Toward a Conceptualization of Organizational Modularity

Citation preview

Toward a Conceptualization of Organizational Modularity Ayberk SOYER Istanbul Technical University, Turkey Umut ASAN Istanbul Technical University, Turkey Özgür Utkan ERİŞ Istanbul Technical University, Turkey

ABSTRACT Organizations are complex systems made up of many different elements interacting with each other. One particular form of organizational design, emerging from the need to deal with organizational complexity and environmental uncertainty, is modularity. There are several studies in the literature dealing with organizational modularity. However, neither a common definition nor a complete measurement model is proposed for this concept. What are the common properties of an organization characterized as highly modular? How are these properties related to organizational design parameters? None of these questions have been adequately addressed in the literature. Thus, this study aims to identify the common properties of an organization which can be considered to be highly modular, and to propose a theoretical conceptualization of organizational modularity based on a framework originally linking Baldwin & Clark’s visible design rules (2000) with Mintzberg’s organizational design parameters (1979) and provides a suggestion for operationalizing the concept of organizational modularity based on the proposed theoretical framework. Keywords: Organizational Structure, Design Rules, Carliss Y. Baldwin, Kim B. Clark, Organizational Design Parameters, Henry Mintzberg, Conceptualization, Operationalization

INTRODUCTION Organizations are complex systems made up of many different elements interacting with each other. Understanding how organizations cope with complexity and adapt to their changing environments has interested organizational theorists and practitioners for a long time. One prevalent way suggested to manage a complex system is to divide it into smaller parts by considering the joints as nature does (Baldwin & Clark, 2000; Alexander, 1964). Different structural configurations (see Mintzberg, 1979) are used in practice to divide the work in an organization and to coordinate its different parts. Some of these configurations helped organizations to manage complexity for quite a long time, at least to some extent. However, organizations of the 21st century need to be more advanced than they were in the late 20th century. Because of advancing technology, environmental uncertainty, and the need to more effectively and efficiently utilize resources, organizations are getting more complex than ever. Increase in size, the Please cite this paper as: Soyer, A., Asan, U. and Eriş, Ö.U., 2018. Toward a Conceptualization of Organizational Modularity, Eds. Doğru, Ç., Handbook of Research on Contemporary Approaches in Management and Organizational Strategy, IGI Global, Hershey, PA, pp. 355-382.

number of different types of activities, the number of units and employees, the diversity of products and services, and the technological interdependences produce greater complexity in organizations. Despite this, a considerable number of organizations still rely on hierarchical rigid structures regardless of their appropriateness for them. Over the last decades improvements in technology has increased the level of competition and allowed firms to enter new markets. For example, advancements in production technology have shorten organizational life cycles, creating pressure for both flexibility and efficiency (D. L. Nelson & Quick, 2013). Developments in information technology have enabled new ways of communicating, allowing people work across organizational boundaries (Cohen & Mankin, 1998). These developments, however, have let increase the complexity in organizations and created the need for increased integration and coordination. The ever-increasing complexity and dynamism of the business environment has also affected organizational processes and design. In an uncertain business environment, it is unlikely that a given organization can do everything that it needs to do excellently. They rather focus on those things they can do uniquely well and outsource those functions that are not directly linked to their competences. Organizations without a proper organizational design will hesitate to expand to new markets or to exploit attractive opportunities because of the complexity related constraints imposed by their structures. The costs and barriers to change the system make these organizations reluctant to take proactive steps to utilize the opportunities in the environment. They are likely to be exposed to higher administrative overheads to manage their processes properly and, depending on the level of complexity; this may become even unaffordable (Andrews & Boyne, 2014). In order to quickly adapt to changing customer requirements and technological advancements, and react to new opportunities in the market, organizations need to review and reconfigure their complex structures and processes. This requires the organizations to gain insight into the interdependencies and interactions among their units. Knowledge of units and their interactions, in turn, helps establishing a set of organizational design rules (c.f. Baldwin & Clark, 2000) that provide a means to deal with organizational complexity. One particular form of organizational design, emerging from the need to deal with organizational complexity and environmental uncertainty, is modularity. Modularity is a special form of design that decomposes a system into highly independent components by defining how these components will interact and thus governs the ways in which system variants can be configured (Sanchez & Mahoney, 1996; Sanchez, 1999; Asan, Polat, & Sanchez, 2008). Many major theories in biology, physiology, genetics, engineering, mathematics and management rely on modular design (Callebaut & RasskinGutman, 2005). The idea of interdependence within and independence across modules allows organizations to make an architectural change in a sub-system, which does not entail a change in another sub-system. This yields greater flexibility for organizations in both scope and scale (Schilling & Steensma, 2001). In addition, an organization will have the opportunity to outsource some of its functions and switch easily between different providers of these functions, which enables responding to different market needs more quickly. However, it is crucial to note that, it is not reasonable to characterize an organization as modular or not, because organizations are all modular to some extent. There are several studies in the literature dealing with organizational modularity. However, neither a common definition nor a complete measurement model is proposed for this concept. The definitions suggested in the literature are rather incomplete or consider only specific aspects of the concept. The current methods for measuring the level of organizational modularity rely on a variety of restrictive assumptions. Most of them attempt to evaluate inter-organizational modularity and ignore important aspects originally rooted in the organization. Moreover, in the context of organizational design, modularity is not an independent and self-contained organizational concept as handled in the literature, but rather consists of several basic ideas that are integral parts of other organizational design concepts. Thus, this paper aims to identify the common properties of an organization that can be considered to be highly modular, and to propose a theoretical conceptualization of organizational modularity based on a framework originally linking Baldwin & Clark’s visible design rules (Baldwin & Clark, 2000) with

Mintzberg’s organizational design parameters (Mintzberg, 1979). This theoretical framework underpins the model suggested for measuring organizational modularity.

THEORETICAL OVERVIEW Although, organizational modularity is a frequently mentioned concept in organization theory, neither a common definition nor a measurement model exists for this concept. In order to address this issue, a conceptual definition together with the common properties of organizational modularity will be introduced based on the relationship established between the organizational design parameters of Mintzberg (1979) and visible design rules of Baldwin and Clark (1997). To serve this purpose, this section provides a comprehensive literature review conducted on organizational modularity and some related concepts, such as, modularity, product modularity, mirroring hypothesis, and organization design.

Modularity One of the most common ways to manage complex systems is dividing them into smaller subsystems and components. In this context, modularity is one of the main approaches that can be used to reduce such interdependencies and manage complexity effectively. The history of modular design, which can be considered as an attempt to understand systems as integrations of partially independent interacting units, dates back to early Mesopotamian era. In that period, by using smaller and simpler stones (i.e., modular entities) the ancient Greeks and Egyptians managed to create more sophisticated (complex) structures, in contrast to the monolithic structure which represented the architectural style of the period and built using large and single pieces of stones (Jones, 2011). One of the first studies not directly using the term modularity but meaning it is the seminal work of Simon (1962). In his paper, Simon investigates the architecture of complexity in biological, physical and social systems and proposes a matrix approach to manage the interdependencies among the units of the systems. Thereafter, the modularity concept was investigated for designing the architecture of products by several researchers, especially in the computer industry (Baldwin & Clark, 2000). Before explaining the modularity concept, it would be useful to define the term module first. Baldwin and Clark (2000) defined a module as “a unit whose elements are highly connected among themselves and relatively weakly connected to the other units’ elements”. Based upon this definition, modularity can be described as a system property that is obtained by minimizing the interdependencies among modules and maximizing the interdependencies within modules, which provides opportunities to create new configurations by means of mixing and matching modules, without losing performance or functionality (Campagnolo & Camuffo, 2009; Baldwin & Clark, 2000). As a special form of design, modular design intentionally creates a high degree of independence or loose coupling between component designs by standardizing component interface specifications (Sanchez & Mahoney, 1996). Thereby, changes to be made in any component of a system will not affect the rest of the system, or at least will affect it at a low level. This feature greatly increases the manageability of the changes and innovations in a module. The implementation of this design approach, which is still valid today, has become more critical, especially in today’s conditions, where products and processes became even more complex. In this context, modularity is a design strategy aiming the effective development of complex products and processes. Modular systems are composed of components that can be designed independently, but function together as a whole (Baldwin & Clark, 1997). Baldwin and Clark (1997) argues that there exists three visible design rules for a modular system design: (i) architecture specifying the modules that will be part of the system and the functions of these modules, (ii) standards for testing the conformity of the modules to the design rules and for measuring module performances relative to the other modules, and (iii) interfaces that describe how the modules will interact (i.e., how they will fit together, connect, and communicate). Langlois (2002) referred to the set of these visible design rules as modularization. Rather

than a design goal that is achieved and stabilized, modularization is an evolutionary process, which advances gradually and can never be fully completed (MacDuffie, 2013). According to Zhang and Gao (2010), modularity is a special case of nearly decomposability, in which the interdependencies between modules are specified by specific design rules, and the degree of modularity is defined by the level of near decomposability -the extent to which inter-module dependencies are specified. Therefore, modularity can be defined as the degree to which a system's components can be separated and recombined (Schilling, 2000). Wolters (2002) summarized the features of a modular system as: 

distinctiveness of components



loose coupling between modules; tight coupling within modules



clarity of mapping between functions and components



standardization of interfaces



low levels of coordination (components are self-organizing; coordination embedded in the architecture)

