey, Flanagan, & Palmer, 2010; Li, 2007; Wu & Wu, 2008), and in any case can be parsimoniously incorporated into the perceived desirability construct, since congruence/incongruence with social norms operates to provide psychic utility/disutility for those seeking social approval. Similarly, for those desirous of upsetting the status quo, as entrepreneurs often are, the incongruence of the proposed new venture with social norms might make the opportunity more desirable to them. 1.5.1

The perceived desirability of entrepreneurship

The perceived desirability of the focal entrepreneurial opportunity depends on the (net) psychic utility expected from the perceived outcomes of that entrepreneurial opportunity. These outcomes are both monetary (i.e. profit) and non-monetary (i.e. job satisfaction and social impact), with innovation to be added as a desirable goal in Chapter 7. Using expectancy-valence theory (Vroom, 1964), we can theoretically evaluate the perceived desirability of any particular outcome as the product of its expectancy and its valence. The expectancy of an outcome is the individual’s expected value of the outcome (measured on an appropriate scale or index), and the valence of the outcome is the importance to the individual of that outcome, which is well captured by their attitude to that outcome. The arithmetic product of the expectancy and the valence provides a measure of the “utility part-worth” of each expected outcome, such that the total worth of the opportunity to the individual is the sum of the utility part-worths of the expected outcomes of that opportunity (see Douglas, 2013; Hair, Black, Babin, Anderson, & Tatham, 2010: 268; Steel & König, 2006). It is fraught to discern personal attitudes by direct questioning of an individual, due to social-desirability bias likely distorting their responses to questions which they might regard as personal issues (such as attitude to profits and to providing social benefit). Fortunately, the use of a conjoint experiment can largely avoid such bias in determining personal attitudes to entrepreneurial phenomena (Hair et al., 2010; Shepherd & Zacharakis, 1999). A conjoint experiment asks respondents to rate on a Likert (e.g. 1–5) scale an opportunity scenario that is composed of a combination of the outcomes (with

Entrepreneurial intention

16

each outcome set as either “high” or “low”), repeating this for different possible outcome combinations, and effectively calculates the utility part-worth for each outcome according to the changes in the scenario rating when one outcome is switched from (e.g.) high to low, with all else equal. Conjoint analysis has been utilised in various studies of entrepreneurial attitudes and intentions subsequently (see, e.g. Douglas, 2013; Douglas & Shepherd, 2000; Fitzsimmons & Douglas, 2011; Monsen, Patzelt, & Saxton, 2010). 1.5.2

The perceived feasibility of entrepreneurial behaviour

The perceived feasibility of entrepreneurial behaviour results from an individual’s introspective evaluation of their self-confidence that they can successfully complete the specific tasks involved in entrepreneurship. It can be parsimoniously measured by ESE which can be gleaned from individuals using multi-item scales in questionnaire surveys (see, e.g. Chen et al., 1998; Zhao et al., 2005). A systematic survey of research on ESE can be found in Newman, Obschonka, Schwartz, Cohen, and Nielsen (2019). Since entrepreneurial tasks can be divided into different task areas (e.g. opportunity recognition, human resource management, financial management) and different stages of the entrepreneurial process (McGee et al., 2009), it seems appropriate to seek the individual’s self-assessment in each of these separate task areas, to allow identification of any task areas where the individual lacks confidence (in their competence), such that they might seek to repair that perceived incompetence or to find one or more partners who are stronger in those deficit areas.

1.6 SUMMARY In this introductory chapter we have laid a foundation for understanding entrepreneurial intention by introducing and clarifying the meaning of important terms and concepts necessary for the discussion of this subject. Entrepreneurship was defined as a behaviour (undertaken by individuals or firms) that can be described in terms of three main vectors, namely proactivity, risk taking, and innovation. Innovation is exhibited by new products (goods and services), new business processes, and/or new business models. The goal of entrepreneurship is to create new wealth, which may accrue totally to stakeholders of the firm (self-oriented or commercial entrepreneurship), or to others external to the firm (social entrepreneurship), or to both (hybrid entrepreneurship). This new wealth may be (net) monetary or non-monetary benefits, which can be aggregated into the common denominator of psychic utility that is derived by individuals from both monetary and non-monetary income. A little reflection will reveal that the other two dimensions of entrepreneurship, namely proactivity and risk taking, are interdependently intertwined with

Introduction

17

innovation. The word innovation derives from the Latin “innovo” meaning introducing something new, and one must be proactive to be first to achieve newness, otherwise the new product, process, or business model would not be innovative (if someone else had already introduced that change). And because the innovation is new, it is necessarily relatively risky, with little or no prior experience to rely on for guidance or information from others that would reduce the uncertainty faced by the innovating entrepreneur or firm. The intention to undertake entrepreneurship is a personal decision formed by individuals after consideration of the perceived desirability and the perceived feasibility of a particular entrepreneurial opportunity. The opportunity is characterised by personal entrepreneurial alertness of a demand–technology nexus – that is, the knowledge that demand exists for a particular new product, process, or business model, and that a technology exists or can be invented that will allow delivery of the new product, process, or business model to those demanding it.

2. Taxonomies of entrepreneurial behaviour 2.1 INTRODUCTION There is considerable heterogeneity on both sides of the individual–opportunity nexus (Gartner, 1985; Shane & Venkataraman, 2000; Venkataraman, 1997). Individuals vary greatly in their preferences and their capabilities, and entrepreneurial opportunities also vary greatly, both in what they require from the individual and what they offer to the individual as outcomes. Using the terms introduced in Chapter 1, this can be restated by saying that different entrepreneurs have different attitudes and different abilities, and that they face different opportunities that provide different profit, psychic-income, and social-benefit outcomes. And of course there are personal and situational differences for individuals that will also impact their choice to become an entrepreneur. Within the wide diversity of entrepreneurial endeavours that are evident there are nonetheless several main clusters, or types of entrepreneurship, which differ noticeably from each other, but within each cluster we observe broadly similar motivations and outcomes. In this chapter we examine this range of entrepreneurial types, and moreover suggest additional types that may be distinguished and which may prove useful for extending entrepreneurship theory. We should expect that heterogeneous intending entrepreneurs will tend to gravitate toward different types of entrepreneurship. Self-determination theory (Deci & Ryan, 1985; Gagné & Deci, 2005; Ryan & Deci, 2000) implies that intending entrepreneurs will seek out a particular type of entrepreneurship, and within that type choose a specific entrepreneurial opportunity that supplies the desired outcomes in a combination that best suits the individual’s preferences for those outcomes, while also suiting their capabilities for the tasks required, as we will discuss in Chapter 3. 2.1.1

Necessity versus opportunity entrepreneurs

A distinction is made in the literature between entrepreneurial opportunities that involve a choice between employment and self-employment, compared to those that offer a choice between self-employment and unemployment. In 18

Taxonomies of entrepreneurial behaviour

19

the latter case, known as “necessity entrepreneurship”, there is no accessible employment opportunity for the individual to earn income, yet the individual needs cash to buy food, clothing, and shelter for themselves and their family, and thus has no choice but to act entrepreneurially to avoid hunger, sickness, and perhaps even death (Brewer & Gibson, 2014). Such dire circumstances are found in poor nations, or in poor regions of nations, and in any nation in the event of a natural disaster, or political or military conflict which impacts the economic infrastructure and causes the economy to be unable to provide enough jobs for the local populace (see, e.g. Block, Kohn, Miller, & Ullrich, 2015; Kautonen & Palmroos, 2010; Margolis, 2014; Reynolds, Camp, Bygrave, Autio, & Hay, 2002). In such circumstances, individuals must undertake entrepreneurship in order to survive. Fishing, catching birds and animals, and picking fruit or harvesting plant crops are means by which the individual can obtain a product (hopefully legally) that can be sold to raise cash. Similarly, making and selling handicraft items can provide cash (or items to barter) that will provide some cash or products needed for the necessities of life. Also, unemployed individuals may be able to provide services to others (who can afford to pay), such as household services, personal services, entertainment services, transportation services, security services, and caretaking services (i.e. minding children, the elderly, or animals), in return for cash, food, or shelter. In extremely dire situations, there will be too few such income-producing opportunities for everyone to create a viable self-employment business venture, and many people may continue to suffer deprivation without respite, unless public funding and/or social entrepreneurship are harnessed to help the needy. A recent paper by Dencker, Bacq, Gruber, and Haas (2019) expands the concept of necessity entrepreneurship to include skilled individuals who have been displaced from their employment position and are unable to secure a replacement role at another organisation. This might arise due to the technological disruption, strategic downsizing, or financial failure of the employing firm. Rapid technological change may cause an imbalance between the supply of and demand for individuals with previously valued (legacy) skills and experience. The shrinkage of demand for such skills and experience by employers forces displaced individuals in need of income to sustain their lifestyle to choose between unemployment and self-employment in roles such as private consulting, or utilising an alternative skill set.1 1 This is an extension of the basic definition of necessity entrepreneurship, which has traditionally been interpreted as meaning that the individual has no possible employment possibilities. One might expect that displaced or disrupted professionals would have other skills, perhaps much more basic, such as painting, cleaning, or caring for animals or children, that would allow them to secure a job somewhere.

20

Entrepreneurial intention

“Opportunity” entrepreneurship, on the other hand, refers to situations where the individual could alternatively find an income-producing job as an employee of an existing organisation, but instead chooses to create a self-employment situation in a new business venture (see, e.g. Block & Wagner, 2010). So opportunity entrepreneurship refers to situations where the individual has the choice between self-employment and wage employment, while necessity entrepreneurship refers to situations where the individual has no choice but to seek self-employment. In both cases the individual potentially may have a choice among more than one entrepreneurial opportunity (Douglas & Shepherd, 2000). The searching, planning, marshalling, and implementing phases of the entrepreneurial process (McGee et al., 2009; Stevenson et al., 1985) will apply to both necessity and opportunity entrepreneurship, probably with more urgency in the former situation (Carsrud & Brännback, 2011; Dencker et al., 2019). 2.1.2

Individual versus organisational entrepreneurship

Entrepreneurship is a behavioural pattern that can be exhibited by both individuals and organisations. Individuals can start entrepreneurial new ventures; or behave entrepreneurially as the owner/manager of an existing firm; or behave entrepreneurially as an employee of an existing firm (which is known as intrapreneurship); or collaborate with others in a management team within an existing firm such that the firm behaves entrepreneurially (which is known as corporate entrepreneurship). Within each of those contexts, individuals might choose to be profit maximisers, profit satisficers, socially oriented, or some hybrid combination of these types of entrepreneurs.

2.2

PROFIT-MAXIMISING ENTREPRENEURS

Classical economics represented all firms as being owner-operated, with individuals each striving to maximise their profit in order to better feed, clothe, and shelter their families (Barreto, 2013; Rashid, 1986; Smith, 1950 [1776]). As firms became larger, to raise capital for expansion entrepreneurs would sell shares in those firms to investors, creating “joint-stock” corporations and the rise of the managerial class, with firms being owned by many people but This extended definition seems to include the proviso that the available jobs are unacceptable to the individual for reasons of insufficient pay, onerous working conditions, or other negative attributes. In this case the employment versus self-employment decision choice outlined by Douglas and Shepherd (2000) seems applicable, and the choice of self-employment is effectively a case of “opportunity entrepreneurship” rather than necessity entrepreneurship.

Taxonomies of entrepreneurial behaviour

21

managed by relatively few hired managers. With this separation of ownership and control, neo-classical economics shifted the focus to the “theory of the firm” where hired managers became the agents of the principals of the firm, with the presumption that these managers should strive to maximise profits for distribution as dividends to the shareholders of the firm (Douglas, 1989; Holmstrom & Tirole, 1989; Jensen & Meckling, 1976). Alongside these larger firms, new and small firms continued to arise as entrepreneurs created new products, new production processes, and new business models. Some of these firms would grow rapidly to become major corporations while others would perish in heavily contested marketplaces. These entrepreneurs were also presumed to want to maximise their profits, since not striving for lower costs and higher prices would place the firm’s survival at risk in monopolistically competitive or oligopolistic markets (see, e.g. Samuelson, 1948: 492). Also, the utility theory of consumer behaviour stressed that individuals would want to maximise their utility, and since shareholders of firms are consumers too, and profits are instrumental to the purchase of utility-providing goods and services, higher profits will be preferred to lower profits. This notion of the profit-maximising entrepreneur continues to be the mainstream perception of commercial entrepreneurs today, as distinct from profit-satisficing entrepreneurs (Simon, 1972) and social entrepreneurs (e.g. Austin et al., 2006). Simon (1972, 1979) argued that under conditions of uncertainty, where much of the information needed to practise profit maximisation is expensive or impossible to find, managers and entrepreneurs must practise “bounded rationality” and make profit-maximising decisions within the boundary of information that is economically rational to obtain. Given the increasing marginal cost of the desired information, information-search activity should not be taken beyond the point where the last dollar spent on information acquisition (e.g. market research) generates only one dollar in additional revenue, since further information search would actually reduce profit. Accordingly, when we say profit maximisation we mean that the entrepreneur (or the firm) seeks to make as much profit as possible subject to information-cost constraints. 2.2.1

Growth entrepreneurs

The motivation to grow the entrepreneurial firm is largely based on the motivation to maximise profit, since replicating the production and selling functions of the firm allows access to economies of scale (i.e. of plant size), economies of scope (due to greater product-line width), purchasing economies (e.g. buying materials in bulk), and specialisation in worker tasks including the more efficient utilisation of discipline-specific managers (Aghion & Howitt, 2008). These sources of cost reduction per unit of output allow greater profit per unit and hence greater profit for the firm overall. In addition to the utility

22

Entrepreneurial intention

to be derived from the use of the profit (distributed as dividends) to purchase goods and services, the entrepreneur may desire growth for its own sake, as a source of pleasure and pride that provides utility directly to the entrepreneur (as a component of job satisfaction). Profit facilitates growth of the firm, since the growing firm must be able to expand its plant and equipment, and hire more employees, in order to produce more output and grow its sales. The existence of accumulated profits allows the firm to fund growth from internal sources, and thus avoid more expensive external debt or the further sale of equity in the firm. But for some firms, growth precedes profitability (Davidsson et al., 2009), in which case the prospect of growth (and future profit) encourages firms to forego short- to medium-term profit and expand the firm’s employment and sales in the pursuit of longer-term profitability. Without sufficient accumulated profit, firms in this situation must access external debt or equity funding to achieve growth, but choose this path to prevent other firms from entering the market and becoming a competitive threat in the future. In markets where firm growth is possible, pioneer firms must grow or they will be outgrown by later entrants to the market who may subsequently pose a competitive (or takeover) threat to the pioneer firm. 2.2.2

Gazelles and unicorns

Some firms enjoy rapid growth and have been called “gazelles” for their ability to accelerate very quickly (see, e.g. Bos & Stam, 2013; Moreno & Casillas, 2007). One definition of gazelles is high-growth firms that increase their sales by at least 50 per cent during three consecutive years (Henrekson & Johansson, 2010). A “unicorn” implies rarity – unicorn firms have been defined as privately held start-up firms that have a current valuation of 1 billion dollars or more (see, e.g. Bryan, Tilcsik, & Zhu, 2017). While all intending entrepreneurs might wish for a gazelle or unicorn outcome, very few actually achieve these outcomes. Thus, some profit-maximising individuals will form the intention to be a “growth-oriented” entrepreneur (Douglas, 2013), seeking to maximise profits over the longer term, which can be operationalised as the maximisation of the net present value of cash flows. Other profit-maximising individuals will form the intention to maximise profit over a shorter time horizon, seeking higher profit in the near future and risking increased rivalry from competitors (and thus suffering lower profits) in more distant years.

Taxonomies of entrepreneurial behaviour

2.3

23

PROFIT-SATISFICING ENTREPRENEURS

Profit-satisficing entrepreneurs settle for a satisfactory level of profits rather than continue to incur search costs in pursuit of profit maximisation (Simon, 1972). Yet such “satisficing” entrepreneurs may not push information-search costs to the point where the marginal dollar of search costs equals the marginal profit of the last unit of output, since they also take into account the non-monetary costs of information-search activity (i.e. the marginal disutility of information search), and that information-search activity has an opportunity cost – it comes at the expense of time that might be spent at leisure or gaining greater job satisfaction. Accordingly profit satisficers seek to make sufficient profit to meet their expenditure needs, and augment the utility derived from that profit with the utility derived from the non-monetary benefits gained as part of the entrepreneurial process. Profit satisficers can be reframed as utility maximisers if they choose the combination of profits and other sources of utility that maximise their total utility. The theory of self-determination (Deci & Ryan, 1985) argues that people choose actions that are expected to maximise their personal wellbeing, which is to say, their psychic utility (Steel & König, 2006). Accordingly, we can imagine profit satisficers to be firms seeking a combination of profit and other outcomes of the entrepreneurial process that are expected to generate, for them personally, the best utility outcome, where utility is derived both from profit (via purchased goods and services) and from non-monetary sources of satisfaction produced by the entrepreneurial process. 2.3.1

Independence-oriented entrepreneurs

Independence entrepreneurs value their decision-making autonomy very highly, and are likely to form entrepreneurial intentions for opportunities that offer large scope for independent decision making (Cooper, Woo, & Dunkelberg, 1989; Douglas, 2013). They expect to derive utility from “being their own boss”, or conversely they value “not having to take orders” from others. Making up their own mind means that they can choose where to conduct business; with whom they conduct business (i.e. which customers, suppliers, and employees); and when they conduct business (i.e. days and hours of operation each week, and scheduling of vacation periods). This proclivity for decision-making autonomy also means that they may be reluctant to take on business partners or equity investors with whom they would need to share decision making, and as a result they may be disinclined to grow their firms to any substantial size.

24

Entrepreneurial intention

Accordingly we expect that a person who has very strong preference for (i.e. attitude towards) decision-making autonomy is likely to form the intention to start a new venture that offers relatively high personal flexibility of decision making, work location, work scheduling, and so on. But as we shall see in Chapter 3, this attitude is but one of several factors that influence the formation of entrepreneurial intentions, and in some cases it might be subjugated to other considerations that collectively provide even greater expected utility to the intending entrepreneur. 2.3.2

Lifestyle entrepreneurs

Lifestyle entrepreneurs are those who conduct a business that is based on their preferred leisure, sporting, or other activities, such as ski instructors, horse-riding instructors, tennis coaches, cooking schools, tour directors, and so on (Barringer & Ireland, 2006). In such businesses they are likely to feel high self-efficacy given their extraordinary knowledge of the production and operations aspects, and are also able to communicate with their customers more effectively. Working with like-minded suppliers, customers, and fellow team members, in a work environment they enjoy, is likely to allow greater job satisfaction and thus greater perceived desirability of self-employment opportunities that are congruent with the individual’s lifestyle preferences. Thus, individuals with a strong preference for particular lifestyle pursuits may be expected to form entrepreneurial intentions for an opportunity that allows that lifestyle activity to be turned into a business, and thereby provide a work environment that they would enjoy more than any other new venture or wage-employment opportunity. 2.3.3

Salary-substitute entrepreneurs

Salary-substitute entrepreneurs are defined as those who work for themselves doing the same work that they would do for others if instead they were employed (Birch, 1987). Accountants, lawyers, personal trainers, plumbers, carpenters, and a host of other professionally trained people might ply their trade in the employ of an existing firm, or alternatively “hang out their shingle” and trade under their own business name (Barringer & Ireland, 2019: 45). Because of their training in a particular technical field they will have relatively high self-efficacy in that task domain and are likely to find attractive an entrepreneurial opportunity that offers them the chance to work in a field in which they can satisfy their need for competence (Deci & Ryan, 1985). Their income and profit objective might be simply to earn as much as they could have earned in that profession as an employee, while at the same time experiencing greater job satisfaction and decision-making autonomy.

Taxonomies of entrepreneurial behaviour

2.3.4

25

Leisure-seeking entrepreneurs

Leisure seekers are those who derive a relatively large amount of psychic utility from not working but instead engaging in leisure pursuits. Yet, unless they are already wealthy, they must work to earn money for food, clothing, and shelter, so they are likely to adopt a satisficing approach to earned income, working only as much is necessary to pay for what they regard as the bare necessities of life (and to afford their leisure pursuits) (see, e.g. Carree & Verheul, 2012). The individual’s tradeoff between leisure and work is an old issue in economics, and basically pits the marginal utility from an extra hour’s work (from the goods and services bought from the income earned) against the marginal utility from an extra hour’s leisure (see, e.g. Kool & Botvinick, 2014; Musgrave, 1974). Leisure-oriented individuals may not wish to work full-time, especially if their expected utility from goods and services is relatively low and their expected utility from leisure is relatively high, and so may therefore prefer part-time work. While part-time employment might require adherence to particular hours of work, some entrepreneurial opportunities may offer part-time work plus the flexibility of work times (e.g. busking, beachcombing, writing, painting) and thereby could be quite attractive to this type of intending entrepreneur.

2.4

SOCIAL-PURPOSE ENTREPRENEURS

Social-purpose ventures conduct business for the primary purpose of providing social benefits to others, external to the firm, who are economically or socially disadvantaged (see, e.g. Austin et al., 2006; Bacq & Alt, 2018; Mair & Marti, 2006; Miller et al., 2012). The legal form of the venture may be “not for profit” or “for profit” but, if the latter, they are not likely to be profit maximising, since they give priority to the generation of social benefits, and greater profit is most likely to be gained at the expense of their primary mission to provide social benefits. For-profit social ventures generally seek a satisfactory level of profit to ensure that they can continue their social mission in subsequent months and years. For some social entrepreneurs it may be satisfactory to have a zero profit (or even a loss as the individual’s personal contribution to funding the social outcomes), while for other social ventures the owners may seek a target (satisfactory) financial return, for example, comparable to the yield on government bonds. Yet others may seek to maximise their profit secondarily, i.e. after having achieved their social impact targets as their primary objective.

26

2.4.1

Entrepreneurial intention

Not-for-profit enterprises

Not-for-profit is not the same as no profit. The latter may happen to profit-seeking entrepreneurs, whether commercially or socially oriented, when their plans to make a profit are not realised. Not-for-profit enterprises have a formal legal status that exempts them from income and some other taxes, as long as they do not distribute any surplus to owners and confine their operations to serving their social mission (Arrow, 2000; Salamon & Anheier, 1992; Shaw & Carter, 2007). They may make surplus income which can be retained within the organisation with a view to spending it subsequently in pursuit of their social mission. Not-for-profit social ventures include redistributive ventures (like the Salvation Army, and various charities) that collect money or goods from members of society and distribute these to members of society who are in need. In some cases these ventures may not be very entrepreneurial at all, simply using known methods of collection and distribution to serve their social mission, while in other cases entrepreneurial methods are required to achieve their revenue and cost objectives (Newbert & Hill, 2014; Shane, Locke, & Collins, 2003; Zahra, Gedajlovic, Neubaum, & Shulman, 2009).

2.5

HYBRID ENTREPRENEURS

Hybrid entrepreneurs are those with both a social and a commercial purpose, and in the best of all possible worlds this would characterise all entrepreneurial ventures. Societal wellbeing would be improved if all commercial entrepreneurs also provided social benefits, and if all social entrepreneurs also earned profit (to allow them to continue their social mission and to encourage investment in social enterprises). This view is now becoming mainstream while the “greed is good” philosophy of profit-seeking commercial entrepreneurs suffers declining social approval. Similarly, the ongoing reliance of social enterprises on philanthropy as a funding source, rather than utilising business methods to sustain their enterprises, is also less socially acceptable now in the face of excess demand for social venture funding. Hybrid-social entrepreneurs are those whose primary purpose is to provide social benefits and whose secondary purpose is to earn profit for their owners (Battilana & Lee, 2014). Oppositely, we could say hybrid-commercial entrepreneurs are those whose primary purpose is self-oriented (i.e. to earn personal profit and non-monetary income) but whose secondary purpose is other-oriented, to generate social benefits for others. The delimiting phrase “primary purpose” creates an artificial dichotomy between commercial and social entrepreneurs that is utilised by researchers who prefer to focus on one of these two types in their scholarly discussions. Unfortunately, it often seems

Taxonomies of entrepreneurial behaviour

27

to imply that commercial entrepreneurs can disregard their corporate social responsibility, thus not contributing to the solution of social and environmental problems that could be eliminated. Similarly, this dichotomy may reduce the pressure on social enterprises to accept what some would regard as their commercial business responsibility to avoid reliance on limited sources of philanthropic and charitable giving, thus crowding out other social ventures who are not able to gain philanthropic support and less able to sustain themselves financially. It is, of course, entirely possible that one very large and profitable primarily commercial-purpose enterprise could provide a greater quantum of social benefits than could many small primarily social-purpose enterprises in total. Similarly one small social enterprise may contribute far more to societal wellbeing than many small salary-substitute or lifestyle businesses. Commercial firms exhibiting “corporate social responsibility” may provide social benefits as an operating cost within their business; or as a profit-reducing non-operating expense (such as a corporate donation); or as donations by the firm’s owners from their dividend income to charities and other social ventures. Social enterprises may exhibit commercial responsibility by being as frugal as possible in spending public and/or philanthropic funding; by seeking revenues from the sale of visitor passes, mementos, and so on; or by seeking the voluntary assistance of social-minded workers; and/or by seeking discounts on supplies from social-minded suppliers.

2.6 INTRAPRENEURSHIP Intrapreneurship refers to entrepreneurial behaviour within an existing firm by an individual employee who does not own a significant share of the firm (Hisrich, 1990; Pinchot, 1987). This could involve the experimental development and introduction of new products or processes, by the employee, which are subsequently integrated into the firm’s operations. In Douglas and Fitzsimmons (2013) we demonstrated that entrepreneurial intention and intrapreneurial intention are separate and distinct constructs, driven by different factors. We found that while self-efficacy is significantly related to both entrepreneurial and intrapreneurial intentions, attitudes to income, ownership, and autonomy relate only to entrepreneurial intentions, while attitude to risk relates only to intrapreneurial intentions. Thus while intending entrepreneurs tend to be more tolerant of risk, intending intrapreneurs tend to be more risk averse, and prefer to let their employer carry all or most of the financial risk associated with innovation. An individual contemplating entrepreneurial behaviour might thus consider intrapreneurship as an alternative to entrepreneurship, particularly if they are more risk averse and do not have particularly strong preference for decision-making autonomy.

28

2.7

Entrepreneurial intention

SPINOFFS AND SPINOUTS

A spinoff is a new company formed by separating out part of an existing company to become a stand-alone company. In most cases the “parent” company would hold a major share of the ownership of the spinoff company and transfer some managers and employees to the new firm. One reason for spinoff companies is that new ventures can be launched as a spinoff firm to pursue innovative opportunities without being constrained by the routines and processes (and possibly conservative mindsets) of the parent company. Spinoffs often involve intrapreneurs from within the parent company who are given the opportunity to transition to an entrepreneurial leadership and part-ownership position in the new venture. Virgin Atlantic is a prime example of a firm that has spun off hundreds of promising new entrepreneurial ventures, giving part ownership and funding support to employees who become the chief executive of the new company (see, e.g. Atienza, 2015; Deprez, Leroy, & Euwema, 2018; Dess & Lumpkin, 2005). Spinouts are a similar situation, in that the new venture is managed by former employees of another company, but where the managers left that company to start their own independent new venture rather than pursue that venture within the “parent” company or as a spinoff of that company. It may be that the intrapreneurs attempted to persuade their bosses to make and sell the new product but were repulsed, or that they left the company with their intellectual property in order to maximise their private gain from their new product idea. A bold example of a spinout is Compaq Computer, which was started in 1982 when the founders of that company were unable to gain corporate approval for a new product (a portable computer) while working for Texas Instruments (an electronic calculator maker), and subsequently resigned from Texas Instruments to start Compaq Computer. This new venture grew to become the largest supplier of personal computers during the 1990s (Davenport, 1998), and in 2002 was sold to Hewlett-Packard for the non-trivial sum of US$25 billion.

2.8

CORPORATE ENTREPRENEURSHIP

Corporate entrepreneurship implies firm-level entrepreneurship, that is, proactive risk-taking innovative behaviour undertaken by the collective decision of the top management team (TMT) of the business entity (Guth & Ginsberg, 1990; Morris, Kuratko, & Covin, 2010; Sharma & Chrisman, 2007). While some would include intrapreneurship in the definition of corporate entrepreneurship, the two constructs can be separated conceptually as follows: intrapreneurship is the practice of entrepreneurship at the individual level by an employee within an organisation, perhaps not congruent with company policy

Taxonomies of entrepreneurial behaviour

29

and perhaps without the approval of the TMT, while corporate entrepreneurship is the practice of entrepreneurship at the firm level of analysis with the awareness and approval of the TMT and intertwined with the firm’s corporate strategy. One of the most celebrated historical examples of corporate entrepreneurship is the 3M Corporation (see, e.g. Garud, Gehman, & Kumaraswamy, 2011; Govindarajan & Srinivas, 2013), which has endured for over 100 years thanks to ongoing product and process innovation guided by its TMTs in each era.

2.9

PAST, PRESENT, AND FUTURE RESEARCH INTO TYPES OF ENTREPRENEURSHIP

In the (distant) past, the classical economist Richard Cantillon introduced the term “entrepreneur” for a person who acquires means of production at certain prices with a view to selling (a product) at uncertain prices. Jean Baptiste Say stated that the entrepreneur’s function is to combine the factors of production into a producing organism – i.e. manage the business firm – while John Stuart Mill solidified the notion that risk bearing was also an intrinsic function of the entrepreneur. Adam Smith distinguished between the capitalist’s return on the use of their capital (i.e. interest) and their return for risk bearing (i.e. profit), see Schumpeter (1954: 555). Moving into the twentieth century Knight (1921) clarified the distinction between risk and uncertainty, and Schumpeter (2000) in 1934 stressed the role of innovation as a source of entrepreneurial profit. As the notion of the entrepreneur continued to bloom, there was no doubt in anyone’s mind that the entrepreneur’s motivation was to maximise profit – in line with neo-classical “utility theory” whereby human behaviour was explained in terms of utility maximisation, since profit was instrumental for the purchase of goods and services from which utility was derived. Thus in the past, the theory of the firm evolved with the conception of the profit-maximising entrepreneur at the helm, and this monolithic perception of the entrepreneur persisted until recent decades. Nowadays, as outlined earlier in this chapter, we distinguish between self-oriented “commercial” entrepreneurs and other-oriented “social” entrepreneurs on the basis of their primary purpose being profit making or social-benefit provision, respectively. We also distinguish between profit-maximising and profit-satisficing commercial entrepreneurs, whose primary purpose is profit seeking or psychic-income seeking, respectively. Thus, at the current time we recognise that entrepreneurs are unlikely to have a single goal, but are more likely to pursue a primary goal (e.g. social impact) with tradeoffs for secondary (e.g. profit) and tertiary (e.g. job satisfaction) goals. Also, we currently recognise the interdependence of these goals within a holistic process of human decision making, rather than envisioning them as discrete, linear-additive elements that independently

30

Entrepreneurial intention

underpin the motivation of the individual to become an entrepreneur (despite the fact that correlational analytical methods treat the antecedent variables as discretely separate causes of entrepreneurship intention). In the future we should surely move to a fully integrated conception of entrepreneurship in which the multiple goals of the individual combine interdependently to determine their formation of entrepreneurial intention for particular types of entrepreneurship. At present we recognise that the three main goals are personal wealth, social impact, and personal wellbeing, which culminate in the three main types of entrepreneur mentioned above. In the following chapters of this book I will add a fourth motivation (and goal) into the mix, namely the motive to innovate in pursuit of the goal of local or global recognition as the pioneer provider of a radically innovative new product, service, or business process. Currently, innovation is treated in the literature simply as an instrument for profit making and/or social-benefit provision. In this book I suggest that innovation is also desired (more or less strongly) by different individuals for its own sake, in the same way that profit, social-benefit provision, and (e.g.) lifestyle outcomes are desired, as a source of psychic satisfaction (utility) for the enterprising individual. Casual empiricism supports the supposition that the prospect of global recognition as a “gazelle” or “unicorn” resulting from achieving domination in the global marketplace surely motivates some entrepreneurs, in addition to their profit, social-impact, and job-satisfaction motivations. Thus the monolithic profit-maximising entrepreneurship construct of the past has morphed into three main types of entrepreneurship in the literature of the present era, namely profit-seeking, psychic income (job satisfaction)-seeking, and social impact-seeking entrepreneurship. Scholars have also noted various “niche” types of entrepreneur, such as mum-preneurs (Duberly & Carrigan, 2013); migrant-preneurs (Levie, 2007); elder-preneurs (Kautonen, Luoto, & Tornikoski, 2010; Watkins-Mathys, 2012); refugee-preneurs (Wauters and Lambrecht, 2008); serial entrepreneurs (Westhead, Ucbasaran, Wright, & Binks, 2005); and international entrepreneurs (Joardar and Wu, 2011). These are likely to be distributed across the three main types as distinct subtypes. For example, some elder-preneurs might be primarily profit seeking, others primarily social-impact seeking, and yet others primarily psychic-income seeking. Similarly, those desirous of intrapreneurship are likely to form a preference for intrapreneurship in a particular type of entrepreneurial firm that offers the intrapreneurial outcomes in a mix that best suits the preferences of the individual. In this book a fourth main goal of intending entrepreneurs is added for future researchers to consider. Four main outcomes of entrepreneurship, viz profit, job satisfaction, social benefits, and innovation, can each be regarded as legitimate goals of entrepreneurship, and we might expect individuals to be moti-

Taxonomies of entrepreneurial behaviour

31

vated to become an entrepreneur by any combination of these four goals. Just as we have argued that the primary objective of the social entrepreneur is to provide social impact, and the primary objective of the commercial entrepreneur is to gain profit, other individuals might be primarily motivated to achieve innovation, or to gain job satisfaction, suggesting four main types of entrepreneur based on their primary motivation. In all four primary types we should expect to find individuals expressing different combinations of a second, third, and fourth priority among the remaining three goals, giving rise to a typology of subtypes of entrepreneurial intention. Further, by investigating within these subtypes at the case level, we might expect different combinations of these goals (i.e. configurations) that operate to drive the individual’s formation of entrepreneurial intention. Future research opportunities include the empirical testing and refinement of a general model of entrepreneurship (as in Chapter 8 here) where heterogeneous individuals are motivated by a combination of four main goals that interdependently motivate the pursuit of four particular types of entrepreneur, namely primarily profit seeking, primarily innovation seeking; primarily social-impact seeking, and primarily job-satisfaction seeking. Within each of these main types there is likely to be several if not dozens of distinct subtypes, including the niche subtypes mentioned just above, that represent alternative pathways to entrepreneurship. It will be argued in Chapter 9 that symmetrical correlational methods (such as multiple regression and structural equation modelling) are unable to uncover much of the finer-grained detail needed to identify these alternative pathways. In that chapter a relatively novel new analytical method – fuzzy-set qualitative comparative analysis – is used to demonstrate how additional information can be gleaned from sample data to reveal alternative pathways to the same entrepreneurial outcome.

2.10 SUMMARY In this chapter we have considered the various types of entrepreneurship that individuals might intend to undertake. The main definitional dichotomy is commercial versus social entrepreneurship – the former is primarily self-oriented and the latter is primarily other-oriented. Commercial entrepreneurs seek profit and psychic benefits for themselves, and may seek to provide social benefits as a secondary purpose, while social entrepreneurs seek to provide social benefits to needy people and causes, and may seek to earn profit as a secondary purpose. Within the category of commercial entrepreneurship, there are profit maximisers, who focus on profit taking, and may offer social benefits if that does not compromise profit maximisation. Similarly, within social entrepreneurship, there are social-impact maximisers, for whom profits

32

Entrepreneurial intention

may be zero if profit taking tends to reduce social impact. Thus we may expect to see not-for-profit social organisations and not-making-any-profit social ventures coexisting with hybrid-social ventures who do seek to take profit as an objective that is secondary to their social purpose. There are also profit-satisficers within the realm of both commercial and social entrepreneurship, who seek to earn a satisfactory (target) level of profit and gain some significant part of their satisfaction from the non-monetary rewards (job satisfaction) associated with the conduct of entrepreneurship. Among commercial entrepreneurs, one such subcategory is independence-oriented entrepreneurs, for whom decision-making autonomy is likely to be very important. Another is salary-substitute entrepreneurs who could do similar work as an employee of an existing firm, but for whom decision-making autonomy and the flexibility of work choices associated with self-employment is likely to be very important. A third subcategory is lifestyle entrepreneurs, who seek to combine their work and lifestyle preferences, not necessarily making a lot of money but being sufficiently compensated by the utility they derive from work enjoyment. Finally we considered leisure-oriented entrepreneurs, who find entrepreneurship the best option for minimising the time they spend working and thus maximising the time and/ or flexibility of time remaining for leisure pursuits. We should expect to find similar submotivations among social entrepreneurs, and there are likely to be hybrid combinations of these main subtypes of profit-satisficing entrepreneurs. Intrapreneurs act entrepreneurially within an employment role in an existing firm or organisation. In reality this may occur with or without the sanction or oversight of their TMT. Intrapreneurs may be given limited autonomy by the TMT to experiment with new products or processes, or alternatively may operate “under the radar” until they are prepared to propose a new product or process to the TMT (or to spin out as a separate new venture). Corporate entrepreneurship involves employed individuals working in teams (either the TMT or subgroups reporting to the TMT) to develop new products, processes, or business models. It is a firm-level behaviour, carried out by individuals acting in concert as part of the firm’s strategy. Social entrepreneurship may also be conducted by intrapreneurs or corporates, of course. Individuals may choose to be employed by an organisation that affords the possibility of (individual) social intrapreneurship or (group) corporate social entrepreneurship, as an alternative to more risky and more difficult social entrepreneurship self-employment opportunities that they perceive. In this chapter, after we considered the various types of entrepreneurial behaviour that represent the current state of the mainstream entrepreneurship intentions literature, we considered what the future of entrepreneurship intention research might look like. It is proposed that we should strive for an integrated general model of entrepreneurship which is broad enough to include

Taxonomies of entrepreneurial behaviour

33

the three main types already identified, as well as a fourth type to be further articulated, namely the intending entrepreneur seeking global pioneership. Individuals would consider the interplay of the four main motivations for entrepreneurship (viz profit, psychic income, social outcomes, and innovation) and choose the type that, overall, best serves their personal wellbeing. We have implied that the attitudes and abilities of individuals will tend to align those individuals with a particular type of entrepreneurship. But as we shall argue briefly in Chapter 3 and more fully in Chapter 9, the determinants of entrepreneurial intention are a fuzzy set of attitudes, abilities, and personal and contextual factors, such that it is the overall total of expected wellbeing (total utility) expected from the available entrepreneurial opportunities that determines which (if any) of those opportunities the individual might form an intention to pursue, rather than any simple driver of entrepreneurial intention.