Combining the above mentioned features, Wolters (2002) made a single definition about modularity as: “a system is modular when it consists of distinct (autonomous) components, which are loosely coupled with each other, with a clear relationship between each component and its function(s) and well-defined, standardized interfaces connecting the components, which require low levels of coordination”. The modularity level of a system decreases as one or more of these features do not exist. Although there are various methods that can be used to design a modular system, such as, axiomatic design (AD), functional modeling, design structure matrix (DSM), modular function deployment (MFD) and variant mode and effects analysis (VMEA) (Ostrosi, Stjepandić, Fukuda, & Kurth, 2014), this process focusing on the design of the architecture, interfaces and standards for a system, may require a quite considerable commitment in terms of time and effort. Additionally, compatibility of system components with each other and their total performance level are among the points of criticism relating to modular design. The high individual performance of the components of a system taken separately, does not necessarily imply that the system will systematically generate a certain level of overall performance (Zirpoli & Becker, 2011). Therefore, specifying the contribution of each individual component to the overall system performance and coupling the system components in a suitable manner that avoids compatibility issues and increases the overall performance level, is not an easy process (Levinthal, 1997).

Organizational Design Parameters The literature on organizational structuring focuses on the mechanisms that can be used by organizations to design their structure .The structure of an organization can be defined simply as the sum total of the ways in which it divides its labor into distinct tasks and then achieves coordination among them (Mintzberg, 1979). In his seminal work on organizational structures, Henry Mintzberg, one of the most prominent organization and management analysts, synthesized the empirical literature currently available on organizational structuring in a comprehensive manner and summarized the basic components of organizational structure into nine design parameters that can be grouped into four basic categories (1979) (see Table 1). The design parameters, in the context of organization theory, are the basic components of organization structure that influence how the organization functions.

Table 1. Design parameters of organizational design (Mintzberg, 1979) Group Design of positions

Design of superstructure

Design Parameter Job specialization Behavior formalization Training and indoctrination Unit grouping

Unit size

Design of lateral linkages

Planning and control systems Liaison devices

Design of decisionmaking system

Vertical decentralization

Horizontal decentralization

Related Concepts Basic division of labor Standardization of work content System of regulated flows Standardization of skills Direct supervision Administrative division of labor Systems of formal authority, regulated flows, informal communication, and work constellations Organigram System of informal communication Direct supervision Span of control Standardization of outputs System of regulated flows Mutual adjustment Systems of informal communication, work constellations, and ad hoc decision processes Administrative division of labor Systems of formal authority, regulated flows, work constellations, and ad hoc decision processes Administrative division of labor Systems of informal communication, work constellations, and ad hoc decision processes

As mentioned before, in this study it is aimed to provide a conceptual definition of the organizational modularity concept by relating organizational design parameters of Mintzberg (1979) and visible design rules of Baldwin and Clark (2000). For that reason, it would be helpful to provide a brief introduction to the fundamental properties of these organizational design parameters (Mintzberg, 1979): •

Job specialization: Concerns the number and breadth of tasks of a given position (horizontal job specialization) and the incumbent's control over these tasks (vertical job specialization). Horizontal job specialization separates work by function whereas vertical job specialization separates the performance of the work from the administration of it. Typically, unskilled jobs are highly specialized in both dimensions, whereas, professional jobs are specialized only horizontally.

•

Behavior formalization: Refers to the extent of formal specification and standardization of work processes through job descriptions, work flows, and rules. Formalization reduces the variability of behavior and hence increases the control over it. Organization structures that rely on formalization of behavior for coordination are referred as bureaucratic.

•

Training and indoctrination: Is the design parameter by which the required knowledge and skills, together with the norms that must be acquired, are standardized through formal instructional programs for each position. This parameter is associated with the internalization of the desired behaviors by all employees.

•

Unit grouping: Deals with the bases by which positions are clustered into units and units into more comprehensive units, until all are clustered together under the strategic apex. In general, positions in an organization can be grouped according to (i) the knowledge and skills of employees; (ii) work processes and functions; (iii) products and services; (iv) clients; and (v)

geographical regions in which the organization operates. By unit grouping, formal authority is established and the hierarchy of the organization is built. Therefore, it is the design parameter by which direct supervision is most importantly effected. •

Unit size: Refers to the number of positions (or subunits) that are grouped into a single unit. Unit size defines the span of control for the managers and also the shape of the organization. If standardization is used for coordination, the unit size increases; on the other hand, if mutual adjustment is used for coordination, the unit size decreases. In other words, the unit size has an impact on coordination mechanisms.

•

Planning and control systems: Are used to standardize the products produced and/or services rendered. There are two different kinds of planning and control systems: (i) action planning and (ii) performance control. Action planning aims to regulate specific actions by predetermining the outputs of specific decisions or actions, while performance control focuses on the regulation of overall performance by measuring the performance of the relevant decisions or actions.

•

The liaison devices: Are the mechanisms by which the organization encourages mutual adjustment within and between units. Liaison positions, task forces that are formed to deal with specific issues, standing committees that work on ongoing organizational issues, integrating managers and matrix structures are among the liaison devices that can be used to facilitate the coordination and communication.

•

Vertical decentralization: Refers to the dispersal of decision-making power down to the chain of line authority, from the strategic apex into the middle line.

•

Horizontal decentralization: Refers to the extent to which decision making power flows informally outside the chain of line authority (i.e., to the nonmanagers, such as, analysts, support staffers, and operators in the operating core). As a result, horizontal decentralization brings the rest of the structure, apart from the strategic apex and middle line, into the power system.

Organizational Modularity As the response of organizations to the changing environmental conditions will be to reorganize themselves by redesigning the relationships of organizational units to each other and to the external units, the need for a modularity approach specific to organizations becomes inevitable. Modular design is not limited to products and processes; it can also be applied to the whole organization to comply with the rapid changes in its environment. The importance of this holistic approach has further increased, as organizations mostly do not have the necessary resources and skills to keep pace with inevitable changes occurring in the business environment. Some of the studies in the literature focusing on organizational modularity or modular organizations investigate the relationship between product modularity and organizational modularity (Colfer & Baldwin, 2010) while some others investigate the meaning of organizational modularity (Hoetker, 2006). Most of the researchers in the organizational modularity field underline that either the conceptualization of the term or the clarification of distinctive properties of a modular organization (i.e., conceptual and operational dimensions) is lacking. Given this argument, it becomes crucial to clarify what organizational modularity is and what the different aspects and properties of organizational modularity are. It is widely accepted in the literature that the organizations characterized by high modularity, can be decomposed into a number of units, which may be mixed and matched in a variety of configurations, or in other words, can be reconfigured. Benassi (2009) indicates that there are two different perspectives regarding the organizational modularity concept. The first perspective takes modularity as a generic property of organizations (A. Nelson & Byers, 2005); while the second perspective considers modular organizations as a distinct organizational form, and argue that modular organizations differ from other types of organizations with respect to a number of characteristics. According to this perspective, some

rules, which are listed below, must be set to investigate the characteristics of modular organizations (Benassi, 2009): 

Level of analysis must be identified.



Distinctive features should be clarified.



Similarities and differences with pre-existing forms should be specified.



The specific advantages and disadvantages should be clarified.



The conditions that are necessary for modular forms to exist and their effects should be evaluated.

The modular approach aims to increase the reconfigurability of the organization as a system readily by substituting functional process variations into the process architecture (Sanchez, 1997). When modularity is discussed in terms of organizations, “autonomy in decision making (for each unit) together with low inter-unit dependencies” stand out as one of the most important design rules. Even if the units of an organization are designed independently with respect to this design rule, and hence have low level of interaction among each other, organizational hierarchy and decision mechanisms may not be suitable for this situation. In other words, such kind of design rules, which reveal the architecture of a system, do not ensure that units are completely independent from each other. Therefore, it is generally not possible to create completely independent modules regarding the autonomy of decision-making. Under these conditions, rather than being considered as a property of an organization that may or may not exist in the organization, modularity should be considered as a property that may exist at different levels. Furlan, Cabigiosu, and Camuffo (2013) stated that organizational modularity is related to the complexity, intensity and frequency of information sharing between organizations. According to another definition taking the inter-organizational relationships, organizational modularity is the degree of firms’ usage of organizational arrangements (e.g., mergers, cooperation or acquisition) with external business units to minimize overall production and transaction costs and increase value (Cheng, 2005). A. Nelson and Byers (2005), who dealt with organizational units and the relationships among these units, claimed that a modular organization is an organization in which administrative interdependence and hierarchical structures are minimized, while cross-unit awareness and bottom-up processes are maximized. Parallel to this definition, Ernst and Kamrad (2000), and Schilling and Steensma (2001) stated that organizations have a high degree of modularity when their units can be disaggregated and recombined into new configurations with little loss of functionality. The units of these organizations are relatively independent of one another and if they are compatible with the overall system architecture, they may be easily recombined with one another (Schilling & Steensma, 2001; Baldwin & Clark, 2000; Sanchez, 1995)). Similarly, Hoetker (2006) indicated that organizational modularity implies that various units in the organization are loosely coupled, can operate autonomously, and can be easily reconfigured. As organizations become modular, "loosely coupled" networks take the place of tightly integrated hierarchies. These loosely coupled organizational forms allow organizational units to be flexibly recombined into a variety of configurations (Schilling & Steensma, 2001; Orton & Weick, 1990). A. Nelson and Byers (2005) regarded organizational modularity as a property focusing on the structures of organizations (i.e., relationships within an organization), rather than inter-organizational relationships. In contrast to this view, Schilling and Steensma (2001) measured the use of modular organizational forms by considering the inter-organizational relationships, such as alliances, use of alternative work arrangements, and contract manufacturing. These loosely coupled organizational forms enable organizations to specialize in their high-return activities and to combine them with other organizations' activities in a variety of configurations. Benassi (2009) also investigated the modular organizations and presented some valuable findings about the properties of modular organizations. Some of the properties mentioned in this study include, high degree of autonomy in decision-making, flatter structures, and managing the relationships with external suppliers directly.

Campagnolo and Camuffo (2009) made a literature review on the concept of modularity and classified the use of modularity in management studies into three main streams: (i) product design modularity, (ii) production system modularity and (iii) organizational design modularity. They emphasized that the drivers of modularization have not been investigated in the studies dealing with modularity in organization design and literature is missing on how and when modular organizational structures should be used. Although, the concept of organizational modularity is frequently mentioned in the literature of organization theory, apart from product modularity, neither a common definition nor a measurement model exists for this concept. Current definitions in the literature are limited in scope, focusing only on certain aspects of modularity. Therefore, in order to address this gap in the literature, a conceptual definition of organizational modularity, based on both the visible design rules of Baldwin and Clark (1997) and organizational design parameters of Mintzberg (1979), is proposed as given: Organizational modularity implies a loosely coupled network of autonomously operating self-contained units, having low level of interaction but high level of awareness among each other through standard interfaces, which can be flexibly recombined into a variety of organizational configurations. According to the definition given above, high organizational modularity is observed if: 

the hierarchical organizational units are replaced with loosely coupled ones, which are surrounded by semi-permeable boundaries (flat structure) and can easily be reconfigured so as to adapt to the changing environmental conditions and market demands (reconfigurability);



the units of an organization can take their decisions independently without substantially affecting each other (autonomy) and maintain their own activities without necessitating other units’ activities (self-containedness);



creating a new unit or maintaining the operations of the existing ones is compatible with the system architecture and standards;



interfaces enabling the communication among units are clearly defined and standardized (low interaction level, awareness, output standards); and



organizations can easily adapt interactions and collaborations with other external business units.

According to the given conceptual definition and explanations the following common properties of organizational modularity are proposed (these properties will be explained in Section 3): Selfcontainedness, Reconfigurability, Low number of hierarchical level, Autonomy, Defined output standards, Awareness, and Low level of interaction. The theoretical framework given in Table 2 represents the relationship between Baldwin and Clark (1997)’s visible design rules and Mintzberg (1979)’s organizational design parameters, and how the proposed properties relate to these rules and parameters.

Table 2. Relationship between properties of organizational modularity, organizational design parameters, and modular design rules Baldwin and Clark (1997)’s modular design rules Architecture

Mintzberg (1979)’s organizational design parameters Unit Grouping Unit Size Vertical Decentralization

Standards Interfaces

Planning and Control Systems Liaison Devices

Properties of organizational modularity - Self-containedness - Reconfigurability - Low number of hierarchical level (Flat structure) - Autonomy (Independent decisions) - Defined output standards - Awareness - Low level of interaction

Mirroring Hypothesis As modularity is one of the most important features of product architecture (Ulrich & Eppinger, 1995), organizational modularity is among the most essential features of organizational structure. In some of the studies on modularity, it is emphasized that designing modular products requires some changes to be made in the structure and processes of the organizations. This relationship between product modularity and organizational modularity, has been tried to be explained by the mirroring hypothesis (Colfer & Baldwin, 2010; Hoetker, 2006). The fundamental principle underlying the mirroring hypothesis is that each module of a system is informationally self-sufficient, and hence can be designed independently from the rest of the system. This implies that even though independent individuals, teams or organizations can separately design different modules, these independent modules can all work together as a whole. Consequently, this one-to-one mapping from designers to modules indicates a structural correspondence between technical and organizational networks (Colfer & Baldwin, 2010; Baldwin & Clark, 2000). Even though, mirroring hypothesis predicts a correspondence between product modularity and organizational modularity, it does not impose a direction of causality. There are a limited number of theoretical and empirical studies in the literature questioning the validity of this assertion. Only a part of these studies concluded that there is a significant relationship between product modularity and organizational modularity (e.g., Baldwin & Clark, 2000; Takeishi, 2002; Tiwana, 2008; MacCormack, Baldwin, & Rusnak, 2012; Nepal, Monplaisir, & Famuyiwa, 2012), whereas some of them did not find any evidence for the hypothesis (Brusoni, Prencipe, & Pavitt, 2001; Brusoni & Prencipe, 2001; Hoetker, 2006; MacCormack, Rusnak, & Baldwin, 2006; Wang, Chen, & Li, 2008). Therefore, it can be concluded that there exists no consensus regarding the relationship between product modularity and organizational modularity. The studies theoretically examining the relationship between product modularity and organizational modularity (e.g., Sanchez, 1995; Sanchez & Mahoney, 1996; Baldwin & Clark, 2000) suggests that rather than exercising managerial authority, standardized component interfaces can be used to achieve embedded coordination. Based on this argument, integrated product designs will generally require development processes carried out in a tightly coupled organization structure that is coordinated by a managerial authority, while, modular product designs are typically carried out by loosely coupled organization structures. In an empirical study performed by Nepal et al. (2012), the relationship between product architecture strategies and supply chain configurations is investigated and it is concluded that switching from an integrated product architecture to a modular product architecture directly affects the supply chain configuration, which is a conclusion that supports the mirroring hypothesis. They also claimed that