3. Why do individuals want to be entrepreneurs? 3.1 INTRODUCTION In this chapter we examine in more depth what motivates individuals to form the intention to act entrepreneurially. In simplest terms, individuals who have discretion over their actions choose to do what they want to do, within the constraints that restrain their actions. Individuals, being basically self-serving, tend to do things that increase their own “psychic satisfaction” or internal sense of wellbeing.1 This psychic satisfaction was called “utility” by the classical economists (see Schumpeter, 1954: 1053–69), who hypothesised that individuals will tend to choose among items and actions in order to maximise the total utility they can obtain from their limited income and other sources of wealth. The classical economists also recognised that some items and actions necessarily involve outcomes which are disliked by the individual, such as the risk that accompanies entrepreneurial profits, and coined the term “disutility” for the psychic dissatisfaction incurred (by the bearing of risk, in this example). The choice of an item or action that generates more than one outcome and potentially negative outcomes thus depends on the net utility expected (ex ante) from the item or action. Individuals want things, first, because they have needs. Whereas early needs theories (Herzberg, 1964, 2005; Maslow, 1943) include physiological needs (e.g. food, shelter, safety), later needs theories focus on psychological needs, arguing that even the physical needs are driven by psychological needs (Deci & Ryan, 1985). These needs enter the individual’s personal objective function, such that fulfilling these needs increases the individual’s sense of wellbeing. As Steel and König (2006: 895) state, “needs represent an internal energy force that directs behaviour towards actions that permit the satisfaction and release of the need itself”. Self-determination theory, as developed by Deci and

1 This is not to exclude social entrepreneurs (or commercial entrepreneurs with corporate social responsibility) who act to increase the psychic satisfaction of others, since it is argued that entrepreneurs may personally gain psychic satisfaction from helping others.

34

Why do individuals want to be entrepreneurs?

35

Ryan (1985), Ryan and Deci (2000), and Gagné and Deci (2005), postulates that individuals seek to maximise their wellbeing by taking actions that are expected to satisfy their innate and situational psychological needs. Thus, we must ask what psychological needs of the individual are served by entrepreneurial behaviour, if we are to apply self-determination theory to the formation of entrepreneurial intentions. In addition to needs, humans also have preferences, and choose among different foods, shelters, and courses of action (that would satisfy their needs) on the basis of “liking” some alternatives more than they “like” other alternatives. By “like” we mean they gain psychic satisfaction (i.e. utility) from the consumption of the particular goods and services that are available to them, and gain greater psychic satisfaction from some alternatives than they do from others. For example, a person may like beer better than they like wine, and like brand-A beer better than they like brand-B beer. Since choices among alternatives are made in advance of consuming (i.e. experiencing) the alternative chosen, the choice must be made on the basis of the anticipated psychic satisfaction, which may or may not be matched by the actual satisfaction experienced, ex post.2 3.1.1

Utility theory of entrepreneurial behaviour

Baumol (1990) argued that the level of entrepreneurial activity in a nation’s economy depended on the reward structure facing the potential entrepreneur, or what he called “the payoff” to entrepreneurship. He argued that when the business and societal environment is more conducive to entrepreneurship (by offering greater wealth, power, and prestige), the number of entrepreneurs will increase, and conversely. Thus Baumol considered not only the profit-seeking motive for entrepreneurship but also alluded to the non-monetary (psychic) benefits of entrepreneurship, without explicitly positing a utility-maximising model, although he was fully aware of classical utility theory. Kihlstrom and Laffont (1979) had earlier argued in an abstract general equilibrium model that risk attitude is an important determinant of entrepreneurial behaviour, being a psychic cost associated with entrepreneurship. They postulated that less risk-averse individuals would opt for self-employment while more risk-averse individuals would opt for employment.

2 Unsatisfactory experience with a product or service may be compensated for by the vendor, but more commonly the buyer bears the disappointment and incorporates the new expected utility information into their stock of knowledge, which may or may not be reliable for the subsequent experience with that product or service. See Darby and Karni (1973) and Nelson (1970).

36

Entrepreneurial intention

Campbell (1992) recognised the negative influence of risk on the intention to be an entrepreneur by incorporating a risk premium into the calculation of the net present value of the projected income stream associated with the entrepreneurial opportunity. Eisenhauer (1995) proposed an expected-utility model instead of the net present value model, whereby the entrepreneur expects utility from both profit and a monolithic variable he called the “working conditions” associated with entrepreneurship. In Douglas and Shepherd (2000) we disaggregated the working conditions into (the utility-generating) income, independence, and perquisite outcomes and the (disutility-generating) risk exposure and work-effort outcomes. Subsequent empirical studies have tested for the statistical association between these hypothetical antecedents and entrepreneurial intention, and will be examined in later chapters. While economists continued to develop the theory of psychic utility into an all-too-elegant mathematical model, across a disciplinary divide the psychology theorists were also trying to explain human behaviour in terms coined by their own intellectual ancestors (Steel & König, 2006). Expectancy-valence theory (Vroom, 1964) argued that human choice among alternatives depends on the arithmetic product of the individual’s expectancy of the outcome and the psychological importance (valence) of that outcome to the individual. We shall see below that the product of the expectancy and the valence is equivalent to the expected utility of the contemplated item or action. Expected utility, estimated by the application of expectancy-valence theory, forms the theoretical foundation for the “perceived desirability” of the entrepreneurial act, which, along with the perceived feasibility of the entrepreneurial act, are the two main pillars of the theory of planned behaviour (Ajzen, 1991; Krueger et al., 2000). Expectancy-valence theory provides a mechanism by which the expected increase in total utility (generated by the contemplated action) can be measured conceptually. Given that self-serving individuals are presumed to be utility maximising, the greater the expected utility of an action, the more desirable the action is perceived to be. Thus, the expected utility of an action is a measure of the perceived desirability of that action, and the individual will be inclined to choose the action that seems most desirable in prospect (Douglas & Shepherd, 2000). Accordingly, in the next main section of this chapter we focus on the perceived desirability of entrepreneurial action. This discussion involves self-determination theory and elucidation of the individual’s innate and situational needs that may be satisfied by the conduct of entrepreneurship. Expectancy-valence theory is applied to demonstrate that the perceived desirability of an entrepreneurial action (such as the formation of entrepreneurship intention) can be measured and tested empirically. In the third main section of the chapter we consider the perceived feasibility of the contemplated entrepreneurial behaviour and the factors underlying that

Why do individuals want to be entrepreneurs?

37

construct. There we will examine more closely the individual’s self-efficacy (Bandura, 1982) for entrepreneurial behaviour and consider scales by which the individual’s entrepreneurial self-efficacy (ESE) can be measured. This leads to consideration of self-regulation theory (Higgins, 1998) and how that theory might apply in the context of entrepreneurship (Brockner, Higgins, & Low, 2004). We explain that the individual’s perceived feasibility of entrepreneurial action might be boosted initially by promotional self-regulation (in the opportunity recognition phase) and later reduced (in the viability screening phase) by preventative self-regulation (Fitzsimmons & Douglas, 2011).

3.2

THE PERCEIVED DESIRABILITY OF ENTREPRENEURSHIP

As Krueger et al. (2000) so eloquently articulated, Ajzen’s (1991) theory of planned behaviour and Shapero’s (1982) theory of the entrepreneurial event can be reconciled into a consistent framework of two main determinants of entrepreneurial intention, namely perceived desirability and perceived feasibility. Ajzen postulated a third element in the theory of planned behaviour model, namely subjective (social) norms, which can be incorporated into the perceived desirability construct – potential entrepreneurs might gain utility (or avoid disutility) by conforming with subjective social norms, or alternatively expect to gain utility by not conforming with the expectations of others and choosing an entrepreneurial action that identifies them as an independent thinker who does not want to conform with societal expectations. Similarly, Shapero’s third element in the theory of the entrepreneurial event model, namely the “propensity to act”, can be incorporated into the perceived feasibility construct, as it is likely to be related to the individual’s self-efficacy that they can successfully achieve the tasks required in starting and operating a (desirable) new venture. Self-determination theory (Deci & Ryan, 1985; Gagné and Deci, 2005; Ryan and Deci, 2000) provides a theoretical basis for the perceived desirability of entrepreneurial behaviour. Deci & Ryan (1985: 230) state “the functional aim of all behaviour can be understood as need satisfaction” and that “people will tend to pursue goals, domains, and relationships that allow or support their need satisfaction”. They argue that humans have three innate needs, namely the need for autonomy, the need for competence, and the need for relatedness, as well as situational needs, which we shall discuss in turn. 3.2.1

Need for autonomy

“Autonomy involves acting with a sense of volition and having the experience of choice” (Gagné & Deci, 2005: 333). Frese (1989) suggests that control over one’s behaviour (i.e. decision latitude) combined with optimal complexity of

38

Entrepreneurial intention

the task leads to optimal wellbeing. Entrepreneurship is noted for the greater decision-making autonomy that it affords the individual, relative to employment, since the entrepreneur can “be one’s own boss” and not have to take orders or follow directions from senior managers as would occur in an employment situation. Yet entrepreneurs may need to follow directives, or at least acquiesce to the demands, of the tax collector, regulators, investors, strident customers, employee organisations, and other suppliers. The degree of autonomy that can be attained might vary from zero (when one is fully controlled by others or by circumstances) to total, where the decision maker has full latitude in decision making. Thus the extent of decision-making autonomy supplied by entrepreneurial new ventures will vary across entrepreneurial opportunities. On the other side of the individual–opportunity nexus, the amount of autonomy preferred by individuals is likely to vary across individuals, being greater or lesser depending on the importance to them of decision-making autonomy. This valence can be encapsulated in the individual’s “attitude to autonomy” – i.e. how they think and feel about having and utilising decision-making autonomy. For some, autonomy might generate a relatively large amount of utility, while for others autonomy might generate a relatively small amount of utility after netting out the negative aspects of autonomy, such as stress and the requirement to deal with difficult personnel matters. Thus, on the surface, we might expect the individual to be attracted towards entrepreneurial opportunities that best match their degree of need for (i.e. attitude toward) autonomy. But individuals have multiple needs, and thus multiple related goals, so we must also consider their other needs, and the satisfaction of those needs, before arriving at a simplistic conclusion. 3.2.2

Need for competence

Individuals have egos, and tend to suffer embarrassment and shame if they are found to be incompetent in a behaviour that they volitionally chose to undertake. Accordingly, people tend to gravitate towards activities at which they expect to do well, and avoid activities at which they are not confident of success, or at least not confident of acquitting themselves reasonably well under the circumstances. Entrepreneurship provides the individual with the opportunity to choose among opportunities and to demonstrate their competence to others at a number of levels – such as creative thinking (to find the entrepreneurial opportunity); organisational competence (to assemble the resources necessary to exploit that opportunity); management competence (to start and operate the new business venture); and so on. The individual’s need for competence underlies their introspective consideration of their self-efficacy before simply proceeding to undertake the most desirable option. Consulting their self-efficacy acts like a “reality check” and

Why do individuals want to be entrepreneurs?

39

may result in the selection of what is a second-best option from the perspective of perceived desirability. 3.2.3

Need for relatedness

Self-determination theory argues that humans are social animals who generally prefer to interact socially with other people. “Humans are active, growth-oriented organisms who are naturally inclined toward integration of their psychic elements into a unified sense of self, and integration of themselves into larger social structures” (Deci & Ryan, 2000: 229). Entrepreneurship offers the opportunity to interact with other members of the top management team (TMT) and other employees within the firm, and with suppliers and customers external to the firm. Entrepreneurs form investment partnerships with investors, and business network relationships with other entrepreneurs who may include suppliers and customers; such business relationships inevitably entail social relationships as well, and we should expect individuals to enjoy (i.e. gain utility from) such social relationships. Some entrepreneurs will place more importance on the need of relatedness than will others. Lifestyle entrepreneurs for example, particularly those who make a business out of their preferred hobby or pastime (e.g. ski instructors, see Barringer & Ireland, 2006) may derive a great deal of utility from interacting with co-workers, customers, and suppliers who share the same passions. Profit-maximising entrepreneurs, on the other hand, may find that socialising with these stakeholders wastes time that they could be spending on making more money (Douglas, 2013). 3.2.4

Situational needs (for income, risk avoidance, and effort avoidance)

Situational needs are those that arise in the particular context of the decision maker (Gagné & Deci, 2005). In the context of entrepreneurship, as in any occupation, the individual must attend to their need for sustenance, clothing, shelter, health services, and entertainment, most of which require earned income as instrumental to the purchase of the goods and services necessary to allay those needs. Thus the first situational need the intending entrepreneur is likely to feel is the need to earn income (and/or profits), in order to be able to afford the necessities of life, if not a more extravagant lifestyle. Second, the situation of being an entrepreneur inevitably requires risk bearing, which serves to generate disutility (Douglas & Shepherd, 2000). Thus the entrepreneur is likely to feel the situational need to avoid risk exposure to a greater or lesser degree (depending on their attitude to risk – i.e. their degree of risk tolerance). Bearing risk (and incurring disutility therefrom) needs to

40

Entrepreneurial intention

be compensated by additional profit or psychic income, of course, to compete with an otherwise equally attractive alternative. Similarly, entrepreneurship may require harder physical and mental work than would an employment situation, and thus the individual may feel the situational need to avoid excessive work effort (Bird & Jellinek, 1988; Douglas & Shepherd, 2000) to mitigate the physical and mental stresses associated with entrepreneurship. The disutility from work effort is likely to vary across individuals, according to their physical health and stamina and their preference for physical or mental activity (or alternatively leisure – see, e.g. Snir & Harpaz, 2002). Workaholism, or the compulsive desire to work long and hard hours, is also a potential contributor to the intending entrepreneur’s attitude to work effort (see, e.g. Douglas & Morris, 2006; Spence & Robbins, 1992). Thus self-determination theory provides a theoretical basis for the innate and situational needs that entrepreneurship generates. Based on prior literature, in Douglas (2013) I argued that the entrepreneur would derive utility from three positive outcomes of entrepreneurship, namely income (or profit), decision-making autonomy, and work enjoyment,3 while deriving disutility from risk bearing and work effort. These five factors align with the entrepreneur’s innate needs for autonomy and relatedness, and the situational needs of income generation, risk avoidance, and work avoidance, respectively. These five needs thus underlie the perceived desirability of entrepreneurial action, while the remaining innate need, the need for competence, is involved on the perceived-feasibility side of the theory of planned behaviour. 3.2.5

Expectancy-valence theory

Expectancy-valence theory (Ajzen, 1991; Ajzen & Fishbein, 1980) provides a framework to measure the perceived desirability of entrepreneurship by the total utility expected to be derived from the outcomes of entrepreneurial behaviour. Expectancy-valence theory, like utility theory, posits that the total utility expected from an action is the linear sum of the “utility part-worths” (UPWs) of the component outcomes of the action. Limiting ourselves to the five salient outcomes posited in Douglas (2013), namely income, autonomy, work enjoyment, risk bearing, and work effort, the “expectancies” of these items are the individual’s estimation of the quantum of each outcome (e.g. how much income will be earned). The “valences” of each outcome are the individ3 Work enjoyment in that paper is not the same construct as job satisfaction in this book, since in that paper some drivers of job satisfaction, notably decision-making autonomy, risk bearing, and work effort, were treated separately. Work enjoyment in that paper referred to the social aspects of interacting with fellow managers, customers, and suppliers.

Why do individuals want to be entrepreneurs?

Table 3.1

Salient outcomes (Expectancies)

41

Evaluating the perceived desirability of an entrepreneurial opportunity Attitudes to the outcomes (Valences)

Utility part-worths (UPWs)

Income

× Attitude to income

= UPW income

Autonomy

× Attitude to autonomy

= UPW autonomy

Work enjoyment

× Attitude to work enjoyment

= UPW work enjoyment

Risk bearing

× Attitude to risk bearing

= UPW risk bearing

Work effort

× Attitude to work effort

= UPW work effort ΣUPW = total utility

ual’s introspective analysis of the personal importance of each outcome, which correspond to the individual’s attitudes to each of the five outcomes. The arithmetic product of the expectancy and the valence for each outcome is the UPW of that outcome, and the sum of the UPWs is the total utility expected from the contemplated entrepreneurial behaviour. In Table 3.1, adapted from Douglas (2013), we summarise the above for the purpose of subsequent discussion. Table 3.1 refers to a particular entrepreneurial opportunity that is evaluated by a particular prospective entrepreneur. The total utility that is introspectively evaluated by the individual is a measure of the perceived desirability of that entrepreneurial opportunity. That individual will likely also evaluate other entrepreneurial opportunities, and perhaps available employment opportunities (Douglas & Shepherd, 2000), and is expected to array these opportunities in descending order of perceived desirability, preparatory to considering the perceived feasibility of the alternative opportunities (to be considered in Section 3.3). 3.2.6

The non-necessity of anything for entrepreneurial intention

It is important to see that no single outcome (expectancy) or attitude (valence) is necessary or sufficient for the formation of perceived desirability, or for that matter, entrepreneurial intention. I stress this point because reviewers have occasionally lectured me that “all entrepreneurs are autonomy seeking” or that “a highly risk-averse individual would not want to undertake a high-risk new venture” (for example). Such misunderstandings are due to a failure to recognise that perceived desirability depends on total utility of the action, not any single UPW, nor the magnitude of any particular expectancy or valence. As I was induced to state explicitly in Douglas (2013: 638): “It is the sum of the five part-worths that is determining, not any particular attitude, salient outcome, or part-worth, since that attitude, outcome, or part-worth might

42

Entrepreneurial intention

easily be outweighed by the combined effect of the others.” Thus, for example, a highly risk-averse individual might choose a high-risk opportunity because the positive UPWs of income, autonomy, and work enjoyment, and high tolerance for hard work, might collectively outweigh the negative (albeit sizeable) UPW of risk bearing. Although it might be shown that in general highly risk-averse individuals shy away from high-risk endeavours, entrepreneurial intention is a person-level phenomenon, and entrepreneurs are heterogeneous, so we must allow for the theoretical possibility that individuals might choose alternatives that seem to contradict one or more of their dispositional attitudes. We revisit this issue in Chapter 9, where the drivers of entrepreneurial intention are considered holistically as a within-person consideration of the interdependence (rather than independence) of multiple causal factors.

3.3

THE PERCEIVED FEASIBILITY OF ENTREPRENEURSHIP

The perceived feasibility of entrepreneurial behaviour results from an introspective evaluation by the individual regarding whether or not they expect to successfully complete the tasks involved in carrying out a specific entrepreneurial behaviour. Thus we must consider what tasks are involved in entrepreneurship and whether the individual can clearly envision these tasks, ex ante, at the point when they are forming entrepreneurial intention (or not). We also need to consider whether the individual’s self-evaluation of their entrepreneurial efficacy is a reliable predictor of success in completing the tasks, since entrepreneurs have been shown to be notoriously overconfident and optimistic (see, e.g. Cassar, 2010; Hayward, Shepherd, & Griffin, 2006; Trevelyan, 2008; Ucbasaran, Westhead, Wright, & Flores, 2010). What does an entrepreneur need to be able to do – or put another way, what abilities and knowledge does one need to be an entrepreneur? To answer this question in the context of entrepreneurship intention we should consider the perspective of a novice entrepreneur, one who is new to entrepreneurship, which perspective will be somewhat naïve in contrast to the perspective of a seasoned entrepreneur (who has probably learned a lot about tasks that are required but were not anticipated a priori). New entrepreneurs often take a simplistic view of entrepreneurship, expecting customers to “beat a path to their door”; that supply chains will fall quickly into place; that employees will be immediately productive and loyal, and so on. This is an issue of management ignorance (Shepherd, Douglas, & Shanley, 2000), but may be compounded by overconfidence and optimism. Experienced entrepreneurs and serial entrepreneurs (see, e.g. Westhead et al., 2005) will have a more realistic view of the skills and knowledge needed to be successful at entrepreneurship.

Why do individuals want to be entrepreneurs?

43

The point to be made here is that individuals will enter the first phases of the entrepreneurial process with varying stocks of the abilities, skills, and knowledge needed to become a successful entrepreneur – as we know, entrepreneurs (and intending entrepreneurs) are heterogeneous (Gartner, 1985; Shane & Venkataraman, 2000). Thus we need a measure of their suitability for the successful conduct of entrepreneurship and that measure is their ESE. Such a measure is likely to be useful for explaining (in part) why some individuals form entrepreneurial intentions; why some intending entrepreneurs never proceed to entrepreneurial action; why some new ventures are more profitable than others; and why some new ventures fail while others succeed. 3.3.1

Entrepreneurial tasks and required abilities

Opportunity recognition An entrepreneur needs a new venture concept, that is, an idea for a new product, new service, new business process, or new business model, before he/she can go very far in the entrepreneurial process. Thus we might say that a requisite ability is the skill to recognise or create (Davidsson, 2015) a new venture opportunity. This is generally considered to be the first phase of the entrepreneurial process, and the first step in the formation of entrepreneurial intention. But note that some individuals may form entrepreneurial intention in a general sense (e.g. “I want to be an entrepreneur”) before they find a specific opportunity to pursue (as noted in Fitzsimmons & Douglas, 2011). Essentially, both the perceived desirability and the perceived feasibility of entrepreneurship should relate to a specific entrepreneurial opportunity, since opportunities are heterogeneous and both the individual’s perceived desirability and feasibility are likely to vary across opportunities. Note also that individuals may perceive “third-person” entrepreneurial opportunities but not regard them as desirable for themselves (i.e. “first-person” opportunities – see McMullen & Shepherd, 2006). Opportunity recognition requires knowledge about customer needs and preferences, on the one hand, and knowledge about technologies that might be employed to supply a new product or service or process, on the other. This combination of knowledge sets (plus imagination and creativity) is likely to endow the individual with “entrepreneurial alertness” (Kirzner, 1999, 2009) and thereby lead to an innovation that could form the basis for a new business venture. Note that the invention of a new technology, and the ideation or creation of the new product opportunity, are antecedents of the entrepreneurial management of the new venture, and are not necessary skills for entrepreneurship. The entrepreneur might licence the technology and/or buy the rights to the new product concept, and then behave entrepreneurially to commercialise the new

44

Entrepreneurial intention

product. Entrepreneurial management will nonetheless require imagination, creativity, and problem-solving skills in setting up the new venture and managing the firm under conditions of uncertainty. Viability screening Viability screening, also sometimes known as feasibility analysis, or business concept assessment or validation, entails an investigation of the entrepreneurial opportunity to discern what issues need to be verified before attempting to start and manage the new business venture. A complete viability screening exercise would require all aspects of the proposed new venture to be subjected to full scrutiny and analysis, and that the business plan is subjected to evaluation by external experts (such as venture capitalists). But initially we simply need a preliminary screening process that “separates the sheep from the goats” and allows us to concentrate on new product or business concepts that seem to offer the greatest promise of market viability. A convenient typology of the feasibility issues is market feasibility, production feasibility, organisation feasibility, and financial feasibility.4 Market feasibility concerns the demand side of the demand–technology nexus – is there a sufficiently large market for the new product (or process)? The concept must first be tested for its desirability in the marketplace. By marketplace we mean the target market, of course, as a new product will not appeal to everyone. So, who are the target customers? Logic dictates that they will be the ones who are most attracted to the value innovations of the new product or process. A second aspect of market feasibility relates to the window of opportunity – is this the right time to introduce the new product? A third aspect of market feasibility relates to target market contestability – how many firms are competing in that market, what is their size distribution, and how rivalrous are they likely to be? A fourth aspect of market feasibility is market size. The larger the market, the smaller the share of that market that a new firm will need to achieve break-even volume of sales. A final aspect of market feasibility is the rate of market growth. In a fast growing market, rivals are very busy keeping up with the growth of their sales, expanding their plant and equipment, managing more people, and so on. They are less likely to notice a new firm nibbling away at the edge of their market, or are likely to ignore the new venture as having little or no apparent effect on their sales. Oppositely, in a slow growth or stagnant market, the incumbent firms will jealously guard their market shares and may attack (via price competition) any new venture that seems to be stealing their customers.

4 It goes without saying that entrepreneurs should not engage in immoral, unethical, and/or illegal ventures, however feasible these may seem.

Why do individuals want to be entrepreneurs?

45

Production feasibility refers to the efficiency of the production process – the proposed production process must be examined for its reliability in producing products of uniformly high quality and devoid of damaging side effects. Quality variability will be damaging to the firm’s reputation and its ability to gain sustainable competitive advantage. Second, cost variations due to material wastage, variations in employee productivity, reworked production to fix defects, and warranty repair costs, for example, will operate to burn cash and may put the new venture in financial jeopardy. Before the new product is launched, it should have gone through several stages of prototype development, during which all production problems will hopefully have been identified and rectified. An alternative for the new venture facing lack of knowledge about the production process is to outsource production to an experienced manufacturer, not necessarily a rival but perhaps an engineering firm that has the manufacturing capability needed. Alternatively, the new venture might enter into a strategic alliance or joint venture to achieve the same objective. Organisational feasibility refers to the capability of the TMT to efficiently operate the new venture, and will depend on their prior education and business experience. The TMT must also have access to the necessary human, financial, and other resources needed to operate the firm, particularly the strategic resources. In many cases the entrepreneur is more technically competent than managerially competent, perhaps being the inventor of the new technology, or is starting a new venture in their area of technical training, without prior business education or business experience. This entrepreneur may need to hire production, marketing, and financial managers in order to build a TMT with the requisite organisational capabilities. Financial feasibility is the final test of new venture feasibility. Does the new venture promise a sufficient rate of return on investment in this particular industry and market situation? First, the entrepreneur has an opportunity cost of his/her time and resources – to proceed ahead the entrepreneur must feel that the risk-adjusted return of the new venture exceeds that of the best employment opportunity. Keep in mind that this return may not be entirely financial, since self-employment also offers non-monetary benefits (like the satisfaction of being one’s own boss) that add to the utility expected from the venture. Next, if the entrepreneur seeks external venture capital to fund the start-up and operation of the new venture, can he/she raise the funds necessary? Bank loans will require clear evidence of credit worthiness and probably also asset collateral (or a guarantor) to pledge to the bank in case the entrepreneur is unable to repay the loan with interest. Equity investors will want evidence of “investor readiness” (Douglas & Shepherd, 2002b) before they will supply venture capital. The financial feasibility of the new venture requires that the investment in the new venture promises a rate of return that exceeds the entre-

46

Entrepreneurial intention

preneur’s opportunity wage and also exceeds the funding party’s opportunity rate of return on the best alternative usage of the investment funds involved. If the new venture passes these initial feasibility analyses it will seem likely to be at least initially successful, and thus survive and potentially make profits. But will the initial competitive advantage of the new venture translate to sustainable competitive advantage as rivals enter to compete with it in ensuing trading periods? To answer this question the TMT must consider the inimitability of its resource base and judge whether the venture can maintain and/or gain strategic resources that are valuable, rare, and inimitable (Barney, 1991). Problem solving Since entrepreneurship involves both innovativeness and uncertainty, there is every likelihood that the hypotheses and “guesstimations” that are necessarily involved in the business plan will not all be validated once the new venture begins operation, thereby inflicting operational problems upon the new venture. These problems must be solved to allow the new venture to “iterate” or “pivot” to modify its plan and continue on its journey (Blank & Dorf, 2012). Accordingly, problem-solving skills are commonly listed as an important capability of successful entrepreneurs (e.g. Buttner & Gryskiewicz, 1993). This skill set is related to creative thinking. Management skills For some purposes it is useful to bifurcate entrepreneurship into two component parts, namely the identification or creation of the opportunity, and the management of that opportunity. As implied earlier, the entrepreneur need not be the inventor of the technology, nor the creator/ideator of the opportunity, but may simply become involved to manage the entrepreneurial opportunity. Effective entrepreneurial management will require marketing, production, people, and financial-management skills. Indeed such skills are also likely to be instrumental for the successful conduct of the above-mentioned opportunity recognition, viability screening, and problem-solving skills. Marketing-management skills are necessary to best align the new product or process with the customer’s needs, and for the effective manipulation of the other main elements of the marketing mix, namely pricing, promotion, and distribution. Production and operations management skills are necessary for quality control and production/cost efficiency. Human resource management, or more broadly, people-management skills, are important for getting the best out of employees and also are useful for dealing with customers and suppliers. Financial-management skills are of course necessary to attract initial equity and debt funding, and to avoid financial illiquidity which would force closure of the new venture.

Why do individuals want to be entrepreneurs?

3.3.2

47

Entrepreneurial self-efficacy

Bandura (1982) proposed the construct of self-efficacy to represent the individual’s self-assessment of their ability to successfully complete tasks in a specific domain. In the entrepreneurship domain we refer to this as ESE (Boyd & Vozikis, 1994). Scales for the measurement of ESE have been developed which include items for all or most of the above-mentioned entrepreneurial abilities and knowledge areas (see, e.g. Chen et al., 1998; De Noble et al., 1999). Very recently, Newman et al. (2019) provided an insightful review of the ESE literature, and suggest an agenda for future research in this area. While prior scholars did posit a variety of specific skills required for entrepreneurship, these tended to load onto a single factor and thus the ESE measure was generally used in empirical studies as a monolithic construct, effectively averaging over the different skills and knowledge areas important for entrepreneurship. McGee et al. (2009) developed a scale to measure ESE in terms of the skills required in several distinct phases of the entrepreneurial process. McGee et al. (2009) argue that an averaged ESE measure could conceal a very low capability area (e.g. financial management) that might prove to be a fatal flaw for the novice entrepreneur. Following Stevenson et al. (1985), they partitioned ESE into four main areas, namely searching, planning, marshalling, and implementing. “Searching” is the set of activities that culminate in the discovery and refinement of the new product or service. “Planning” is the set of activities that include choice of the business model and the strategies to be advocated by the new venture’s business plan. “Marshalling” (resources) is the set of activities that are needed to assemble the funding, plant and equipment, and people required for the start-up of the new venture. “Implementation” refers to skills required for the effective management of the firm after the start-up, and focuses mostly on people-management and financial-management skills. Table 3.2 shows the items and their loadings in the McGee et al. (2009) study. Note that the “implementing” items loaded onto two distinct factors, representing people-management and financial-management skills, which is not too surprising given that these two subsets of management skills involve quite different skill sets (and probably are best managed by different kinds of people). We will return to the McGee et al. (2009) scale in later chapters. Self-regulation theory We suppose that the individual contemplating entrepreneurship (and having high perceived desirability) would use the perceived feasibility as a reality check, asking themselves the question “Yes, this opportunity is very attractive to me, but can I successfully achieve the outcome?” Low ESE scores for an initially desirable opportunity might render that opportunity undesirable if low ESE causes the risk of failure to be uncomfortably high.

Entrepreneurial intention

48

Table 3.2

Item loadings for the McGee et al. (2009) entrepreneurial self-efficacy scale

How much confidence do you have in your ability to:

Searching

Planning

Marshalling

α = 0.84

α = 0.84

α = 0.80

Implementing

Implementing

(people)

(financial)

α = 0.91

α = 0.84

Brainstorm (come up with) a new idea for a product or

0.80

service? Identify the need for a new product or service?

0.79

Design a new product or service that will satisfy

0.79

customer needs and wants? Estimate customer demand for a new product or service?

0.81

Determine a competitive price for a new product or

0.80

service? Estimate the amount of start-up funds and working capital necessary to start my

0.72

business? Design an effective marketing/advertising campaign for a new product

0.70

or service? Get others to identify with and believe in my vision and

0.77

plans for a new business? Network – i.e. make contact with and exchange

0.76

information with others? Clearly and concisely explain verbally/in writing my business idea in everyday

0.75

terms? Supervise employees?

0.82

Recruit and hire employees?

0.81

Delegate tasks and responsibilities to employees in my business?

0.81

Why do individuals want to be entrepreneurs?

How much confidence do you have in your ability to:

Searching

Planning

Marshalling

α = 0.84

α = 0.84

α = 0.80

49

Implementing

Implementing

(people)

(financial)

α = 0.91

α = 0.84

Deal effectively with day-to-day problems and

0.80

crises? Inspire, encourage, and motivate my employees? Train employees?

0.78 0.78

Organise and maintain the financial records of my

0.82

business? Manage the financial assets of my business? Read and interpret financial statements?

0.81 0.76

Self-regulation theory (Higgins, 1998; Shah & Higgins, 1997) suggests that individuals regulate their behaviour by applying a positive, or conversely a negative, perspective to their decision making. According to Brockner et al. (2004), entrepreneurs are likely to apply promotional (positive) self-regulation when searching for the entrepreneurial opportunities, effectively adopting a more optimistic stance and discounting contradictory observations. But as they approach a positive decision they tend to apply preventative (negative) self-regulation and adopt a more wary and pessimistic stance. In the context of entrepreneurial intention, the implication is that after evaluating an opportunity as highly desirable, the individual would adopt preventative self-regulation and critically assess whether or not he/she possessed the entrepreneurial efficacy needed to successfully conduct entrepreneurship. In Fitzsimmons and Douglas (2011) we found a negative interaction effect between perceived desirability and perceived feasibility in the regression equation purporting to explain entrepreneurial intention of MBA students in Australia, China, India, and Thailand, supporting the notion of preventative self-regulation in the formation of entrepreneurial intention. Subsequently, we have tested for but not found an interaction effect between perceived desirability and perceived feasibility in other studies (e.g. Douglas, 2013). Overconfidence and optimism Entrepreneurs are notoriously overconfident, typically showing greater ESE than employed managers, who in turn are more overconfident than their employees (Bernardo & Welch, 2001; Busenitz & Barney, 1997; Camerer & Lovallo, 1999; Forbes, 2005; Hayward et al., 2006; Palich & Bagby, 1995;

50

Entrepreneurial intention

Simon & Houghton, 2003). And what a good thing this is for societies and nations, even if it is a curse for individual entrepreneurs whose ventures fail because of exaggerated assessment of their capabilities. Without such bold individuals who venture into the unknown risking their future, new wealth would depend on new resource discovery and/or serendipity. Many whose ventures succeed may not have ventured if they had not been so overconfident. So, just as we applaud those who venture and succeed, we should applaud those who venture and fail (assuming they were “investment ready” (Douglas & Shepherd, 2002b) when they launched) because they might have succeeded had the dice rolled differently. Entrepreneurs also tend to exhibit excessive optimism, which causes them to subconsciously inflate their prior predictions of the financial outcomes of their new ventures (Cassar, 2010; Dushnitsky, 2010; Hmieleski & Baron, 2009; Trevelyan, 2008; Ucbasaran et al., 2010). This may be associated with promotional self-regulation (Bandura, 1991; Brockner et al., 2004), influencing them to look more favourably upon their own ideas and to ignore contrary evidence, at least initially. As with overconfidence, optimism augurs for a greater quantum of entrepreneurial start-ups and thereby contributes to societal and national objectives. In Douglas (2009), with help from Norris Krueger, I proposed (mainly as a teaching device) that entrepreneurs look at the world through a series of “lenses” that serve to clarify or distort their perception of a proposed new venture. The “clear lens” is the optically correct lens that allows everything that is known to exist to be seen accurately. This analogy refers to the best information that has been obtained by the entrepreneur about the reality of the entrepreneurial opportunity, and this may be better or worse information than is known to bystanders including investors. But the entrepreneur is likely to “flip down” one or more additional lenses over the clear lenses, thus changing his/her perception of reality. The “rose-lens” effect (Simon et al., 2000) refers to overconfidence, which manifests as failure to know the limits of one’s knowledge and the (false) illusion of control (Boyd & Vozikis, 1994). Entrepreneurs recognise the risks involved in the venture, but conclude that they can handle them, not recognising that they “do not know what they do not know” and overestimating their ESE. The “blue-lens” effect is like sunglasses that restrict the amount of light (i.e. information) entering the entrepreneur’s consciousness, such that using this lens is tantamount to choosing to ignore information that would be available given information-search activity. For example, some entrepreneurs simply assume that customers will buy the product rather than undertaking market research to test their assumptions about customer needs and preferences. The “yellow-lens” effect is about the entrepreneur’s extrinsic motivation, and refers to the entrepreneur’s attitude to income (or degree of preference

Why do individuals want to be entrepreneurs?

51

for income) relative to their attitude to risk (or degree of risk aversion). If the former is relatively strong and the latter is relatively weak, this may manifest what seems to be irrational boldness in the entrepreneur that induces them to take the risks required to capture the profit. The “purple-lens” effect is about the entrepreneur’s intrinsic motivation, and refers to the entrepreneur’s attitude to risk relative to their attitude to the psychic rewards of entrepreneurship. If the UPW of the psychic benefits is relatively high (e.g. due to the promise of achieving self-actualisation or fame, in conjunction with a high preference for those outcomes) and the UPW of risk is not very negative, then the pursuit of those outcomes might seem reckless rather than rational. So, in summary, we do not expect what seems to others to be fully rational behaviour from intending entrepreneurs. Their emotions influence their actions. There is a considerable literature on the impact of emotions on entrepreneurial behaviour which extends the above account considerably, and in the interests of space here the interested reader is referred to that literature (see, e.g. Cardon, Wincent, Singh, & Drnovsek, 2009; Foo, 2011; Podoynitsyna, Van der Bij, & Song, 2012; Welpe, Spörrle, Grichnik, Michl, & Audretsch, 2012). Solo versus team entrepreneurial intention A common perception of the entrepreneur is as a heroic individual who stands alone against the uncertainty of customer demand, supplier cooperation, rival reactions, employee productivity, legislative change, and other unpredictables. Thus we tend to expect the intending entrepreneur to have relatively high ESE in order to be capable of successfully completing all entrepreneurial task areas. Yet we sometimes find individuals with relatively high entrepreneurial intention but relatively low ESE. In Fitzsimmons and Douglas (2011) we took a mean split of perceived desirability (PD) and perceived feasibility (PF) to create a 2 × 2 matrix of entrepreneurial intentions. We found that entrepreneurial intention (EI) was associated with both high and low PD scores, due to the interdependence of PD and PF. If PF was high, a low PD was associated with high EI for some individuals. Similarly, EI was associated with both high and low PF scores – if PD was high, a low PF was associated with EI for some individuals. That paper presented a taxonomy of entrepreneurial intentions as shown in Table 3.3. “Natural” entrepreneurs are perhaps the ideal type, being both highly motivated and highly confident of their skills, and are perhaps most likely to succeed at entrepreneurship. The potential for “accidental” and “inevitable” entrepreneurs requires further comment. Accidental entrepreneurs (i.e. those with low PD but with formed entrepreneurial intention) may be envisaged as people whose interest in entrepreneurship was historically low (like science professors), prior to some life-changing event, such as discovering a new technology with substantial commercial promise. Others may have suddenly

Entrepreneurial intention

52

Table 3.3

Natural, accidental, and inevitable entrepreneurs Perceived desirability HIGH

Perceived desirability LOW

“Natural” entrepreneurs

“Accidental” entrepreneurs

“Inevitable” entrepreneurs

Non-entrepreneurs

Perceived feasibility HIGH Perceived feasibility LOW

ideated a new service, or won the lottery, or gained an inheritance. The prospect of commercial (and/or social) gains may raise the individual’s PD and they are likely to be “pushed” into entrepreneurship by family, friends, and associates. Bhave (1994: 223) called this “externally-stimulated opportunity recognition”. Inevitable entrepreneurs have strong PD but weak PF, and tend to be “pulled” as if by a magnet (Smilor & Feeser, 1991) into entrepreneurship. From the perspective of individual entrepreneurship it would seem highly likely that these intending entrepreneurs would experience venture failure, unless they enter extremely forgiving markets and learn quickly from their experiences. Or alternatively, perhaps the missing variable is their expectation of assistance in the operation and management of the new venture from self-employed parents, from partners on the TMT with complementary expertise, from participation in a new venture incubator, and/or from members of an advisory board.