modular product architectures are much more flexible than integrated product architectures, and often result in lower total supply chain costs. Similarly, MacCormack et al. (2012) explored this relationship in the software industry for five different types of applications (i.e., financial management, word processing, spreadsheet, operating system, and database applications) developed by both commercial software firms and open source software communities. In this study, they used the Design Structure Matrix (DSM) technique to compare the design of products developed by loosely coupled and tightly coupled organizations, and concluded that loosely coupled organizations tend to develop more modular products than tightly coupled organizations for all the cases they studied. Although, a direct relationship between product modularity and organizational modularity is reported in the studies mentioned above Hoetker (2006), Wang et al. (2008) and Benassi (2009), among others, stated that there is no direct relationship between these concepts. For example, Hoetker (2006) indicated that increased product modularity enhances reconfigurability of organizations, which is considered as one of the several properties of organizational modularity; but does not allow organizations to move activities out of their hierarchy, which is also considered as another property of organizational modularity. When the structures and processes of the organizations that design and manufacture modular products are examined, it can be seen that some of these organizations are not modular. Similarly, Press and Geipel (2010) concluded that modularity fosters the development of independent units (i.e., disintegration), but on the other side, benefits that can be obtained from modularity or disintegration, depend on the level of entry and exit for the relevant industry. Accordingly, Press and Geipel (2010) emphasized that there exists a conditional relationship between product modularity and organizational modularity. Especially for the industries that are associated with high levels of entry and exit, as new industry entrants commonly imitate the top performer, industry level product quality increases while differentiation decreases. In an another study, Takeishi (2002), indicated that for automobile companies, which perform regular projects that do not involve new technologies, architectural knowledge is more important rather than componentspecific knowledge, as such kind of knowledge is supposed to be provided by the suppliers. Conversely, component-specific knowledge is more important than architectural knowledge for the projects involving new technology. Eventually, Takeishi (2002) concluded that there exist a conditional relationship between product modularity organizational modularity, as Press and Geipel (2010) did. This result, indicating a conditional relationship, also had been observed in other studies that were performed in different industries (e.g., Prencipe, 1997; Chesbrough & Kusunoki, 2001) Brusoni and Prencipe (2001) conducted a study to explore aircraft engine manufacturers and indicated that these systems integrator firms outsource the detailed design and manufacturing processes to specialized suppliers, while they maintain in-house concept design and systems integration capabilities to coordinate the work. And consequently, they concluded that there is no direct relationship between product modularity and organizational modularity. MacCormack et al. (2006), compared the architectures of the Linux operating system and the Mozilla Web browser and have reached similar results with Brusoni and Prencipe (2001)’s study. Linux is an open source software that has numerous developers in many different locations all around the world. Despite the fact that Mozilla also is an open source software, there is a collocated team of developers responsible from the development process. Therefore, Linux is supposed to have higher organizational modularity when compared to Mozilla. This finding seems to support the hypothesis that assumes a direct relationship between product modularity and organizational modularity. In 1998, as a result of managerial choice, a group of Mozilla's core developers carried out a redesign effort with the aim of making Mozilla more modular. In their study, MacCormack et al. (2006) compared the redesigned version of Mozilla with a comparable version of Linux by using the metrics defined to compare the structures of different designs, and concluded that the redesign effort had succeeded in making Mozilla more modular than Linux. Consequently, they have ascertained that product modularity can be increased, though the organization structure remain unchanged, and managerial actions have a significant impact in adapting a design's structure. The results of the above mentioned studies investigating the relationship between product modularity and organizational modularity are summarized in Table 1. As it can be seen in Table 1, for the theoretical

studies in the literature there is a general tendency to assume a direct relationship between product modularity and organizational modularity. However, empirical investigations generally do not strongly support a causal relationship and state that the relationship is much more complex than the theoretical assumptions. Consequently, product modularity is not treated as a property or direct cause of organizational modularity; rather it is treated as a facilitator for organizational modularity. Table 3. Studies investigating the relationship between product modularity and organizational modularity Study Sanchez (1995) Baldwin and Clark (2000) Sanchez and Mahoney (1996) Benassi (2009) MacCormack et al. (2012) Nepal et al. (2012) Tiwana (2008) Takeishi (2002) Press and Geipel (2010) Brusoni et al. (2001) Brusoni and Prencipe (2001) Hoetker (2006) MacCormack et al. (2006) Wang et al. (2008)

Type of Study Theoretical Theoretical Theoretical Theoretical Empirical Empirical Empirical Empirical Empirical Empirical Empirical Empirical Empirical Empirical

Relationship Between Product Modularity & Org. Modularity Strong relationship Strong relationship Strong relationship No direct relationship Strong relationship Strong relationship Strong relationship Conditional relationship Conditional relationship No direct relationship No direct relationship No direct relationship No direct relationship No direct relationship

PROPERTIES OF ORGANIZATIONAL MODULARITY Hoetker (2006), in his study, has claimed that modular organizations do not organize their production around internal or long-term suppliers, since modularity reduces the requirement for internal suppliers and allows quickly switching to new suppliers. Although, long-term relationships with external suppliers increase the risk of losing knowledge about both individual components and the overall system, the effect of switching to new suppliers is expected to be relatively small for modular organizations. In this context, in order to measure the organizational modularity concept, Hoetker (2006) used two fundamental properties given below: 

moving from tight hierarchies to ‘loosely coupled networks’ (organization of production around external suppliers)



creating ‘reconfigurable organizations’ (ability to add new suppliers and drop existing suppliers)

Schilling and Steensma (2001), examined three properties of modular organizations (i.e., ‘contract manufacturing’, ‘alternative work arrangements’, and ‘alliance formation’), which are supposed to reduce the inertia that cause an organization to adapt more slowly to the changing environmental conditions, and increase flexibility. Some other properties of modular organizations, such as, autonomous business units and other network forms are not considered in their study. A. Nelson and Byers (2005), explored the intra-organizational relationships between the technology transfer and entrepreneurship education units at Stanford University, and they used ‘autonomy in decision making’ and ‘cross unit awareness’ properties to measure the organizational modularity of these units. Worren, Moore, and Cardona (2002), developed a conceptual model for investigating the relationship between modular product architectures, strategic

flexibility and firm performance, in which innovation orientation and organizational architecture concepts were taken as mediating variables. Modular structure and modular processes were identified as subdimensions of organizational architecture and in order to measure the modular structure property, items about the existence of ‘small, autonomous units’ and ‘rotation of personnel’ were used. Karim (2006) compared the reconfiguration of internally developed and acquired units of an organization and concluded that the acquired units are reconfigured sooner than the internally developed units are. The reason of this result can be the organizations’ lack of familiarity with acquired units’ resources and the motivation to generate greater benefits from acquisitions. Properties of organizational modularity concept identified in the theoretical and empirical studies published in the literature are summarized in Table 4. Units of modular organizations can communicate with each other and with the external units only through standardized interfaces within a standardized architecture; and when compared to the integrated organizational forms, they can communicate with other external units much more easily. The main issue that the organizational architecture deals with is the design of the relationships of organizational units with each other and external units. Therefore, relationships with external organizations are considered more important for the design of modular organizations, when compared to the design of traditional organizations. The reason for this is that modular organizations typically outsource their non-core activities. From this point of view, Benassi (2009)stated that modularity at an intra-organizational level is used mostly by analogy with inter-organizational modularity, and Schilling (2000) derived a model of interfirm product modularity. In his study, Wolters (2002) distinguished between intra- and interorganizational processes and claimed that they possess different features. According to Wolters (2002), processes refer to the technologies and procedures used to produce or deliver products or services within an organization while a supply chain refers to the choice and design of the network surrounding an organization. Therefore, processes describe the ways products are produced, while supply chains describe who does it and how they interact with each other. Both concepts are obviously interconnected, but separated as well. When the empirical and theoretical studies mentioned above are taken into consideration, it is clearly observed that modular organizations may have various properties. These properties considered in previous studies were used as a basis to identify the properties of organizational modularity given below: 

Self-containedness



Reconfigurability



Low number of hierarchical level (Flat structure)



Autonomy (Independent decisions)



Defined output standards



Awareness



Low level of interaction



Outsourcing capacity



Supplier switching capacity



Alternative work arrangement capacity



Alliance formation capacity

When these properties are examined in terms of their scope, it can be seen that the first seven properties (i.e., self-containedness, reconfigurability, low number of hierarchical level, autonomy, defined output standards, awareness, and low level of interaction) concern the interactions between organizational units,

while the last four properties (i.e., outsourcing capacity, supplier switching capacity, alternative work arrangement capacity, and alliance formation capacity) concern the interactions with external units. In the following sections, a brief introduction to each of these properties is given. Table 4. Properties of organizational modularity Study

Type of Study

Sanchez (1997)

Theoretical

Sanchez and Mahoney (1996)

Theoretical

Baldwin and Clark (2000)

Theoretical

Langlois (2002)

Theoretical

Hoetker (2006)

Empirical

A. Nelson and Byers (2005)

Empirical

Worren et al. (2002)

Empirical

Schilling and Steensma (2001)

Empirical

Cheng (2005)

Empirical

Helfat and Eisenhardt (2004)

Empirical

Karim (2006) Benassi (2009)

Empirical Empirical

Todorova and Durisin (2009)

Empirical

Properties of Organizational Modularity Self-managing Autonomy Reconfigurability Loosely coupling Embedded coordination Defined output Nearly decomposability Little managerial authority. Low interaction Low interdependency Decomposability Low interdependency Decomposability Reconfigurability Loosely coupled networks Autonomy Awareness Autonomy Standardization of work processes across organizational units Alliances Contract manufacturing Alternative work arrangements Alliances Contract manufacturing Alternative work arrangements Decentralization Reconfigurability Reconfigurability Loosely coupling Autonomous decisions Self-contained units Autonomous units Standardization of work processes across organizational units

Properties of Organizational Modularity Concerning the Interactions between Organizational Units This section provides an overview of the organizational modularity properties concerning the interactions between organizational units.