3.4 SUMMARY In this chapter the conceptual model was established that individuals will want to be entrepreneurs if they expect entrepreneurship to deliver to them greater expected utility than would any other employment of their time, energy, and other resources (Douglas & Shepherd, 2000). Self-determination theory was invoked to argue that entrepreneurship serves the individual’s innate needs for autonomy, competence, and relationships, as well as the situational needs of income, risk avoidance, and effort avoidance. These needs align with the two main determinants of planned behaviour (perceived desirability and perceived feasibility), with attitudes to the five main outcomes of income, autonomy, work enjoyment, risk, and work effort operating to determine perceived desirability, and ESE being an effective measure of perceived feasibility. Expectancy-valence theory was introduced to demonstrate that perceived desirability can be measured as the sum of the UPWs of the outcomes of entrepreneurship. Importantly, no single attitude, nor any single outcome, is either necessary or sufficient for the formation of entrepreneurial intentions. It is conceptually possible for a very risk-averse person to choose a high-risk opportunity, or for a person with low tolerance for work effort to choose an

Why do individuals want to be entrepreneurs?

53

opportunity that provides sufficient utility from other outcomes that more than compensate for the undesirable aspects of the opportunity. ESE is an appropriate reflection of the innate need for competence because self-serving individuals will be attracted towards tasks and occupations at which they feel confident of successfully completing the required tasks. Accordingly, perceived feasibility can be measured by the individual’s introspective assessment of their personal competence in the tasks associated with the conduct of entrepreneurship. These tasks will include opportunity creation, viability assessment, problem solving, and the management skills required to organise and control the operations of the new venture. A variety of scales to measure ESE have been utilised by researchers, and the scale of McGee et al. (2009) was outlined as an exemplar. The possibility of an interaction effect between perceived desirability and perceived feasibility was discussed briefly, based on the findings of Fitzsimmons and Douglas (2011) where this negative moderating influence was attributed to preventative self-regulation of intending entrepreneurs. The legendary overconfidence and optimism of entrepreneurs was also noted, which, along with emotions, may lead to intention decisions that may appear to be irrational (to observers) but in the mind of the intending entrepreneur reflect the culmination of their preferences and biases regarding entrepreneurship. Finally we consider the empirical finding that an underqualified person (i.e. exhibiting low ESE in one or more management skill areas thought necessary for the successful conduct of entrepreneurship) nonetheless intends to become an entrepreneur. We acknowledge that this decision might be dependent on the concurrent expectation of a team approach to entrepreneurship, with other members of the TMT being charged with responsibility for the deficit-skill areas of the respondent. Alternatively, anticipated reliance on self-employed parents and their social and business networks could explain what seems to be a self-destructive intention from the perspective of individual entrepreneurship.

4. Issues for entrepreneurial intention research 4.1 INTRODUCTION The foregoing chapters have been largely concerned with the theoretical underpinnings of entrepreneurial intention. This theory has been developed using qualitative (case-study) analytical research methods to build propositions, and quantitative (sample-based) analytical research methods to test hypotheses which concern the antecedent variables that are associated with the formation of entrepreneurial intention by individuals. Empirical support for the theory of entrepreneurial intention has a relatively “spotted” history, with hypothesised relationships being strongly argued theoretically but weakly supported empirically, and with contrary findings for particular antecedent variables being found in subsequent studies. Such disparity of empirical results is due to a number of issues that arise in empirical studies, as noted in meta-analyses of entrepreneurial intention (e.g. Liñán & Fayolle, 2015; Schlaegel & Koenig, 2014). Before addressing some of these studies in the following chapters, it is instructive to consider several issues that arise in the empirical study of entrepreneurial intention (and other entrepreneurial phenomena). These include issues of construct definition, measurement issues, and methodological issues.

4.2

CONSTRUCT ISSUES

4.2.1

Broad versus narrow entrepreneurial intention constructs

Initially, entrepreneurial intention was seen as a monolithic construct relating to the intention to start any kind of new business venture. This was despite the widespread acceptance of the heterogeneity of the individual–opportunity nexus (Shane & Venkataraman, 2000) and the recognition that some entrepreneurs are primarily oriented towards profit and growth outcomes (e.g. Davidsson et al., 2009) while others are apparently profit satisficing while seeking other non-monetary sources of utility, such as lifestyle entrepreneurs (e.g. Barringer & Ireland, 2006), not to mention those that are social 54

Issues for entrepreneurial intention research

55

purpose-oriented (e.g. Austin et al., 2006). The implicit assumption may have been that the expected utility of starting a new venture was unrelated to the expected utility of operating that new venture, or alternatively, that starting a new venture served the national goal of increasing employment, and beyond that it was up to the entrepreneur to survive and thrive, if they could. Particularly where public funds are utilised to encourage entrepreneurial start-ups, it is myopic to presume that all intending or nascent entrepreneurs are equally likely to want to grow their ventures to boost national employment numbers and company income-tax collections. Using public funds to subsidise low-employee or no-employee lifestyle and salary-substitute ventures is very hard to justify economically, although perhaps not politically (Shane, 2009). In Douglas (2013) two distinct subtypes of commercial entrepreneurial intentions were hypothesised, these being profit and growth-oriented intentions and independence-oriented intentions. Separate scales were developed and these two were found to be separate and distinct intention constructs with acceptably high reliability. Similarly, social entrepreneurial intention has been shown to be a separate and distinct construct from commercial entrepreneurial intention. In addition to these three main subtypes of entrepreneurs, scholars have considered other more narrowly defined subgroups of entrepreneurs, such as mum-preneurs (e.g. Duberly & Carrigan, 2013), elder-preneurs (e.g. Kautonen et al., 2010; Watkins-Mathys, 2012), migrant-preneurs (e.g. Levie, 2007), refugee-preneurs (Wauters & Lambrecht, 2008), serial entrepreneurs (Westhead et al., 2005), and international entrepreneurs (Joardar & Wu, 2011). While these groups are likely to have much in common (e.g. attitude to autonomy) across these subgroups, within each subgroup they are likely to have more in common with the other members of their subgroup, such that there will be less unexplained variance within the subgroup, allowing the researcher to achieve a more fine-grained explanation of the differences in their entrepreneurial behaviour vis-à-vis other subtypes of entrepreneur (Baum & Locke, 2004). Given the heterogeneity of entrepreneurial individuals, and of entrepreneurial opportunities, it is constructive to more narrowly define the subtypes of entrepreneurial intention and to test for multiple subtypes of entrepreneurial intention, allowing respondents to reveal which subtype they identify with most strongly. In Douglas (2013), respondents completed a scale for both growth-oriented entrepreneurial intentions and independence-oriented entrepreneurial intentions, subsequently revealing themselves as having greater intention for one or the other, and allowing the statistically significant antecedents of each type of entrepreneurial intention to be estimated. Overaggregation of subtypes in a sample of entrepreneurs weakens the explanatory power of the analysis because minority groups tend to nullify each other and leave only the dominant relationships that apply more commonly

56

Entrepreneurial intention

across the sample to prevail as significant in the regression equation. While this allows identification of the general antecedents of entrepreneurial intention (broadly defined), it does not contribute much to our understanding of the individual’s process of forming entrepreneurial intention, and since forming that intention is an individual decision, it is important to more narrowly define entrepreneurial intention in terms of “what kind” of new venture is intended. 4.2.2

Latent variable versus indicator variable

In many studies of entrepreneurial intention, the perceived desirability of entrepreneurial behaviour has been measured by the individual’s “attitude toward the act” (Bandura, 1982) where the “act” is a monolithic construct of “entrepreneurial behavior” with unspecified outcomes. The attitude toward the act is a latent variable representing an empirical phenomena that cannot be directly observed, but it may be reliably assessed by reflective measures which are observable and measurable (Diamantopoulos, Riefler, & Roth, 2008). Early studies which simply asked the respondent to espouse their attitude to the act of entrepreneurship and then conducted sample-wide regression analysis on the data effectively assume that all entrepreneurial opportunities have the same mix of outcomes, and/or that the mix of and magnitude of those outcomes are not important to the individual’s formation of entrepreneurial intention. Since there is heterogeneity on both sides of the individual–opportunity nexus, implicitly aggregating all outcomes operates to subsume the relative importance of the different outcomes and offers less explanatory power for individual entrepreneurial intention. This approach effectively correlates entrepreneurial intention with the average income, autonomy, risk, and other outcomes of entrepreneurship, rather than with the expected outcomes that are specific to the type of entrepreneurial behaviour that is envisioned (and appreciated more or less) by the specific intending entrepreneur. Since the Douglas and Shepherd (2002a) study, entrepreneurial intention studies have tended to treat attitude toward the act (of forming entrepreneurial intention) as the latent variable, and sought to find indicator variables that constitute a reflective model of the attitude toward the act, by estimating respondents’ attitudes toward specified outcomes of the act, such as profits, autonomy, risk exposure, work enjoyment, and work effort required (as in Fitzsimmons & Douglas, 2011). In this way perceived desirability is measured using multi-item measures reflecting the individual’s attitudes toward the multiple outcomes of entrepreneurial behaviour, rather than the individual’s attitude toward entering the state of entrepreneurship without throwing any light on the relative importance of the various outcomes of entrepreneurial behaviour.

Issues for entrepreneurial intention research

4.3

MEASUREMENT ISSUES

4.3.1

Construct clarity

57

Suddaby (2010) stresses the importance of construct clarity. Any given theoretical construct (e.g. entrepreneurial intention) should be defined and utilised in a clear and consistent manner across subsequent studies if we are to test and build theory using empirical evidence. Different definitions of constructs may end up measuring different things. Similarly, different multi-item scales purporting to measure the same construct may measure different things, and thus not provide strictly comparable results. Thus, wherever possible, existing measurement scales, if properly designed, tested, and validated, should be utilised rather than developing ad hoc new measurement scales, assuming that the existing scale reliably measures the construct under examination. This allows replication and extension of prior studies and builds more robust theory. This is not to say that better scales cannot be devised, but the onus is on the researcher to demonstrate that a revised or different scale is indeed superior to the earlier scale, and this should probably require comparing the impact of both scales on the focal research question. 4.3.2

Common-method bias

Collecting data for all the dependent variables and/or for both the dependent and independent variables, at the same time using the same method, tends to bias data observations since the respondent is likely to recollect responses to prior questions and subconsciously allow these to influence subsequent responses (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). Surveying respondents in multiple “waves” of surveys should serve to reduce common-method variance. Care must be taken to link each individual’s surveys together, preferably using a secret code known only to the respondent and thus allowing them to feel anonymous to the researcher and to each other. Common-method variance may not be completely avoidable in single-survey cross-section studies, but various defences against it are suggested (Podsakoff et al., 2003) and should be utilised and revealed when publishing results. 4.3.3

Espoused versus revealed data

Survey respondents may feel self-conscious about their responses to survey items, and may, as a consequence, consciously or subconsciously distort their answers to avoid embarrassment, or to avoid revealing what they consider personal and private information, or to give a more favourable impression to

Entrepreneurial intention

58

the interviewer. For example, feelings that others might regard as reflecting selfishness, greed, lack of social orientation, or poor character might induce respondents to understate or overstate their responses to questions about these items. These espoused responses to the survey items are likely to introduce “social-desirability bias” or “interviewer bias” in the data collected. An alternative data-collection method that is particularly useful when seeking to measure personal attitudes is a “conjoint” experiment (Green & Srinivasan, 1978; Shepherd & Zacharakis, 1999). This method reveals the valence to the individual of items when presented conjointly in a series of hypothetical scenarios each containing multiple outcomes, where the outcome values are set variously at “high” or “low” in different combinations in each subsequent hypothetical scenario. The respondent is asked repetitively to rate (on a Likert scale) the perceived desirability of each scenario. This avoids asking the respondent to espouse his/her valence of any particular outcome in isolation, and thus preserves a sense of privacy about that valence. Using regression analysis, the analyst can isolate the individual’s valence for each of the outcomes under examination.1 Conjoint analysis has been utilised to elicit “revealed” attitudes in several studies of entrepreneurial phenomena (see, e.g. Choi & Shepherd, 2004; Douglas, 2013; Douglas & Shepherd, 2002a; Fitzsimmons & Douglas, 2011; Haynie, Shepherd, & McMullen, 2009; Monsen et al., 2010). Experimentally, in some data sets we have collected both espoused measures of attitudes as well as those revealed by conjoint analysis, and found that the two correlate very poorly with each other, such that the espoused attitudes generate much poorer regression results at the sample level. Accordingly there seems likely to be substantial measurement error in the attitude data if these values are espoused rather than revealed. The same is probably true of espoused entrepreneurial self-efficacy (ESE) data, although no studies that I am aware of have utilised conjoint analysis to estimate individual ESE. 4.3.4

Sex versus gender

Differences between the entrepreneurial proclivity of males and females is an enduring point of discontent in the entrepreneurship domain, largely because of measurement error. While it is not difficult to measure the physiological 1 A regression equation is calculated for each respondent, where the dependent variable is the Likert-scale desirability score given to each scenario, and the independent variables are binary variables for each of the (independent) variables in the conjoint experiment, these set to 0 for “Low” and 1 for “High”. The regression coefficient to each independent variable reveals the individual’s valence of the high outcome for each independent variable.

Issues for entrepreneurial intention research

59

sex of individuals, there is a clear distinction between sex and gender identity. Gender identity is a socially constructed concept referring to feminine and masculine traits and behaviours (see, e.g. Ahl, 2006; Fischer & Arnold, 1994; Gupta, Turban, Wasti, & Sikdar, 2009; Marlow & Patton, 2005). While a person is born male or female, their gender traits develop over time as normative social pressures dictate and guide their self-perception, intentions, and actions. Whereas sex is a physiological distinction that describes what people are, gender relates to their cognitions and behaviours, i.e. what they think and do. The assumption of “biopsychosocial equivalence” of sex and gender, meaning that being male correlates highly with masculine gender traits and that being female correlates highly with feminine gender traits, is widely rejected in the social-constructionist feminist literature (see, e.g. Harding, 1987; West & Zimmerman, 1987). Entrepreneurship scholars argue that entrepreneurial behaviour depends on entrepreneurial cognitions (Baron, 2007; Carter, Brush, Greene, Gatewood, & Hart, 2003; Dutta & Thornhill, 2008; Rauch & Frese, 2007, 2014). Given the fact that both men and women become entrepreneurs, entrepreneurial cognitions are clearly shared by some (but not all) men and by some (but not all) women. Notwithstanding the observation in some studies that men may be more likely to become entrepreneurs than women (Fay & Williams, 1993; Gupta et al., 2009; Gupta, Turban, & Pareek, 2013; Hisrich & Brush, 1984; Shinnar, Giacomin, & Janssen, 2012), the physiological sex (per se) of individuals has not been shown conclusively to have a definitive impact on entrepreneurial cognitions. There is no compelling argument that physiological differences due to one’s sex (such as men tending to be on average taller or stronger) will have either a positive or negative effect on ESE. Instead we expect ESE to depend more on one’s knowledge of technical and managerial aspects pertinent to the focal venture, and on one’s cognitions regarding innovation, strategy, and risk taking, for example. These cognitive differences are more likely to vary with gender characteristics rather than with sex characteristics.2 Unfortunately, in the entrepreneurship literature many researchers have used the term “gender” to describe whether a person is male or female, such

2 There may be businesses where the entrepreneur’s sex is more important than gender traits, due to the nature of the business and the clientele of the business being primarily of one sex or the other. For example, if the clientele of flower shops, hair and nail salons, and some personal services are largely female, or where the employees are predominantly female, we might expect that being female might be a significant correlate of starting such businesses, predominantly due to job-satisfaction reasons. Similarly, males may be more likely to start ventures in industries where the customers and suppliers are predominantly male.

60

Entrepreneurial intention

that the distinction between sex and gender is typically lost (Ahl, 2006; Bird & Brush, 2002; Fischer & Arnold, 1994; Gupta et al., 2009, 2013). While many studies have looked specifically at entrepreneurship (start-up) intentions, growth intentions, and performance of women entrepreneurs, sometimes relative to men (e.g. Manolova, Carter, Manev, & Gyoshev, 2007; Marlow & Patton, 2005), almost all have been restricted to the observation that the respondent is physiologically either male or female, with little or no consideration of their gender characteristics (but see Gupta et al., 2009). The sex of the respondent is not likely to adequately reflect the entrepreneurial cognitions of the respondent, which would better inform the research question. For example, as well as masculine gender traits and behaviours, males may also exhibit what have been characterised as feminine gender traits and behaviours, such as being caring and supportiveness. Similarly, as well as feminine traits and behaviours, females may also exhibit what have been labelled masculine gender behaviours, such as assertion and aggression (Ahl, 2006; Bem, 1981; Gupta et al., 2009). As these examples suggest, both males and females can possess both masculine and feminine gender traits, and both masculine and feminine traits can be argued to facilitate entrepreneurial behaviour and success (Bird & Brush, 2002; Mueller & Dato-On, 2008). Accordingly, it is important to use “sex” to distinguish males from females, and “gender” to distinguish between cognitions and behaviours that reflect socially constructed masculine and feminine traits. Researchers have called for more work on gender rather than sex differences in the context of entrepreneurship (e.g. Ahl, 2006; Eddleston & Powell, 2008; Lewis, 2006; Marlow & Patton, 2005), but there have been relatively few studies to address the influence of masculinity and femininity (e.g. Gupta et al., 2009, 2013).3 It is particularly important to examine critically the notion that entrepreneurship is a male-gendered phenomenon, and that males are likely to be more suited to entrepreneurship than are women, when we readily observe both successful female entrepreneurs and unsuccessful male entrepreneurs. The perpetuation of this notion may tend to make men perceive that they are more likely to be successful, and women to perceive that they are less likely to be successful, in the pursuit of entrepreneurship. This would inevitably result 3 It is of course not pejorative to say that men exhibit feminine traits or that women exhibit masculine traits, unless you don’t understand them (the traits, I mean). Gender traits are socially conditioned cognitive and behavioural patterns that, still today, are imbued into children and persist into adulthood. Fortunately, this legacy of the past is receding in most cultures, such that we may soon be able to remove the word “gender” (and its sly implications for men or women) from the term “gender traits” and instead simply refer to these as human traits.

Issues for entrepreneurial intention research

61

in an oversupply of underqualified males, and an undersupply of qualified females, into the ranks of entrepreneurs (Ahl, 2006; Ahl & Marlow, 2012) resulting in personal and societal loss.

4.4

METHOD ISSUES

4.4.1

Non-representative samples

The disparity of results generated by entrepreneurial intention studies is surely due in some large part to the disparity of the samples utilised. In any given sample the predominant motivations of the respondents will emerge to characterise the sample, and thus the research results will be sample-dependent. Entrepreneurs (and intending entrepreneurs) are heterogeneous, and vary regionally and internationally, such that a truly random global sample (and thus comparable samples for international studies) cannot be expected. And so we attempt to collect random samples of the relevant “population”, be these the national wood-turners association (Baum & Locke, 2004); US-based nascent entrepreneurs (Reynolds et al., 2002); Spanish firms (Liñán & Chen, 2009); Swedish firms (Delmar & Davidsson, 2000); Chinese firms (Lau & Busenitz, 2001); and so on. A random sample is only really important if we are attempting to test empirically hypotheses that are theorised to apply in general across members of a population. For example, does formal entrepreneurship education positively impact the survival of entrepreneurial new ventures, or their profitability, or their longevity? The answer will be “yes” for some individuals, “no” for others, but a dominant tendency is likely to be found, and if statistically significant, can be used to support the theory that entrepreneurship education positively impacts engagement with and the success of entrepreneurship. This finding will be specific to the population sampled and may not generalise across other populations of individuals. But if the purpose is to explain, at the individual level of analysis, why person X intends to start a new venture, or why person Y does not intend to start a new venture, such sample-based results offer little explanation. Perhaps person X had no prior business and entrepreneurship education, while person Y had five years of prior relevant education. Even if “prior relevant education” is included in the data collection, it may not be found to be a significant precursor of entrepreneurial intention if the majority of respondents did not have prior relevant education but nonetheless had formed entrepreneurial intention. The individual’s decision to start (or not start) a new venture is made holistically within the mind of the individual, and is probably not influenced by the general (sample-wide) relationships between the independent variables and the dependent variables (Magnusson & Torestad, 1993; Muñoz & Dimov, 2015).

62

Entrepreneurial intention

Moreover, the antecedent variables are likely to combine interdependently, rather than independently, to cause the focal outcome in the individual’s case. We return to this issue later in this chapter when the analytical method is discussed. Student samples are often used to test newly hypothesised relationships between antecedent variables and the behaviour under examination, and this choice of (convenience) sample is often excoriated by journal reviewers. The issue of course is whether the students sampled are representative or not of the population they are purported to represent. Is it reasonable to purport that they represent a population that includes individuals who may form entrepreneurial intention? And since our interest in entrepreneurial intention is that it may lead to subsequent entrepreneurial behaviour, do some students from this population actually start new ventures subsequent to the completion of their education? These questions are relatively easy to investigate and answer, and thereby support the use of a particular sample. Scholars have argued that business students, particularly those that are older, better trained, and have more business experience, such as MBA students, constitute populations that contain potential entrepreneurs (e.g. Shepherd & DeTienne, 2005). Since entrepreneurial intention is a thing of degree, rather than a binary variable, samples are likely to contain individuals with different degrees of entrepreneurial intention (including none), and may allow empirical studies to support new theory development. Arguing this, in defence of your choice of sample, becomes imperative in view of the reactionary response by some reviewers to the use of a student sample. Other scholars have argued that samples for the study of entrepreneurial intention should include only individuals who have taken discrete steps toward starting a new venture, such as writing a business plan, developing product prototypes, and so on. These are known as “nascent entrepreneurs” (Reynolds, Carter, Gartner, & Greene, 2004), and of course this is the relevant population if one is investigating the behaviour of nascent entrepreneurs. In Douglas (2013: 637) I argued that the process of entrepreneurial intention formation includes the nascency phase, since entrepreneurial intention is the intention to start a new venture, which happens at the end of the nascency phase. During the nascency phase the individual may write, iterate, and pivot their business plan (Blank & Dorf, 2012), and may proceed to assemble the resources necessary for the launch of the new venture, but retain the real option to not launch the new venture. The intention to start a new venture may be strengthened or weakened on a daily basis by the discovery of new information about the feasibility of starting that new venture. The phase of nascent entrepreneurship may be abandoned, or may be prolonged for months or years while the individual remains “still trying” to start the new venture (Carter, Gartner, & Reynolds, 1996). The formation of entrepreneurial intention is a process that starts with

Issues for entrepreneurial intention research

63

an awareness of, and initial interest in, behaving entrepreneurially (e.g. starting an entrepreneurial venture) and culminates in starting that new venture. By this logic, nascent entrepreneurs represent a subset of intending entrepreneurs who are in the advanced stages of entrepreneurial intention formation (but are not necessarily close to the end of that process). This begs the question of how strong one’s entrepreneurial intention is at the point in time when that person is interviewed or sampled. The strength of entrepreneurial intention may be conceptualised as ranging from relatively weak to total commitment, and may increase or decrease in strength from day to day as more information is received and processed. Confining our sample to nascent entrepreneurs does serve to exclude those whose intention is not yet strong enough to motivate them to begin “taking steps” to start their business, and potentially excludes from the sample the “tyre kickers” who would simply add unexplained variance in empirical studies. 4.4.2

Cross-sectional versus longitudinal studies

Because we believe that entrepreneurship is a planned behaviour, the causal effect of the independent variables on the dependent variable cannot be asserted in cross-sectional studies if the data on both independent and dependent variables are collected at about the same time – data which is hypothesised to be causal must be collected substantially prior to the collection of the outcome data. In effect, the use of cross-sectional data means regressing current outcome data with current (rather than antecedent) causal data. To establish causality, the current antecedent data must be considered against future outcome data, with appropriate control variables, in a longitudinal study, to establish causality (or not). Longitudinal analysis thus requires collection of data in waves separated by months or years to allow the process of causality to work through and be captured in a subsequent wave of data collection. Separate waves of data collection are also advocated for cross-sectional studies to avoid or reduce common-method variance. The intervening periods between waves of data collection (with distracting events) operate to erase or dim respondents’ recollections of their specific responses to earlier survey items or experiments, and thereby serve to ensure a nearly independent evaluation of the next wave of survey items. 4.4.3

Symmetrical correlational analysis

Most prior studies of entrepreneurial decision making have utilised symmetrical correlational methods, such as multiple regression and structural equation modelling, that analyse and identify which independent variables are statis-

64

Entrepreneurial intention

tically significant in explaining the dependent variable at the sample level. Each coefficient in this single dominant “net-effects” model is effectively the sample average of the relationships between each independent variable and the dependent variable, holding constant the influence of all other independent variables. These sample-level results are useful to discover the main drivers of entrepreneurial behaviour at the macro level, and facilitate the development and empirical testing of entrepreneurship theory. They also provide useful information for entrepreneurship policy, practice, and education. But finding which independent variables are significantly associated on average with the dependent variable across the entire sample of individuals leaves undiscovered a great deal of fine-grained information about the decision making of individual entrepreneurs. The single dominant prescription for the dependent variable (provided by the regression equation) presumes that the focal decision depends on the independent effects of the antecedent variables, but entrepreneurs most likely use “holistic” decision making (Magnusson & Torestad, 1993) to arrive at decisions after considering the interplay of personal and contextual factors that are salient to their personal wellbeing. If so, sample-level correlational methods, by averaging the relationships between variables across the respondents, are incongruent for the analysis of the holistic decision-making process of individuals who make decisions for their own particular reasons (Magnusson & Torestad, 1992; Woodside, 2013; Zyphur, 2009). Moreover, the single dominant prescription for the relationship between the independent variables and the dependent variable, offered by the regression equation, belies the observable fact that there are often multiple paths to entrepreneurship – these are not limited to the one prescription implied by the combination of significant antecedent variables indicated in the regression equation. For example, the regression analysis may conclude that being male is significantly associated with the formation of entrepreneurial intention, yet being female is clearly part of an alternative pathway to entrepreneurship. Similarly, the regression analysis might infer that high ESE is associated with entrepreneurial intention, yet in Fitzsimmons and Douglas (2011) we found that low ESE was also associated with entrepreneurial intention for what we called “inevitable entrepreneurs”. A third issue with symmetrical correlational methods is that they are based on a linear-additive model of causality, and investigate the relationship between independent variables and the dependent variable discretely (i.e. holding constant the impact of all other independent variables). While interactive terms can be employed to test the interdependency of independent variables and their combined impact on the dependent variable, the number of such interactive variables is limited by considerations of sample size and degrees of freedom. Moreover, symmetrical correlation methods are deductive methods,

Issues for entrepreneurial intention research

65

such that the independent variables (and interactive variables) included in the regression model should be hypothesised a priori, based on theoretical arguments – there should be no fishing expeditions (Gelman & Loken, 2014) to look for possible significant associations with post hoc rationalisation of that finding. It is relatively easy to hypothesise why two-way and three-way interactions operate to drive the outcome of interest, but it is more difficult to rationalise higher-way interactions that may exist in the data. To investigate the potential interdependency of higher-way interactions one needs to conduct inductive case-based analysis to discover these and subsequently hypothesise the generality of these interactions for testing using a deductive method. Fourth, symmetrical correlational methods presume data symmetry, namely that the frequency distribution of the data is (approximately) symmetrical around the mean value for each variable. If there is substantial skew in the data for a variable it must be removed by a transformation that renders the distribution approximately symmetrical. The presence of outliers in the data, that is, observations that are more than three standard deviations from the mean and also distant from any other observation, will distort the results, such that outliers must be removed and treated as meaningless aberrations. But entrepreneurs, particularly those that create radical and disruptive new products and business processes, are often regarded as outliers. In entrepreneurship research it may be “throwing the baby out with the bathwater” to disregard the outliers. Fifth, these regression and structural equation modelling methods assume symmetry of relationships, meaning that the relationship between an independent variable and the dependent variable is hypothesised to be either positive or negative. If the majority of respondents express a positive relationship between these variables, the sign of the relationship in the regression equation is positive, and the fact that it was negative for others is hidden, reflected only in suppressing the confidence level for that independent variable. The fact that there may be (at least) two different pathways to the focal outcome, one involving a positive relationship and another involving a negative relationship, for subgroups within the sample, is not investigated by symmetrical correlational methods. Thus, symmetrical correlational quantitative methods have limitations for the empirical study of entrepreneurial intention, and potentially a variety of other forms of entrepreneurial behaviour (Douglas, Shepherd, & Prentice, 2019). While these analytical methods have been instrumental and indispensable in building entrepreneurship theory over the formative decades of the field, the above-mentioned limitations indicate that they are unsuitable to examine issues of individual motivation, multiple pathways, interdependency of antecedents, asymmetric data, and asymmetric relationships, such that we need to consider other analytical methods that can incorporate these issues that are observed in entrepreneurial phenomena, and thereby provide additional

66

Entrepreneurial intention

finer-grained information about entrepreneurial behaviour, thereby complementing the traditional analytical methods. 4.4.4

Asymmetrical qualitative comparative analysis

Qualitative comparative analysis (QCA – Ragin 1987, 2000, 2008) is a case-based asymmetrical analytical method that overcomes these restrictions. Fuzzy-set qualitative comparative analysis (fsQCA) software provides a new tool for the entrepreneurship researcher’s toolkit that can provide new and more granular information about entrepreneurial behaviour at the case level – i.e. individuals, or firms, or nations – that supplements information gained at the aggregate level by symmetrical correlational methods. FsQCA examines the within-case relationships between and among the variables, and characterises cases by the combination of antecedent conditions (known as configurations) that the cases associate with the focal outcome (Ragin, 2000). QCA can be viewed as a combination of qualitative and quantitative methods (Ragin, 2008), and is congruent with the holistic theory of individual decision making. It seeks to discover the factors common to cases who take a particular pathway, as distinct from those who take other pathways, to a given entrepreneurial outcome. Thus fsQCA aggregates cases from a heterogeneous sample into relatively homogeneous subgroups, and provides an empirical basis for the development of theory specific to each of the subgroups. FsQCA is thus complementary to traditional symmetrical analytical tools, as it adds fine-grained detail to our knowledge about the drivers of entrepreneurial phenomena and provides an empirical basis for new theory building about those phenomena. So, while symmetric correlational methods are the mainstay of our research, fsQCA provides a complementary analytical tool that can (i) align with the holistic theory of human decision making to explore the relationships among the data at the case level; (ii) identify multiple equally effective pathways to the same outcome, if they exist; (iii) recognise the interdependence of a set of antecedent variables when it exists; (iv) accommodate data asymmetry; and (v) accommodate asymmetrical relationships. The latter facility allows the researcher to conduct counterfactual analysis – in this case, where the presence of some antecedent variables is associated with entrepreneurial intention in some pathways to entrepreneurship, while simultaneously the absence of that antecedent variable is also associated with entrepreneurial intention in one or more other pathways to entrepreneurship. For example, both being male, and not being male (i.e. being female), can be found to be instrumental to entrepreneurial behaviour. In Chapters 5 to 7 we review empirical studies that have utilised symmetrical correlational methods for commercial entrepreneurs, intrapreneurs, and social entrepreneurs, respectively, before applying fsQCA in Chapter 9 to one

Issues for entrepreneurial intention research

67

of these studies to demonstrate the additional nuances that QCA can provide for understanding the antecedents of entrepreneurial intention.

4.5 SUMMARY In this chapter we have traversed a variety of issues that arise in the empirical study of entrepreneurial intention and the building of new theory in the entrepreneurship domain. Different treatments of these issues contribute to the disparity of findings across empirical studies. First we considered the importance of not overaggregating the possible types of entrepreneurial intention into a monolithic measure of that construct. More useful information about the drivers of entrepreneurial intention in large samples can be derived by testing for multiple types of entrepreneurial intention in the same study, to find how the drivers of entrepreneurial intention differ across the types of entrepreneurship that may be contemplated by the respondent. Next, we noted that the latent variable “attitude to entrepreneurship” is not easily measurable and is subject to social-desirability biases. A multi-item reflective measure of the attitudes to the salient outcomes of entrepreneurial behaviour is likely to produce a more reliable measure of the individual’s attitude to entrepreneurship. We then considered measurement issues, starting with the necessity of construct clarity. If the construct is viewed ambiguously by respondents they will be reporting on different phenomena. The multiple items in scales purporting to measure the construct must be pre-tested and post-tested for validity and reliability, and then should be used for subsequent replications of the focal relationships, unless validated improvements to those scales can be argued. Common-method variance must be avoided to the greatest extent possible in cross-sectional studies, since this also distorts the accuracy of data either consciously or subconsciously. Espoused measures of attitudes are demonstrably inferior to revealed measures of attitudes, and the latter can be derived using relatively simple conjoint experiments. The use of sex as a proxy for the entrepreneurial proclivity of males and females was also cited as being prone to measurement error, since entrepreneurship is typically much less a physical activity than it is a mental or cognitive activity. The assumption of “biopsychosocial equivalence” is widely rejected in the sociology and psychology literatures, and should be more widely rejected in the entrepreneurship field as well, because both men and women exhibit both “masculine and feminine gender” traits that may be conducive to entrepreneurial intention. Finally we considered analytical methods, noting that the mainstream symmetrical correlational methods (such as multiple regression analysis and structural equation modelling) have limitations when dealing with holistic decision-making practice of individuals. We also need to consider the potential

68

Entrepreneurial intention

for multiple pathways to the same outcome, the interdependence of antecedent conditions, asymmetric data, and asymmetric relationships between variables. The advent of QCA and fsQCA software allows researchers to dig deeper into the data to reveal additional information about the antecedents of entrepreneurial behaviour that complements the general sample-wide findings revealed by traditional quantitative methods. We shall return to this analytical method in Chapter 9, after building an integrated model of entrepreneurial intention in Chapter 8 which illuminates the need for a case-based analytical method to test empirically for the presence of multiple pathways and asymmetric data relationships in the context of entrepreneurial intention.

5. Commercial entrepreneurial intention 5.1 INTRODUCTION In this chapter we examine some empirical studies of entrepreneurial intention which focus on the intention to start a “commercial” new venture – that is, a new venture where the primary purpose is to provide profit and psychic benefits to the owners of the firm. This “self-oriented” purpose is distinct from a “social” new venture, where the primary purpose is “other-oriented” – i.e. to provide social benefits to individuals external to the firm (Dees & Elias, 1998). We begin with “generic” commercial entrepreneurial intention, meaning no distinction is made between or among subtypes, such as growth-oriented, independence-oriented, lifestyle, salary-substitute, or subsistence entrepreneurial intention (Barringer & Ireland, 2006; Douglas, 2013). Subsequently we recognise that a monolithic entrepreneurial intention construct lacks construct clarity (Suddaby, 2010), due to the overaggregation of what are essentially different types of entrepreneurial intention, and discuss the Douglas (2013) paper which demonstrated that growth-oriented intention (GOI) and the catch-all category of independence-oriented intentions (IOI) are separate and distinct constructs, driven by differing antecedent variables. We first clarify that, although it is tempting to call commercial entrepreneurship “profit-seeking entrepreneurship”, that term cannot be used synonymously with commercial entrepreneurship, as social entrepreneurs may also seek to earn profit (see, e.g. Dees, 1998; Miller et al., 2012; Wry & York, 2017). The profit of social enterprise may or may not be distributed (at all, or entirely) to the firm’s owners, and may be used to extend the scale and reach of the social enterprise in order to serve further the primary purpose of creating social impact. These social entrepreneurs have been called hybrid-social entrepreneurs to distinguish them from non-profit-seeking social entrepreneurs.1 More recently it has become understood that social entrepreneurs also need 1 There is an alternative literature that uses the term “hybrid entrepreneurs” to identify individuals who practise entrepreneurship while simultaneously employed – see, e.g. Folta, Delmar, and Wennberg (2010) and Solesvik (2017). In this book we shall use the term to refer to the hybrid that is part-social, part-commercial, as it is commonly used in the social-entrepreneurship literature.

69

70

Entrepreneurial intention

to be profit seeking to extend or maximise their social impact, since governmental and philanthropic funding is limited. Social entrepreneurial intention is considered in Chapter 7. This chapter is not intended to be a comprehensive review of the prior literature on commercial entrepreneurial intention – such reviews have been done well already (see, e.g. Fayolle & Liñán, 2014; Liñán & Fayolle, 2015; Schlaegel & Koenig, 2014; Shook, Priem, & McGee, 2003). Rather, the purpose is to outline the antecedents of commercial entrepreneurial intention that have received theoretical and/or empirical support in prior studies.