Self-containedness From the organization design perspective, Mintzberg (1979) mentioned about unit grouping based on workflow interdependencies, in which the whole workflow is grouped into a single unit, under a unit work-flow supervisor and refers to a study of Thompson (1967). According to Thompson (1967), the organizational units are formed according to reciprocal interdependencies, where the work flows between tasks. If there exists no reciprocal interdependency, sequential interdependencies are taken into consideration, and if neither of them exist, then organizational units are formed according to the shared resources. From the modularity perspective, Miller and Elgard (1998), who tried to clarify the module, modularity and modularization concepts, indicated that a unit could only be treated as a module when it contains selfcontained functionality. Accordingly, Baldwin and Clark (2000) stated that the modules that will be a part of a system and the functions of these modules are specified by the architecture, and they emphasized that the modularity is a design rule, which deals with intra- and inter-module dependencies. By definition, since a system is composed of a set of components that act and interact together, organizations, which are referred to as complex social systems, are composed of discrete business units in which there exists individuals performing functional tasks. Therefore, a distinctive property of a modular organization is consisting of units, which are self-contained. These units should maintain their own activities without necessitating other units’ activities.

Reconfigurability As Mintzberg (1979) stated, given a set of positions in an organization, to be able to design organizational structure, firstly, it should be decided on how to group these positions into units (i.e., unit grouping, which is one of the nine organizational design parameters). Unit grouping has an important impact on coordination, resource sharing, and performance measures. In this context, Karim (2006) defined reconfigurability as adding new units to the organization, deleting units from the organization, and recombining units within the organizations. Reconfiguration has the potential to create synergy, which in turn improves the productivity of each unit and enhances innovativeness. It is worth to mention here that reconfigurability term is used in a broader sense including also the recombination term. In his study, Hoetker (2006) mentioned some of the advantages of organizational modularity and one of them was ease of reconfigurability, which allows an organization to select the best supplier for a given component at a given time. Waard and Kramer (2008) investigated the deployment strategy of the Dutch Armed Forces and found that their organizational structure allows them to reconfigure 90 different functional components into a variety of constellations. They also claim that, modularly designed parent organizations (e.g., Dutch Armed Forces) can mobilize different kinds of temporary organizations through the reconfiguration of its own organizational parts. Therefore, as reconfiguration of organizational structure increases the flexibility of organizations, these organizations are better able to quickly respond to the changing market conditions.

Low number of hierarchical level Low number of hierarchical level, i.e., flat structure corresponds to a shorter chain of command from the strategic apex to the operating core. On the other hand, long chain of command corresponds to more managerial workload and more complexity and interaction. When organizational modularity is high, autonomous decisions take place in modular designed units. Autonomous decisions are related to the organization structure and characteristics of the tasks. Units are expressed as flat hierarchies in organizational modularity, and this flat structure does not only include low number of hierarchical levels but also includes participation of organizational units to the decision making process (Benassi, 2009). Mintzberg (1979) had stated two fundamental assumptions describing the effects of unit size to the organization structure: (i) increasing standardization for coordination will result in bigger working groups

(ii) more reciprocal relationships require organizations to have smaller unit size (due to interdependencies among complex tasks). In a modular designed organization, coordination activities are usually more standardized. Thus, interdependencies are grouped into modular designed units and decisions are taken independently. As unit size increases, the number of hierarchical levels decreases. In vertical hierarchies, as the number of employees reporting to each manager tends to be smaller, managers usually have the opportunity to supervise and monitor employees, and satisfy the security needs of them. However, in this type of organizations, employees, especially the ones who ask for autonomy, can feel that they are under tight control (Porter & Lawler, 1964). Ivancevich and Donnelly (1975) study of salespersons found that salespersons in flat organizations perceive more satisfaction with respect to self-actualization and autonomy, and lower amounts of anxiety-stress when compared to those working in tall organizations. Thus, autonomous decisions, which is a fundamental aspect of organizational modularity, is directly related to the hierarchical structure of the organizations. In general, decisions that are more autonomous are expected to be taken in less hierarchical structures.

Autonomy (Independent decisions) Self-managed teams are responsible for their internal activities, can set their priorities and have decisionmaking power over their own functional areas (Vaccaro, Jansen, Van Den Bosch, & Volberda, 2012). Mintzberg (1979) defined vertical decentralization as delegating decision-making power to lower level staff down the chain of authority. Parallel decentralization in the vertical dimension eliminates the decision interdependencies especially for the units that are grouped based on market or industry. Each organizational unit is decoupled from the others and delegated with the decision-making power for those decisions affecting its own products, services, or geographical areas. Therefore, parallel vertical decentralization is the only way to delegate market-based organizational units the power that is needed to act freely (i.e., to function in a quasi-autonomous manner). Benassi (2009) stated that modular organizations are based upon small organizational units, which (i) have a specific domain; (ii) correspond to a specific product, a sub-system, a component or a part of a process; (iii) can make decisions on various business issues; (iv) deal directly with suppliers; (v) are characterized by a flat structure, and (vi) can make autonomous decisions to accomplish their tasks. Similarly, A. Nelson and Byers (2005) indicated that if units of an organization retain autonomy in decision-making and are not dependent on other units to complete their goals, they are said to be modular. N. Argyres and Bigelow (2011) investigated the relationship between modularity and vertical deintegration with reference to the mirroring hypothesis, which claims that organizational design tends to reflect product design; and consequently, they claimed that modularity has a positive and significant impact on vertical disintegration. Jacobides (2005) stated that new generation CEOs do not believe in integration as they think decentralized organizations run more efficiently than centralized ones, and they emphasized that decentralization within organizations paves the way for disintegration.

Defined output standards If standardization and customization are considered to be on the opposite ends of a continuum, customized standardization (i.e., modularization) can be placed in the middle of the continuum (Lampel & Mintzberg, 1996). Miller and Elgard (1998) described modularity as a means for balancing standardization and customization. In this respect, modularity implies the customization of the standardized modules, and necessitates that the resulting customized modules conform to the visible information behind the organizational design, namely, the architecture, interfaces and defined output standards. Therefore, output standards, which can be applied in order to reduce variety and facilitate compatibility between organizational units and/or across organizations, is a fundamental property for enabling modularity (Johansson, 2013). In modular organizations, each organizational unit delivers its outputs to other organizational units by means of interfaces, which describe how the modules will

interact. Therefore, defining the output standards ensures that the outputs are produced in accordance with the predetermined standards and expected performance levels. Organization design literature suggests two ways for organizations to control the output: (i) action planning, which is used to determine the required decisions or actions sooner, and (ii) performance control, which is used to measure the results of the actions taken and to make necessary changes (Mintzberg, 1979). In this context, action planning and performance control should be regarded as a means of standardizing outputs.

Awareness In some cases, direct supervision and three forms of standardization (i.e., standardization of work processes, outputs and skills) remain insufficient to satisfy the coordination requirement of organizations (Mintzberg, 1979). Wolters (2002) stated that having an architecture that provides embedded coordination reduce the level of coordination required, and can be treated as one of the properties of modular organizations. As interactions among organizational units increase, complexity of the work processes performed in these units also increase, and consequently, out goes the need for a managerial coordination and an organizational structure incorporating an authority hierarchy. However, for modular organizations, which have standardized interfaces between its units, the need for managerial authority and authority hierarchy decreases, and these organizations can be guided by embedded coordination (Sanchez, 1995). It was observed in most of the studies in the literature that even though the interdependencies among the modules are tried to be eliminated in modular design, typically, it is not possible to eliminate them completely. Therefore, instead of trying to eliminate these interdependencies, organizations must concentrate on the management of them as they emerge (Staudenmayer, Tripsas, & Tucci, 2005). One of the points criticized in the literature about organizational modularity is that, it enables organizational units to advance in their own operations, and consequently, the knowledge of each organizational unit is limited to their own work processes. In connection with this, Hahn and Turowski (2005) indicated that ‘information hiding (among the units of an organization and across organizations)’ is a characteristic of modular organizations, which, potentially, limits the awareness of modules. Therefore, modular design is regarded as particularly suitable for incremental innovations rather than for radical innovations. In this respect, Brusoni and Prencipe (2006) stated that, the changes in unit configurations should be shared among all organizational units in a coordinated and timely manner, since these changes are mostly associated with new connections both within and across organizational units. In other words, awareness of modules should be increased to make way for radical change processes to be designed. Similarly, Baldwin and Henkel (2012) also indicated that sharing general architecture knowledge and operational knowledge is a requirement for innovation. Therefore, to manage intra- and inter-module dependencies, each module in a system should communicate directly to each other by means of interfaces, and should be aware of the changes occurring in other modules, which are related to the operations of the relevant module and general architecture of the system. In general, organizations have a tendency to use internal units instead of external suppliers, especially in case of large measurement costs (i.e., the agency costs incurred when something is purchased from the market). For that reason, organizations more often prefer hierarchical organizations to contracting with (semi-) autonomous external agencies in order to overcome inherent transaction costs and bargaining problems, and to have control over all activities. (N. S. Argyres, 1999). Developing a mechanism for creating and/or increasing awareness of modules, provides easier access to information contained each organizational unit, helps to decrease the measurement costs, and remove the barriers complicating outsourcing arrangements. In their study, Colfer and Baldwin (2010) stated that there exists three different levels of transparency that collaborating units can achieve: (i) material transparency denoting the disclosure of information, (ii) conceptual transparency by which all units access information and make sense of it, and (iii) actionable transparency by which all units access information, make sense of it, and can act on the design itself. All kinds of transparency is important with respect to organizational modularity, as they increase the

awareness of each unit. Although, all these arguments seem to conflict with some of the fundamental requirements of modularity, as A. Nelson and Byers (2005) stated, a synergy between organizational units must be developed so that each unit will be aware of other units.

Low level of interaction Ethiraj and Levinthal (2004) stated that organizational units often depend on each other for inputs, what Thompson (1967) termed reciprocal dependence. As mentioned in previous sections, one of the ultimate aims of modular design is minimizing the inter-module dependencies, which could be possible only when a simplified interface is designed in between modules. For this purpose, tools, such as, design and task structure matrices, could be used to identify and eliminate the interactions among modules (Baldwin & Clark, 2000). Modularity increases the range of manageable complexity by limiting the scope of interaction between organizational units, and thereby reduces the effort and time spent on design or production processes (Baldwin & Clark, 2000). The degree of loosely coupling, which is among the properties of organizational modularity, is a function of the strength of inter-unit interactions. If there exists no interactions between organizational units, then the organization is said to be fully decoupled, or modular. In this context, Simon (1962) underlined that, it is important to highlight who interacts with whom, rather than, who locates closer to whom, for defining social hierarchies. Regarding the level of interaction between organizational units, Mintzberg (1979) suggested that liaison devices could be used, especially when, work processes are highly complex, horizontally specialized and highly interdependent. However, there are also some other mechanisms that would allow handling such kind of situations (e.g., increasing awareness of modules).

Properties of organizational modularity concerning the interactions with external units This section provides an overview of the organizational modularity properties concerning the interactions with external units.

Outsourcing capacity To be able to outsource some activities or units, there should be “thin crossing points” between these activities/units and the remaining ones in the organization (Baldwin & Clark, 2006; Hoetker, 2006). When crossing points become thicker, the level of vulnerability of the remaining activities/units increases. In order to replace the regulation function of hierarchies for internal units with outsourcing contracts, contractual governance will be useful where transfers in crossing points can be identified, measured and compensated (Blair, O’Hara O’Connor, & Kirchhoefer, 2011). Here, it is assumed that a more modular organization will have a higher outsourcing capacity. Notice, however, that an organization with high level of outsourcing could be less modular than an organization with more internal suppliers. In some cases, organizations that use outsourcing are not capable of changing their supplier because of their non-modular design. If organizations could not change their decisions according to the new conditions, then it is not possible to mention about the options they have. Therefore, it is important to measure the capacity of outsourcing instead of actual outsourcing. This is an important assumption underlying the role of outsourcing in organizational modularity. Capacity can tell us whether it is possible to take an action or to give up when it is necessary. Thus, the capacities should be considered in order to understand whether an organization is flexible to adapt to the changing environmental conditions and able to change the decisions when it is required. Note that the same idea applies to supplier switching, alternative work arrangements and alliance formation.

Despite the advantages of outsourcing, organizations should ensure that organizations are still holding the intra-organizational control power and inter-organizational control power (transaction cost power, quality control, technology control, etc.) (Wu & Park, 2009). Moreover, to utilize outsourcing efficiently, it should be considered that external units (e.g., suppliers) are capable to adapt this kind of work style. They should have the abilities to take the ownership of the corresponding module’s activities and deliver its expected output with the determined specifications on time. The design and management capacity of the external unit should be carefully considered.

Supplier switching capacity This property can serve as a measure of the effect of past relationships with the suppliers for a new deal. Relationship with the suppliers can be managed with contracts if the activities, which is performed by the supplier, is clear. In this case, the firm can switch between contract suppliers that perform different activities, and the contract supplier can similarly work for different firms. Thus, as firms in a given industry begin resourcing organizational units that lie outside of organization boundaries for activities that were once conducted in-house, the system becomes increasingly modular, and this particular type of organizational design helps decreasing the number of interactions within the organization and between organizations, and thereby, decreases the importance of past relationship with the suppliers. However, most of the long-term relationships with the suppliers are largely uncertain and open-ended. In this case, trust becomes the organizing principle for managing the complex relationships (Sinha & Van de Ven, 2005). Relying on trust, however, to manage the relationships with a supplier is risky. It complicates both changing the supplier for an existing activity and eliminates the objectivity in order to evaluate the alternative suppliers for new deals. In addition, a supplier with a long-term relationship gains the communication advantage. A common language is developed in order to discuss the technical details. This type of language is especially important for implicit knowledge transfer. However, increased standardization and less need for communication will decrease the importance of implicit knowledge transfer and using common language. Thus, past relationships with suppliers is expected to be less important for higher level of organizational modularity (Hoetker, 2006).

Alternative work arrangement capacity Alternative work arrangements, i.e., recruiting contract based or temporary employees in organizations, allows organizations to increase or decrease their scope and capacity in different situations (Schilling & Steensma, 2001). Using temporary employees is an advantage, in terms of cost and flexibility. However, employee loyalty and hands on experience is low when compared to the long-term employees (Sinha & Van de Ven, 2005). It is important to define the processes and skills required for the tasks and standardize the activities in order to utilize alternative work arrangements. Capacity for alternative work arrangements is an important indicator of organizational modularity, since it helps understanding the capability of responding to demand changes in the market. Alternative work arrangement is possible when organizations convert the implicit knowledge to explicit nature. In this way, it is possible to split the information into pieces regarding the proprietary information and to assign the tasks to the contract-based employee with the limited proprietary information without sharing all details. It also facilitates finding and hiring process of alternative labor force.

Alliance formation capacity Organizations can use alliances to access to the critical capabilities that they do not have and/or they cannot create quickly. Usually, these critical capabilities have the potential to provide flexibility to adapt to the changing environmental conditions (Schilling & Steensma, 2001). Managers are engaged in activities related to alliance formation, such as, regulating the resources and knowledge transfer between

units, specifying required organizational arrangements (Mingo, 2013). Spending less effort of managers and less resource for these activities are related to the organizational design. It is expected that more modular organizations are easier to adapt alliance formations. Tiwana (2008) researched the effects of modularity, process control and output control on the alliance performance, and found a positive relationship between organizational modularity and outsource performance. Alliance formation capacity is an important indicator for an organization whether its processes and units are designed in modular fashion or not. Easily and quickly forming an alliance with other parties is critical to respond to the new opportunities in the environment; especially, when internal reconfiguration is not economically feasible and/or not capable of satisfying all the necessary requirements.