5.2

COMMERCIAL ENTREPRENEURIAL INTENTION STUDIES

Early studies of entrepreneurial intention took for granted that the main motive for entrepreneurship was commercial, based on the classical economists’ notion of the profit-seeking entrepreneur as a self-serving hedonistic actor in the economy. The entrepreneur’s attitude to profit was axiomatically assumed to be positive, and their attitude to risk was axiomatically assumed to be negative, necessitating a risk premium (i.e. additional profit commensurate with the degree of risk) to induce them to become entrepreneurs (see, e.g. Kihlstrom & Laffont, 1979). Baumol (1990) argued that, historically, entrepreneurial behaviour had waxed and waned according to the profit, power, and prestige afforded to entrepreneurs, in effect implying that in addition to risk aversion, two other non-monetary cognitions were also at play, namely the entrepreneur’s attitudes to power and to prestige. These cognitions may be restated, in the terminology of Douglas and Shepherd (2000), as the attitude to decision-making autonomy (or independence) and the attitude to the perquisites (or work enjoyment) associated with entrepreneurship, respectively. In positing a utility model of entrepreneurial motivation, Eisenhauer (1995) rolled all such non-monetary sources of utility (and disutility) into a catch-all variable he called “working conditions”. What these working conditions are had been under discussion for some time. Eden (1973) found that self-employed individuals work longer hours and report working harder. The expectation of this hard work is expected to have a negative effect on entrepreneurial intention, of course, as it would generate expected disutility. Later Bird and Jellinek (1988), Hammermesh (1990), and Kolvereid (1996a, 1996b) agreed that the self-employed tended to work harder in their own business than they would in an employment situation. Eden (1973) also found the self-employed to experience greater freedom and autonomy at work, which is supported by Katz (1992), and also by Baumol’s (1990) argument that entrepreneurs tend to value power, presumably manifesting as decision-making power. Kolvereid (1996b) also found that potential entrepre-

Commercial entrepreneurial intention

71

neurs valued autonomy and authority. Eden (1973) also mentioned the greater job satisfaction of the self-employed. A grounded research paper by Kolvereid (1996a) found the reasons given by respondents for preferring self-employment included (i) economic opportunity (i.e. income or profit); (ii) authority; (iii) autonomy; (iv) challenge; (v) self-realisation; and (vi) ability to participate in all aspects of the business (perhaps interpretable as job satisfaction). Conversely, reasons for preferring employment were (i) job security (i.e. risk aversion); (ii) social environment; (iii) easier workload; (iv) avoid responsibility; and (v) career/promotion reasons. Note that the negative of most of this latter list may be a reason to prefer self-employment. 5.2.1

Antecedents of commercial motivation

Based on these earlier fragments of a theory of self-employment choice, in Douglas and Shepherd (2000) we argued in a conceptual model that the main conditions motivating entrepreneurship would be the expected income (i.e. profit); decision-making autonomy (i.e. independence); risk exposure; work effort required; and a catch-all “net perquisites” construct encompassing all positive and negative elements of job satisfaction not otherwise accounted for. Subsequently, in Douglas and Shepherd (2002a), we conducted an empirical test of that model (excluding net perquisites) and found that individuals did indeed consider income, independence, and risk, but not work effort, when evaluating career alternatives. Moreover, their attitude to independence, and their attitude to risk (i.e. tolerance of risk) was positively related to entrepreneurial intention. Surprisingly, given the accepted wisdom at the time that entrepreneurship is motivated by expected income and demotivated by expected hard work, both attitude to income and attitude to work effort were found to be not significantly related to entrepreneurial intention, although they exhibited the hypothesised signs (positive and negative, respectively). Apparently the sample (who were alumni of an undergraduate business degree programme contacted 2–10 years after they had graduated) contained a mixture of individuals with both high and low attitudes to income and to hard work, such that the net effect of the regression analysis was that this variable was not significant, sample-wide. This did however suggest that some may have intended a less hard-working entrepreneurial pursuit (perhaps a lifestyle venture) while others may have been intending a venture involving harder work (perhaps as a growth venture). Shortly thereafter, Shook et al. (2003) published a review of prior works and presented a conceptual model of the enterprising individual, which they felt represented the state of the art of the entrepreneurial process at that time. They proposed that this process comprised the formation of entrepreneurial inten-

72

Entrepreneurial intention

tion, opportunity discovery, the decision to exploit, and exploitation activity, in that order. They postulated that psychological factors (including personality factors, beliefs, values, attitudes, needs, and traits); individual characteristics (including demographics, education, prior experiences, and abilities); and cognitions (including knowledge structures, biases, and heuristics) operated to drive those four main components of the entrepreneurial process. They noted that psychological variables were hypothesised to cause some individuals to prefer entrepreneurship (over employment) due to personality factors such as a greater need for achievement (McClelland, 1961); and stronger locus of control and risk-taking propensity (Brockhaus, 1980). Related to locus of control, the individual’s desire for autonomy was suggested as a potential driver of entrepreneurial motivation (Herron & Sapienza, 1992; Kolvereid, 1996a) as were attitudes toward workload and security (Kolvereid, 1996a). Demographic factors such as sex, age, education, and entrepreneurial parents were also cited as antecedents of entrepreneurial motivation. These variables were generally mentioned in the context of the theory of planned behaviour (TPB; Ajzen & Fishbein, 1980) and/or the theory of the entrepreneurial event (Shapero, 1975; Shapero & Sokol, 1982) as determinants of either perceived desirability or perceived feasibility. Shook et al. (2003) reported that (at that time) the bulk of empirical work had focused on individual perceptions of feasibility, desirability, and social support as determinants of entrepreneurial intention (see, e.g. Krueger, 1993; Krueger & Brazeal, 1994; Krueger & Carsrud, 1993; Kreuger et al., 2000). Shook et al. (2003: 385) stated that although the role of individual cognitions had received theoretical attention, “empirical work has yet to be done”. The Douglas and Shepherd (2002a) empirical study of the impact of individual attitudes on entrepreneurial intention was not included in their comprehensive review, most likely because their survey included papers explicitly based on the TPB. While our earlier conceptual paper (2000) did nominate both the attitudes and the abilities of the entrepreneur as being drivers of entrepreneurial intention, the empirical study (2002a) focused on the perceived desirability of entrepreneurship (through the lens of utility theory which posits that individuals will choose the action expected to generate the maximum utility, making it the most desirable option). More recently, Liñán and Fayolle (2015) reviewed 409 papers published in the decade 2004 to 2013 inclusive, seeking to elucidate the subfields of entrepreneurial intentions research. They identified five main cluster areas and a sixth “emerging research” cluster. In the cluster concerning personal-level variables that influence entrepreneurial intention, they note papers citing risk tolerance, by Segal, Borgia, and Schoenfeld (2005); career anchors, by Lee and Wong (2004); prior family business exposure, by Carr and Sequeira (2007); prior entrepreneurial activity, by Gird and Bagraim (2008); the influence

Commercial entrepreneurial intention

73

of university studies, by Guerrero, Rialp, and Urbano (2008); and the role of social capital, by Liñán and Santos (2007). Also during that period, Lee et al. (2011) had examined the impact of job satisfaction on entrepreneurial intention. Affirmations and extensions of the TPB during that decade included Van Gelderen et al. (2008) who affirmed the general applicability of the TPB for entrepreneurial intention, and Schlaegel and Koening (2014) who conducted a meta-analytic test and integration of the TPB and Shapero’s theory of the entrepreneurial event, concluding that both work well but the TPB exhibits somewhat greater explanatory power. Kolvereid and Isaksen (2006) conducted a longitudinal study of the TPB using Norwegian nascent entrepreneurs, and confirmed that entrepreneurial intention does indeed predict subsequent entrepreneurial behaviour. In Fitzsimmons and Douglas (2011), based on research in other domains that had indicated an interaction effect between perceived desirability and perceived feasibility, we hypothesised preventative self-regulation on the part of the intending entrepreneur (see Higgins, 1987; Shah & Higgins, 1997; Brockner et al., 2004) and demonstrated that a measure of perceived desirability (i.e. attitude to owning the venture) and a measure of perceived feasibility (i.e. entrepreneurial self-efficacy (ESE)) did indeed interact negatively, on average, for our sample.

5.3

PROFIT-MAXIMISING VERSUS PROFIT-SATISFICING ENTREPRENEURIAL INTENTION

In the decade covered by the review by Liñán and Fayolle (2015), virtually all entrepreneurial intention studies had used a generic construct of entrepreneurial intention, as if there was only one type of (commercial) entrepreneurship to be pursued, that being profit-seeking entrepreneurship.2 The generic monolithic construct of commercial entrepreneurial intention has been used in empirical studies despite the common observation that two main types of profit-seeking entrepreneurs are empirically evident – Ronstadt (1984) had much earlier distinguished between small-lifestyle ventures and high-growth ventures. Entrepreneurship textbooks, like Allen (1999) and Barringer and Ireland (2006), inform students of small businesses that focus on lifestyle benefits or salary substitution for their owners, in contrast to high-growth new ventures that focus on profitability and growth. Shane (2009) argued that public funding

2 An exception is a paper by Carey et al. (2010) who examined student intention for small-lifestyle, high-income, and high-growth ventures, and concluded that individuals have differing prior preferences for these subtypes of entrepreneurship.

Entrepreneurial intention

74

should be limited to firms with high-growth potential, rather than supporting “salary-substitute” new ventures that do not grow and consequently do not provide societal benefits of additional employment and company tax revenues. Based on these societal objectives, it is important that we seek a more fine-grained investigation of entrepreneurial intention, and consider “how much profit and growth” intending entrepreneurs hope to generate. 5.3.1

Profit maximisation

The implicit assumption, carried over from classical economics, is effectively that intending entrepreneurs would want to pursue profit maximisation. The inclusion of non-monetary benefits and costs in linear-additive regression (and utility) models of entrepreneurial intention effectively implied that the non-monetary (psychic) rewards and costs of entrepreneurship were independent of and additive to the monetary rewards. Yet we know that the achievement of greater profit will likely entail more risk, harder work, less autonomy, and/ or less work enjoyment. These interdependencies among the explanatory variables for entrepreneurial intention imply that tradeoffs must be made between additional profit and the other four sources of utility which underlie the perceived desirability of entrepreneurship. These tradeoffs mean that the extreme pursuit of profit maximisation would result in reduced levels of the perceived desirability of entrepreneurship – that is lesser levels of expected utility, compared to stopping somewhere short of profit maximisation at the point where the impact on total utility (i.e. marginal utility) of additional profit is just balanced by the marginal disutility that would be incurred from having less work enjoyment, harder work, less autonomy, and/or greater risk exposure. 5.3.2

Profit satisficing

Thus, strictly, profit maximisation would only be pursued by individuals who expect no utility or disutility from the non-monetary ancillaries of profit seeking. This does not seem to characterise very well entrepreneurs operating new and small ventures. Conversely, profit maximisation might be pursued by a large firm that has many owners (stockholders), with each having a miniscule share of the total ownership of the firm and who also have no empathy for the managers and employees of that firm. Instead we should expect entrepreneurs to be profit satisficing (Simon, 1972), ex ante, whereby they strive to earn a satisfactory level of profit (but may not succeed ex post) that is sufficient to meet their needs, and supplement that profit with the non-monetary psychic income of entrepreneurship. Simon argued that individuals try to maximise profit (and utility) by limiting information-search costs to the point where additional search cost is just compensated for by additional revenue generated,

Commercial entrepreneurial intention

75

such that information search will be truncated well short of a state of “full information”, such that there remains residual uncertainty for the entrepreneur. Thus firms must operate in a situation of uncertainty, and make decisions on the basis of the limited information that they have, which Simon (1972) called “bounded rationality”. Simon’s concept of profit satisficing relates to the impossibility of profit maximisation under conditions of uncertainty, due to the increasing marginal cost of information. Instead he argued that individuals would seek a satisfactory level of profit, rather than extend search costs further and thereby reduce profit to an unsatisfactory level. Here, in the context of the intention to start a new firm, we are adding to this information-cost argument (for stopping short of the extreme point of profit maximisation) a psychic-cost argument based on the tradeoffs between profit and other sources of utility (i.e. non-monetary rewards) that are available in the entrepreneurial process. The amount of profit to be sacrificed in order to gain additional (net) non-monetary rewards will differ across intending entrepreneurs, and will be dependent on their attitudes to profit, relative to their attitudes to the non-monetary rewards. At one extreme will be those intending to “almost maximise” profit (under conditions of bounded rationality) while gaining relatively low net utility from non-monetary rewards, while at the other extreme there will be those intending to gain only the minimum profit sufficient to meet their needs, while gaining relatively high net utility from non-monetary rewards. In all cases the profit level targeted can be regarded as the satisfactory profit level for that individual, and thus we can say that the entrepreneurs intend to be profit satisficing. Profit-satisficing entrepreneurs will likely include most lifestyle-, salary-substitute, and subsistence entrepreneurs (Barringer & Ireland, 2006), as well as mum-preneurs (Duberly & Carrigan, 2013); elder-preneurs (Kautonen et al., 2010; Watkins-Mathys, 2012), migrant-preneurs (Levie, 2007), and refugee-preneurs (Wauters & Lambrecht, 2008). On the other hand, serial entrepreneurs (Westhead et al., 2005) and international entrepreneurs (Joardar & Wu, 2011) might more likely be profit-maximising entrepreneurs. In Douglas (2013) I hypothesised that GOI is a separate and distinct construct to IOI and that these alternative types of entrepreneurship would attract different types of people. Rather than intend to start a new venture and separately consider whether they want the venture to grow or not, it was hypothesised that potential entrepreneurs would intend, from the outset, to start either a profit and growth-oriented venture or an independence-oriented venture, and that these separate intentions would be driven by different antecedent variables. The following section is drawn largely from that paper.

76

5.4

Entrepreneurial intention

GROWTH-ORIENTED VERSUS INDEPENDENCE-ORIENTED ENTREPRENEURIAL INTENTIONS

The individual–opportunity nexus (Shane & Venkataraman, 2000) exhibits heterogeneity on both sides of the nexus – individuals are different and the opportunities they seek to exploit also differ from each other. Davidsson and Wiklund (2001) argued that to understand entrepreneurial intention one must consider both the characteristics of the individual and the characteristics of the new venture that they intend to start. Studies that attempt to explain the antecedents of entrepreneurial intention but neglect the heterogeneity on the opportunity side (in effect assuming all opportunities to be the same), must confound the influence of opportunities and individuals, such that intentions attributable to individuals might depend on the specific opportunity envisioned by the individual. Accordingly, researchers must specify a particular new venture scenario, or a specific type of new venture opportunity, in order to derive reliable results on the antecedents of entrepreneurial intention for that particular opportunity or type of opportunity. In Douglas (2013) I specified two main types of commercial entrepreneurship, namely one relating to a “profit and growth” opportunity, where entrepreneurs would seek to maximise their profits, and one relating to an “independence” opportunity, where a profit-satisficing individual would pursue a satisfactory level of profit augmented by the non-monetary benefits associated with the pursuit of a limited-growth entrepreneurial opportunity. Growth of the firm allows the firm to increase its profit, up to a point, so profit maximisers will want to grow their firm to the point where further growth would reduce profit (due to diseconomies of firm size). Profit satisficers, on the other hand, may grow their firm to the size necessary to achieve their target profit level, but thereafter would not seek to grow the firm because further profit and growth would reduce their non-monetary rewards and hence reduce their total utility from entrepreneurship. These non-growth-oriented profit-satisficing individuals were dubbed “independence-oriented” (for want of a better name), on the presumption that their decision-making independence would be a substantial source of their non-monetary rewards. For “salary-substitute” entrepreneurs this might well be true, as many of these would enjoy being their own boss while practising their trade (e.g. plumbing or carpentry) or profession (e.g. accounting or architecture) under their own banner, in preference to a similar job working for an employer. For “lifestyle” entrepreneurs, working with customers and suppliers who share the same passion for a pastime (such as skiing, sailing, and horse-riding instructors), the major part of the non-monetary rewards might

Commercial entrepreneurial intention

77

derive from the work enjoyment resulting from interacting with like-minded enthusiasts. For “subsistence” and part-time entrepreneurs, who might work only long enough to earn the bare minimum income needed to provide for their needs (such as buskers, itinerant travellers, and retirees), the major part of their non-monetary rewards might derive from the avoidance of work effort, or conversely, the enjoyment of leisure. Thus, the term “independence-oriented” entrepreneurs used in that paper simply provides a collective term for a variety of individuals who share the desire to pursue a satisfactory (rather than a maximising) profit target while working for themselves in a non-growth venture. 5.4.1

The hypotheses

The 2013 paper sought to ascertain whether different people chose different types of entrepreneurial behaviour for different reasons. Yet it also recognised that a person might screen opportunities of both types before fixing on a particular entrepreneurial opportunity that was available at the time. The theoretical lens for the paper was “self-determination theory” (Deci & Ryan, 1985; Gagné & Deci, 2005; Ryan & Deci, 2000) which argues that individuals make choices and undertake actions that they believe will best serve their personal needs. Self-determination theory argues that individuals have three innate needs and other situational needs that they seek to fulfil, and here we are positing that the pursuit of a particular subtype of entrepreneurship may best fulfil their needs. The innate needs of humans are said to be the need for competence, the need for autonomy, and the need for relatedness (Deci & Ryan, 1985). The need for competence is manifested in the desire of individuals to be “good at” the actions they choose to take, and so they would be reluctant to attempt actions where they are not confident of success. In the language of the TPB this translates to their perceived feasibility of successfully completing the tasks required, which we have earlier defined as the individual’s ESE, and of course this is likely to vary across individuals and across types of entrepreneurship for any particular individual. Scholars have argued that growth-oriented entrepreneurship requires greater management ability than non-growth-oriented entrepreneurship (e.g. Covin & Slevin, 1997; Knockaert, Foo, & Erikson, 2011). Accordingly, it was hypothesised that ESE would be positively related to both GOI and to IOI, but ESE would be more-positively related to GOI than to IOI. The need for autonomy refers to the need to be free, to make one’s own decisions, to be self-directed, rather than enslaved or directed excessively by others. This need is manifested in the individual’s attitude to independence (or decision-making autonomy) and is also likely to vary across individuals and across types of entrepreneurship for any particular individual. It can

78

Entrepreneurial intention

be argued that the scope for independence and decision-making autonomy is greater in the pursuit of non-growth entrepreneurship than in the pursuit of growth-oriented entrepreneurship, due to the added complexities of the latter and the decision-making encroachment of other individuals and entities (including regulatory authorities, major customers, and suppliers). Accordingly, it was hypothesised that the individual’s attitude to independence would be positively related to both GOI and to IOI, but this attitude would be more strongly related to IOI than to GOI. The need for relatedness refers to the need for social relationships with other people. In the context of entrepreneurial intention this may be translated as the individual’s attitude to job satisfaction (Lee et al., 2011). In the model tested in the 2013 paper, this construct is truncated to include only the attitude to social interaction with co-workers, customers, and suppliers in the work environment, and called “work enjoyment”, due to other major aspects impacting job satisfaction (like risk exposure, decision-making autonomy, and expectations of hard work) being accounted for separately. Attitude to work enjoyment is likely to vary across individuals and across types of entrepreneurship for any particular individual. An argument can be made that the more one prefers (and acts to attain) profit maximisation, the less time remains for social interaction, such that social interaction has the opportunity cost of foregone profit. Thus, the more the individual acts to pursue profit and growth, the less they must value social interaction in the workplace. Accordingly it was hypothesised that an individual’s attitude to work enjoyment would be positively related to IOI, but negatively related to GOI. The situational needs occasioned by the pursuit of entrepreneurship can be modelled as the need for income, for risk avoidance, and for work-effort avoidance. Income is needed to provide the funds to hire employees, and to buy equipment, raw materials, and components, not to mention to buy food, clothing, and shelter for the entrepreneur and family. Preference for income (or the strength of the need for income) will vary across heterogeneous individuals, and different entrepreneurship opportunities will provide different prospects for income generation. The tradeoff at the margin between additional income, additional work enjoyment, additional independence, reduced risk, and reduced work effort (that is likely to happen in the pursuit of entrepreneurial opportunities) means that the more one pursues profit income the less one can experience work enjoyment and independence and the less one can avoid the disutility of additional risk and work effort. Accordingly it was hypothesised that attitude to income would be positively associated with GOI but negatively associated with IOI. Risk avoidance is needed to lessen the risk of financial, physical, and emotional damage to the individual that could eventuate in entrepreneurship. Individuals will prefer to avoid intolerable risk situations and seek to reduce

Commercial entrepreneurial intention

79

risks, where they can, to levels they feel are tolerable. Risk tolerance (or the strength of the need to avoid risk) varies across heterogeneous individuals, and different entrepreneurship opportunities will provide different exposure to risk for the individual. It can be argued that since risk and profit potential tend to be positively correlated, the pursuit of profit and growth requires greater risk tolerance than does the pursuit of non-growth entrepreneurship, such that those more intolerant of risk would incline toward non-growth entrepreneurship while those more tolerant of risk would tend to prefer growth-oriented entrepreneurship. Accordingly it was hypothesised that tolerance for risk would be positively related to GOI and negatively related to IOI. Work-effort avoidance is needed to eliminate the possibility of physical and mental exhaustion, and to reduce the disutility-inducing work-related feelings of tiredness, muscle pain, emotional stress, and so forth to tolerable levels. Work-effort tolerance (or the strength of the need for additional work effort) varies across heterogeneous individuals, and different entrepreneurship opportunities will provide different work-effort requirements for the individual. As argued above, the quantum of work effort would ideally be adjusted such that the marginal disutility of additional work effort would be just balanced by the marginal utility of the income generated by that additional work effort. For the totally lazy, or for those incapacitated in some way, the marginal disutility of work effort may be very high (in absolute terms) from the outset, but for most entrepreneurs we would expect work-effort tolerance to be relatively high initially and to reduce monotonically at the margin, such that marginal disutility of work effort rises the longer are the hours worked and the greater is the intensity of that work. This implies that the more intolerant the individual is of work effort, the less work effort they will prefer to supply and thus the less income (and growth) they would choose to pursue. Accordingly, it was hypothesised that tolerance for work effort would be positively associated with GOI and negatively associated with IOI. 5.4.2

The empirical study

The model for the empirical study was that both GOI and IOI would depend on ESE (representing perceived feasibility) and the attitudes to income, autonomy, work enjoyment, risk exposure, and work effort (representing perceived desirability). Socio-demographic variables were collected, namely sex, age, prior business degree (or not), and extent of prior work experience, since prior studies have indicated that they may provide additional explanatory power for GOI and IOI. The sample was 106 second-year MBA students at the Sasin Graduate School of Management in Thailand, and the data were collected in waves over several weeks to minimise common-method bias. The order of data collection was ESE first, using De Noble et al.’s (1999) 23-item scale; then

Entrepreneurial intention

80

Table 5.1

Scales for growth-oriented and independence-oriented entrepreneurial intention

How likely is it that you would want to start a new business venture that:

GOI

Exploits a new technology that promises to have very good prospects for

0.6036

IOI

long-term growth and eventual profitability? Is based on the gamble that a particular change in the laws will happen, and

0.6193

that you will be in a position to capitalise on that change? Will not be very profitable at first, requiring several rounds of external funding

0.5297

as it grows, before it eventually becomes highly profitable? Involves a high risk of failure, but is expected to be extremely profitable quite

0.5936

quickly if it takes off? Will slowly build up sales and eventually become a very large business with

0.4985

potentially thousands of employees around the world? Allows you to spend plenty of time away from work at rest or undertaking

0.6332

recreational activities? Allows you to earn enough money by doing the things you like best?

0.5965

Focuses on a recreational or other pastime that you really enjoy being involved

0.5980

in? Does not require you to work long hours every day, such that you can spend

0.6894

plenty of time at home and/or take part in social activities? Does not require you to work at high levels of intensity, which might cause

0.5466

fatigue, stress, or other undesirable effects? Capitalises on your knowledge and enthusiasm for a particular hobby, sport, or

0.6245

other recreational pastime that you really enjoy doing? Allows you to close the business and take time off, for holidays, whenever you

0.7237

choose? Allows you to work at a job that is more like “play” than work, since you

0.6582

really enjoy the type of work you would be doing? Cronbach’s alpha

0.7105

0.8385

the socio-demographic data; then the conjoint experiment to reveal respondent attitudes; and finally the intentions data. Separate multi-item scales were developed and pre-tested to measure each of GOI and IOI. The final items are reproduced in Table 5.1. Factor analysis revealed five items indicating profit and growth intentions and eight items indicating IOI. Loadings less than 0.4 are not shown. GOI and IOI were positively but weakly correlated with each other (r = 0.175), which was marginally significant at the 10 per cent level. This indicated that respondents were, in general, inclined towards both growth-oriented and independence-oriented entrepreneurship, but almost universally had a stronger intention for one or the other.

Commercial entrepreneurial intention

81

To ascertain respondents’ attitudes to income, autonomy, work enjoyment, risk exposure, and work effort, a conjoint experiment was conducted, following prior explication and application of this method by Shepherd and Zacharakis (1999), Douglas and Shepherd (2002a), Monsen et al. (2010), and others. In a conjoint experiment respondents rate the desirability of the outcomes of an action (e.g. entrepreneurial behaviour) conjointly (i.e. in a series of scenarios characterised by combinations of the outcomes set at “high” or “low”). This not only allows the interdependencies of these outcomes (i.e. between and among income, autonomy, work enjoyment, risk exposure, and work effort) to be incorporated into their personal rating of the scenarios, but also avoids the individual having to contemplate or reveal their attitude to any of these outcomes individually, thus avoiding (or at least reducing) the social-desirability bias that might impact their response if asked to espouse each of the attitudes separately. Conjoint analysis allows the separate attitudes to the five salient outcomes to be isolated and thus revealed to the researcher indirectly, rather than the less reliable espoused attitudes that might be collected from responses to single- or multi-item survey questions. The individual’s attitudes are revealed to the researcher by regressing the Likert-scale (1–7) rating (for all the scenarios presented to individuals) against binary variables set to 0 for “low” and 1 for “high” in different combinations in each of the scenarios presented (with definitions provided for what is meant by “high” and “low”). The coefficients in the regression equation represent the average change in the Likert score (representing the change in expected utility) when an outcome is high, rather than low, all other outcomes being the same, thus revealing the respondent’s attitude to having more of that outcome. The explained variance of the regression equation for each respondent (i.e. the R2 statistic) indicates the care and attention paid to the rating of the scenarios by each respondent. For example, if two scenarios have the same combination of high and low outcomes for four of the five outcomes, while the fifth outcome (e.g. autonomy) is high in the first scenario and low in the second scenario, the first scenario should logically be rated higher than the second scenario, since more autonomy is expected to be better than less autonomy. Alternatively, if a respondent does not like autonomy, he/she should respond in the negative consistently across scenarios. We cannot expect R2 = 1, however, due to the intransitivity which may result from interdependencies among the attitudes (to income, autonomy, work enjoyment, risk, and work effort). Accordingly the rule was followed to eliminate from the analysis any respondents with R2 < 0.5, implying that the simple correlation between the “high” setting and the attitude was at least r = 0.7 on average.

Entrepreneurial intention

82

5.4.3

The results

The regression equations for both GOI and IOI are shown in Table 5.2. It can be seen that GOI is positively and highly significantly associated with ESE; negatively and significantly associated with attitude to work enjoyment; and positively and marginally significantly associated with being male, with all other independent variables being insignificant in this equation. For IOI, only attitude to risk (i.e. risk aversion) is significant, being negatively associated with IOI. Notably, for all independent variables except age and prior business degree, their coefficients show opposite signs for GOI and IOI, and although these are not significant in most cases, indicate that GOI and IOI might be polar opposites for many individuals in this sample. Of course, the regression coefficients are effectively the sample averages of the respondents’ coefficients between the dependent variable and each of the independent variables, showing the dominant net effects, and therefore do not reveal individuals for whom other independent variables were significantly associated with GOI or IOI, nor individuals for whom the relationship is the opposite of the dominant net effect. In Chapter 9 we use fuzzy-set qualitative comparative analysis to find that most of the antecedent variables (in a different sample) do indeed drive entrepreneurial intention for at least some subgroups of individuals within that sample.3 It is notable that the explanatory power of the regression equation for IOI is much lower (R2 = 0.0938) than for GOI (R2 = 0.1554), and that the IOI equation is not significant (p = 0.2854). The greater unexplained variance for the IOI equation indicates the greater heterogeneity of “independence-oriented” entrepreneurial intention. That catch-all subtype is a combination of profit-satisficing subtypes who might be hypothesised to trade off profit for different main reasons – that is, salary-substitute entrepreneurs may trade off profit mainly for the sake of their autonomy; lifestyle entrepreneurs may do it mainly for the sake of work enjoyment; and subsistence entrepreneurs may do it mainly to avoid excessive work effort (or to enjoy greater leisure). 5.4.4

Subtypes of profit-satisficing commercial entrepreneurs

In addition to the two main types of commercial entrepreneurs identified empirically in Douglas (2013), three main subtypes of profit-satisficing entrepreneurs are suggested, namely independence-seeking salary-substitute

3 In Douglas et al. (2019) we do apply fsQCA to the data of the Douglas (2013) study, and find that separate subgroups can be identified with different configurations that lead to both GOI and IOI.

Commercial entrepreneurial intention

Table 5.2

83

Regression coefficients for the antecedents of GOI and IOI

Variable

GOI model

IOI model

Constant

2.9343***

5.352***

Entrepreneurial self-efficacy

0.6551***

−0.0447

Attitude to income

0.0406

−0.0405

Attitude to autonomy

−0.0089

0.3042

Attitude to work enjoyment

−0.2926**

0.1617

0.2362

−0.3429**

Attitude to work effort

0.1743

−0.0586

Age

−0.0114

−0.0124

Sex (male = 1)

0.32108

−0.1860

Prior business degree (yes = 1)

0.1895

0.0711

Pseudo R2

0.1554

0.0938

RMSE

0.9694

0.8871

0.0260**

0.2854

Attitude to risk

Probability

Notes: ** p < 0.05; *** p < 0.01; N = 106.

entrepreneurs, lifestyle entrepreneurs, and subsistence entrepreneurs. As noted earlier, other entrepreneurship subtypes may include mum-preneurs, elder-preneurs, migrant-preneurs, serial entrepreneurs, international entrepreneurs, and of course social entrepreneurs. It is anticipated that these subtypes would be either profit maximisers who seek profit and growth (perhaps including most international entrepreneurs), or profit satisficers, who seek satisfactory levels of profit in combination with (net) non-monetary benefits that serve to maximise their expected utility (perhaps including most mum-preneurs and elder-preneurs). This is an empirical question, of course, and no doubt earnest arguments could be developed to hypothesise the sign and effect size of variables theorised to underlie the perceived desirability and the perceived feasibility of each of these subtypes of entrepreneurship. Such studies of entrepreneurial subtypes are interesting and theory building if they (i) hypothesise and demonstrate that new, previously unconsidered variables significantly affect the formation of entrepreneurial intention for the subtype under consideration, or (ii) hypothesise and demonstrate the opposite directional impact on entrepreneurial intention, compared to prior studies, of a particular antecedent variable. For example, in Douglas (2013), risk tolerance was hypothesised to have a negative relationship with entrepreneurial intention for independence-oriented individuals, but a positive relationship for growth-oriented individuals. Similarly, socio-demographic variables, such as being married, may have a positive relationship with entrepreneurial intention for independence-oriented individuals, but a negative relationship for

84

Entrepreneurial intention

growth-oriented individuals. Or, the respondent’s number of children may have a positive relationship with entrepreneurial intention for mum-preneurship intention, but a negative relationship for elder-preneurship intention. These are empirical questions which may have important implications for policy, and which may be answered by future studies. In Chapter 9 we use fuzzy-set qualitative comparative analysis to identify specific combinations of antecedent variables for individuals and groups of individuals who have different reasons for wanting to behave entrepreneurially, often running counter to the dominant net-effects prescription identified by the regression equation.

5.5 SUMMARY In this chapter we have examined prior research investigating the antecedents of commercial entrepreneurial intention. Commercial entrepreneurs were defined as those whose primary purpose is to gain private benefits (profit and non-monetary benefits), as distinct from social entrepreneurs, whose primary purpose is to provide social benefits to others. Commercial entrepreneurs may be profit maximisers or profit satisficers. Profit maximisation implies little or no tradeoff between profit and the other non-monetary outcomes of entrepreneurship (namely, autonomy, work enjoyment, risk exposure, and work effort) because these outcomes tend to be interdependent – more profit is likely to mean less autonomy, less (time for) work enjoyment, bearing more risk, and harder work. In the utility model of behavioural motivation, the individual introspectively considers what combination of the outcomes is expected to generate maximal utility. The profit maximiser maximises utility by a combination of more profit, less autonomy, less work enjoyment, more risk, and harder work. Profit satisficing implies that the individual foregoes greater levels of profit, maximising utility via a combination of less profit, more autonomy, more work enjoyment, less risk, and/or less work. It follows that in a study of entrepreneurial intention using a generic, monolithic measure of entrepreneurial intention, the regression coefficients to the independent variables would depend on the relative proportions of profit maximisers and profit satisficers in the sample. If the former group predominates, the effect size for attitude to profit (for example) would tend to be higher and that variable would be more strongly significant, compared to if the latter group predominates. Or, if the two groups were of roughly even size, they might cancel each other out, such that any particular independent variable is reported as insignificant in the regression equation. The antecedents of entrepreneurial intention were argued to be ESE representing perceived feasibility, and the attitudes to income, autonomy, work enjoyment, risk, and work effort representing the perceived desirability. While attitudes to each of these outcomes of entrepreneurship can be theorised to

Commercial entrepreneurial intention

85

drive entrepreneurial behaviour for at least some individuals, when studied at the collective level, a particular attitude may not be found statistically significant for the sample as a whole. Conversely, while ESE may be significantly associated with entrepreneurial intention at the sample level, this dominant net effect may not be mirrored by some individuals within the sample. Similarly, for any hypothesised socio-demographic antecedent of entrepreneurial intention, the regression equation may report as significant only the dominant relationship between the independent and the dependent variable, ignoring other respondents in the minority where the relationship was the opposite. Thus regression analysis has limitations as an analytical method when seeking to understand the reasons why a particular individual would decide to undertake entrepreneurship. We return to this issue in Chapter 9 where we utilise a complementary analytical technique called fuzzy-set qualitative comparative analysis to reveal additional more nuanced information about commercial entrepreneurial intentions.

6. Intrapreneurial intention 6.1 INTRODUCTION Entrepreneurship has earlier been defined as a behaviour which involves proactive risk-taking innovation, in pursuit of new wealth via the introduction of new products, services, and/or business processes.1 Such behaviour can be conducted at the level of the individual, the firm (or other organisation), and the nation. In this chapter we are concerned with entrepreneurial behaviour by individuals as employees within firms, which was initially known as “intracorporate entrepreneurship” (e.g. Collins & Moore, 1970; Susbauer, 1973) and later “internal entrepreneurship” (e.g. Schollhammer, 1982; Vesper, 1980) before becoming known as intrapreneurship following Pinchot (1985, 1987). Intrapreneurship can be regarded as “employed entrepreneurship” as distinct from self-employed entrepreneurship, and contributes to corporate entrepreneurship at the firm level (see, e.g. Antoncic & Hisrich, 2001; Burgelman, 1983; Honig, 2001; Lumpkin & Dess, 1996; Morris et al., 2010; Parker, 2011). In this chapter we are concerned with the antecedent drivers of intrapreneurial intention, and whether these are different from the drivers of (self-employed) entrepreneurial intention. Since both individual and corporate entrepreneurship are desirable for national economic growth and global competitiveness, it is important to understand both types of entrepreneurship behaviour (Burgelman, 1983; Honig, 2001; Parker, 2011). As Honig (2001: 22) states, “establishing differences between nascent entrepreneurs and nascent intrapreneurs is... of considerable theoretical and empirical interest”. By and large, the prior literature had implicitly considered that entrepreneurs and intrapreneurs are basically the same, motivated by additional income and the psychic benefits of entrepreneurial behaviour, but demotivated by the risk and the hard work involved (Douglas & Shepherd, 2000, 2002a). Other researchers had generally assumed that entrepreneurs and intrapreneurs are essentially similar in terms of their human capital (Menzel, Aaltio, & Ulijn, 2007: 734; Parker, 2011: 27), and that their risk attitudes and cognitive styles

1 This chapter is based on a previously published article, Douglas and Fitzsimmons (2013).

86

Intrapreneurial intention

87

are also rather similar (Hisrich, 1990; Hitt, Ireland, & Hoskisson, 2001). Antoncic and Hisrich (2001) and later Antoncic (2003), identified that personal risk aversion was a negative influencing factor that operates to induce the choice of intrapreneurship (over entrepreneurship). Intrapreneurs are typically protected from risk to some large extent by their employer, whereas entrepreneurs are at greater risk of job loss and wealth loss if the new venture were to fail. Monsen et al. (2010) tested the hypotheses that intrapreneurs are positively motivated by financial gain, finding supportive evidence for the direct effect of profit sharing on the desire to participate in corporate venturing. They also found the direct effect of profit sharing was moderated positively by expectations of project success, and moderated negatively by pay risk and job risk, and that work effort required had a direct negative effect (rather than the negative moderating effect hypothesised). Many previous studies have examined the individual’s choice between self-employment and employment (e.g. Carter, Gartner, Shaver, & Gatewood, 2003; Douglas & Shepherd, 2002a; Evans & Leighton, 1989; Katz, 1992; Kolvereid, 1996a, 1996b; Reynolds, 1997; Shaver, Gartner, Crosby, Bakalarova, & Gatewood, 2001). Such studies have generally proceeded on the basis of this simple “employed vs. self-employed” dichotomy, and have failed to recognise the intermediate case where the individual can behave entrepreneurially as an employee within a corporate context. For example, Carter, Gartner, et al. (2003) compared the reasons for career choice of nascent entrepreneurs and non-entrepreneurs, but chose to exclude those who were in the process of starting a new venture in the context of their employer. Further, while much has been written about how managers might motivate increased intrapreneurial behaviour from existing employees in the corporate context (e.g. Hamel, 2000; Morris et al., 2010), little is known about how employers might best select employees who have a propensity for intrapreneurship. Insight into the cognitive and human capital differences of those intending employment as an intrapreneur might allow more efficient matching of individuals with corporate roles that require intrapreneurial behaviour, compared to other roles that are largely administrative with little scope for individual initiative or creativity (Stevenson, 1983). Moreover, firms seeking intrapreneurial employees may need to understand that hiring individuals with strong (self-employment) entrepreneurial intention may not be the best strategy, as such employees may not dwell long in the corporate context before leaving to become self-employed. Policy makers should also understand that a different policy approach may be required to encourage individuals to introduce innovations as distinct from encouraging firms to do this. Public policy to nurture and support fledgling firms may be pushing individuals into entrepreneurial new ventures where the person is better suited to intrapreneurship, and thereby wasting public

88

Entrepreneurial intention

resources (Shane, 2009), not to mention increasing the personal cost of entrepreneurial failure (Shepherd, 2003; Shepherd, Wiklund, & Haynie, 2009). There are also implications for entrepreneurship education – institutions that teach entrepreneurship in one subset of courses, while teaching management, marketing, production, operations, and so forth in the corporate context, need to understand that potential intrapreneurs may not be best served by either of these polar perspectives. In Douglas and Fitzsimmons (2013) we investigated whether individuals regard entrepreneurship and intrapreneurship as different career paths, and this chapter relies on that paper. The first research question was whether self-employed entrepreneurship and corporately employed intrapreneurship are viewed by individuals as separate and distinct entrepreneurial behaviours. The second research question was whether the antecedents of intrapreneurial intention are the same as, or are different from, the antecedents for individual entrepreneurial intention. In the following sections we overview the theory building and hypothesis development, then discuss the sample and research method, and report the analysis and results.

6.2

DEVELOPMENT OF HYPOTHESES REGARDING INTRAPRENEURIAL INTENTION

We use the theoretical lens of the theory of planned behaviour to argue that intrapreneurship behaviour is best predicted by intentions toward that behaviour. In line with the analysis in prior chapters, the intention to undertake intrapreneurial behaviour is expected to depend on the perceived desirability and the perceived feasibility of intrapreneurship. The perceived desirability of this behaviour will depend on the individual’s attitudes towards the outcomes of intrapreneurship (Robinson, Simpson, Huefner, & Hunt, 1991), and the more positive are these attitudes, the greater will be the intention to engage in intrapreneurship (Ajzen, 1985, 1991; Fishbein & Ajzen, 1975). A strong positive attitude to an outcome implies that the individual expects to gain substantial psychic satisfaction from experiencing that outcome, and this militates in favour of the individual subsequently pursuing that action (Douglas & Shepherd, 2000). The perceived feasibility of intrapreneurial behaviour depends on the individual’s introspective perception that he/she can successfully complete the tasks associated with intrapreneurship. Thus we need to specify the outcomes and the tasks associated with intrapreneurship. For the purposes of comparing and contrasting the individual’s attitudes and abilities as applied to either intrapreneurship or entrepreneurship, we assume that a specific wealth-creating opportunity is considered from each of these two perspectives – i.e. it might be exploited either by the individual forming a new business entity (i.e. entrepreneurship) or by the individual

Intrapreneurial intention

89

working within an existing business firm (i.e. intrapreneurship). Thus the outcomes and the tasks are held the same, such that any difference in the individual’s intention (between the two modes of exploitation) will be due to differences in the individual’s perceived share of the outcomes (e.g. profit and risk) in the two different contexts, and to differences in their perceived responsibility for the tasks (e.g. sole versus shared) in the two different contexts. That is, the proportion of each outcome accruing to the individual, and the proportion of each task that must be accomplished by the individual, may be perceived to be higher or lower when a given opportunity is to be exploited as an entrepreneur, as compared to exploiting the same opportunity as an intrapreneur, due to different weights they place on the salient outcomes. For example, if the individual is highly risk averse – and if the intrapreneurial exploitation alternative would expose the individual to less risk – this would augur for a preference for the intrapreneurial alternative. The salient outcomes, or the “payoffs” to entrepreneurial activity (Baumol, 1990), include both monetary and intrinsic benefits and costs associated with entrepreneurial activity (Wiklund, Davidsson, & Delmar, 2003). In Douglas and Shepherd (2000) we specified the intrinsic costs as those relating to risk taking and providing work effort, and specified the intrinsic benefits as relating to decision-making autonomy and a catch-all category of all other (net) perquisites. Regarding these intrinsic benefits, McClelland (1961) had earlier argued that a major benefit of entrepreneurial activity relates to a sense of achievement, and other scholars had considered a range of intrinsic benefits that are (in prospect) or were (in retrospect) expected to be attained as a result of becoming an entrepreneur (see, e.g. Birley & Westhead, 1994; Gatewood, Shaver, & Gartner, 1995; Kolvereid, 1996a, 1996b; Scheinberg & MacMillan, 1988). Subsequently Shaver et al. (2001: 8) mention the “fundamental elements of independence and ownership” as salient outcomes of entrepreneurship, and in Fitzsimmons and Douglas (2011) we found that attitude to ownership of the firm is a distinctly separate construct to attitude to independence. Underlying these attitudes might be the pride of achievement, the possession of discretionary power, and/or the recognition by others that one is a business owner (in entrepreneurship) or the psychological owner of a project (in intrapreneurship). In Douglas and Shepherd (2000) we implicitly argued that the perceived desirability of a career opportunity is a higher-order construct encompassing attitudes to income, autonomy, risk exposure, work effort, and other net perquisites, where no particular attitude is either necessary or sufficient for perceived desirability. Rather, it is the combined impact of these, in conjunction with the salient outcomes of the entrepreneurial opportunity that determines perceived desirability (i.e. greater utility) or not. The individual’s attitudes toward the salient outcomes of an entrepreneurial opportunity (i.e. the valences) form the weights attached to the quantum expected to be provided of

90

Entrepreneurial intention

each salient outcome (i.e. the expectancies) by the entrepreneurial opportunity (Van Eerde & Thierry, 1996; Vroom, 1964). The opportunity is perceived to be most desirable (i.e. utility maximising) if the weighted sum of the salient outcomes (i.e. the expected utility of the opportunity) exceeds that of all other accessible employment and self-employment opportunities. An individual’s entrepreneurial self-efficacy (ESE) has also been argued to be a significant driver of entrepreneurial intention (Krueger, 1993; Krueger & Brazeal, 1994; Markman, Balkin, & Baron, 2002; McMullen & Shepherd, 2006; Newman et al., 2019). Self-efficacy is the strength of an individual’s belief that they can successfully accomplish a specific task or series of related tasks. It is related to self-confidence and individual capabilities, which are dependent on prior experience, vicarious learning, social encouragement, and physiological issues (Bandura, 1982). ESE relates to the individual’s confidence that they can successfully accomplish tasks associated with individual entrepreneurship, and ESE has been found to be related to the formation of entrepreneurial intention (Boyd & Vozikis, 1994; Chen et al., 1998; De Noble et al., 1999; Markman et al., 2002; McGee et al., 2009). In this chapter we ask whether ESE is also related to the formation of intrapreneurial intention. Monsen et al. (2010) investigated the intention to participate in intrapreneurship among current employees, and found that the prospective willingness to participate in intrapreneurship depends negatively on both the amount of risk exposure and the work effort expected, and positively on the share of profits (or bonus) they expect to receive for successful projects. These results align with the predictions of Douglas and Shepherd (2000) for entrepreneurship motivation, although subsequently we found that attitude to work effort was not significantly related to entrepreneurial behaviour (Douglas & Shepherd, 2002a). Parker (2011) hints at the influence of the individual’s cognitions on entrepreneurial and intrapreneurial intention when he notes that the “independence-seeking [individual] is more likely to engage in [entrepreneurship rather than intrapreneurship]” (Parker, 2011: 22), and that “one might expect” more risk-averse individuals to be more likely to engage in intrapreneurship (than entrepreneurship) since existing firms might be expected to provide a “forgiving environment within which to try a start-up” (Parker, 2011: 31). On the basis of this prior literature, we developed a series of hypotheses regarding the relationship between attitudes and ESE and the formation of the intention to become either an entrepreneur or an intrapreneur. 6.2.1

Is intrapreneurial intention a distinct construct?

The first research question to be tested is whether entrepreneurial intention and intrapreneurial intention should be treated as distinct and separate constructs.

Intrapreneurial intention

91

We argued that they should be, because the profiles of the salient outcomes are typically perceived to be substantially different in these alternative modes of exploitation. First considering income, the individual entrepreneur (as majority owner) will be the residual claimant of the majority of the firm’s profits, whereas the intrapreneur (as an employee, possibly with a minority-ownership share) may be awarded a relatively small portion of the firm’s profits as a performance bonus or dividend but will expect that most of the profits will accrue to the majority shareholders. Thus, for a given opportunity the expected income will be substantially greater for the individual if the opportunity is exploited as an entrepreneur as compared to exploiting it as an intrapreneur, other things being equal. Second, decision-making autonomy for the individual exploiting an opportunity will be substantially greater if the individual were to exploit that opportunity as an entrepreneur rather than as an intrapreneur. As owner-manager of the new venture the entrepreneur must make all strategic and operational decisions (albeit potentially advised and/or restricted by mentors, investors, bankers, and possibly others) whereas as an intrapreneur the individual should expect to defer to senior management and/or corporate policies on at least some decisions that would have been made by the individual if he/she were an entrepreneur. Third, it was argued that ownership of the firm in which one works will confer greater psychic benefits to the individual as an entrepreneur than would non-ownership of a firm to the same person when exploiting the same opportunity in an employment position within that firm. Shaver et al. (2001) suggest that ownership is an important driver of entrepreneurial behaviour, and Fitzsimmons and Douglas (2011) found empirically that the psychic benefits of business ownership are separate from the psychic benefits of decision-making independence, and that these are indeed separate constructs. The expected benefits of business ownership relate to the psychic utility an owner might derive from pride, sense of achievement, power to set strategic direction and tactics, power to hire and fire employees, and so on. Majority ownership may simultaneously expose the individual to psychic costs associated with additional levels of responsibility and stress associated with managing people, making individual decisions, having a wider span of control, and concern for employee wellbeing (Wiklund et al., 2003). While non-ownership employment situations may also offer pride, prestige, and power, these psychic benefits are likely to be greater in a self-employment majority-ownership situation, other things being equal. The psychic costs of the non-owning intrapreneur do not include some that are borne by the owner-manager entrepreneur but the former is also likely to have lesser authority to control or eliminate irksome working conditions. Thus we conclude that majority ownership of the firm in which

92

Entrepreneurial intention

one works will confer greater net psychic benefits than would a non-ownership employment position within the firm, all else equal. Fourth, the individual’s exposure to risk will be viewed as greater in the role of entrepreneur as compared to intrapreneur, other things being equal, since the former is the majority owner of the firm that may be bankrupted by undertaking the new venture, while the latter is not. The risk of bankruptcy will be perceived as higher for the entrepreneurial alternative (Shepherd et al., 2000; Stinchcombe, 1965) than it would be for the intrapreneurial alternative, since an established firm will have other products or services already generating revenues. The risk of losing one’s job is likely to be greater for the individual entrepreneur as well, since the intrapreneur might expect to be reassigned somewhere else in the organisation if the new venture does fail. Within an existing organisation the intrapreneur can expect to seek help from more senior managers, who might be expected to provide advice and additional resources if the individual gets into difficulty, which would also cause the individual to perceive intrapreneurship as less risky than entrepreneurship, other things being equal. Finally, work effort required is probably perceived to be significantly higher in the entrepreneurship alternative than in the intrapreneurship alternative. Stories abound of entrepreneurs having to work long and hard to keep their new venture alive and to grow in the face of unexpected crises and setbacks (see, e.g. Bird & Jellinek, 1988). As an intrapreneur, the individual might expect to seek higher-level approval for the reassignment of other employees or the recruitment of additional employees to help cope with the workload. The foregoing analysis outlining the difference in the expected salient outcomes associated with entrepreneurship compared with intrapreneurship therefore suggests the hypothesis that entrepreneurship and intrapreneurship are viewed as distinctly different career alternatives. 6.2.2

Attitudes to the outcomes of intrapreneurship

Next, we consider the individual’s attitudes to each of the five salient outcomes and note that attitudes are dispositional and likely to be the same across different choice alternatives. Starting with income, we note that the individual’s attitude to income is derived from their desire for goods and services that can be purchased from income. Basic economic theory teaches that the individual’s desire for goods and services is effectively unlimited, and that fulfilment of those desires provides psychic wellbeing. Individuals with stronger desires for goods and services will tend to desire greater incomes, and since both entrepreneurship and intrapreneurship are generally expected to generate higher incomes than ordinary employment, we expect that the formation of both entrepreneurial and intrapreneurial intention will be positively related to

Intrapreneurial intention

93

the individual’s attitude to income. We note that Monsen et al. (2010) found that intrapreneurial intention was positively related to the share of profits or bonus that respondents expected to receive for successful projects. This suggests that the stronger is the attitude to income, the more positive will be both entrepreneurial intention and intrapreneurial intention. Next concerning the individual’s attitude to decision-making autonomy, or independence, we know from self-determination theory that the need for autonomy is an innate psychological need (Deci & Ryan, 1985). Prior research in the entrepreneurship domain has indicated that entrepreneurs value independence and decision-making control (e.g. Wiklund et al., 2003), and the individual’s attitude to independence is likely to be positively related to the intention to become an entrepreneur (e.g. Douglas & Shepherd, 2002a; Shane, 2003: 106–8), but will this also hold for intrapreneurial intention? An intrapreneur is likely to be allowed more decision-making independence within the organisation than are other employees who are not involved in corporate entrepreneurship projects. Accordingly, individuals are likely to expect greater autonomy in both entrepreneurship and intrapreneurship as compared with ordinary employment, so the more they value decision-making autonomy the more they are likely to want to become either an entrepreneur or an intrapreneur. This suggests the hypotheses that the more positive is the attitude to independence, the more positive will be the entrepreneurial intention, and similarly for intrapreneurial intention. Now considering the individual’s attitude to the net psychic benefits associated with majority ownership of the firm, we note that as well as increased decision-making independence, ownership of the firm also serves to satisfy the individual’s need for achievement, power, and recognition. Wiklund et al. (2003) found that the growth motivation of small business owners was positively related to attitude for employee wellbeing, indicating that ownership of the firm tends to satisfy the entrepreneur’s need to take care of his/ her employees. Shaver et al. (2001) argued that independence and ownership are the two main reasons for entrepreneurial behaviour, and Fitzsimmons and Douglas (2011) found that entrepreneurial intention was positively related to the individual’s attitude to majority ownership of the firm, so we expect to find the same here. But we do not expect attitude to majority ownership to be positively related to intrapreneurial intention – instead we expect that these variables will be negatively related, since the stronger one’s preference for majority ownership the less satisfying intrapreneurship would be, other things being equal. Thus we expect that attitude to majority ownership will be positively related to individual entrepreneurial intention and negatively related to intrapreneurial intention, other things being equal, since the individual could less likely be able to claim responsibility for firm achievements, and would less likely be recognised publicly for his/her achievements, and would likely

94

Entrepreneurial intention

have less control of decisions that affect employee wellbeing. This suggests the hypotheses that the more positive is the attitude to majority ownership, the stronger will be entrepreneurial intention but the weaker will be intrapreneurial intention. The individual’s attitude to risk has been shown to generate mixed results in the context of entrepreneurial intention and behaviours. While Douglas and Shepherd (2002a) found risk tolerance to be positively related to entrepreneurial behaviour, many others have found it to be unrelated to entrepreneurial behaviours (e.g. Brockhaus, 1980; Busenitz & Barney, 1997; Palich and Bagby, 1995). These mixed results indicate that the relationship between attitude to risk and entrepreneurial behaviour is more complex. Palich and Bagby (1995) and Busenitz and Barney (1997) argue that risk is often misperceived by nascent entrepreneurs due to overconfidence, the use of heuristics, or optimistic framing of the opportunity (see also Douglas, 2009; Krueger & Dickson, 1994). At the intentions stage of the entrepreneurial process we are concerned with the individual’s perception of risk in two alternative modes of opportunity exploitation – one acting as entrepreneur and the other as intrapreneur. Our question here is whether one’s attitude to perceived risk will have a different impact on intentions in the entrepreneurship versus the intrapreneurship mode of exploitation. Note that in Fitzsimmons and Douglas (2011) we found attitude to risk to be unrelated to entrepreneurial intention, while Monsen et al. (2010) found prospective willingness to participate in intrapreneurship depends negatively on risk exposure. In light of these prior findings, we expect individuals to be risk averse, and to recognise that entrepreneurship carries more risk than does intrapreneurship, other things being equal. This suggests the hypotheses that the more positive is tolerance for risk, the more positive will be the entrepreneurial intention; and that the less positive is tolerance for risk, the more positive will be the intrapreneurial intention. Finally, concerning attitude to work effort, Bird and Jellinek (1988) noted that entrepreneurs often need to work very hard. In Douglas and Shepherd (2000) we proposed that the individual’s attitude to work effort would be negatively related to entrepreneurial behaviour, but in a subsequent paper (Douglas & Shepherd, 2002a) we failed to find a significant empirical relationship between attitude to work effort and entrepreneurial activity. Wiklund et al. (2003) found that attitude to workload was significantly positively related to growth motivation of small business owners in a very large Swedish sample but this relationship was only significant in one of the three smaller subsamples. Since work effort has physiologically deleterious effects after some point, we assume that individuals generally are averse to greater work effort, other things being equal.

Intrapreneurial intention

95

Thus we expect that work-averse individuals will tend to avoid occupations where the work-effort demands are heavier, other things being equal. We also expect that an individual contemplating working as either an intrapreneurial employee or as a self-employed entrepreneur would consider the likely work -effort requirements of these alternatives in conjunction with the strength of his/her own attitude towards work effort, and we have argued earlier that entrepreneurship is likely to be perceived as requiring greater work-effort demands than intrapreneurship. This suggests the hypotheses that the more positive is the tolerance for work effort, the more positive will be the entrepreneurial intention; and that the less positive is the tolerance for work effort, the more positive will be the intrapreneurial intention. 6.2.3

The impact of entrepreneurial self-efficacy on intrapreneurial intention

The impact of ESE on the intention to behave entrepreneurially has been well discussed in the context of individual entrepreneurship (e.g. Chen et al., 1998; De Noble et al., 1999; McGee et al., 2009), but little if any research has investigated the relationship between ESE and the intention to engage in intrapreneurship. As noted earlier, greater ESE implies greater confidence to successfully complete entrepreneurial tasks, while lesser ESE implies lesser confidence to successfully complete such tasks, and we might expect the less confident individual to want the advice and guidance of others. Moreover we have argued above (in the context of decision-making autonomy) that individual entrepreneurship is, in general, likely to constitute a more difficult set of tasks than does intrapreneurship. As an intrapreneur the individual can more readily seek advice and direction from more senior managers, but as an entrepreneur the individual needs to be much more self-reliant. Moreover, seeking advice and direction from others takes time and may incur financial and psychic costs. Accordingly, we expect that the greater the individual’s ESE the less likely they will be seeking advice from others as to the appropriate action, and the more likely they are to prefer entrepreneurship to intrapreneurship, other things being equal. We nonetheless expect a positive relationship between ESE and intrapreneurial intention because intrapreneurship is an outlet for the individual’s ESE and an opportunity to demonstrate competence. This suggests the hypotheses that the more positive is ESE, the more positive will be entrepreneurial intention; and similarly that the more positive is ESE, the more positive will be intrapreneurial intention.

96

6.3

Entrepreneurial intention

SAMPLE AND METHOD

Our sample was 414 MBA students surveyed at the beginning of their entrepreneurship course in Australia (n = 46), or China (n = 39), or India (n = 204), or Thailand (n = 125) in 2003–2004. These students might reasonably be considered potential entrepreneurs and/or intrapreneurs, since they were approaching a career decision point at which they might either enter into entrepreneurship, intrapreneurship, or some other employment (Shepherd & DeTienne, 2005). In their course they were required to form groups to write a formal business plan for a new venture in the context either of a new firm (entrepreneurship) or an existing firm (intrapreneurship). The sample for each country was generally similar in characteristics such as age, work experience, and prior educational background which allowed us to focus on other potential determinants of their intentions. We measured intentions using a seven-point scale ranging from very unlikely (1), to very likely (7), over seven items measuring intentions to engage in a range of entrepreneurial behaviours including items related to entrepreneurial activity and intrapreneurial activity (see Table 6.1). We employed principal components analysis to investigate the underlying structure of these items, with oblique rotation. To obtain measures for the entrepreneurial attitudes of the individuals in the sample we used a conjoint experiment. Respondents were asked to evaluate hypothetical scenarios that offered combinations of income, autonomy, ownership, risk, and work effort. Each of the five salient outcomes was set as “high” or “low” (clearly defined initially) and respondents were asked to rate the attractiveness of each scenario on a seven-point Likert scale anchored by very low attractiveness (1), and very high attractiveness (7). Respondents were led through an initial scenario so that they understood the process (data were not collected) and then were advised to proceed at their own pace but to continue working forward through the scenarios and not turn back to check earlier responses. To reduce the number of scenarios to a manageable number we used a fractional factorial design that nonetheless maintains orthogonality among the outcome levels and the part-worth estimates (Hair et al., 2010). In line with the approach of Douglas and Shepherd (2002a), each of 16 different scenarios was repeated once during the experiment to allow a test-retest measure of reliability. All respondents viewed the same 32 scenarios but, to avoid order effects in the data, the order of the scenarios was reversed in two versions of the survey and the ordering of the salient outcomes in each scenario was reversed in two versions, providing four different orderings of the same scenarios and salient outcomes. Further details on the conjoint experimental method can

Intrapreneurial intention

97

be found in Green and Srinivasan (1978), Shepherd and Zacharakis (1999), Douglas and Shepherd (2002a), Monsen et al. (2010), and Hair et al. (2010). For self-efficacy, we used the ESE scale developed by Chen et al. (1998). This scale consists of 22 items measuring an individual’s confidence in their ability to perform entrepreneurial tasks, with each item measured on a five-point Likert scale ranging from completely unsure (1), to completely sure (5). Following Chen et al. (1998) we calculated the total ESE score by taking the average of scores on the 22 items. We included as control variables the respondent’s age, prior income level, total prior work experience, sex, three levels of prior education (bachelor’s, master’s, and doctorate), whether they were previously self-employed, and country of origin (with Australia being the base case).

6.4

ANALYSIS AND RESULTS

6.4.1

Separate and distinct constructs?

The first research question is whether entrepreneurial intention and intrapreneurial intention are viewed as distinct and separate entrepreneurial activities. We asked how likely it was (1–7 scale) that respondents would engage in each of seven types of entrepreneurial activity, which are shown in Table 6.1. We employed principal components analysis with oblique factor rotation and found two factors with eigenvalues above 1.00, which accounted for 73.3 per cent of the cumulative variance. The two distinct factors were identified as those relating to entrepreneurial intention (four items, α = 0.79) and those related to intrapreneurial intention (three items, α = 0.77). Accordingly, we found support for the first hypothesis and concluded that self-employed entrepreneurship and corporately employed intrapreneurship were viewed by the sample respondents as distinctly separate career options. 6.4.2

Regression analysis

Subsequently, two regression models were considered, one to explain entrepreneurial intention and one to explain intrapreneurial intention. We used the average score for the four items constituting the entrepreneurial intention factor, and the average score for the three items constituting the intrapreneurial intention factor, as the dependent variables in the two regression models. The independent variables were the same for each model, namely the human capital control variables, ESE, and the attitudes toward the salient outcomes of income, autonomy, ownership, risk, and work effort. We utilised “seemingly unrelated regression” analysis within the STATA package to minimise the overall variance across the two models. The descriptive statistics and intercor-

Entrepreneurial intention

98

Table 6.1

Factor analysis results: entrepreneurial versus intrapreneurial intention

Item How likely is it that you would want to be self-employed within

Entrepreneurship

Intrapreneurship

intention

intention

0.83

two years after graduation, assuming you had a good new business opportunity and you could raise the funding necessary to start your own business? How likely is it that you would want to be self-employed at some

0.80

later point in the future, assuming you had a good opportunity and could raise the funding necessary? How likely is it that you would want to start your own business

0.79

to exploit a radical innovation? 0.89

How likely is it that you would want to manage (within your employer’s business) a new division (or branch) that is set up to exploit a radical innovation? How likely is it that you would want to start your own business

0.71

to introduce a new variant of an existing product or service? 0.84

How likely is it that you would want to manage (within your employer’s business) a new division set up to introduce a new variant of an existing product or service?

0.65

How likely is it that you would want to manage (within your employer’s business) a new division (or branch) set up to introduce an existing product into a new market? Cronbach’s alpha

0.79

0.77

Note: Factor loadings less than 0.50 suppressed.

relations for the sample are shown in Table 6.2 and the regression coefficients for both models are shown in Table 6.3. As is evident in Table 6.3, for the entrepreneurship model we found significant positive relationships between entrepreneurial intention and the attitudes to income, independence, and ownership, while attitudes to risk and to work effort were found to be insignificant. In addition, we found ESE to be positively and significantly related to entrepreneurial intention. The following human capital variables were also found to be significant: age and prior self-employment were positively related to entrepreneurial intention, while individuals with more prior work experience were significantly less likely to form the intention to start a new venture. Regarding country differences, only the Indian students showed a significant relationship, indicating that they were significantly negatively inclined towards entrepreneurship as compared to the Australians.

Intrapreneurial intention

Table 6.2

99

Descriptive statistics and intercorrelation matrix

1. Entrepreneurship intention

Mean

S.D.

5.38

1.26

1

2

3

4

5

6

2. Intrapreneurship intention

5.58

1.12

0.08

3. Income

2.74

0.91

0.03

−0.02

4. Independence

1.10

0.72

0.11*

0.14**

−0.35**

5. Ownership

0.33

0.63

0.39**

−0.13*

−0.25**

0.00

6. Risk tolerance

−0.51

0.60

0.13**

−0.06

−0.17**

0.29**

0.15**

7. Work effort

−0.13

0.62

−0.01

0.08

−0.19**

0.23**

−0.00

0.33**

3.79

0.53

0.21**

0.16**

0.04

0.12*

0.01

0.09

8. Entrepreneurial self-efficacy

7

0.13*

Notes: * p < 0.05; ** p < 0.01.

The results of the intrapreneurship model are shown in the third column of Table 6.3. Similar to findings from the entrepreneurial intention model, we find the individual’s self-efficacy to be significant and positively related to intrapreneurial intention. Not similar, however, was the impact of attitudes to income, independence, and ownership, where we find no evidence of a relationship between these particular attitudes and an individual’s intention to be an intrapreneur. But we did find a significant negative relationship between tolerance for risk and intrapreneurial intention, in contrast to the entrepreneurial intention model where we found no evidence of a relationship between attitude to risk and entrepreneurial intention. This suggests that, other things being equal, individuals who are more risk averse will tend to seek the shelter of a corporate environment to conduct their desired entrepreneurial behaviour, as implied by Parker (2011: 31). Concerning the control variables, we found that individuals with a prior doctoral education were less likely to form intrapreneurial intention, and that prior bachelor and master degrees had no significant effects on either entrepreneurial or intrapreneurial intention. Regarding country differences, the only significant relationship was for the Thai students who showed a significant negative relationship with intrapreneurial intention, indicating that they were negatively inclined towards intrapreneurship as compared to the Australians.

6.5 DISCUSSION In this chapter we have examined the attitudinal and self-efficacy antecedents of the intention to behave in an entrepreneurial manner, where this intention can be directed towards actuality either as self-employed entrepreneurship or as corporately employed intrapreneurship. Contrary to the previous retrospec-

Entrepreneurial intention

100

Table 6.3

Regression results for entrepreneurial and intrapreneurship intention Entrepreneurship

Intrapreneurship

intention

intention

Constant

0.56

5.55***

Age

0.07*

−0.01

Gender (1 = female)

−0.09

−0.13

Education_2

0.35

−0.73

Education_3

0.09

−0.79

Education_4

−1.12

−3.1**

Prior income

−0.02

0.05

Previously self-employed

0.42*

0.02

Total years’ work experience

−0.08***

0.01

Entrepreneurial self-efficacy

0.53***

0.21* −0.01

Attitude to income

0.23***

Attitude to independence

0.34***

0.10

Attitude to ownership

0.83***

−0.12

Attitude to risk

0.15

−0.21*

Attitude to work effort

−0.06

0.07

Dummy: China

−0.27

−0.43

Dummy: India

−0.47*

0.19

Dummy: Thailand

−0.16

−0.42*

R2

0.30

0.16

Notes: * p < 0.05; ** p < 0.01; *** p < 0.001; N = 373.

tive studies of the reasons entrepreneurs give for taking the self-employment path (e.g. Carter et al., 2003; Shaver et al., 2001), our study was prospective and used a conjoint experiment to discern the revealed (rather than espoused) attitudes toward the various outcomes of the entrepreneurial experience. Further, we utilised multi-item factors to discern the willingness of respondents to engage in entrepreneurial or intrapreneurial activity, unlike most prior studies at that time. Consistent with a previous study (Douglas & Shepherd, 2002a) we found that individuals who prefer more income and more independence have higher entrepreneurial intention. In contrast to Douglas and Shepherd (2002a), however, but in accord with Brockhaus (1980), Palich and Bagby (1995), and Busenitz (1999), we find no evidence that more-risk-tolerant individuals will have higher entrepreneurial intention. But we found that lower risk tolerance is significantly associated with intentions for intrapreneurship. This difference

Intrapreneurial intention

101

may explain the conflicting results in previous studies of the dependence of entrepreneurial intention on attitude to risk (or risk propensity) – it seems that the impact of risk attitude depends importantly on the type of entrepreneurial behaviour contemplated by the individual. Accordingly we wonder whether past studies may have confounded the relationship of attitude to risk and entrepreneurial intention by not distinguishing between self-employed entrepreneurial and commercially employed intrapreneurial intention. As in Fitzsimmons and Douglas (2011), we found attitude to majority ownership to be significantly and positively related to entrepreneurial intention, but our hypothesis that intrapreneurial intention would be negatively related to attitudes to majority ownership was not supported at an acceptable significance level, although the sign was negative. This is nonetheless an interesting result, since it demonstrates that the benefits of firm ownership are either not highly valued by intending intrapreneurs, or that the psychic costs associated with firm ownership (such as stress) do not significantly drive individuals to intrapreneurship in preference to entrepreneurship, or that the psychic benefits and costs of majority ownership more or less offset each other for those with relatively strong intentions to become an intrapreneur. Our result does provide further evidence that attitude to independence and attitude to ownership are distinctly separate constructs, as implied by Shaver et al. (2001) and as demonstrated by Fitzsimmons and Douglas (2011). No significant relationship was found between entrepreneurial intention and attitude to work effort, similar to Douglas and Shepherd (2002a) and supporting the overall conclusion from the replicated studies of Wiklund et al. (2003). And contrary to Monsen et al. (2010), who found a significant negative relationship between intrapreneurial intention and the expected level of work effort, we found no relationship between attitude to work effort and intrapreneurial intention. Significant positive relationships were also found between ESE and both entrepreneurial and intrapreneurial intention, which provides further evidence for the importance of self-efficacy and its relationship to intentions to engage in entrepreneurial behaviours. We note that ESE is significant at a higher level of significance, and has greater effect size, for entrepreneurial intention as compared to intrapreneurial intention. Thus we speculate that individuals who perceive themselves as having greater ESE may be more likely to form the intention to engage in self-employed entrepreneurial behaviour rather than corporately employed intrapreneurial behaviour, but those with weaker self-efficacy might nonetheless form the intention to be an intrapreneur within the more protective confines of an existing organisation. Regarding the control variables, we find age influences only entrepreneurial intention, with older individuals having greater intention to engage in entrepreneurship, but caution the reader that the age range in the sample of MBA stu-

102

Entrepreneurial intention

dents was relatively narrow and hardly representative of the general population in each country. Similarly, the result for prior PhD education should be treated with caution, since the negative preference for intrapreneurship expressed by those with prior doctorates (almost certainly in non-business areas) probably reflects a preference to move from a technical to a managerial (rather than intrapreneurial) role within a firm, post-MBA. One other control variable, previous self-employment experience, exhibited a relatively large effect size and was significantly related to entrepreneurial intention but was not significantly related to intrapreneurial intention. This may reflect respondents’ preference for gaining relevant work experience within a firm, post-MBA, before selecting entrepreneurship at a later stage, and may have confounded the impact of age on intrapreneurial intention. Finally, it is interesting to note that while a significant and positive bivariate correlation was found between risk tolerance and entrepreneurial intention (see Table 6.2), this relationship was not significant in the multivariate analysis when the impact of the control variables, self-efficacy, and other attitudes was considered. Similarly, intrapreneurial intention was significantly correlated with attitudes to independence and to ownership in the simple bivariate analysis but was unrelated to these variables in the multivariate analysis. This illustrates the danger of jumping to conclusions on the basis of simple correlations.

6.6 SUMMARY The study of intrapreneurial intention extends the entrepreneurial intention literature beyond the typical approach that focuses on the intention to start an independent new business venture. Instead we note that many individuals choose to act entrepreneurially within existing firms as intrapreneurs, and that individuals might form intention to be either an entrepreneur or an intrapreneur. We also offer evidence to refute the general presumption that nascent intrapreneurs are basically similar in attitudinal makeup and self-efficacy to nascent entrepreneurs, perhaps choosing to work as an intrapreneur within an organisation because they have not yet discovered a viable new business opportunity, or not yet raised enough money, and/or not yet gained enough experience to go out on their own. Our findings reveal that nascent intrapreneurs tend to be different from nascent entrepreneurs in their cognitive makeup, having lesser ESE and greater risk aversion, and consequently they anticipate that intrapreneurship will be more attractive to them than entrepreneurship, and thus they form stronger intentions for intrapreneurship than for entrepreneurship. As a postscript to that study, we now appreciate more fully that the single dominant net-effects prescription provided by multiple regression analysis reflects the sample-average relationship between each independent variable

Intrapreneurial intention

103

and the dependent variable. It necessarily suppresses individual differences, and is likely to suppress any asymmetric data relationships (e.g. some intending entrepreneurs being highly intolerant of risk while others are highly tolerant of risk). It also tends to hide subgroups of respondents who arrive at the decision to form entrepreneurial (or intrapreneurial) intention via a different configuration of antecedent variables. These issues will be investigated in the context of entrepreneurial intention in Chapter 9.

7. Social entrepreneurial intention 7.1 INTRODUCTION Social entrepreneurs are typically defined as entrepreneurs whose primary purpose is to address social problems and provide social benefits to those in need, including socially and economically disadvantaged members of the local and global communities (see, e.g. Austin et al., 2006; Mair & Marti, 2006; Peredo & McLean, 2006). As entrepreneurs they are expected to be proactive, risk taking, and innovative (Lumpkin & Dess, 1996), and to use business methods to ensure the ongoing viability and sustainability of their organisation (Miller et al., 2012). This does not necessarily mean that they are profit seeking, as they may obtain the revenues they need (for operations) from government grants, charities, philanthropic organisations, and/or altruistic individuals. But increasingly social entrepreneurs are encouraged to be “hybrid” entrepreneurs, that is, a cross between a “self-serving” commercial entrepreneur and an “others-serving” social entrepreneur, earning at least some profit to avoid repetitive approaches to governments and other sources of social funding for additional grants to continue their social mission (Battilana & Lee, 2014). Social entrepreneurs usually address social issues that are not profitable for commercial entrepreneurs to undertake. Profit-maximising firms will neither undertake business or social opportunities that are expected to be unprofitable, nor opportunities which have profitability lower than other opportunities they might undertake with the same funds. This is called the “market failure problem” (Santos, 2012; Weisbrod, 1975) because the market mechanism (a euphemism referring to the incentive effects of profit to induce supply of wanted goods and services) fails to induce profit-maximising firms to supply the needed goods and services. Accordingly, these social issues may persist because no profit-maximising firm wishes to make smaller profit by incurring losses in one area of their business. Profit-satisficing firms might choose to undertake unprofitable actions, subsidising this with revenues from other markets, and thereby exhibit social entrepreneurship and corporate social responsibility. Traditionally, due to the market failure problem, the alleviation of social problems is an expected responsibility of government, using tax revenues col104

Social entrepreneurial intention

105

lected from individuals and firms to pay for the social benefits provided to the needy. But governments may be asked to resolve hundreds of social issues, in thousands of locations, and cannot possibly resolve them all, tending to focus on those that have the greatest political consequences, since politicians are also self-serving humans seeking to be re-elected, or at least to have their party returned to power. Thus, “government failure” also occurs for many social problems, on top of market failure, such that social entrepreneurs are needed to tackle the social problems that persist (Santos, 2012).

7.2

THE “THREE PILLARS” OF SOCIAL ENTREPRENEURSHIP

Newbert and Hill (2014) proclaim that the three main pillars of social entrepreneurship are pro-social motivation, profit, and innovation. Like a commercial entrepreneur, social entrepreneurs are, by generally accepted definition, risk-taking, profit-seeking innovators, but are distinguished from the former by their primary goal being to provide social benefit (rather than to benefit the owners of the firm). Echoing the lament of others (e.g. Arend, 2013; De Drue & Nauta, 2009; Lumpkin, Moss, Gras, Kato, & Amezcua, 2013; Renko, 2013) Newbert and Hill argue that relatively little attention has been paid to the social entrepreneur’s motivation to make profit or to innovate. Profit is important for social entrepreneurship because it may be reinvested in the venture to provide ongoing social benefits, obviating the need to seek funding repetitively from limited public and philanthropic sources. Moreover, the prospect of dividends and capital gains from a social enterprise would serve to attract venture capital and private investment into the realm of social enterprise and lead to the greater provision of social benefit. Of course, many hybrid social entrepreneurs do simultaneously pursue pro-social and profit goals (Battilana & Lee, 2014; De Drue & Nauta, 2009; Dees, 1998; Haigh, Walker, Bacq, & Kickul, 2015; Williams & Nadin, 2011). Their motivation for earning profit is not necessarily to distribute profit to owners but may be to reinvest the financial surplus to enable continuation and extension of their social mission (Bacq & Janssen, 2011). The lack of research attention to the social entrepreneur’s motivation to innovate is problematic because innovative new production and distribution processes may be necessary to solve the market failure problem (by reducing cost or increasing revenue). Moreover, the government failure problem, which occurs when myriad competing demands for public funding exceed the government’s capacity to supply that funding, may be alleviated if innovative new solutions to social problems are judged to be a “better investment” of those public funds. Similarly, there is a limited capacity of philanthropic foundations and charities to provide the funding that would allow supplier firms to

106

Entrepreneurial intention

overcome the market failure problem, such that they too must make allocative choices. The motivation to innovate is seldom discussed in the social entrepreneurship literature, nor indeed in the commercial entrepreneurship literature where it is axiomatic that entrepreneurs are innovative (following Schumpeter, 1934), while non-entrepreneurs are not. It is understood that innovation is instrumental to profit making and/or social-benefit provision, but the direct reward – that is, the psychic utility accruing to the individual’s innovativeness – is generally neglected. The innovativeness of individuals is not a binary variable – it is likely to vary among individuals and depend on the characteristics of the opportunity and contextual variables. While many social entrepreneurs are demonstrably innovative, others exhibit little or no innovativeness (e.g. Newbert & Hill, 2014; Shane et al., 2003; Zahra et al., 2009). Short, Moss, and Lumpkin (2009) state that innovation is a key theme in social entrepreneurship, and call for more theory-based research on innovativeness among social entrepreneurs. If social entrepreneurial intention (SEI) is almost totally motivated by the desire to provide pro-social outcomes, with little or no motivation to generate financial surplus or to innovate, the social outcomes are likely to be suboptimal compared to pro-social behaviour that is buttressed by a substantial measure of profit motivation and innovation motivation. The latter two motivations for social entrepreneurship, added to pro-social motivation, should lead to superior outcomes for society. Together these three motivations should be expected to drive greater efficiency in production and distribution systems; generate financial surpluses for reinvestment and scaling up of social ventures; and (via innovation) solve social problems that were previously unsolvable due to market and government failure. Thus, focusing on pro-social motivations alone may not lead to the socially optimal supply of social entrepreneurs and their efforts to reduce or eliminate ongoing social problems. It is apparent that there is considerable variation among social entrepreneurs across these three dimensions. Martin and Osberg (2007: 30) observe that the study of social entrepreneurship has become an “immense tent into which all manner of socially beneficial activities fit”. Newbert and Hill (2014) have called for the study of social entrepreneurship in “small tents” where the respondents are more alike rather than being widely diverse. Accordingly, to better understand social entrepreneurship it is important to analyse the motivation for all three pillars of social entrepreneurship. Liñán and Fayolle (2015) note that SEIs have received very little attention from entrepreneurship scholars, except for Renko (2013) and Hockerts (2017), but these scholars did not address the motivation for all three pillars of social entrepreneurship.