A Suggestion for Operationalizing the Concept of Organizational Modularity Studies in the literature are highly limited and differ considerably in the dimensions they consider to operationalize the organizational modularity concept, as it can be concluded from Table 5. There is no study critically examining these differences and suggesting a complete measurement model. In this respect, this study provides a valuable contribution for the first steps of the operationalization of the concept of organizational modularity. Table 5. Dimensions used for measuring organizational modularity Study Schilling (2000); Schilling and Steensma (2001) Worren et al. (2002) A. Nelson and Byers (2005) Karim (2006) Hoetker (2006) Todorova and Durisin (2009) Furlan et al. (2013)

Dimensions > Contract manufacturing > Alternative work arrangements > Alliances > Modular structure > Modular processes > Cross-unit dependence > Awareness > Reconfigurability > Preference for Internal suppliers > Prior transactions with suppliers > Modularity of organizational structure > Modularity of organizational processes Buyer–supplier information sharing related to: > new product development > contract and price negotiation > logistics

Operationalization of any concept first requires conceptualizing the concept by identifying its meaning in theoretical terms, which has been done for the concept of organizational modularity in Section 2.3. This definition provides the means for measuring organizational modularity. The common properties described in the preceding subsections help understanding the different conceptual dimensions of organizational modularity. Typically, these dimensions are then connected to observations by means of operational indicators, but this is beyond the scope of this paper. Indicators allow observing or measuring the value of the determined dimensions and, thereby, the value of the concept. As a further research, the suggested model can be empirically tested by employing a confirmatory factor analysis. The suggested model is depicted in Figure 1.

Figure 1. Organizational modularity measurement model

CONCLUSION Although, organizational modularity concept is frequently mentioned in the literature of organization theory, neither a common definition nor a measurement model exists for this concept. Nevertheless, according to the results of the literature review conducted in this study, it is observed that there exist two main streams of thought considering the definitions and characteristics of organizational modularity. One stream extends the definition of product modularity to organizations, while the other stream takes interorganizational relationships as a basis. The theoretical framework introduced in this study to conceptualize organizational modularity grounds on the internal characteristics of an organization, however, takes also the characteristics related to inter-organizational relationships into account. To be more precise, the organizational modularity concept is tried to be described with respect to the link developed between visible design rules and organizational design parameters, while regarding the relationships across organizations. Thus, originally, and differently from the current literature, this study conceptualizes the organizational modularity concept within the frame of organization theory, and reveals the common properties of modular organizations. The suggested theoretical framework and common properties of organizational modularity underpin the model developed for measuring organizational modularity. It is worth to mention that, by definition, all organizations are open systems that exchange material, energy, people, information, etc., with its environment in order to survive. In this respect, since organizations differ from products, the concept of organizational modularity should incorporate interorganizational relationships. Therefore, simply extending the definition of product modularity to organizations with respect to the mirroring hypothesis is not sufficient. Moreover, recent studies have shown that the relationship between product design and organizational structure depends on the external conditions and factors.

FUTURE RESEARCH DIRECTIONS This study can be considered as a theoretical attempt to conceptualize organizational modularity. Therefore, as a future study, properties of organizational modularity can be operationalized by identifying their relevant indicators, and thereafter the suggested model can be empirically tested. This would allow examining the content of organizational modularity empirically, and revealing the properties, which are statistically significant. Furthermore, the measurement model would provide the opportunity to test the mirroring hypothesis, and thereby, the relationship between product modularity and organizational modularity. Finally, the measurement model could be used to investigate the effect of organizational modularity on financial and/or non-financial performance.

ACKNOWLEDGMENT This work was supported by Research Fund of the Istanbul Technical University (Project Number:37966)

REFERENCES Alexander, C. (1964). Notes on the synthesis of form. Cambridge, MA: Harvard University Press. Andrews, R., & Boyne, G. A. (2014). Task complexity, organization size, and administrative intensity: The case of UK universities. Public Administration, 92(3), 656-672. Argyres, N., & Bigelow, L. (2011). Innovation, modularity, and vertical deintegration: Evidence from the early U.S. auto industry. Organization Science, 21(4), 842-853. Argyres, N. S. (1999). The impact of information technology on coordination: Evidence from the B-2 “stealth” bomber. Organization Science, 10(2), 162-180.

Asan, U., Polat, S., & Sanchez, R. (2008). Scenario-driven modular design in managing market uncertainty. International Journal of Technology Management, 42(4), 459-487. Baldwin, C. Y., & Clark, K. B. (1997). Managing in an age of modularity. Harvard Business Review, 75(5), 84-93. Baldwin, C. Y., & Clark, K. B. (2000). Design rules: The power of modularity. Cambridge, Massachusetts: The MIT Press. Baldwin, C. Y., & Clark, K. B. (2006). Architectural innovation and dynamic competition: The smaller “footprint” strategy. Retrieved from Boston, Massachusetts: Baldwin, C. Y., & Henkel, J. (2012). The impact of modularity on intellectual property and value appropriation. Retrieved from Boston, Massachusetts: Benassi, M. (2009). Investigating modular organizations. Journal of Management & Governance, 13(3), 163-192. Blair, M. M., O’Hara O’Connor, E., & Kirchhoefer, G. (2011). Outsourcing, modularity, and the theory of the firm. Brigham Young University Law Review, 2001(2), 263-314. Brusoni, S., & Prencipe, A. (2001). Unpacking the black box of modularity: technologies, products and organizations. Industrial and Corporate Change, 10(1), 179-205. Brusoni, S., & Prencipe, A. (2006). Making design rules: A multidomain perspective. Organization Science, 17(2), 179-189. Brusoni, S., Prencipe, A., & Pavitt, K. (2001). Knowledge specialization, organizational coupling, and the boundaries of the firm: Why do firms know more than they make? Administrative Science Quarterly, 46(4), 597-621. Callebaut, W., & Rasskin-Gutman, D. (2005). Modularity: Understanding the development and evolution of natural complex systems. Cambridge, Massachusetts: MIT Press. Campagnolo, D., & Camuffo, A. (2009). The concept of modularity in management studies: A literature review. International Journal of Management Reviews, 12(3), 259-283. Cheng, L. C. (2005). Drivers of organizational modularity in supply chains- A cross-sectional study of U.S. manufacturing industries. (Ph.D), University of Maryland. Retrieved from http://hdl.handle.net/1903/3222 Chesbrough, H. W., & Kusunoki, K. (2001). The modularity trap: Innovation, technology phase shifts and the resulting limits of virtual organizations Managing Industrial Knowledge: Creation, Transfer and Utilization (pp. 202-231). London: SAGE Pub. Ltd. Cohen, S. G., & Mankin, D. (1998). The changing nature of work: Managing the impact of information technology. In S. A. Mohrman, E. E. Lawler, & J. R. Galbraith (Eds.), Tomorrow's organization: Crafting winning capabilities in a dynamic world. San Francisco: Jossey-Bass. Colfer, L., & Baldwin, C. Y. (2010). The mirroring hypothesis: Theory, evidence and exceptions (15565068). Retrieved from Boston, Massachusetts: Ernst, R., & Kamrad, B. (2000). Evaluation of supply chain structures through modularization and postponement. European Journal of Operational Research, 124(3), 495-510. Ethiraj, S. D., & Levinthal, D. (2004). Modularity and Innovation in complex systems. Management Science, 50(2), 159-173. Furlan, A., Cabigiosu, A., & Camuffo, A. (2013). When the mirror gets misted up: Modularity and technological change. Strategic Management Journal, 35(6), 789-807. Hahn, H., & Turowski, K. (2005). Modularity of the software industry: A model for the use of standards and alternative coordination mechanisms. International Journal of IT Standards and Standardization Research, 3(2), 29-42. Helfat, C. E., & Eisenhardt, K. M. (2004). Inter-temporal economies of scope, organizational modularity, and the dynamics of diversification. Strategic Management Journal, 25(13), 1217-1232. Hoetker, G. (2006). Do modular products lead to modular organizations? Strategic Management Journal, 27(6), 501-518. Ivancevich, J. M., & Donnelly, J. H. (1975). Relation of organizational structure to job satisfaction, anxiety-stress, and performance. Administrative Science Quarterly, 20(2), 272-280.