Social entrepreneurial intention

7.2.1

107

The pro-social motivation of social entrepreneurs

The primary motivation of social entrepreneurs is variously ascribed to “other-centred” attitudes and feelings, such as altruism, empathy, moral judgement, caring, compassion, and guilt (see, e.g. Austin et al., 2006; Bacq & Alt, 2018; Bornstein, 2007; Choi & Majumdar, 2014; Dacin, Dacin, & Tracey, 2011; Dees, 1998; Mair & Marti, 2006; Mair & Noboa, 2006; Miller et al., 2012; Murphy & Coombes, 2009; Peredo & McLean, 2006; Santos, 2012; Shaw & de Bruin, 2013; Sullivan Mort, Weerawardena, & Carnegie, 2003; Zahra et al., 2009, 2014). For parsimony we follow Grant and Berry (2011) and refer collectively to these as the higher-order construct of “pro-social attitude”. There is a continuing healthy debate among scholars regarding the cognitive antecedents of pro-social attitude. Scholars have suggested and championed various “other-centred” cognitions, as mentioned in the above paragraph, as the main antecedent variable underlying pro-social attitude. Altruism, defined as the “willingness to benefit another with disregard or no regard for rewards for self, or even at a cost to one-self” (Batson & Shaw, 1991: 107) was previously the favoured antecedent, but seems incongruent with the notion that individuals might gain utility from pro-social behaviour. But note that altruism can be accommodated within a utility-maximising model of perceived desirability, since the utility derived from social entrepreneurship might be either positive, zero (i.e. no benefit to oneself), or negative (i.e. disutility for oneself). Lately, empathy may be the strongest contender for the main antecedent of pro-social attitude (see, e.g. Bacq & Alt, 2018), perhaps followed by compassion (e.g. Miller et al., 2012). It is not my purpose here to enter that discussion – rather my purpose is to discuss the addition of profit motivation and innovation motivation to the objective function of the intending social entrepreneur, to evaluate the proposition that all three “main pillars” are involved in the formation of SEI.1 7.2.2

The profit motivation of social entrepreneurs

Individual social entrepreneurs may or may not be highly motivated by the prospect of profit. Rather than a binary issue, the question is “how much profit” they desire, if any. They may desire a positive profit, or zero profit, 1 A recent paper by myself and Catherine Prentice (Douglas & Prentice, 2019) examines the pro-social, profit, and innovation motivations of social entrepreneurs, using a large and representative sample of the Australian public. While there are parallels between the discussion in that paper and Chapters 7 and 9 in this book, the (student) sample utilised in this book is different, so as not to impinge on that paper’s findings.

108

Entrepreneurial intention

or even negative profit (for altruists and those wishing to give away their wealth). This question invokes the profit-satisficing discussion of Chapter 5. It is theoretically possible that a social entrepreneur, with the primary purpose of social-benefit provision, might set a target for the desired level of social-benefit provision, and beyond that may seek to maximise profit from the venture. This we might call conditional profit maximisation. Conversely, the social entrepreneur might set a profit target (i.e. profit satisficing) and beyond that attempt to maximise social-benefit provision from the venture, assuming that the primary purpose of the venture remains the provision of social benefit (to stay within the common definition of the social entrepreneur). Strictly, an entrepreneur seeking zero profit (or a loss) can be regarded as a profit satisficer, if that is the satisfactory target they are aiming for. Such individuals might seek to maximise social-benefit provision by running profit down to the minimum level that avoids degeneration of the owners’ equity in the venture (and avoids the market failure problem). Indeed, a social entrepreneur with sufficient previously accumulated wealth (or the wish to “give their wealth away”) might set a negative profit target for their social venture to allow the provision of a greater magnitude of social benefits to others. Note that a utility-maximising model of the perceived desirability of social entrepreneurship does not preclude the individual having a negative attitude to any of the three pillars of social entrepreneurship. Indeed, a negative attitude (valence) to profit, for example, multiplied by a negative profit outcome (expectancy) provides a positive utility part-worth (UPW) for that component of total utility. 7.2.3

The innovation motivation of social entrepreneurs

Innovative activity by social entrepreneurs is effectively presumed by the definition of the entrepreneur as a proactive, risk-taking innovator. The presumption is also made by many that the social entrepreneur’s attitude to innovation will be, or should be, positive, since innovation is likely to be instrumental for achieving the social mission of the social venture. The attitude to innovation is surely a continuous variable, and may be found empirically to be relatively high or low in individual cases, and also may be negative for risk-averse individuals, since innovation carries with it the risk that the innovation will not, ex post, deliver the social outcomes expected of it, ex ante. Some social enterprises may be observed to not be innovative (see, e.g. Zahra et al., 2009), nor proactive, instead conducting a reactive social venture with a conventional business model, such as a soup kitchen for the homeless that utilises volunteer labour and philanthropic financial support. The purist may quibble that lack of innovation means that this case is not strictly social entrepreneurship but is instead social enterprise (Chell, 2007). But note that

Social entrepreneurial intention

109

in a multiple-goal model, non-innovation is also consistent with the possibility that other goals may dominate the decision to be a social entrepreneur. If the focal opportunity provides little or no expectancy of successful innovation outcomes, but has high prospects of pro-social and profit outcomes, the latter two motivations may cause it to be the utility-maximising alternative. In this situation, the UPW of innovation is zero or minimal, and thus we should not expect to see individuals “going out of their way” to be innovative while pursuing their primary purposes of social-benefit provision and secondary purpose of profit making.2

7.3

AN EMPIRICAL STUDY OF SOCIAL ENTREPRENEURIAL INTENTION

In Douglas and Prentice (2019) we examine the dependence of SEI on the individual’s attitudes toward the three pillars of social entrepreneurship, namely social impact, profit, and innovation. That study is effectively under the embargo and copyright of the journal concerned, so a different study is reported here. In the following study the three main outcomes of social entrepreneurship are hypothesised to be social benefit, profit, and psychic income. These three outcomes can be argued to be more or less desirable by heterogeneous intending social entrepreneurs (see, e.g. Stevens, Moray, & Bruneel, 2015). Thus, both the regression model, and the sample used here, are different from the 2019 paper. For this study I collected entrepreneurial intentions data for three types of entrepreneurship, namely SEIs, growth-oriented (entrepreneurial) intention (GOI) and independence-oriented (entrepreneurial) intention (IOI). These constructs are not orthogonal – an individual will simultaneously have separate intentions for pro-social outcomes, profit outcomes, and for psychic income outcomes. As a result I was able to regress all three intentions variables against the same set of perceived desirability, perceived feasibility, and control variables, and thereby compare and contrast the drivers of each intention. The scales for GOI and IOI were taken from Douglas (2013), and the scale for SEI is from Douglas and Prentice (2019).

2 This theme will be further developed in later chapters, where the four goals of profit, pro-social impact, psychic rewards, and innovation are considered simultaneously. Since there is heterogeneity of both individuals and the opportunities they seek to undertake, any subset of the four motivations may be sufficient for utility maximisation.

Entrepreneurial intention

110

7.3.1

The model

Utilising the theory of planned behaviour (Ajzen, 1991; Ajzen & Fishbein, 1970) we follow extant theory to argue that the formation of entrepreneurship intentions depends on the perceived desirability and the perceived feasibility of the entrepreneurial behaviour (Krueger, 1993; Krueger & Brazeal, 1994; Krueger et al., 2000). Perceived desirability depends on the individual’s attitudes to the expected outcomes of the entrepreneurial behaviour. The arithmetic product of the individual’s outcome expectations (the expectancies) and the individual’s attitudes to those outcomes (the valences) generate a series of UPWs that sum to the expected utility of the behaviour (see Douglas, 2013; Steel & König, 2006). Individuals are expected to select the behavioural option that maximises their expected utility (subject to considerations of perceived feasibility). The UPWs for the three outcomes were derived from a conjoint experiment similar to that discussed in Chapter 5 but involving these different outcome variables. The perceived feasibility of entrepreneurial behaviour was measured by the 19-item entrepreneurial self-efficacy (ESE) scale of McGee et al. (2009) which allows subsets of managerial capability to be self-assessed by the respondent. The behavioural options in this model are social entrepreneurship, profit-maximising entrepreneurship, or profit-satisficing entrepreneurship, for which individuals will form SEI, GOI, and IOI. Other things being equal, we expect individuals with a strongly positive pro-social attitude to form stronger SEI; individuals with a strongly positive attitude to profit to be more likely to form stronger GOI; and individuals with a strongly positive attitude to psychic income to form stronger IOI. These expectations are not only due to the above attitudes (valences) but are also due to the expectancies, since greater social outcomes are expected in social entrepreneurship opportunities; more profit is expected in profit and growth opportunities; and more psychic income is expected in profit-satisficing opportunities. The perceived desirability of undertaking a particular entrepreneurial opportunity is of course the sum of the UPWs (valence x expectancy) of the three expected outcomes for that opportunity. Turning now to the perceived feasibility of entrepreneurial behaviour, measures of ESE effectively and parsimoniously measure the individual’s ability to accomplish the tasks associated with entrepreneurship (e.g. Boyd & Vozikis, 1994). Self-efficacy is a measure of the individual’s confidence that they can successfully accomplish the tasks in a specific domain (Bandura, 1982). Scholars have devised and tested scales to measure ESE (e.g. Chen et al., 1998; McGee et al., 2009), and ESE has been positively associated with the formation of entrepreneurship intention in a variety of previous studies

Social entrepreneurial intention

111

of commercial entrepreneurship (e.g. Douglas, 2013; Douglas & Shepherd, 2002a; Fitzsimmons & Douglas, 2011; Krueger et al., 2000). Each of these ESE scales specifies a series of tasks that the entrepreneur, defined broadly, is likely to encounter in the entrepreneurial process. McGee et al. (2009) argue that there are four main task areas, which they called “searching” (for the opportunity), “planning” (the business model for the new venture), “marshalling” (resources to launch the new venture), and “implementing” (the business plan, via effective management of people and financial resources). In their empirical study they found that “implementing” broke into two distinct self-efficacy factors, namely people management and financial management. McGee et al. (2009) argued that individuals may be more self-confident about some aspects of the management of the new venture than about others. Since the different types of entrepreneurship may require greater or lesser expertise in particular management skill areas (see Douglas, 2013), for example financial management, it is important to use this multifaceted ESE scale in this study of multiple types. At the time of this empirical study there was little or no prior research on the relationship between ESE and social entrepreneurship.3 A priori, we expect that like commercial entrepreneurs, social entrepreneurs will be more likely to be motivated to start a new venture the greater is their overall ESE. Of great interest is whether differences exist for the management subdiscipline components of ESE according to whether the individual is primarily profit-oriented, primarily psychic income-oriented, or primarily socially oriented. While one might argue that different self-efficacy scales may be required for different types of entrepreneurship, the McGee et al. scale measures the self-confidence the respondent has in successfully completing generic management tasks (e.g. marketing management, people management, and financial management) and one can safely say that these will be required for all three types of entrepreneurship (perhaps to differing degrees).4 Control variables were included in the regression models due to prior reasoning and empirical evidence that these variables might exert an influence on the formation of entrepreneurial intention. Age of the respondent is included

3 More recently Bacq and Alt (2018) published an informative paper on this issue, and also propose a new ESE scale for social entrepreneurs. At the time of this study I was unaware of that social ESE scale, but in any case sought to use a common ESE scale across three types of entrepreneurial intention, and chose the McGee et al. scale because it examines different aspects of ESE separately. 4 The McGee et al. scale delivered only these three distinct factors when applied to this sample’s data. We have also found this three-factor solution when applying that scale to other MBA student samples, as well as for a large representative (non-student) sample of the Australian public – see Douglas and Prentice (2019).

112

Entrepreneurial intention

on the basis that it is likely to be influential in individual cases – for example, younger people may be more idealistic and such idealism may militate for greater social intentions, or older people may be less aspirational which may militate against wishing to maximise profit. In general, people tend to change their thinking and their behaviour as they age, so age is effectively a proxy for a variety of missing variables that may correlate with age. The respondent’s sex is included as an explanatory variable due to previous findings that males tend to have higher entrepreneurial intentions than females. Since physiological sex does not hinder or enhance performance in the exploitation of many, perhaps most, entrepreneurial opportunities, we must also look at socially conditioned gender traits (Ahl & Marlow, 2012; Bem, 1981; Gupta et al., 2009). Accordingly we will not only use the binary sex variable but also use measures of masculinity and femininity (Bem, 1981) to better explain the gender differences in entrepreneurial intentions across the types of entrepreneurial behaviour. Prior business education is also likely to influence entrepreneurial intention. The prior familiarity with business concepts and practices that one assimilates from business education is likely to be a form of vicarious learning and thereby bolster one’s ESE, if not also one’s attitudes toward profit and the psychic income gained from entrepreneurial behaviour. On the other hand, perhaps prior business education serves to brainwash individuals in favour of commercial entrepreneurship and against social entrepreneurship. Thus we need to include this variable as a potential determinant of entrepreneurial intention (Boyd & Vozikis, 1994; Kolvereid, 1996a, 1996b; Krueger, 1993). Having self-employed parents and growing up in a family business environment is also likely to increase the propensity for sons and daughters to intend to be self-employed in their own right. This family situation is likely to provide a learning environment where younger family members benefit from experiential learning that, like business education, is likely to augment both the perceived desirability and the perceived feasibility of entrepreneurial behaviour for the next generation (see, e.g. Bosma, Hessels, Schutjens, Van Praag, & Verheul, 2012; Chlosta, Patzelt, Klein, & Dormann, 2012). Next, one’s marital status (formal or informal) should be expected to negatively influence entrepreneurial intention, since “togetherness” tends to reduce one’s flexibility to relocate, to work long hours, to invest money in risky ventures, and other actions that being entrepreneurial might require. On the other hand, a spouse who actively supports the entrepreneur physically, emotionally, and/or financially would militate in favour of entrepreneurial intention (see, e.g. Blenkinsopp & Owens, 2010; Hsu, Wiklund, Anderson, & Coffey, 2016; Kirkwood, 2009). There is likely to be an asymmetric data relationship here and it is not clear what will be the dominant net effect of the regression analysis.

Social entrepreneurial intention

113

Children add an extra dimension to consider for an individual who is contemplating entrepreneurial behaviour, since parents must allocate time, energy, emotion, and funds to fulfil their responsibilities toward their children. Because children require parental attention, we might expect a negative influence of having children on the formation of entrepreneurial intention. But conversely, some individuals might intend entrepreneurship as a means of gaining more free time for their family, or in search of a larger income to support their family. Thus having children may contribute positively to the formation of entrepreneurial intention in some cases and negatively in others (see, e.g. Holtz-Eakin, Joulfaian, & Rosen, 1993; Schindehutte, Morris, & Brennan, 2003). 7.3.2

The sample and method

The sample comprised 82 MBA/EMBA students who were taking an elective course in entrepreneurship and new venture planning at the Sasin Graduate School of Management in Thailand. While this sample size is considered small for regression analysis, it is sufficient to allow usable results to be attained for our purposes here. In defence of a student sample, it can be argued that by enrolling in an MBA, these students have effectively signalled their interest in improving their knowledge and credentials for a potential change of employment status (Shepherd & DeTienne, 2005), and by selecting an entrepreneurship elective have signalled their interest in behaving entrepreneurially (Hockerts, 2017). The MBA students were between 24 and 49 years old (average 30.5) at the time of data collection. The sample was (fortuitously) 50 per cent male and 50 per cent female. All had undergraduate degrees, 40 per cent of which were business degrees. Their average business experience was 6.06 years. Thirty-four per cent had prior or current self-employment experience; and 73 per cent had a self-employed parent or parents in a family business. The latter two facts, and the high tuition cost, should alert you to the fact that this sample is relatively wealthy and not typical of the Thai population in general, although it is typical of the students attending that business school which emphasises new venture creation by its graduates. The data was hand collected in three waves, with each wave separated by three to four days to minimise common-method variance. Each respondent’s data were aggregated using their choice of a secret identifying code (being the first three letters of their mother’s name and four numbers representing her birth day and month). The first wave involved a conjoint experiment to ascertain respondent attitudes to three outcomes of the entrepreneurial process, which were specified as profit income, psychic income, and social-impact provision. Respondents were presented with a series of scenarios, which were

114

Entrepreneurial intention

described as “either employment or self-employment situations providing these outcomes”, systematically presented in combinations with each outcome set as either “high” or “low”. Definitions of these levels were provided, and respondents were asked to rate each scenario on a scale of 1–7 where 1 represents “very low attractiveness” and 7 represents “very high attractiveness”. Since there were only 23 = 8 combinations of the three salient outcomes at two settings, we utilised a fully replicated design to present 16 scenarios to respondents, with each scenario being repeated later (with a different code name) within the experiment. Respondents were told that there were no correct answers; that they should respond intuitively to each scenario; that their responses would be confidential; and that results would be reported only in aggregated format. Separate regressions were run for every respondent, with the dependent variable being the attractiveness score on each scenario, and the independent variables being dummy variables set to “1” for high outcomes and “0” for low outcomes for each of the three outcome variables. For each respondent, the coefficients to the dummy variables in their particular regression equation provide an estimate of the UPW (and hence the attitude) for each of the pro-social, profit, and psychic income outcomes. Following the conjoint experiment, socio-demographic data were also collected from respondents. The second wave utilised the McGee et al. (2009) ESE survey, comprising 19 items which respondents rated on a 1–7 Likert scale. After elimination of items with loadings less than 0.4 or with cross-loading greater than 0.4, factor analysis with varimax rotation revealed three factors with eigenvalues exceeding 1 and explaining 99.11 per cent of the cumulative variance. As shown in Table 7.1, these factors corresponded to ESE marketing, ESE people, and ESE financial and had Cronbach alpha scores of 0.661, 0.784, and 0.839, respectively. We were unable to replicate McGee et al.’s five factors in this sample, but did replicate their findings that the “implementation” tasks separated into two distinct factors representing people-management and financial-management skills. The somewhat lower reliability (α = 0.661) for ESE marketing may be due to the admixture of the searching, planning, and marshalling tasks into a composite factor which, after deletion of items loading below 0.4, evoked marketing, promotion, and communication skills. The final survey wave involved a 15-item entrepreneurial intentions scale with five items each for profit-oriented entrepreneurship intention, independence-oriented entrepreneurship intention, and SEI. The items for profit-seeking (GOI) and independence-seeking (IOI) entrepreneurship were taken from the Douglas (2013) study, and had Cronbach-alpha reliability scores of 0.81 and 0.70, respectively. Five items for SEI came from the SEI scale published in Douglas and Prentice (2019) where the Cronbach alpha was 0.90. The bivariate correlations for the data set are shown in Table 7.2.

Social entrepreneurial intention

Table 7.1

115

Factor analysis to identify the subcomponents of entrepreneurial self-efficacy

I have a great deal of confidence in my ability to:

ESE

ESE

ESE

finance

people

marketing

Manage the financial assets of my business

0.7856

Read and interpret financial statements

0.7311

Estimate start-up funds and working capital required

0.6695

Organise and maintain the financial records of my business

0.7931

Recruit and hire employees

0.5580

Supervise employees

0.6279

Train employees

0.6629

Delegate tasks and responsibilities to employees in my business

0.6668

Network – make contact and exchange information with others

0.5238

Brainstorm (come up with) a new idea for a product or service

0.4578

Estimate consumer demand for a new product or service

0.4190

Design an effective marketing/advertising campaign for a new

0.7246

product or service Design a new product or service that will satisfy customer needs

0.6485

and wants Reliability (Cronbach’s alpha)

0.839

0.784

0.661

Eigenvalues

3.171

1.856

1.416

Cumulative variance explained

0.396

0.760

0.991

7.3.3

Regression analysis

Since the formation of entrepreneurial intentions (for the three types of entrepreneurial opportunity) are made interdependently and simultaneously within each person, is it appropriate to regress all three intentions against the independent variables simultaneously using “seemingly unrelated regression” to minimise the unexplained variance across the three regression equations (Zellner, 1962). The regression equations are presented in Table 7.3. For each of the three types of entrepreneurial intention, the socio-demographic variables are entered in a first step, and the perceived desirability and feasibility variables are entered as the second step. Starting with profit- (and growth-)seeking commercial entrepreneurship intention in the left side of the table (which we shall call growth-oriented intention to align with the discussion in Chapter 5), we observe that when the control variables are entered first, having a business degree and not being married are the only control variables significantly associated with GOI. Adding the perceived desirability items (i.e. the UPWs for profit, psychic

1

1.00

0.35

0.12

0.08

−0.11

0.29

0.09

0.11

0.05

−0.33

0.00

−0.05

0.26

−0.03

−0.05

−0.04

0.32

0.11

0.07

Variable

1. GOI

2. IOI

3. SEI

4. Age

5. Male

6. Business degree

7. Years experience

8. Prior Ent

9. Parent Ent

10. Married

11. Children

12. Femininity

13. Masculinity

14. UPW profit

15. UPW psych

16. UPW social

17. ESE marketing

18. ESE people

19. ESE finance

0.01

0.01

0.41

0.01

0.14

−0.22

0.37

0.08

−0.13

−0.12

0.17

0.26

0.01

0.06

−0.06

−0.00

0.14

1.00

2

−0.11

0.04

0.25

0.42

0.08

−0.29

−0.02

0.09

−0.02

−0.25

0.35

−0.13

−0.12

0.08

−0.24

−0.19

1.00

3

−0.07

0.16

0.12

0.06

0.11

−0.06

0.28

0.10

0.15

0.31

−0.32

0.20

0.95

−0.22

−0.01

1.00

4

0.28

−0.14

−0.12

0.02

−0.06

0.14

−0.06

0.01

−0.11

0.08

−0.13

0.09

−0.04

−0.04

1.00

5

0.09

0.06

0.14

0.06

0.02

0.01

0.32

0.05

0.01

−0.16

0.19

−0.02

−0.21

1.00

6

−0.20

0.15

0.08

0.15

0.10

−0.10

0.33

0.08

0.14

0.29

−0.29

0.19

1.00

7

0.08

0.05

0.09

−0.05

−0.04

0.04

0.29

0.07

−0.07

0.09

0.05

1.00

8

SEI, GOI, and IOI data correlation matrix

Table 7.2

0.03

0.14

0.07

0.10

0.17

0.07

0.15

−0.03

−0.00

−0.17

1.00

9

−0.05

0.12

−0.20

−0.17

0.04

0.26

0.03

0.04

0.19

1.00

10

−0.25

−0.15

0.08

−0.06

−0.04

0.15

0.05

−008

1.00

11

0.09

0.26

0.28

0.14

−0.04

−0.12

0.12

1.00

12

0.03

0.34

0.30

0.18

0.06

−0.25

1.00

13

0.18

−0.01

−0.30

−0.50

−0.52

1.00

14

−0.16

−0.12

0.16

−0.14

1.00

15

−0.14

0.11

0.05

1.00

16

0.07

0.12

1.00

17

0.23

1.00

18

1.00

19

116 Entrepreneurial intention

Social entrepreneurial intention

Table 7.3

Age Male Business degree Prior

117

Regression models for three types of entrepreneurial intention Profit

Profit

Psychic

Psychic

Social

Social

intention

intention

intention

intention

intention

intention

(GOI)

(GOI)

(IOI)

(IOI)

(SEI)

(SEI)

0.033

0.029

−0.014

−0.016

−0.003

−0.006 −0.379**

−0.160

−0.112

−0.075

−0.039

−0.362*

0.514**

0.452**

−0.189

−0.231

−0.027

0.110

0.199

0.150

0.293

0.240

−0.222

−0.111

−0.076

−0.148

0.136

0.174

0.693***

0.815***

self-employment Parent self-employment Married

−0.890***

−0.874***

−0.191

0.023

−0.400

−0.033

Children

0.036

−0.003

−0.119

−0.208

0.011

−0.012

Femininity

−0.105

−0.182

0.034

−0.009

0.212*

0.097

Masculinity

0.131

0.108

0.334***

0.316***

−0.061

−0.242**

UPW profit

0.089

−0.055

UPW psych

−0.069

−0.005

0.072

UPW social

−0.133

−0.111

0.708***

0.349**

0.415***

0.306**

ESE people

0.148

−0.208

−0.059

ESE finance

−0.005

−0.032

ESE marketing

“Rsqd” Change “Rsqd”

0.2590

0.3237 0.0647

0.2183

0.3549 0.1366

0.007

0.008 0.2530

0.4800 0.2270

Notes: * 0.1; ** 0.05; *** 0.01; N = 82.

income, and social-benefit provision) and the perceived feasibility items (i.e. the average score on the ESE items for each of marketing management, people management, and financial management) in the second step reveals that only ESE marketing is statistically significant in the GOI model. That is, none of the other antecedents of GOI exhibit a dominant net effect that is significant at the sample level. The pseudo R2 statistic indicates that the regression model explains a little over 32 per cent of the variance in GOI. In the middle two columns of Table 7.3 are displayed the stepwise regression model for psychic income-seeking (profit-satisficing) commercial entrepreneurship, which we shall call independence-oriented intention to align with the discussion in Chapter 5 and Douglas (2013). Here we see that among the control variables, only the measure of masculinity is significant, being positively related to IOI, with a relatively large effect size. Adding the remaining independent variables shows that, again, ESE marketing has a dominant net

118

Entrepreneurial intention

effect on IOI as well. This regression model explains almost 35 per cent of the variance in IOI. In the right-hand-side columns of Table 7.3 are displayed the results of regression analysis of the determinants of SEI. Considering first the control variables, note that being male is negatively associated with SEI, meaning that being female is associated with SEI, and that socially conditioned feminine traits (not necessarily in women) are positively associated with SEI. Parents being self-employed is also highly significant. Adding the remaining independent variables in the second step reveals that pro-social attitude (UPW social) is a highly significant determinant of SEI and that ESE marketing is again positive and significant. Interestingly, the introduction of UPW social appears to have mediated out the femininity variable and strengthened both the significance and the effect size of the “male” and the “masculinity” variables, as well as the parent-self-employed variable. This regression model explains about 48 per cent of the variance in SEI, much more than it did for GOI and IOI. 7.3.4

Contrast between the different types of entrepreneurial intention

The regression analysis reveals that GOI, IOI, and SEI are, in general at the sample level, driven by essentially different antecedent phenomena, except for ESE marketing which is common to all three. Considering the main effects shown in Table 7.3 we note that, except for ESE marketing, the independent variables that are significant in one model are insignificant, and are sometimes reverse-signed, in the other two. Age is insignificant in all three models, perhaps due to the limited range of age found among these MBA students (mean 30.5, S.D. 6.06) – they are all essentially of the same vintage, so no major age-dependent differences should be expected on average. Sex was only significant in the SEI model, where being female was significantly associated with pro-social intention, consistent with the findings of Hockerts (2017) and Ernst (2018). Yet femininity was insignificant in all three models, indicating it is not the socially conditioned “feminine traits” (i.e. seeing oneself as warm, gentle, affectionate, sensitive to others’ needs, tender, and sympathetic, from Bem’s (1981) short scale) that drive SEI. Rather, socially conditioned masculinity traits were negatively related to SEI (i.e. seeing oneself as a leader, defending one’s own beliefs, having leadership abilities, being dominant, having a strong personality, and making decisions

Social entrepreneurial intention

119

easily (Bem, 1981)). That is, not being “macho” in these ways was significantly associated with SEI, on average.5 Having a prior business degree drives only GOI, on average, where it had a relatively large effect size. A business degree is designed to equip individuals with the knowledge and terminology to better understand what running a profit- and growth-oriented business requires, and was more highly correlated with GOI (r = 0.29) than with IOI (r = 0.06) or SEI (r = 0.08). Perhaps a business degree, in general, tends to acculturate students to be profit-oriented and self-centred, and/or conversely, acculturates against profit satisficing and pro-sociality? Having parents self-employed drives only SEI, on average. This is unexpected, as various studies have shown that prior exposure to self-employment was associated with commercial entrepreneurship intention, but it may be that this variable was mediated out of the regression equation by ESE marketing and perhaps other variables, and/or that its influence was captured by non-significant variables that act to reduce the unexplained variance. This might also explain the non-significance of the respondent’s own prior self-employment experience. Being not married drives only GOI, on average. This may reflect the general understanding within the sample that growth-oriented entrepreneurship competes with marriage for time, energy, and resources, whereas independence-oriented and social entrepreneurship may be considered less demanding of personal time, energy, and resources. Having children is (marginally) negatively significant for IOI, but insignificant for the other two types of entrepreneurial intention, on average. Perhaps the ego-centricity of the profit-satisficing IOI entrepreneur envisions that having children will reduce the psychic benefits that they can derive from that kind of entrepreneurial behaviour. Now considering the attitudinal drivers (i.e. perceived desirability) of entrepreneurship, the attitude to profit (UPW profit), and the attitude to psychic income (UPW psychic), are found insignificant for all three types of intended entrepreneurship, on average. This is a surprising result, as we might at least expect attitude to profit to be generally associated with GOI and attitude to psychic income to be generally associated with IOI. For some individuals these expectations would undoubtedly be true, but on average across the sample, it is

It should be noted that being male was not highly correlated with masculinity (r = nor with femininity (r = −0.005) using the short version of Bem’s (1981) gender scales with this sample. All other bivariate correlations were less than 0.6 except for that between age and years work experience (r = 0.95), and the latter was excluded from the analysis. Apart from age, married, and having children showing correlations between 0.5 and 0.6, no other correlation exceeded 0.41. 5

−0.06)

120

Entrepreneurial intention

not the dominant net effect. Perhaps it is mediated out by the plethora of other independent variables in each regression equation, which, although they might not be statistically insignificant, tend to capture chance correlations with the dependent variable. Pro-social attitude (UPW social) is strongly and positively associated with the SEI, which is taken as almost axiomatic in the social entrepreneurship literature. For ESE, we find that all three types of entrepreneurial intention are positively associated with ESE marketing, but are insignificantly associated with the other two subfields of management capability that we expect are required for successful entrepreneurial behaviour. Intending to become an entrepreneur without self-confidence in the requisite skills would seem to be fraught with danger, exacerbating the risk of new venture failure. But this may be rationalised in three ways. First, respondents may envision teaming with others to start a venture, and seeking complementary skills on the top management team such that their own skill deficits are covered by a partner. Second, respondents may envision seeking the guidance and advice of their self-employed parents and a board of advisors (or directors) in their deficit areas. Third, respondents may envision outsourcing the people-management and financial-management tasks to professional accountants and employment services. Importantly, each of the regression coefficients represents the sample-wide average of the respondents’ coefficients between a particular independent variable and the dependent variable. They are the “dominant net effect” sample-wide for each independent variable, and inevitably will conceal individuals for whom the direction of effect and/or the level of significance is different. So within the sample there may be individuals for whom other independent variables are indeed responsible for their decision to form entrepreneurial intention of a particular type. Moreover, regression analysis considers the antecedent variables discretely (i.e. independently of every other independent variable), and no interdependencies between and among the antecedent variables have been considered in the present analysis. We shall consider these issues in Chapter 9, which utilises the same sample.

7.4 SUMMARY In this chapter we sought to extend the understanding of SEI by including as motivational drivers of SEI other sources of utility that might appeal to intending social entrepreneurs. It was argued that, in addition to pro-social motivation, the motivation for profit (UPW profit) and for psychic income (UPW psychic) would enter the utility function of some or perhaps many social entrepreneurs. Since the motivation for psychic income has been argued by many in the context of commercial entrepreneurship, it was considered important to investigate whether intending social entrepreneurs also envision psychic

Social entrepreneurial intention

121

benefits (due to enjoying independence and relationships with co-workers, for example) in the context of social entrepreneurship that are distinct from the pro-social motivation (i.e. UPW social) for social entrepreneurship. We found that for a sample of MBA students a relatively large portion of the variance in SEI (R2 = 0.48) was explained by a model that included socio-demographic variables (age, sex, prior business degree, prior self-employment, parents self-employed, being married, having children, and femininity and masculinity scores reflecting social-conditioned gender roles); as well as perceived desirability variables reflecting attitudes to profit, psychic income, and pro-social outcomes; and perceived feasibility variables reflecting self-confidence in marketing management, people management, and financial management. A cogent argument can be made that each of these independent variables might encourage or discourage entrepreneurial intention for specific individuals in specific contexts (such as social entrepreneurship), but multiple regression analysis found statistical significance at the sample level for only some of these variables. In general, SEI was found to be driven by being female, having parents self-employed, not having masculine gender traits, having a pro-social attitude, and having self-efficacy in marketing management. By comparison, profit- and growth-oriented (entrepreneurial) intention (GOI) was driven by none of those antecedent variables, except for ESE marketing. GOI was instead significantly associated with respondents having a prior business degree, not being married, and having high ESE marketing. IOI was similarly driven by high ESE marketing, but the other determinants were the absence of children and having masculine gender traits. These findings are dependent on a relatively small sample of Thai MBA students, which may not generalise well either within Thailand or internationally. But context does matter (Johns, 2006) and an argument exists that MBA students more generally are in a position to change careers upon graduation, and on that basis many prior studies have utilised MBA students as proxies for intending entrepreneurs. More importantly, these findings are derived using symmetric correlational analytical methods and show the sample-average relationships rather than purporting to explain the decision-making behaviour of any individual. To do that requires a case-based method, where the “within-person” relationships between and among the variables, rather than the sample-wide relationships, are examined in greater detail to reveal the finer nuances of the antecedents of entrepreneurial intention. This we will address in Chapter 9.