Jacobides, M. G. (2005). Industry change through vertical disintegration: How and why markets emerged in mortgage banking. Academy of Management Journal, 48(3), 465-498. Johansson, M., (Year). Value configurations and implications for business model innovation. In Proceedings of The 17th EURAS Annual Standardisation Conference, Brussels. Jones, G. (2011). Modular systems in design. Retrieved from http://www.gethynjones.co.uk/pac/modular-systems-in-design/ Karim, S. (2006). Modularity in organizational structure: The reconfiguration of internally developed and acquired business units. Strategic Management Journal, 27(9), 799-823. Lampel, J., & Mintzberg, H. (1996). Customizing customization. Sloan Management Review, 38(1), 2130. Langlois, R. N. (2002). Modularity in technology and organization. Journal of Economic Behavior and Organizations, 49(1), 19-37. Levinthal, D. A. (1997). Adaptation on rugged landscapes. Management Science, 43(7), 934-950. MacCormack, A., Baldwin, C., & Rusnak, J. (2012). Exploring the duality between product and organizational architectures: A test of the “mirroring” hypothesis. Research Policy, 41(8), 13091324. MacCormack, A., Rusnak, J., & Baldwin, C. Y. (2006). Exploring the structure of complex software designs: an empirical study of open source and proprietary code. Management Science, 52(7), 1015-1030. MacDuffie, J. P. (2013). Modularity-as-property, modularization-as-process, and ‘modularity'-as-frame: Lessons from product architecture initiatives in the global automotive industry. Global Strategy Journal, 3(1), 8-40. Miller, T. D., & Elgard, P., (Year). Defining modules, modularity and modularization: Evolution of the concept in a historical perspective. In Proceedings of The 13th IPS Research Seminar, Fuglsoe. Mingo, S. (2013). The Impact of acquisitions on the performance of existing organizational units in the acquiring firm: The case of an agribusiness company. Management Science, 59(12), 2687-2701. Mintzberg, H. (1979). The structuring of organizations: A synthesis of the research. Englewood Cliffs, N.J: Prentice-Hall. Nelson, A., & Byers, T. (2005). Organizational modularity and intra-university relationships between entrepreneurship education and technology transfer University Entrepreneurship and Technology Transfer: Process, Design, and Intellectual Property (Vol. 16, pp. 275-311): Emerald Group Pub. Ltd. Nelson, D. L., & Quick, J. C. (2013). Organizational behavior: Science, the real world, and you. United States: Cengage Learning. Nepal, B., Monplaisir, L., & Famuyiwa, O. (2012). Matching product architecture with supply chain design. European Journal of Operational Research, 216(2), 312-325. Orton, J. D., & Weick, K. E. (1990). Loosely coupled systems: A reconceptualization. The Academy of Management Review, 15(2), 203-223. Ostrosi, E., Stjepandić, J., Fukuda, S., & Kurth, M., (Year). Modularity: New trends for product platform strategy support in concurrent engineering. In Proceedings of 21st ISPE International Conference on Concurrent Engineering, Beijing, China. Porter, L. W., & Lawler, E. E., III. (1964). The effects of "tall" versus "flat" organization structures on managerial job satisfaction. Personnel Psychology, 17(2), 135-148. Prencipe, A. (1997). Technological competencies and product's evolutionary dynamics a case study from the aero-engine industry. Research Policy, 25(8), 1261-1276. Press, K., & Geipel, M. M. (2010). Vanishing hands? On the link between product and organization architecture. Industrial and Corporate Change, 19(5), 1493-1514. Sanchez, R. (1995). Strategic flexibility in product competition. Strategic Management Journal, 16(S1), 135-159. Sanchez, R. (1997). Preparing for an uncertain future: Managing organizations for strategic flexibility. International Studies of Management & Organization, 27(2), 71-94.

Sanchez, R. (1999). Modular architectures in the marketing process. Journal of Marketing, 63(Fundamental Issues and Directions for Marketing), 92-111. Sanchez, R., & Mahoney, J. T. (1996). Modularity, flexibility, and knowledge management in product and organization design. Strategic Management Journal, 17(S2), 63-76. Schilling, M. A. (2000). Toward a general modular systems theory and its application to interfirm product modularity. Academy of Management Review, 25(2), 312-334. Schilling, M. A., & Steensma, H. K. (2001). The Use of modular organizational forms: An industry-level analysis. Academy of Management Journal, 44(6), 1149-1168. Simon, H. A. (1962). The architecture of complexity. Proceedings of the American Philosophical Society, 106(6), 467-482. Sinha, K. K., & Van de Ven, A. H. (2005). Designing work within and between organizations. Organization Science, 16(4), 389-408. Staudenmayer, N., Tripsas, M., & Tucci, C. L. (2005). Interfirm modularity and its implications for product development. Journal of Product Innovation Management, 22(4), 303-321. Takeishi, A. (2002). Knowledge partitioning in the interfirm division of labor: The case of automotive product development. Organization Science, 13(3), 321-338. Thompson, J. D. (1967). Organizations in action: Social science bases of administration. New York: McGraw-Hill. Tiwana, A. (2008). Does interfirm modularity complement ignorance? A field study of software outsourcing alliances. Strategic Management Journal, 29(11), 1241-1252. Todorova, G., & Durisin, B. (2009). Do firms innovate through integrated modular designs? Disentangling the effects of types of modularity on types of innovation (KITeS Working Papers (013)). Retrieved from Milano, Italy: Ulrich, K., & Eppinger, S. D. (1995). Product design and development. New York: McGraw-Hill. Vaccaro, I. G., Jansen, J. J. P., Van Den Bosch, F. A. J., & Volberda, H. W. (2012). Management innovation and leadership: The moderating role of organizational size. Journal of Management Studies, 49(1), 28-51. Waard, E. J., & Kramer, E. H. (2008). Tailored task forces: Temporary organizations and modularity. International Journal of Project Management, 26(5), 537-546. Wang, F., Chen, G., & Li, D. (2008). The formation and operation of modular organization: A case study on Haier’s “market-chain” reform. Frontiers of Business Research in China, 2(4), 621-654. Wolters, M. J. J. (2002). The business of modularity and the modularity of business. (Ph.D), Erasmus University Rotterdam. Retrieved from http://hdl.handle.net/1765/920 (ERIM PhD Series;EPS2002-011-LIS) Worren, N., Moore, K., & Cardona, P. (2002). Modularity, strategic flexibility, and firm performance: A study of the home appliance industry. Strategic Management Journal, 23(12), 1123-1140. Wu, L., & Park, D. (2009). Dynamic outsourcing through process modularization. Business Process Management Journal, 15(2), 225-244. Zhang, G., & Gao, R. (2010). Modularity and incremental innovation: The roles of design rules and organizational communication. Computational and Mathematical Organization Theory, 16(2), 171-200. Zirpoli, F., & Becker, M. C. (2011). The limits of design and engineering outsourcing: performance integration and the unfulfilled promises of modularity. R&D Management, 41, 21-43.

ADDITIONAL READING Ron Sanchez, (2012), Architecting Organizations: A Dynamic Strategic Contingency Perspective, in Ron Sanchez, & Aimé Heene (Eds.) A Focused Issue on Competence Perspectives on New Industry Dynamics (Research in Competence-Based Management, Volume 6) (pp. 7-48), Emerald Group Publishing Limited.

Foss, N. J. (2002). New organizational forms-Critical perspectives. International Journal of the Economics of Business, 9(1), 1-8. Frenken, K., & Mendritzki, S. (2012). Optimal modularity: a demonstration of the evolutionary advantage of modular architectures. Journal of Evolutionary Economics, 22(5), 935-956. Ulrich K. (1994) Fundamentals of Product Modularity, in Sriram Dasu, & Charles Eastman (Eds.) Management of Design (pp. 219-231). Dordrecht: Springer. O'Grady, P. (1999). The age of modularity: Using the new world of modular products to revolutionize your corporation. Iowa, USA: Adams & Steele Pub. Albert, D. (2018). Organizational Module Design and Architectural Inertia: Evidence from Structural Recombination of Business Divisions. Organization Science. Tee, R., Davies, A., & Whyte, J. (2018). Modular designs and integrating practices: Managing collaboration through coordination and cooperation. Research Policy. Peng, G., & Mu, J. (2018). Do modular products lead to modular organisations? Evidence from open source software development. International Journal of Production Research, 1-15.

KEY TERMS AND DEFINITIONS Architecture: The structure specifying which modules will be part of a system together with their functions. Conceptualization: The process involving the identification of the meaning of a concept, or construct, (i.e., ideas that represent the phenomenon) in theoretical terms. Interfaces: The coupling protocols among modules of a system that describe how they will interact. Mirroring Hypothesis: The claim that the structure of an organization mirrors the architecture of the products it develops. Modularity: A special form of design that decomposes a system into highly independent components by defining how these components will interact and thus governs the ways in which system variants can be configured (Sanchez & Mahoney, 1996; Sanchez, 1999; Asan et al., 2008) Organizational Design Parameters: The basic elements defined by Henry Mintzberg for designing organizational structures within which work is divided and coordinated to establish stable patterns of behavior. Organizational Modularity: A loosely coupled network of autonomously operating self-contained units, having low level of interaction but high level of awareness among each other through standard interfaces, which can be flexibly recombined into a variety of organizational configurations. Standards: The procedures used for testing the conformity of the modules to the design rules and for measuring module performances. Visible Design Rules: The global, unhidden assumptions that affect design decisions by defining the components of a system, the interaction of these components with each other, and their performance.