8. An inclusive model of entrepreneurial intention 8.1 INTRODUCTION There is a wide variety of commercial entrepreneurial ventures (Gartner, 1985), and also social entrepreneurial ventures (Defourny & Nyssens, 2010; Martin & Osberg, 2007), which vividly exemplifies the heterogeneity on both sides of the individual–opportunity nexus of entrepreneurship (Shane & Venkataraman, 2000). Observing this diversity, several scholars have lamented the lack of an integrating model of entrepreneurship (e.g. Choi & Majumdar, 2014; Newbert & Hill, 2014). In the preceding chapters we have considered how commercial entrepreneurship includes both profit-maximising and profit-satisficing entrepreneurs, with the former likely to be growth-oriented while the latter are typically not. We argued that profit-satisficing entrepreneurs are effectively trading off additional profit outcomes in favour of (i) non-monetary benefits that deliver psychic income; or (ii) social-benefit outcomes; and/or (iii) innovation outcomes. These four outcomes of entrepreneurship can be modelled as the potential goals of intending entrepreneurs, with individuals considering all, or some subset of, these four goals when forming the decision to become an entrepreneur. Depending on their attitudes to each of these four outcomes of entrepreneurship, individuals will care more or less about each of the profit, social-impact, innovation, and psychic-income outcomes expected to be associated with prospective entrepreneurial opportunities. Similarly for social entrepreneurial intentions, we should expect to see a variety of pro-social, profit, innovation, and psychic-income outcomes, depending on the relative strength of the individual’s attitudes to those four outcomes. The extant literature typically treats commercial and social entrepreneurship separately, dichotomising entrepreneurs into those whose primary purpose is social-benefit provision versus those whose primary purpose is private-benefit acquisition (Dees & Elias, 1998; Stevens et al., 2015; Tan, Williams, & Tan, 2005). In my view this simple dichotomy does not best serve the development of entrepreneurship theory or empirical research, since it leads researchers to focus on a single goal of the intending entrepreneur (either profit or social 122

An inclusive model of entrepreneurial intention

123

impact), and to treat any other goals as an amorphous remainder and not worth investigating separately. The extent to which social entrepreneurs wish to pursue profit making (and the extent to which commercial entrepreneurs wish to pursue social impact) tends to be glossed over, and the extent to which they wish to be innovative has been largely ignored in the entrepreneurship literature so far. Innovation has typically been treated as instrumental to profit making or social-benefit provision, rather than as a separate goal of the entrepreneur, despite anecdotal evidence that some entrepreneurs appear to be highly motivated by the prospect of becoming the local or global pioneer with a new technology or disruptive product, service, or business model, that with luck, might result in global domination and “unicorn” status. We first investigated the extent to which commercial entrepreneurs wish to pursue profit, in Chapter 5. We argued that profit making will be limited by the extent to which the entrepreneur also seeks to enjoy (i.e. gain psychic utility from) the entrepreneurial process, these benefits arising from decision-making independence, choice of perquisites, and social relationships with co-workers, customers, and suppliers, among other things (Campbell, 1992; Gimeno et al., 1997). The intending entrepreneur may also wish to avoid the psychic costs associated with greater intensity of profit seeking, such as additional risk bearing, mental stress, longer and more intense working hours, and less time to satisfy the innate relationship needs of the individual (Deci & Ryan, 1985). The net result of these psychic benefits and costs is expected to be a motivating factor for individuals contemplating entrepreneurship (Douglas & Shepherd, 2000). For expositional purposes it is convenient to call this net-psychic-income goal the “job-satisfaction” goal, following Lee et al. (2011). Note that attitude to job satisfaction is a higher-order construct, underlying which are the individual’s attitudes to autonomy, risk, hard work, and work enjoyment (Douglas, 2013). We argued earlier that some, perhaps many or most, intending entrepreneurs would truncate profit seeking before the point of profit maximisation, to increase job satisfaction and thereby maximise their total utility (TU; from both profit and job satisfaction) provided by the entrepreneurial process. We characterised this as profit satisficing, implying the pursuit of both their profit and job-satisfaction goals simultaneously and interdependently. In Chapter 7 we considered the extent to which intending entrepreneurs would wish to jointly pursue the provision of social-benefits, profit, and innovation outcomes. Social entrepreneurs, whose primary mission is to provide social benefits, will necessarily be profit satisficers as they must (at the margin) truncate profit seeking in favour of greater provision of social benefits. Their profit target might be higher or lower – even zero or negative – depending on the tradeoffs between the other three goals, which in a utility-maximising model involves the equality of the marginal utilities (MUs) of all four goal

124

Entrepreneurial intention

outcomes at the solution point. In reality, the pursuit of the four goals of entrepreneurship would be considered holistically and interdependently within the mind of the intending entrepreneur.1 In this current chapter we build a conceptual model that considers the strength of the profit, social-impact, innovation, and job-satisfaction motivations simultaneously and interdependently, consistent with the holistic theory of human decision making (Magnusson & Torestad, 1993). In principle, we should expect to find four primary types of intending entrepreneur. The primarily commercial-purpose and the primarily social-purpose entrepreneurs will still be in evidence, but it is theoretically possible that they will be joined by primarily innovation-seeking and primarily job-satisfaction-seeking entrepreneurs. It is also theoretically possible that each of these four main types will encompass multiple and different configurations of the non-primary motivations in conjunction with the primary motivation that illustrate different pathways to the formation of the same primary type of entrepreneurial intention. The conceptual model will be developed in this chapter, and the evidence for multiple configurations and asymmetric data relationships will be investigated empirically in Chapter 9, using fuzzy-set qualitative comparative analysis. These four goals of entrepreneurship have been recognised by prior scholars, of course. In the context of social entrepreneurship, Newbert and Hill (2014: 256) call social impact, profit, and innovation the “three pillars” of social entrepreneurship, and note that “while all definitions [of social entrepreneurship] recognize that pro-social behaviour is central to social entrepreneurship, only some note the importance of novelty [i.e. innovation], fewer still highlight the role of profit, and none include explicit reference to all three dimensions”. At about the same time, Lepoutre, Justo, Terjesen, and Bosma (2013) considered all three dimensions in a typology of social entrepreneurship, based on a binary classification of whether firms are, or are not, primarily social purpose, profit seeking, and innovative. Most recently, in Douglas and Prentice (2019), we considered all three pillars in a study of social entrepreneurial intentions using continuous attitude measures. Newbert and Hill (2014: 246) call for new theory that informs all organisations. Choi and Majumdar (2014) state that a theoretical model that reconciles the cognitions of social and private entrepreneurs would serve to move social entrepreneurship research forward. I would argue that an integrated model 1 The multiple regression analysis method that was utilised to investigate empirically the “importance” of these separate motivations treated them as independent and discrete, rather than as interdependent causal antecedents. Although a limited number of two-way and three-way interactions may be incorporated in multiple regression analysis, these are limited by sample size and multicollinearity restrictions. We will rectify this incongruity in Chapter 9 by the use of fuzzy-set qualitative comparative analysis.

An inclusive model of entrepreneurial intention

125

would also move commercial entrepreneurship research forward. An integrative model that explains why entrepreneurs may place a greater or lesser emphasis on the provision of social benefits, a greater or lesser focus on profitability, and adopt a more or less innovative business model is necessary to coalesce the divergent streams of research in this area. A model of commercial entrepreneurship that incorporates pro-social motivation would provide a more nuanced understanding of the motivation for the corporate social responsibility of commercial entrepreneurs, and help policy makers and educators induce and nurture pro-social behaviour by commercial entrepreneurs. Similarly, an integrated model would help policy makers and educators induce and nurture profit making and innovation by social entrepreneurs, and thereby facilitate the more efficient allocation of scarce government and philanthropic funding for social ventures. A key factor in this general model of entrepreneurial intentions are the interdependencies (or tradeoffs) between and among the four goals. Attaining greater profit outcomes is likely to mean less social impact, less job satisfaction, and less innovation (due to less spending on research and development). Similarly, more pro-social activity is likely to mean less job satisfaction (due to stress, fatigue, and emotional drain) and less innovation (due to the reluctance to abandon a legacy system in favour of a risky new alternative). And more innovation is likely to be accompanied by less job satisfaction, because innovation is risky, and we expect individuals to prefer less risk for any particular level of profit or social outcomes. Other tradeoffs are possible, but to avoid getting hung up on such a debate, the integrated model proceeds on the basis simply that there will be a negative tradeoff between job satisfaction and each of the other three outcomes of entrepreneurial behaviour. This will allow the model to be depicted as a three-dimensional model (depicting profit, social, and innovation outcomes) with the fourth dimension (job satisfaction) being implicit as the reason that the first three goals are not maximised in order to attain utility maximisation. Accordingly, this chapter presents an inclusive model of entrepreneurship that integrates both commercial and social entrepreneurship, rather than regarding these as separate subfields. By modelling that all entrepreneurs may receive utility from profit, social-impact, innovation, and job-satisfaction outcomes, a conceptual model of entrepreneurship is developed that considers not only primary-purpose commercial (or social) entrepreneurs, but also those for whom profit (or social benefit) is a second, third, or fourth priority. Similarly, profit maximisation, profit satisficing, and not for profit are alternative outcomes of the model, which recognises that some individuals seek more profit than do others. Finally, some entrepreneurs wish to be more innovative than do others. The preferences of entrepreneurs are treated as a continuum in the social-benefit, profit, innovation, and job-satisfaction dimensions, rather than

Entrepreneurial intention

126

as binary variables, such that the heterogeneity of the individual–opportunity nexus is embodied in the model in finer detail. This single model of entrepreneurial intention can be partitioned into 16 subcategories to provide a typology of entrepreneurial types based on combinations of “high” and “low” attitudes to social-benefit, profit, innovation, and job-satisfaction outcomes.

8.2

ELEMENTS OF THE CONCEPTUAL MODEL

As in previous chapters, entrepreneurship is defined as proactive, risk-taking, innovative behaviour that involves the creation of new products, services, and/or business processes that are intended to create wealth for the individual and/or society. The wealth created may be both monetary and non-monetary, and this wealth may be internalised as private benefits and/or externalised as social benefits. Thus, both commercial and social entrepreneurship fit within this definition of entrepreneurship, understanding that some part of the wealth generated may be in the form of social benefits for persons or entities that are external to the firm’s owners. The individual’s rewards from entrepreneurship may be totally monetary (i.e. profit maximising), something less than profit maximising (i.e. profit satisficing) but augmented by “psychic benefits” (Gimeno et al., 1997), which we will call job satisfaction here, or entirely non-monetary where profit is not intended and no dividend income or capital gain would accrue to any owners. 8.2.1

Theoretical lens

The overarching theoretical lens is self-determination theory (Deci & Ryan, 1985; Gagné & Deci, 2005), which suggests that individuals make behavioural decisions based on their expectation that the decision alternative selected will best serve their innate and situational needs and thus maximise their personal wellbeing (also known as psychic utility). Accordingly, the model will be explicated in terms of a utility-maximising model of human behaviour. The theory of planned behaviour (Ajzen, 1991; Ajzen & Fishbein, 1970) argues that an individual’s chosen behaviour is preceded by the formation of the intention to undertake that behaviour, and that this intention depends in turn on their attitudes toward the outcomes of that behaviour. Newbert and Hill (2014: 251) argue that “it is essential to define the [social entrepreneurship] construct by what [social entrepreneurs] intend to do a priori rather than what they may or may not have been able to do post hoc”. Entrepreneurial intentions models have been studied extensively in the context of commercial entrepreneurs (see, e.g. Douglas, 2013; Hessels, Van Gelderen, & Thurik, 2008; Krueger et al., 2000; Lee et al., 2011), but there have been relatively few studies of social entrepreneurial intentions (e.g. Douglas & Prentice, 2019; Hockerts, 2017;

An inclusive model of entrepreneurial intention

127

Mair & Noboa, 2006). Intentions models often use expectancy-valence theory (Vroom, 1964) to argue that the envisioned value of a prospective behaviour can be modelled as the product of the expected outcome of that behaviour (the expectancy) and the attitude (the valence) that the individual holds toward that outcome, as detailed in Chapter 3. Utility theory is consistent with the above psychological theories of human behaviour (Steel & König, 2006), and argues that individuals will undertake actions expected to maximise their personal utility. Utility theory has been used to explain and predict entrepreneurial behaviour previously (e.g. Douglas, 2013; Douglas & Shepherd, 2000; Eisenhauer, 1995; Fitzsimmons & Douglas, 2011; Kihlstrom & Laffont, 1979; Levesque & Minniti, 2006; Monsen et al., 2010; Segal et al., 2005; Simon, 1959; Wiklund et al., 2003). As we have noted in preceding chapters, the utility expected from a prospective action is the sum of the utility part-worths of the various expected outcomes of that action, where each utility part-worth is the product of an expected outcome and the individual’s attitude to that outcome. Thus, the model will depict entrepreneurs as making the decision to form entrepreneurial intention in order to maximise the TU expected from the profit, social-benefit, innovation, and job-satisfaction outcomes associated with a specific new venture opportunity. The four goals (and motivations) are now considered in some detail. 8.2.2

Profit as a goal of entrepreneurial behaviour

Individuals may derive monetary income (both salary and profit) from entrepreneurial behaviour, and can use these funds to purchase goods and services for personal and family consumption to satisfy personal needs and wants. While some social entrepreneurs may act in a voluntary capacity and seek neither salary nor profits (such as Jimmy Wales of Wikipedia), others will seek a salary which may be more or less than their opportunity cost in an alternative employment situation, and others will seek both salary and profits. Profit-seeking entrepreneurs may seek to maximise profit or alternatively seek satisfactory profit, in both cases in conjunction with non-monetary benefits (i.e. job satisfaction) derived from their entrepreneurial behaviour. Some psychic benefits may be derived, and psychic costs incurred, independently of profit outcomes (e.g. independent decision making and risk taking) but other psychic benefits, and the avoidance of some psychic costs, will involve the sacrifice of incremental profit. Thus, at the margin, we expect that the individual will trade off incremental profit in favour of incremental sources of (net) psychic income. In terms of MU, the profit making will be pursued to the point where the MU of the next dollar of profit is just equal to the MU of the job satisfaction that would need to be foregone by seeking that marginal dollar.

128

Entrepreneurial intention

Heterogeneous individuals will have differing preferential preferences for profit, which will depend on their attitude to material goods and services, this attitude commonly being called materialism. In psychology and cognate fields such as consumer behaviour, materialism identifies the extent to which individuals create and retain their self-identity through the acquisition of goods, services, experiences, and relationships (Shrum et al., 2013). Materialism has been associated with social status (Farid, 2007); individual identity (e.g. Richins & Dawson, 1992; Swann & Bosson, 2010); and wellbeing (e.g. Burroughs & Rindfleisch, 2002; Roberts & Clement, 2007; Shrum et al., 2013). The influential paper by Belk (1985) identified three main subconstructs of materialism, namely possessiveness, envy, and non-generosity. In Belk’s schema, generosity to others (a construct similar to pro-sociality) may be perceived as the polar opposite of materialism, but in the current model the attitude to profit is independent of the attitude to social benefits (although these may interact interdependently to drive TU and thus entrepreneurial intention). There is likely to be a tradeoff between the pursuit of profit and the pursuit of job satisfaction because the time and other resources spent pursuing one of these outcomes will have an opportunity cost in terms of the other. An entrepreneur seeking to maximise profit might receive minimal, zero, or even negative job satisfaction as a complement to that profit, due to increased levels of stress, work effort, and perhaps guilt (see Miller et al., 2012) that pursuit of profit maximisation may involve. A social entrepreneur might feel that zero profit is a satisfactory level, or alternatively might wish either to maximise profit (after achieving a pro-social target) or to gain a satisfactory level of profit, and in the latter case will be conscious of the tradeoff involved at the margin between the profit foregone and the social benefits provided (and the job satisfaction gained by that provision). For primarily social entrepreneurs in general, the job satisfaction from social benefits provided is expected to constitute a relatively large part of the TU derived, other things being equal. The model also assumes diminishing MU of profits – i.e. the MU accruing to the entrepreneur due to incremental dollars of profit will after some point decline progressively (Tang, 1995). More and more income from profit cannot buy everything that human beings desire, particularly intangibles such as emotional peace, self-respect, and the friendship and respect of co-workers and customers. Moreover, the more one seeks profit outcomes, the more one must trade off job satisfaction associated with the entrepreneurial process, due to the opportunity costs imposed by time and energy constraints. Entrepreneurs who seek a satisfactory level of profit, as a complement to the job satisfaction derived from their entrepreneurial behaviour, are presumed (by the assumption of utility maximisation) to balance the MU they gain from an extra hour of profit seeking with the marginal disutility they incur from the foregone job satisfaction that they might have otherwise derived in that hour. These arguments

An inclusive model of entrepreneurial intention

129

suggest that heterogeneous individuals will gain differing amounts of utility from any given quantum of profit earned, and after some point individuals will experience diminishing MU from additional profit income in any particular time period. 8.2.3

Social benefit as a goal of entrepreneurial behaviour

In the integrated model of entrepreneurial intention it is argued that entrepreneurs may serve the needs of their customers for both selfish reasons, i.e. to gain utility from the resultant profit and job satisfaction derived from the transaction, and for unselfish reasons, i.e. to gain utility from providing social benefits to others less fortunate than themselves. Unselfishness typically requires a personal sacrifice of income or psychic wellbeing from the consumption or production activity that is foregone in order to serve the interest of others (Becker, 1976). However, unselfishness also provides a positive psychic reward in that people generally “feel good” about helping, and/or avoiding doing harm to, others (Andreoni, 1990; Becker, 1976; Margolis, 1982). As Newbert and Hill (2014: 250) say, “to the extent that [they] are motivated by the gratification they feel or the admiration they receive from engaging in these transactions, they obtain private benefits as a result”. A construct commonly used in the social entrepreneurship literature to encapsulate unselfishness is pro-sociality, which has been defined as the willingness to help other people. It may be based on altruism, where there is no expectation of reward, or indeed the willingness to make personal sacrifice by helping others (Becker, 1976; Patzelt & Shepherd, 2011; Penner, Dovidio, Piliavin, & Schroeder, 2005). Researchers have also looked at the “compassion” construct for its other-orientation and as a driver of pro-social motivation (Miller et al., 2012). However, compassion is an emotion whereas pro-sociality is a motivational state (Batson & Shaw, 1991). Although pro-sociality is widely agreed to motivate social entrepreneurship, it has received relatively little attention in the commercial entrepreneurship contexts. Any investigation into an individual’s pro-social behaviour should distinguish conceptually between the desire of an individual to enhance their own welfare, as distinct from the desire to enhance the welfare of others. Thus, in this model the utility derived from pro-social behaviour (i.e. the utility derived from the generation of social benefits) is separated conceptually from the utility derived from the profit, innovation, and job satisfaction associated with those acts. Accordingly, pro-sociality is defined here more broadly than altruism to be the individual’s attitude toward pro-social behaviour, with the expectation that the entrepreneur may gain utility from pro-social behaviour. In this model the utility derived by the entrepreneur from generating social benefits is not related to the quantum of utility derived by the recipients, who enjoy those

130

Entrepreneurial intention

social benefits independently of the entrepreneur’s utility. Thus the model does not involve interdependent utility functions, nor a social welfare function. It is assumed that the entrepreneur’s MU from pro-social outcomes will after some point tend to decline as additional pro-social behaviour, such as helping one more homeless person, is conducted during any specific time period. The next pro-social act is postulated to generate lesser psychic satisfaction (i.e. MU) than did the preceding act, after some point, because the physical and emotional energy of individuals are finite and they are likely to become physically tired and/or emotionally upset as they increase the frequency of their pro-social behaviour in any given time interval. Moreover, progressively more unselfishness must equate with progressively less selfishness, as time and energy are limited, and people will wish to attend to their own and family needs as well. The foregoing argument suggests the proposition that after some point individuals will experience diminishing MU from additional social benefits generated in any particular time period. 8.2.4

Innovation as a goal of entrepreneurial behaviour

Innovation is instrumental to the entrepreneurial process, by definition of entrepreneurship as proactive, risk-taking, innovative behaviour (e.g. Miller, 1983). In this model we posit that in addition to its indirect impact on TU via profit and social-impact outcomes, innovation will have a direct impact on TU. It is posited that individuals will have a net positive or net negative attitude to innovation, per se, and the degree of this preference or aversion will differ according to a variety of antecedent factors. Some will derive utility from innovation due to its conferral upon them of the prestigious titles of innovator, pioneer, or first mover. Some will derive utility from the novelty of the innovation situation, and/or from the predicted sense of achievement associated with successfully meeting the challenges associated with innovation. Some will also have to engage in more challenging forms of innovation, such as bricolage, in resource-constrained environments (Baker & Nelson, 2005; Senyard, Baker, Steffens, & Davidsson, 2014), and gain utility from meeting that challenge. Others may be averse to innovation because their lack of knowledge of the new technology or new market causes negative affect relating to risk and uncertainty regarding the reactions of customers, competitors, and legislators. They may fear legal challenges to their claims for intellectual property, or be concerned that defending these will be very expensive. The individual’s attitude to innovation will be the net of these positive and negative effects. Note that a negative attitude to innovation does not necessarily mean the individual will not innovate – in a multigoal model, the utility from the other goals might more than compensate for the disutility from innovation (even for an innovation-averse individual).

An inclusive model of entrepreneurial intention

131

On balance, some will hold a negative attitude to innovation (and are thus less likely to innovate) while others will hold a positive attitude to innovation and will seek to innovate in pursuit of additional utility. Their attitude to innovation will exhibit degrees of preference or aversion to innovation, which will be manifested in the business model chosen. The firm’s business model includes the choice of products and services, business processes, target market, differentiation, and growth aspirations (Morris, Schindehutte, & Allen, 2005). The social venture may be more or less innovative, ranging from very low to relatively high levels of innovation in the products, services, and/or business processes involved in their business models. Particularly where the lack of a product or service is due to market failure (Santos, 2012), innovation by social entrepreneurs is perhaps more likely to be business process innovation, rather than the creation of new products or services. The attitude of individuals to innovation will be a composite of their attitudes to the various outcomes that are expected to be associated with their innovation in a particular context. Beneficial (utility-generating) outcomes may include a sense of creative achievement (McClelland, 1961) associated with creating a new product or business process; satisfaction associated with overcoming technical or managerial challenges; and pride in having the courage to risk their time and money in the implementation of a new venture. Adverse outcomes associated with innovation will include enhanced risk exposure, harder work, and increased mental stress that is due to the innovation – i.e. incremental to what the new venture would have without the innovation. As with the MU of profit and of social benefits, the MU of innovation outcomes (from the entrepreneur’s perspective) is expected to decline after some point as the increasing quantum of innovation requires greater work intensity, greater uncertainty, and greater personal stress, and as the risk of the innovation increases. The foregoing arguments suggest that heterogeneous individuals will gain differing amounts of utility from any given quantum of innovative outcomes generated, and that after some point individuals will experience diminishing MU from additional innovative outcomes generated in any particular time period. 8.2.5

Job satisfaction as a goal of entrepreneurial behaviour

Following Eisenhauer (1995), in Douglas and Shepherd (2000) we argued that entrepreneurs will expect to gain (net) utility from the non-monetary outcomes of entrepreneurship. Gimeno et al. (1997) called this “psychic income” and here we will call it “job satisfaction” following Lee et al. (2011). As mentioned previously, this psychic income is derived from various non-monetary sources, such as decision-making independence, flexibility, prestige, work enjoyment, and so on. On the other hand, individuals may expect to incur psychic costs

132

Entrepreneurial intention

associated with entrepreneurial behaviour, such as risk bearing, hard work, and mental stress. Above, I have argued that additional job satisfaction is the opportunity cost of achieving more profit, and/or more social impact, and/or more innovation. Accordingly, it simplifies the exposition of the model somewhat to include the job-satisfaction goal as the opportunity cost of each of the other three goals, and thereby focus on what Newbert and Hill (2014) called the three pillars of social entrepreneurship (namely, social impact, profit, and innovation). To integrate the theories of both commercial and social entrepreneurship, we propose here that social impact is also potentially relevant for commercial entrepreneurial intention, and that attitude to innovation should be included in the model as a potentially motivating goal for both the commercial and social entrepreneur, rather than simply being instrumental to the achievement of profit and social outcomes.

8.3

DEVELOPMENT OF THE GENERAL UTILITY MODEL

Thus, the model posits that heterogeneous individuals will have continuous attitudes to profit, social benefits, innovation, and job satisfaction, rather than a binary attitude (such as “agree/disagree” or “applies/does not apply”) to any of these outcomes. Accordingly, the model posits that individuals exhibit degrees of preference or aversion for the four goals proposed in the model, and assumes that individuals will prefer to start a new venture that maximises their expected utility from a venture’s “production” of profit, social-benefits, innovation, and job-satisfaction outcomes. The model is at first an unconstrained utility-maximising model – that is, there is no budget constraint applied – it is effectively a “wish list” of how an individual would prefer to design a new venture to maximise utility from the four salient outcomes if not subject to any budget constraint. A budget constraint would apply if the entrepreneur is unable to raise the desired amount of start-up capital, with the consequence that the venture (and its four main outcomes) would need to be scaled back or otherwise modified. With subsequent survival, profit, and growth, the venture may be able to scale up and diversify in pursuit of their longer-term objectives, but this is well beyond the intentions stage of the new venture process so will not be considered here. As in any utility-maximising model, rather than go too far in the direction of any one source of utility, such that its MU falls below the MU of another source of utility, or becomes negative, individuals will choose more of whichever utility source offers the highest MU, and less of another that offers a lesser MU, thus tending to equilibrate the MU from all sources. In the context of a prospective new venture, the individual will be prepared to forego some units of one utility source (e.g. from profit) if they can gain greater MU from another

An inclusive model of entrepreneurial intention

133

utility source (e.g. from providing social benefits), such that TU is increased by that tradeoff. In an unconstrained utility-maximisation model, the four sources of utility will be pursued until their MUs are equal. In a constrained utility model the decision-making process will culminate in the solution where the MUs of all utility sources are equal at a value greater than zero. If this were not true, a tradeoff of one source of utility for another would increase the individual’s TU. In practice, of course, it is unlikely that any entrepreneurial opportunity’s “supply” of the profit, social-impact, innovation, and job-satisfaction outcomes would exactly match the individual’s “demands” for those outcomes. Instead, the individual would search for and choose the entrepreneurial opportunity that best fits their preferences for (i.e. attitudes toward) these outcomes, as well as their self-efficacy (i.e. abilities) for the tasks involved in that entrepreneurial opportunity, in line with self-determination theory (Deci & Ryan, 1985), the theory of planned behaviour (Ajzen, 1991), and expectancy-valence theory (Vroom, 1964). 8.3.1

Graphical depiction of the model

It is useful, particularly for more “visual” readers, to depict the model graphically to demonstrate the interdependence of the three main pillars in providing the utility-maximising solution for the intending entrepreneur. For expositional reasons, we begin with a simple two-dimensional model (social benefits and profit) as shown in Figure 8.1, to allow initial focus on the profit-making and pro-social aspirations of intending entrepreneurs. Indifference curves for a particular individual, relating preference for social-benefit outcomes to preference for profit outcomes, are depicted as the curved lines representing combinations of profit and social-impact outcomes that generate the same level of TU to the individual. The negative and convex slope of indifference curves reflects positive but diminishing MU of both variables, with higher curves preferred to lower curves due to the assumption of transitive preferences (see any microeconomics text). Note that each axis of Figure 8.1 depicts a scale of 0–100 with 100 representing the maximum possible on that scale, namely the maximum possible social benefits and maximum possible profit, respectively. Indifference curves are negatively sloping only as long as the MUs are positive. At higher levels of social-benefit and profit outcomes the declining MUs may become negative and cause the indifference curves to transition to a positive slope, requiring less of one or both variables to remain at the same level of TU. As shown in Figure 8.1, at high levels of both variables the indifference curves adopt positive slopes and become ellipses. The point of maximum utility (X) is found at the dot representing the centre of the smallest ellipse, where pursuit of any further profit or social impact would cause TU to

134

Figure 8.1

Entrepreneurial intention

The utility-maximising combination of social benefits and profit

decline (due to the job satisfaction that must be sacrificed), as that combination would lie on an adjacent ellipse which necessarily indicates a lower level of TU. At point X, the MUs of the three outcomes are therefore equal. The dotted line emanating from the origin shows the utility-maximising path that an individual might hypothetically follow, from the least preferred combination (at the origin) to the most preferred combination (X) at the centre of the smallest ellipse. By choosing the profit–social combination X the individual maximises utility with relatively high social-benefit outcomes (S1) and relatively high profit income (P2). This individual does not maximise profit, nor social benefits, but rather maximises utility via a combination of the social benefits provided, the profit earned, and the job satisfaction derived. The individual stops short of both profit maximisation and social-benefit maximisation first due to the negative tradeoff between these two outcomes but also because of the negative tradeoff between job satisfaction and each of the other two outcomes. A similar two-dimensional representation (not shown) can be imagined for the profit and innovation choices of the intending entrepreneur. Again, diminishing MU would cause the utility-maximising combination to occur at

An inclusive model of entrepreneurial intention

135

something less than profit maximisation and something less than an extremely innovative business model, due to the tradeoffs between these two variables and the job satisfaction which must be sacrificed to reach higher levels of profit and innovation. The third two-dimensional combination of the outcome variables (not shown) would be the choice between social-benefit outcomes and innovation outcomes. Also in this case, diminishing MU would cause the utility-maximising combination to occur at something less than the maximum social benefits possible and something less than an extremely innovative business model, due to tradeoffs between these two (if any) and/or due to the job satisfaction that would be sacrificed to achieve these extreme outcomes. We now proceed to consider the interdependency of all four variables simultaneously in an Edgeworth–Bowley box diagram (Edgeworth, 1881; Humphrey, 1996), as in Figure 8.2.

Figure 8.2

Indicative paths to utility-maximising choice of new venture characteristics

Figure 8.2 depicts social-benefit outcomes on the X-axis, profit outcomes on the Y-axis, and innovation outcomes on the Z-axis of the box diagram. Again, each dimension is measured on a scale of 0–100 with 100 representing the maximum possible on that scale. The preferences of the person depicted earlier

136

Entrepreneurial intention

in Figure 8.2, call this person A, are indicated by the heavy dotted line on the left-hand side of the box diagram, with profit and social outcomes P2 and S1 as before, and assuming that person A gains no utility from innovation. The underlying three-dimensional indifference surfaces are omitted in Figure 8.2 to allow focus on the different pathways to utility maximisation followed by different individuals. Two other hypothetical individuals are also depicted in Figure 8.2 – they each start from the origin (corner 1) and move consistently in the direction of the greatest incremental gain in TU by choosing to produce more of the outcome for which the MU is highest. Person B is represented by the unbroken curved line, which depicts combinations of the four outcomes that are expected to generate progressively higher levels of TU for that person. The convexity of this curve demonstrates that the MUs of the three variables are declining progressively throughout, until the point where MU is equal for all four outcome variables, and thus TU is maximised. For this individual, the preferred new business venture would produce a very high level of social benefits (S3); a relatively high level of profit (P1); a relatively low level of innovation in the business model (I1); and a relatively low level of job satisfaction (implied by the location of the solution inside the box). A third individual, person C, is represented by the dashed curve: this person would maximise utility with relatively high social-benefit production (S2); relatively high profit (P2); relatively high level of innovation in the business model (I2); and relatively low job satisfaction. Other examples can be imagined. An extremely selfish person who does not care for social-benefit production would want to be located somewhere on the “floor” of the box, with the exact location dependent on the preferred combination of profit, innovation, and job satisfaction. Yet another individual, with high pro-sociality and who does not care for profit, would want to be located on the “back wall” of the box, operating a not-for-profit social venture more or less innovatively, depending on their relative preference for pro-social, innovation, and job-satisfaction outcomes. Thus, it is postulated that individuals choose the design of their new venture according to their personal cognitions (valences) regarding social benefits, profit, innovation, and job satisfaction, in conjunction with their expectancies of what the new venture offers in those dimensions. Note that consideration of the job-satisfaction goal operates to forestall extreme demands for the other three sources of utility by the focal entrepreneur in this model. Thus, an individual’s utility might be maximised at any particular point within the box, or on the outside surface of the box. Such “corner” solutions are possible if the person derives either zero utility from one of the outcomes, or if MU does not decline, or does decline but remains positive at extreme values for that source of utility. The individual “stops” at a point within the box when the MUs of

An inclusive model of entrepreneurial intention

137

all four variables have declined to the same level; given declining MU of all four variables there is no point choosing higher levels of any variable because that would cause TU to decline. Thus, the utility-maximising solution for any individual is likely to be somewhere in the interior of the box, and will depend on the individual’s relative strength of preference for profit, social-benefit, innovation, and job-satisfaction outcomes.

8.4

A TYPOLOGY OF COMMERCIAL VERSUS SOCIAL ENTREPRENEURIAL INTENTION

To reduce the infinite number of points representing potential combinations of profit, social-benefits, innovation, and job-satisfaction outcomes, that is evident inside the box, to an empirically useful number of entrepreneur (or venture) types, and to adhere to the convention that entrepreneurs might be either primarily social or primarily commercial, a median split may be used to dichotomise each attitude as either “high” or “low” and thereby identify 23 = 8 possible combinations of attitudes to these four goals of entrepreneurship. These eight separate intention types are shown in Table 8.1, coded to reflect the relevant corner of the box diagram depicted in Figure 8.2. The exemplar ventures shown in the last column of Table 8.1 represent generalisations and may be debatable in specific cases, but serve to indicate occupations or business situations that might be found in each quadrant of the box diagram as a result of different founder attitudes to the four main outcomes. Figure 8.3 depicts the typology in a different way, stepping through the three main axes (pillars) of entrepreneurial intention included in this model, following Lepoutre et al. (2013). The first step is the primary-purpose dichotomy, the second step is the profit-maximisation versus profit-satisficing dichotomy, and the third step is the innovation (high versus low) dichotomy. In Figure 8.4 the box diagram is again shown, this time partitioned (according to the median splits of the attitudes to the three pillars) into eight smaller quadrant boxes, each anchored at one of the corners of the box diagram. Each quadrant box represents a particular combination of above the median, that is high (H) or below the median (L) on each of the three attitudes. Thus, for example, quadrant 6 represents HHH, or above-median attitude to each of social benefits, profit, and innovation. All eight types are shown in Figure 8.4 except for type 4 (LLH) which is hidden at the back corner of the larger box. 8.4.1

The value of typologies for theory building

Delbridge and Fiss (2013: 329) argue that “typologies present a particularly attractive form of theorizing”, and Snow and Ketchen (2014: 231) argue that “the organization sciences would benefit from an increased emphasis on

Entrepreneurial intention

138

Table 8.1

Case

Social

Typology of entrepreneurs based on social-benefits, profit, and innovation attitudes Profit

Innovation

High

High

Description of type

Indicative examples

Primarily social-purpose,

Social-purpose profit-seeking firms that

profit-maximising, more

innovate to serve social needs, such as

innovative firm

Virgin and DuPont.

Primarily social-purpose,

Social-purpose profit-seeking firms with

profit-maximising, less

more conventional business model, such as

innovative firm

commercially operated retirement homes.

Primarily social-purpose,

Social-purpose profit-satisficing firm

profit-satisficing, more

with innovative business model, such as

innovative firm

Wikipedia.

Primarily social-purpose,

Social-purpose profit-satisficing firm with

profit-satisficing, less

conventional business model, such as soup

innovative firm

kitchens for the homeless.

Primarily commercial,

Commercial-purpose profit-seeking firm

profit-maximising,

with innovative business model, such as

more innovative firm

high-growth technology ventures.

Primarily commercial,

Commercial-purpose profit-seeking firm

profit-maximising,

with more conventional business model,

less innovative firm

such as “salary-replacement” retailing firms.

Primarily commercial,

Commercial-purpose profit-satisficing, more

profit-satisficing,

innovative firm, such as those based on

benefits 7

6

3

2

8

5

4

1

High

High

High

High

Low

Low

Low

Low

High

Low

Low

High

High

Low

Low

Low

High

Low

High

Low

High

Low

more innovative firm

personal expression, such as artists.

Primarily commercial,

Commercial-purpose profit-seeking firm

profit-satisficing, less

with more conventional business models,

innovative firm

such as “lifestyle” small businesses.

typology-driven theorizing” particularly in emerging topic areas where it is “important that the typology be accompanied by a theoretical framework upon which future theorizing can build”. This is still relevant for commercial entrepreneurship and is certainly a propos for the emerging social entrepreneurship literature. While others have attempted to categorise social entrepreneurs (e.g. Lepoutre et al., 2013; Zahra et al., 2009), these eight types provide a more inclusive typology of entrepreneurial types based on a theoretical model relating the attitudes of the entrepreneur to the profit, social-benefit, innovation, and job-satisfaction outcomes to be produced by the new venture. The eight categories of entrepreneurial types include four types for whom the primary purpose is the production of social benefit and the secondary purpose is the production of profit. Similarly, there are four types for whom profit is the primary purpose and social-benefit production is a secondary purpose.

An inclusive model of entrepreneurial intention

139

Note: * Profit satisficing includes individuals for whom zero or negative profit is a satisfactory profit, and reflects the psychic-benefit motivation.

Figure 8.3

A typology of intending entrepreneurs (and their firms)

Figure 8.4

A typology of entrepreneurs in social-benefit, profit, and innovation space

140

Entrepreneurial intention

At the extremes, but within the model, are not-for-profit, profit-maximising, non-social, and non-innovative firms and organisations, where the preference for one or other of the three main sources of utility is zero or negative. Condensing the myriad possible combinations into eight types is potentially useful for research questions that pertain to a particular subset of entrepreneurs rather than the entire universe of entrepreneurs, since it serves to reduce the amount of unexplained variance due to extraneous and/or missing variables using correlational methods. 8.4.2

Empirical evidence of eight subtypes

It is of interest whether the eight subtypes of entrepreneur are equally likely to be observed, or whether some subtypes occur more frequently than do others. A simple test is provided by the sample utilised in Chapter 7, comprising 82 MBA students, who may be regarded as a suitable proxy when the research question concerns entrepreneurial intentions (McGee et al., 2009). They were subjected to a conjoint experiment to reveal their attitudes to the social-benefit, profit, and innovation outcomes of entrepreneurship. As argued earlier, the conjoint method is likely to provide more reliable attitude measures than survey questions due to social-desirability bias in espoused responses (Shepherd & Zacharakis, 1999). Respondents rated a series of job options (which could be either employed or self-employed) in terms of the ability of those jobs to provide social-benefit, profit, and innovation outcomes. For each job the outcome variables were set as “high” or “low” with these levels being defined as relatively large and relatively small, respectively. This allowed the computation of their utility part-worths for each of the three desirable outcomes of entrepreneurial behaviour. The scores of each individual were then compared to the sample median for each attitude and rated as “high” if above the median and “low” if below the median. The individuals were then categorised, for example, as “High, High, Low” (i.e. HHL) if they were above the sample mean for their attitudes to both social benefits and profits but below the sample mean for attitude to innovation. Table 8.2 shows the distribution of the student sample across the eight categories. Some of these differences are notable. Relatively few respondents (4.9 per cent) were found at the extreme case of HHH, which is indicative of the “full-on” hybrid-social entrepreneur, and similarly at the other extreme case of LLL (the low-innovation, profit-satisficing, commercial entrepreneur) with only 2.4 per cent of the sample. For those who are primarily social purpose-seeking, profit-satisficing intentions (35.4 per cent) substantially exceeded profit-maximising intentions (14.7 per cent). Less innovative social-purpose intentions (33 per cent) substantially exceeded more innovative social-purpose intentions (17.1 per cent). For those with commercial

An inclusive model of entrepreneurial intention

Table 8.2

141

Empirical frequency of the eight types of entrepreneur

Code

Type

Percentage

HHH

7

Social-purpose, profit-seeking, more innovative

4

HHL

6

Social-purpose, profit-seeking, less innovative

8

4.9 9.8

HLH

3

Social-purpose, profit-satisficing, more innovative

10

12.2

HLL

2

Social-purpose, profit-satisficing, less innovative

19

23.2

LHH

8

Commercial-purpose, profit-seeking, more innovative

15

18.3

LHL

5

Commercial-purpose, profit-seeking, less innovative

12

14.6

LLH

4

Commercial-purpose, profit-satisficing, more innovative

12

14.6

LLL

1

Commercial-purpose, profit-satisficing, less innovative

2

2.4

Totals

82

100

(private-purpose) intentions, profit-maximising intentions (32.9 per cent) substantially exceeded profit-satisficing intentions (16.6 per cent). Finally, more innovative private-purpose intentions (32.9 per cent) substantially exceeded less innovative private-purpose intentions (16.8 per cent). These percentages are derived from a specific (convenience) sample and it is not claimed that they will generalise to other samples. But they do provide food for thought and require explanation for the apparent differences. They paint a broad picture of the heterogeneity of both social and commercial entrepreneurs, and an unequal distribution across the eight-way typology when it comes to profit, social impact, and innovation motivation. They suggest that primarily social-purpose entrepreneurs tend to be less innovative and less profit seeking than commercial-purpose entrepreneurs, which, in general, is not contradicted by casual observation. Future studies might attempt to gain larger samples within each of the eight types, to reduce unexplained variability and to discover the finer nuances (including missing antecedent variables) of what motivates individuals to intend entrepreneurship of these different types. Utilising such “more homogeneous” samples would allow a more fine-grained analysis of entrepreneurship intention in future studies.

8.5

EXTENDING THE MODEL TO THE FOURTH DIMENSION

In the above three-dimensional model we explicitly considered profit, social impact, and innovation, while implicitly considering job satisfaction as the force preventing an individual from “going too far” in the direction of any of these three dimensions, due to the opportunity cost, in terms of job satisfaction foregone, of so doing. An alternative specification of this “four-goals” model would add job satisfaction as a separate desirable goal, alongside the other

142

Entrepreneurial intention

three drivers of perceived desirability, and I suspect some readers will be pleased that I am unable to represent it graphically! We expect that entrepreneurs would choose to pursue these four goals differentially, depending on their expected utility part-worths for each of the four sources of utility, and on their entrepreneurial self-efficacy (ESE) for the tasks involved. Given the heterogeneity of intending entrepreneurs, we expect that most intending entrepreneurs are likely to have one predominant (or primary) motivation and three lesser (possibly even zero or negative) motivations, that combine in interdependent conjunction with each other, and with their ESE, to cause the individual to intend to undertake a particular entrepreneurial opportunity. Consideration of the four main motivations underlying entrepreneurship intention leads to a taxonomy of primary types of entrepreneurial intention as indicated in Figure 8.5.

Note: Any number of configurations within each of these main types may be discovered empirically.

Figure 8.5

Schematic of primary-entrepreneurial-intention typology

An inclusive model of entrepreneurial intention

143

It is clear that expanding the motivating goals from three to four expands the number of combinations of high (above median) and low (below median) from 23 = 8 (as in Figure 8.3) to 24 = 16 distinctly different configurations of the perceived desirability of entrepreneurship intention. It is not necessary to add extra pages to this book to illustrate this – it can be imagined on the basis of Figure 8.3. And when different combinations of the three identified subtypes of ESE are added to the mix, there would potentially be a multitude of potential combinations that culminate in the formation of entrepreneurial intention. What matters is which of the myriad possible combinations are actually observed in practice. This we shall consider in Chapter 9.

8.6 CONCLUSION In this chapter we developed a conceptual model of entrepreneurial intention to integrate several issues that have largely been discussed separately in the entrepreneurship literature. The separation of the social and commercial entrepreneurship literatures, and the partitions within both the social entrepreneurship literature (e.g. for profit versus not for profit; innovative versus not innovative) and the commercial entrepreneurship literature (e.g. growth oriented versus independence oriented), have served to inhibit theoretical and empirical research in both social entrepreneurship and commercial entrepreneurship. The variety of definitions and the lack of an underlying general theory of entrepreneurship have been lamented by many scholars, as has the overreliance on dichotomous categorisations of social-purpose versus commercial-purpose firms, of profit-maximising versus profit-satisficing firms, non-profit versus for-profit social enterprises, and so on. Such binary categorisations serve to confuse and fragment research into less effective tribal discussions that need to be interpreted and reconciled. In this chapter we first argued that entrepreneurial intention depends conjuncturally on the individual’s attitudes toward the profit, social-benefit, innovation, and job-satisfaction outcomes of entrepreneurship, and that attitudes to these four outcomes are continuous rather than binary variables. This model recognises that both social entrepreneurs and commercial entrepreneurs may want to earn profit and provide social benefits, and allows for these as separate and independent continuous variables, rather than one being the complement of the other along a bipolar scale. This model also makes explicit that innovation is not just something that entrepreneurs do, it is something they may want to do because they derive psychic satisfaction from achieving innovative outcomes per se (in additional to deriving satisfaction from the profit and social outcomes for which innovation is instrumental). It also makes clear that pursuit of each of the three pillars of entrepreneurship is limited by the interdependen-

144

Entrepreneurial intention

cies among these three outcomes as well as by the opportunity cost of foregone job satisfaction that the other three goals require. The four-goal model, incorporating both the “three-pillars” and “job-satisfaction” considerations, introduces the notion that “pioneering” may be a separate and important goal for some intending entrepreneurs. Extending our perception of entrepreneurial innovation, beyond it being instrumental for profit and social-impact outcomes, to being a source of utility for the intending entrepreneur, serves to put innovation on the same footing as profit, social benefit, and job satisfaction as desirable outcomes of entrepreneurial behaviour. Moreover, it hypothesises that the pursuit of innovation, per se, may be revealed in subsequent empirical testing as a separate and distinct goal of at least some entrepreneurs, particularly those who have been acculturated into the desirability of becoming recognised as a global pioneer responsible for a high-growth new business venture. Incubators and three-day start-ups are replete with individuals who seem obsessed with being first to market, harbouring visions of global dominance in the mould of “unicorns” like Microsoft, Apple, Amazon, Google, Facebook, YouTube, Twitter, and so on. The integrated four-goal model addresses the “extent of innovation” question that has been sidestepped for so long, offering a theory for why some entrepreneurs are more innovative than are others.2 To the extent that the innovation motivation is found empirically to be distinct from profit, social-impact, and job-satisfaction motivation, it should provide additional explanatory power to models of entrepreneurial intention and subsequently allow us to build better entrepreneurship theory. There are myriad combinations of these four goals that might be valued by heterogeneous intending entrepreneurs. To reduce this multitude to a more manageable number for analysis and effective discourse, we divided the universe of intending entrepreneurs into eight types based on their being above or below the median on each of the three pillars of entrepreneurship. These eight types include entrepreneurs whose preference for social impact varies from relatively high to relatively low, that is, ranging from being the primary purpose of the new venture to being a secondary or lesser purpose. Similarly, the preference for profit ranges from being relatively high (profit maximising) to being a satisfactory (lesser) target level. The dichotomisation of entrepreneurs as primarily social purpose, or primarily commercial purpose, still exists in the model, but represents only a first cut at the complexity of both social and commercial entrepreneurship. Although the continued use of this simple

2 The extant theory is that innovation is greater when it allows greater profit or social outcomes, and thus innovation is desired indirectly. Here we argue that innovation is also desired directly for its generation of utility for the entrepreneur.

An inclusive model of entrepreneurial intention

145

dichotomy will accommodate many research questions, a more fine-grained analysis of commercial and social entrepreneurs is possible by including the innovation motivation and the job-satisfaction motivation as well. In each of the resultant eight quadrants of the four-goals model, individuals will be found whose motivations for entrepreneurship are more alike those of fellow quadrant members than they are alike the motivations of others in different quadrants. It may be useful for a variety of research questions to sample only one quadrant of the population, such as HLL (i.e. social-purpose, profit-satisficing, and less innovative) individuals, for example. Sample selection on this basis would facilitate theory building using correlational analytical research methods. This would reduce the amount of unexplained variance inherent in the researcher’s sample, and allow more fine-grained analysis of other issues and the introduction of (missing) new variables into the analysis. We then considered more explicitly the hidden fourth dimension, namely the job satisfaction that the individual expects to derive from the entrepreneurial process in pursuit of a particular entrepreneurial opportunity. Bringing this to the forefront of the analysis should serve to identify intending entrepreneurs whose primary motivation is to improve their quality of life, giving this greater weight in their entrepreneurial decision making than they do for the profit, social-impact, and innovation outcomes. The prior identification of independence-oriented entrepreneurs, leisure-oriented, salary-sacrifice, and lifestyle entrepreneurs, as well as mum-preneurs, elder-preneurs, migrant-preneurs, displaced professionals, and so on, are probably best considered as proponents of the “primarily job satisfaction-seeking” type of entrepreneur, distinctly different in their primary motivation to the other three primary-motivation types. Moving from the primarily social versus primarily commercial dichotomy of entrepreneurship, to the “three pillars” of entrepreneurship, and in turn to the “four-goal” model of entrepreneurship, allows the development of an inclusive model of entrepreneurship that encompasses all combinations of profit, pro-social, innovation, and job-satisfaction motivation that may jointly underlie the formation of entrepreneurial intention (for some type of entrepreneurship or another). This four-goal model can potentially provide a more nuanced theoretical answer for the questions such as “why do (particular) individuals undertake entrepreneurship?” and “why do different individuals take different pathways to entrepreneurship?” It argues that individuals choose to pursue a particular pathway to a particular type of entrepreneurship because they anticipate that a particular combination of profit, social-benefit, innovation, and job-satisfaction outcomes (of their focal entrepreneurial opportunity) will maximise their personal wellbeing. It integrates the intensity of profit seeking, including profit-maximisation, profit-satisficing, and not-for-profit motivation, as a continuous variable within a single model, with the entrepre-

146

Entrepreneurial intention

neur’s choice of profit-seeking intensity being dependent upon their attitude to for-profit outcomes, moderated by their attitudes towards social benefits, innovation, and job satisfaction. The integrated model depicts the interdependent interaction of attitudes to profit, social benefit, innovation, and job satisfaction as joint and interdependent determinants of entrepreneurial intention, consistent with the holistic theory of individual decision making, rather than modelling them as independent and discrete determinants, as in linear-additive utility models (e.g. Douglas & Shepherd, 2000) and linear regression models (as in the great majority of the prior empirical literature). On that note, we are now ready for Chapter 9, where we depart from the linear-additive models of entrepreneurship intention and instead consider the determinants of entrepreneurship intention operating in interdependent conjunction, using fuzzy-set Qualitative Comparative Analysis to provide additional fine-grained information about the heterogeneity of entrepreneurial intention.

9. The holistic approach to entrepreneurship intentions 9.1 INTRODUCTION In preceding chapters we have considered “correlational” studies of entrepreneurship intention, whereby a sample of individuals is analysed using symmetrical analytical methods (multiple regression analysis and structural equation modelling) to test deductive hypotheses about the underlying antecedents of entrepreneurial intention. However, the symmetrical correlational approach has limitations for the study of individual entrepreneurial behaviour, because those methods do not align with the phenomena (i.e. individual decision making) that they seek to analyse. Whereas individuals make up their own minds “holistically” via a “within-person” introspective consideration of all the personal and situational factors that impact their decision to be made (Magnusson & Torestad, 1993), correlational methods analyse the “within-sample” relationships between the variables at the aggregate level and derive the sample-average relationships between the antecedent (independent) variables and the focal outcome (dependent) variable, to present a dominant net-effects prescription for the phenomenon which is unlikely to closely represent any individual in the sample.1 In this chapter we first consider the misfit of symmetric correlational methods with the holistic theory of individual decision making, before discussing “qualitative comparative analysis” (QCA), and the software tool fuzzy-set qualitative comparative analysis (fsQCA), which provide a different way to look at the data and thereby uncover fine-grained detail about behaviour at the individual level. After outlining how to use fsQCA, we re-analyse the data set introduced in Chapter 7 to demonstrate how this case-based method can extract additional fine-grained information from the same data set to enrich our understanding of the antecedents of entrepreneurship intention, in the contexts of

1 It is extremely unlikely that any individual’s coefficients (relating each antecedent variable to the dependent variable) would align closely, for every antecedent variable, with the sample average of those coefficients, unless the sample was homogeneous or nearly so, yet entrepreneurs are famously heterogeneous.

147

148

Entrepreneurial intention

profit and growth-oriented intention (GOI), independence-oriented intention (IOI), and social entrepreneurship intention (SEI).

9.2

THE LIMITATIONS OF SYMMETRICAL CORRELATIONAL METHODS

Symmetrical correlational methods, such as multiple regression and structural equation modelling, seek to find relationships among each of multiple independent variables (IVs) and the dependent variable (DV) that are statistically significant at the aggregate (sample) level of analysis. The relationships (regression coefficients) it finds for each IV are the sample average of the relationships existing between that discrete IV and the DV for all respondents in the sample, where “discrete” means that the impact of each IV is analysed in isolation of the impact of all other IVs, except for some two-way (and possibly also three-way) interaction variables that are the multiplicand of two (or three) IVs that may be argued to interactively moderate the relationships between discrete IVs and the DV. Even then, the regression coefficients to the moderator variables reflect the sample average, and thus represent the hypothetical sample-average respondent rather than any specific respondent. The individual relationships between the antecedent variables and the DV are subsumed in correlational analysis, being treated as deviations from the sample-mean relationship, and their diversity is reflected only in the significance (or probability) levels for each IV, and in the coefficient of determination (R2), both of which will be higher if the individuals are more homogeneous. But we know that entrepreneurs are heterogeneous (Gartner, 1985; Martin & Osberg, 2007; Shane & Venkataraman, 2000), so the presumption of a relatively small deviation from the mean (for each IV or moderator) is likely to be unrealistic. The entrepreneur is generally expected to be not an average person but a person who deviates from the herd with an innovative new product or process. So, the first issue is that the sample-wide perspective ignores the fact that the relationships between the variables (in the minds of individuals within the sample) for the great majority of respondents will in all probability not be closely approximated by the average relationship of all individuals in the sample (since probably no one will exhibit the sample-mean coefficients for every one of the IVs). Second, correlational methods require symmetry of data, such that the mean, median, and mode for each variable should be about the same value, or at least be closely similar. If there is a noticeable skew in the data, the researcher must transform the data in some way, for example using logarithms, to bring the data distribution closer to symmetry. Outliers, that is, a single or a few observations that are distant from the mean and next closest observation, may need to be deleted to achieve acceptable data symmetry. But many highly successful

The holistic approach to entrepreneurship intentions

149

entrepreneurs have been considered outliers, so it may be pertinent to retain outliers in our sample for further analysis, rather than to exclude them. Third, the single dominant net-effects expression represented by the regression equation shows which IVs are significantly related to the DV for most respondents. For example, two thirds of the sample may exhibit a positive relationship between a particular IV and the DV, and the remaining one third may exhibit a negative relationship between these variables, but the minority group remains invisible within the sample (except for reducing the significance level for the coefficient and the R2), effectively dominated by the relationship exhibited by a larger group within the sample. Fourth, correlational analytical methods expect symmetrical relationships between any IV and the DV – that is, the same direction of the relationship, either positive or negative, is assumed for all respondents. A positive relationship for some respondents and a negative relationship for others would reflect an asymmetrical relationship. But in entrepreneurship we might expect asymmetrical relationships – being married (for example) might be positively related to new venture survival in some cases (where the spouse helps out in the business, and/or earns a separate income), while in other cases marriage is negatively related to venture survival (where the spouse will not agree to remortgage the house to raise money, or demands a better work–family balance). The IV “married” would be a positive and significant determinant of venture survival if the former dominated; a negative and significant determinant if the latter dominated; or would be insignificant if these two opposing relationships effectively cancelled each other out (or if there was no impact of marriage on venture survival for the majority of the sample). Further, suppose there were three distinct subgroups of similar size within the sample, such that the IV’s relationship with the DV is positive, zero, and negative, in the three groups, respectively. In this case none would be dominant, and that IV would be found to be insignificantly related to the DV at the aggregate level, despite being potentially causal for each of the three groups within the sample. So correlational methods tend to either report the dominant group’s result (while ignoring minority groups), or report an insignificant result if there are two or more subgroups within the sample who effectively cancel each other out. Fifth, correlational methods treat the IVs as “discrete” variables (i.e. separate and independent of each other), and find the impact of each IV (on the DV) while holding constant the impact of all other IVs on the DV. But in many entrepreneurial situations the IVs may co-vary with each other. For example, the perceived desirability of entrepreneurship may depend positively on expected profit and negatively on expected risk of venture failure, but expected profit may co-vary with risk – i.e. profit is higher when risk is higher. In regression analysis the procedure is to add an interactive term (i.e. profit × risk) as an additional (moderating) variable in the regression model, and check for the

150

Entrepreneurial intention

statistical significance of that new variable. There is no problem with that, as long as two rules are not broken. First, the correlation coefficient (r) between the moderating variable and either of the “parent” variables must not exceed 0.7 (as a rule of thumb) to avoid what is deemed to be excessive collinearity of the variables which would render the regression results unreliable. Also, after adding the moderating variable, there must remain sufficient degrees of freedom (i.e. sufficiently more respondents in the sample than there are variables in the model) to avoid unreliable results. Correlational methods can handle two-way interactions (such as the above) and three-way interactions, as long as these two rules are respected, although higher-way interactions become very hard to hypothesise with supportive theory, as is required by deductive analysis. In entrepreneurship we suspect that there may often be multiway interactions among antecedent variables (e.g. age, married, education, work experience, entrepreneurial self-efficacy (ESE), and risk tolerance) when they jointly culminate in an entrepreneurship decision (such as to form entrepreneurship intention). Thus there is the sense of “forcing a square peg into a round hole” when using sample-level correlational methods to examine individual-level decision making. This is not to say that we have not learned a great deal about entrepreneurial behaviour, and developed robust entrepreneurship theory, thanks to multiple regression analysis and structural equation modelling. But by looking at the data from a different perspective, at the case level rather than the sample level, we can discover fine-grained detail and develop a more nuanced theory of entrepreneurial behaviour that will further build entrepreneurship theory.

9.3

CASE-BASED ANALYSIS OF ENTREPRENEURSHIP INTENTION

Case-based qualitative analysis is an inductive method of investigating phenomena at the individual (person, firm, or country) level. It is most useful when there is little knowledge about the precise causes of the observed phenomenon of interest, and involves detailed observation and/or interrogation of the decision maker(s) to isolate the antecedent variables of the focal phenomenon, and the relationships between these variables, that apparently operate to cause the phenomenon. A single case study may require several hours of interviewing the subjects involved, followed by transcribing the recorded interview, analysing the transcript, coding the responses, and so on (see Yin, 2006). Since one case cannot be presumed to be the same as all others, multiple cases are usually studied in an attempt to find the point of “saturation” – that is, the point where no new information about the antecedents and the relationships between variables is gained by conducting any further case studies (Fusch & Ness, 2015). Yet “saturation” presumes that there exists a dominant explanatory model

The holistic approach to entrepreneurship intentions

151

within the population that can be discovered by repeated single-case analyses, and implicitly that the cases exhibit random variations around the population (or sample) means between each of the IVs and the DV, focusing on the similarities across the cases rather than distinctive differences between the cases, when the latter is often the purpose of our investigation. Happily, Ragin (1987, 2000) introduced QCA – a hybrid of quantitative and qualitative methods that investigates the relationships among the antecedent variables (known as “conditions”) at the individual-case level, and aggregates similar cases into subgroups within the larger sample that exhibit the same or a highly similar combination of conditions (known as a “configuration”). This allows a focus on subgroups (individuals with the same configurations) that are more empirically relevant (i.e. have greater “coverage”) and which more reliably exhibit the same relationships between the set of conditions and the focal outcome variable (i.e. have greater “consistency”). 9.3.1

Fuzzy-set qualitative comparative analysis

FsQCA software, made freely available by Ragin (2008), is now used increasingly in the management disciplines (see Kraus, Ribeiro-Soriano, & Schüssler, 2018; Misangyi et al., 2017) and has also been adopted by entrepreneurship researchers (e.g. Beynon, Jones, & Pickernell, 2016; Devece, Peris-Ortiz, & Rueda-Armengot, 2016; Douglas & Prentice, 2019; Muñoz & Dimov, 2015; Muñoz & Kibler, 2016). In Douglas et al. (2019) we provide a comprehensive research agenda for the use of fsQCA to develop new theory in the entrepreneurship domain. The fsQCA method is based on the three main principles of conjunction, equifinality, and causal asymmetry (Misangyi et al., 2017). Conjunction means that the conditions within a configuration operate in conjunction with each other, that is, interdependently rather than discretely, to cause the outcome of interest.2 Equifinality means that there may be more than one configuration that results in the same outcome, and thus there may be multiple causal pathways to the same phenomenon. Causal asymmetry means that configurations that culminate in a given outcome might contain the presence of a particular

2 Regression models allow for moderating variables, usually limited to two-way or three-way interactions, of course, but the number of interactions is limited by degrees of freedom and multicollinearity considerations. More importantly, the coefficients to these interactive terms will reflect the sample-average relationship, rather than identifying the within-person relationships for specific individuals.

152

Entrepreneurial intention

condition (for some individuals), while another configuration may contain the absence (i.e. the negation) of that condition (for other individuals).3 Hence fsQCA software can identify multiple pathways to the same focal outcome (i.e. multiple different configurations, also known as “recipes”). When the level of analysis is the person (as in entrepreneurial intention) these pathways or recipes can be best understood as “personas” – the way a person thinks and acts – and depends upon their within-person (introspective) analysis of the personal and environmental issues (i.e. the conditions) that influence their thinking and behaviour at the individual level. The software also conducts counterfactual analysis to accommodate possible asymmetric relationships – thus it indicates whether the “absence” (i.e. negation) of a condition is fundamental to any configuration (i.e. which conditions if any within a particular configuration militate against the formation of entrepreneurship intentions). Further, fsQCA is based on set-theoretic relationships within configurations, and thus examines the interdependencies of all conditions, rather than the discrete impact of individual conditions and a limited number of interactions between those conditions. Moreover, fsQCA does not require data symmetry, relationship symmetry, degrees of freedom, large samples, deletion of outliers, and the avoidance of multicollinearity and heteroscedasticity, all of which constrain symmetrical correlational methods. In summary, fsQCA is an inductive method that reveals which antecedent conditions, in what configurations, are consistently associated with the outcome of interest. As a result it reveals empirical evidence of multiple pathways to the same outcome (if they exist in the data), asymmetrical relationships within the data (if they exist), and counterfactual information, which may be of great interest to entrepreneurship researchers. By providing evidence of individual-level relationships among the variables, fsQCA is complementary to correlational analytical methods (which reveal general relationships at the sample level of analysis), and provides an empirical basis for further theory building. 9.3.2

Using fuzzy-set qualitative comparative analysis

FsQCA is an inductive analytical method which can be used iteratively (Greckhamer, Furlani, Fiss, & Aguilera, 2018) to reveal whether or not 3 The word “causal” is used liberally in fsQCA without asserting that the relationship has been hypothesised and tested empirically for bona fide causality. FsQCA reveals configurations that are consistently associated with the outcome without explaining how or why they interact with each other. Hypotheses need to be proposed and tested subsequently to establish true causality. Thus the word “causal” should be read as “apparently causal”.

The holistic approach to entrepreneurship intentions

153

particular conditions operate in conjunction with others to culminate in the focal outcome for different groups of individuals within a wider sample. “Iteratively” means that the researcher might add or subtract conditions to/ from the configural model to improve the consistency and coverage of the model. But fsQCA should not be used as a “fishing expedition” (Gelman & Loken, 2014) – each condition entered into the model by the researcher should be justified on the basis of extant theory, new theoretical propositions, or at least prior empirical evidence from a different sample (Greckhamer et al., 2018). The purpose of fsQCA is to maximise the coverage and consistency of the solution, while minimising the complexity of the solution (i.e. the number and complexity of configurations that culminate in the focal outcome). There are 2k possible combinations of conditions, where k is the number of conditions included in the model – so, for example, if there are nine conditions, there are 29 = 512 possible configurations. Therefore the researcher must be parsimonious in adding additional conditions and in setting the thresholds for empirical relevance and consistency (discussed below), to achieve the above-mentioned coverage, consistency, and complexity goals. Accordingly, to use fsQCA the first task is to identify the outcome of interest and the conditions that appear to have a theoretical basis for inclusion in the configural model. In the context of this book, the focal outcome is the individual’s decision to form entrepreneurial intention. The extant theory of entrepreneurial intention (supported by prior correlational studies) is that entrepreneurial intention depends on measures of perceived desirability and perceived feasibility, as well as socio-demographic variables such as sex, age, business education, parents self-employed, business experience, and prior self-employment experience (for surveys of prior research, see Liñán & Fayolle, 2015; Schlaegel & Keonig, 2014; Shane, 2003: ch.4). Constructs and measures The second task is to define the constructs clearly (Suddaby, 2010) and establish the measures to be used to represent those constructs. As indicated in earlier chapters, the outcome (entrepreneurship intention) is either stated generically or is qualified to relate clearly to a specific subtype of entrepreneurial intention, such as GOI, IOI, or SEI. Alternatively, researchers might be interested in other subgroups of entrepreneurs, such as mum-preneurs (Duberly & Carrigan, 2013), elder-preneurs (Kautonen et al., 2010; Watkins-Mathys, 2012), migrant-preneurs (Levie, 2007), refugee-preneurs (Wauters & Lambrecht, 2008), serial entrepreneurs (Westhead et al., 2005), and international entrepreneurs (Joardar & Wu, 2011). The third task is to select the measures to represent the constructs, taking care to avoid measurement error which would confound the results. To represent the perceived desirability of entrepreneurship scholars have used meas-

154

Entrepreneurial intention

ures of the individual’s attitudes to the outcomes of entrepreneurial behaviour (such as profit, social benefits, decision-making autonomy, work enjoyment, risk exposure, and hard work) in conjunction with their perception of the quantities of those outcomes that are expected to be provided by the focal entrepreneurial opportunity. To represent the perceived feasibility of entrepreneurship we have used a measure of ESE, which arises from an introspective self-assessment by the individual regarding their capability to successfully complete the tasks involved in the envisioned entrepreneurial opportunity. Personal characteristics, such as age and sex, should be technically accurate (if reported truthfully) but it is not clear what these variables are measuring in the context of entrepreneurship intention. Is age a convenient proxy for knowledge, good judgement, social responsibility, and other personal qualities that grow with age, or is age negatively related to risk taking, innovativeness, counterfactual thinking, and other issues that tend to be avoided as one ages? Sex is a physiological characteristic that divides humans into two main categories and is generally used as a proxy for socially conditioned gender traits (see Ahl, 2006; Bem, 1981; Gupta et al., 2009). Yet not all men and not all women conform to the traditional masculine and feminine stereotypes of behaviour and social interaction. Thus we should not expect the variables age and sex to be devoid of measurement error, but instead represent a general tendency of individuals to think and behave in a particular socially conditioned manner when observed in a population (or a sample from that population) in the context of an entrepreneurial phenomenon, without this relationship necessarily applying to any particular individual. The same is true for many other socio-demographic variables that might be used in correlational studies of entrepreneurship intentions, such as married, with children, work experience, parents self-employed, prior self-employment experience, and so on. Case-level analysis, such as fsQCA, examines the particular behaviour of particular individuals in a particular context, and may or may not find that particular individuals conform to the sample average for any antecedent variable. Calibration The raw data collected to measure the antecedent conditions and the outcome of interest must be “calibrated” for analysis using Boolean algebra. FsQCA is based on set theory, and the researcher must make initial judgements about what value of the raw data would make it highly likely that a respondent was in the set (e.g. of persons intending to behave entrepreneurially) and what value of the raw data would make it highly likely that a respondent is not in the set (i.e. not intending to behave entrepreneurially). For binary data, such as sex, married, or prior start-up experience, “fully in” would be calibrated as 1 and “fully out” would be calibrated as 0. For continuous data and multivalue sets (e.g. Likert scales) the researcher must select a “fully-in” score, a “fully-out”

The holistic approach to entrepreneurship intentions

155

score, and a point of maximum ambiguity, in the raw data, for each set (condition) included in the configural model. These calibration decisions must be based on the researcher’s theoretical and contextual knowledge, and/or reference to data distributions observed previously. For continuous data the point of maximum ambiguity is usually set to the median, to accommodate asymmetrical data. Similarly, for Likert-scale data (e.g. 1–7) the point of maximum ambiguity might be 4, the fully-in score might be 5, 6, or 7; and the fully-out score might be 1, 2, or 3, based on the researcher’s theoretical and contextual knowledge and judgement. The fully-in, maximal ambiguity, and fully-out points provide reference points for the fsQCA software to assign calibrated scores for the raw scores between the two thresholds and around the point of maximum ambiguity. Choice of these points must be done thoughtfully, as it directly affects the coverage and consistency of the configurations identified. As fsQCA is an iterative inductive method, different calibrations may be used to test for the robustness of the solution when these calibration points are varied. Minimum frequency, coverage, and consistency A configuration comprises a combination of some (rarely all) the conditions that are included in the configural model. This may include combinations of two, three, four, and so on up to k conditions, where k is the total number of conditions in the model. There are 2k possible configurations – e.g. for eight conditions there are 28 = 256 potential configurations. The software identifies which of these are not observed in the data, and those configurations are called “counterfactuals” or “logical remainders”. The researcher must set the minimum frequency to at least 1 to exclude the logical remainders, and for large samples may wish to set the minimum frequency at 2, 3, or higher to exclude configurations that are empirically trivial (i.e. observed only once or a few times in a large sample), but should ensure that at least 80 per cent of the observations are not excluded by this cut-off frequency and are retained in the subsequent analysis (Ragin, 2008). Related to frequency, “coverage” indicates the “empirical relevance or importance” (Ragin, 2008) of each configuration, and “solution coverage” indicates what proportion of the sample is explained collectively by all the configurations identified. Thus the coverage statistic for each configuration is analogous to the coefficient of determination (R2) in correlational analysis. A “configuration coverage” of 1 would indicate that the entire sample shares the same combination of conditions, and a “solution coverage” of 1 would indicate that all members of the sample are included in the (multiple) configurations reported by the software. Some configurations will be more heterogeneous than are others, and so the researcher must set a “consistency” threshold for the configurations to be

156

Entrepreneurial intention

reported by the fsQCA software. Consistency assesses the degree to which the combination of conditions consistently produces the focal outcome (across the individuals represented by a configuration), and is analogous to statistical significance in correlational analysis. Setting consistency at 1 would ensure that all respondents sharing a configuration have exactly the same relationships among the conditions and the outcome – i.e. they are homogeneous in respect of those conditions. This would be useful for some purposes, but the convention is to set the consistency threshold at 0.8 or higher (following Fiss, 2011), to ensure that individuals sharing the configuration are highly similar with each other in terms of the interdependencies between and among the variables, while exhibiting relatively minor variations. Higher consistency thresholds, such as 0.85 or 0.9, may be set when the model includes many conditions, in order to serve the fsQCA objective to maximise coverage and consistency while minimising complexity.4 Necessity and sufficiency analyses The fsQCA software can identify conditions that are necessary for the focal outcome to occur. Correlational studies in entrepreneurship may mislead readers to think that some IVs (conditions) are necessary for the DV (outcome) to occur, because that IV is statistically significant and may have a relatively large effect size in the regression equation. Examples may include the belief that a strong positive attitude to decision-making autonomy is necessary for the formation of entrepreneurship intentions, or that a strong pro-social attitude is necessary for the formation of SEI. (Neither is true, as we shall see.) The fsQCA software tests each condition to find to what degree that condition is consistently present when the focal outcome occurs, for all members of the sample. For a condition to be deemed a necessary condition it must exhibit consistency exceeding the 0.9 threshold (Schneider, 2018; Schneider & Wagemann, 2012) and have non-trivial coverage. Finally, sufficiency analysis identifies the combinations of conditions (i.e. configurations) that are sufficient for the outcome to occur, given the calibration, minimum frequency, and minimum consistency thresholds set by the researcher. The fsQCA software generates three solutions at different levels of complexity. The “complex” solution is the most complicated and does not 4 The “proportional reduction in inconsistency” (PRI) measure (Greckhamer, 2016) provides a stronger test of consistency. With fuzzy sets, the same condition may be a subset of the outcome and also a subset of the negation of the outcome, thus creating an empirical paradox. The PRI consistency measure eliminates the influence of the cases that exhibit such data asymmetry. Greckhamer et al. (2018) state that PRI scores below 0.5 indicate excessive inconsistency, while higher thresholds (0.65–0.70) are considered best practice.

The holistic approach to entrepreneurship intentions

157

take into account whether each configuration is, or is not, observed in the data. The unobserved configurations (logical remainders) are factored into the counterfactual analysis to arrive at the “intermediate” and the “parsimonious” solutions. The intermediate solution considers those logical remainders that represent “easy” counterfactuals – that is, those that are consistent with theoretical knowledge and prior empirical evidence (although not observed in the present sample). The parsimonious solution produces the most concise results by considering also the “difficult” counterfactuals – those that may be consistent with (prior) empirical observation but not with theoretical knowledge. The intermediate and parsimonious solutions allow the researcher to identify the “core” and the “peripheral” (also known as “supporting”) conditions. The core conditions are those “for which the evidence indicates a strong causal relationship with the outcome of interest. In contrast, peripheral conditions are those for which the evidence of a causal relationship with the outcome is weaker” (Fiss, 2011: 398). Core conditions will have higher consistency scores, while peripheral conditions will have lower consistency (but still above the consistency threshold set by the researcher). Core conditions remain part of the solution when all counterfactuals are included, and are thus part of both the intermediate and parsimonious solutions, while the peripheral conditions are eliminated in the parsimonious solution, appearing only in the intermediate solution. In some solutions, two or more configurations will share the same core conditions and differ only in their peripheral conditions, and these are called “neutral permutations” (Fiss, 2011), differing only in the lesser details.

9.4

FUZZY-SET ANALYSIS OF ENTREPRENEURIAL INTENTION DATA

To demonstrate the additional information that fsQCA can add to a correlational study of entrepreneurship intentions, we will use it to analyse the same data set analysed using regression analysis in Chapter 7. We shall see that fsQCA reveals patterns in the data that correlational approaches cannot reveal because correlational analysis takes place at the sample level, whereas fsQCA identifies patterns at the case level that are common across subgroups within the sample. FsQCA may also reveal that some of the antecedent variables (conditions) have an asymmetric relationship with entrepreneurial intention (i.e. a positive relationship for some individuals and a negative relationship for others). Thus fsQCA has the potential to demonstrate that within the sample there may be alternative and unconventional pathways to the sample outcome, thus providing an empirical foundation for new entrepreneurship theory building.

Entrepreneurial intention

158

Table 9.1

Regression results for growth, independence, and social intentions

Dependent variable =>

GOI

IOI

SEI

Age

0.029

−0.016

−0.006

Sex (male = 1)

−0.112

−0.039

−0.379**

Business degree

0.452**

−0.231

0.110

0.150

0.240

−0.112 0.815***

Prior self-employment Parents self-employed

−0.148

0.174

Married

−0.874***

0.023

−0.033

Children

−0.003

−0.208*

−0.012

Femininity

−0.018

−0.009

0.097

Masculinity

0.108

0.316***

−0.242**

UPW profit

0.089

−0.055

0.007

UPW psychic

−0.069

−0.005

0.072

UPW social

−0.133

−0.111

0.707***

ESE marketing

0.349**

0.415***

0.306** −0.059

ESE people

0.148

−0.208

ESE finance

−0.0050

−0.032

0.008

“R-sqd”

0.3237

0.3459

0.4800

Probability

0.0006

0.0001

0.0000

Notes: * 0.1; ** 0.05; *** 0.01; N = 82.

To set the base level to allow illustration of the additional information that can be provided by fsQCA, the multiple regression results found in Chapter 7 for the n = 82 sample of Thai MBA students are repeated here as Table 9.1. The statistically significant variables in these regression models highlight the dominant net effects, and deflect attention from the variables that are not statistically significant, despite the fact that, for example, some men do undertake social entrepreneurship, and being married is for some people a positive driver of entrepreneurial intention. FsQCA can find these counterexamples within the data set, if they exist. To apply fsQCA using this data, we must first specify the configural model. To align the fsQCA study with the regression study, you might expect that we should include the same antecedent conditions in our configural model as in our regression model. But the regression model included 15 IVs, including nine control variables, three perceived desirability variables, and three perceived feasibility variables, such that there are 215 = 32,768 possible combinations of the 15 conditions, which would generate an extremely complex solution with relatively low consistency and coverage. Given that the goal of fsQCA is to maximise consistency and coverage while

The holistic approach to entrepreneurship intentions

159

minimising complexity of the resulting configurations, it is necessary to specify a simpler model with fewer antecedent conditions. This will allow the fsQCA software to cope, and will raise the consistency and coverage levels of the configurations while substantially reducing the complexity of the solution. Among the personal conditions (control variables) it was decided to exclude age, sex, prior business degree, prior self-employment, and having children, for the following reasons. First, conditions should be included in the configural model only if their influence on the focal outcome can be clearly hypothesised based on theory and/or strong prior evidence of a causal relationship. For age, it is not clear within the relatively narrow range (excluding a few outliers) of this MBA student sample that age should materially influence these entrepreneurial intentions on a “consistent” individual basis. Sex was excluded because there is no compelling argument that physiological sex, per se, should be a determinant of entrepreneurial intention, and because the gender trait variables seem to have accounted for gender differences in the regression equations. Prior business degree was excluded because many entrepreneurial ventures are observed to be started following non-business prior education. Prior self-employment experience was excluded because it is likely to include measurement error, being subject to social-desirability bias, and moreover is a binary variable with no qualitative measurement involved. Having children was excluded because it has only weak arguments for supporting or hindering the formation of entrepreneurial intentions. Among the perceived desirability conditions, there is a reasonably strong theoretical argument for utility part-worth (UPW) profit influencing GOI; for UPW psychic influencing IOI, and for UPW social influencing SEI, but not for each of these attitudinal variables to influence the other two outcomes. Accordingly, two of the attitude variables were dropped from the configural model for each of the intention models. For the three perceived feasibility variables, ESE marketing was significant in all three regression models while the other two were not. Marketing self-efficacy is retained in all three configural models because it is fundamental to the recognition of an entrepreneurial opportunity. The other two were excluded because it can be argued that those with weak ESE people and/or weak ESE finance might reasonably expect to recruit partners with the specialist skills required and undertake entrepreneurship on a team basis rather than individually. Thus, for a relatively simple fsQCA analysis for the purposes of this chapter, the number of antecedent conditions in the configural model for each of the focal outcomes was reduced to six. It is of course possible to iterate these models to progressively more complex models, but a relatively simple configural model here will serve to illustrate what additional information fsQCA can add to the regression results. For baseline comparison purposes, the regression equations for this more limited model are shown in Table 9.2. By compar-

Entrepreneurial intention

160

Table 9.2

Parent

Reduced regression models for entrepreneurial intention Profit

Profit

Psychic

Psychic

Social

Social

intention

intention

intention

intention

intention

intention

GOIa

GOIb

IOIa

IOIb

SEIa

SEIb

−0.128

−0.210

0.184

0.165

0.743***

0.909***

self-employed Married

−0.846***

−0.861***

−0.223

−0.098

−0.451*

−0.105

Femininity

−0.092

−0.151

0.041

−0.032

0.209*

0.076

Masculinity

0.314***

0.298**

0.316***

0.238***

−0.093***

−0.260***

UPW profit

0.224*

UPW psych

0.101

UPW social

0.692***

ESE marketing “Rsqd” Change “Rsqd”

0.418** 0.1883

0.2643 0.0760

0.380*** 0.1654

0.2512 0.0858

0.345*** 0.2043

0.4325 0.2282

Notes: *** p