Firm size innovation : the influencing effects of organizational slack and structural inertia

(1)

Firm Size and Innovation: The influencing effects of

Organizational Slack and Structural Inertia

Student: Emiel de Greef / Student No 11420685

Date of submission: June 22nd_{2018 (Final version)}

MSc. in Business Administration - Strategy track 

University of Amsterdam, Faculty of Economics and Business

(2)

2

Statement of originality

This document is written by Student Emiel de Greef who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents

(3)

3

Abstract

This paper aims to contribute to the ongoing debate whether organizational size is positively or negatively related to innovation by examining the influence of organizational slack and structural inertia both separately and in combination. Current literature often mentions the role of organizational slack and structural inertia on the size-innovation relationship but is not in a consensus about the potential mediating effect of both.. Additionally, by combining both constructs, this study intends to explore an inverted U-shape relation between size and innovation. By bringing organizational slack and structural inertia into a broader perspective, both the voluntaristic view of a behavioral theory of the firm and the deterministic view of the population ecology theory are incorporated in this paper, which may shed light on an

enhancement or an inhibition of innovation. Data from publicly listed US manufacturing firms is used for several regression analyses in order to find support for the proposed hypotheses. The findings suggest that organizational slack mediates the relation between organizational size and innovation. Additionally, strong inertial pressures inhibit organizations to be

innovative. This research provides more insights into the influence of organizational slack and structural inertia and therefore strengthens arguments with regard to innovation within the behavioral theory of the firm and the population ecology theory. Future research can improve the findings by incorporating non-financial measurements, considering innovation from several dimensions and setting up a longitudinal research approach.

Key words: Strategic renewal, Organizational size, Innovation, Organizational Slack, Behavioral Theory of the Firm Structural Inertia, Population Ecology

(5)

5

1. Introduction

‘’Size is perhaps the most powerful explanatory organizational covariate in strategic analysis’’ – Stanislav D. Dobrev & Glenn R. Carroll (2003, p. 541). This quote reflects the fact that organizational size has been on the research agenda for several decades now. A considerable amount of research has been conducted associated with the relation between firm size and organizational outcomes. As such, according to the studies of Volberda, Baden-Fuller & van den Bosch (2001) and Damanpour (1992), organizational size may be an important determinant of innovation. Nowadays large multi-unit firms are operating in a world that requires both stability (exploitation) and flexibility (exploration) (Volberda et al., 2001). Especially the ability to adapt an organization for tomorrow might be a real challenge (Volberda et al., 2001). Additionally, organizations are operating within environments in where rapid changes are common due to technological developments and globalization.

1.1 Topic

Current literature recognizes the importance of organizational size as antecedent of innovation outcomes (Camisón-Zornoza, Lapiedra-Alcamí, Segarra-Ciprés, & Boronat-Navarro, 2004; Damanpour, 1992; Hadjimanolis, 2000; Josefy, Kuban, Ireland, & Hitt, 2015; Wolfe, 1994). Several meta-analyses as well as review studies have been conducted in order to examine a clear outcome of the size-innovation relationship. (Camisón-Zornoza et al., 2004; Damanpour, 1992; Josefy et al., 2015). These studies predominantly proved that there still is an ongoing debate whether size is positively (Aiken & Hage, 1971; Dewar & Dutton, 1986; Ettlie, Bridges, & O’Keefe, 1984) or negatively (Aldrich & Auster, 1986; Ettlie et al., 1984; Hage, 1980; Kelly & Amburgey, 1991) related to innovation.

Studies concerning the positive relationship clarify their results by mentioning that large firms might have access to more diverse & complex resources (Nord & Tucker, 1987; Sirmon, Hitt, Arregle, & Campbell, 2010). As such, these firms seem to have an advantage to possess

(6)

6

organizational slack that they can use in order to experiment with product development. The negative relationship, in turn, can be explained, because large firms might not be able to operate as flexible as small and medium-sized firms (SMEs). In general, large firms seem to be structural inert which results in more formalization and bureaucracy. Therefore these organizations may face constraints regarding innovation (Hitt, Hoskisson, & Ireland, 1990).

Thus far, the literature devoted much attention to prove either a positive or negative relationship between organizational size and innovation. A possible reason for these mixed results is that scholars seem to incorporate one underlying theory to explain the relationship between organizational size and innovation. Additionally, researchers tend to emphasize diverse elements of either an organization or an environment (Zajac & Kraatz, 1993). For instance, studies that advocate organizational slack seem to put its focus mostly on factors that an organization can use in order to adapt to its environment, whereas studies related to structural inertia tend to emphasize factors that put pressure on an organizations’ ability to innovate (Zajac & Kraatz, 1993).

Elaborating further on the mixed results regarding the size-innovation relationship, one can divide the arguments for either a positive or a negative relationship into a theoretical perspective and a methodological perspective. From a theoretical perspective, scholars that show a positive relationship seem to make arguments from the perspective of the behavioral theory of the firm (Aiken & Hage, 1971; Dewar & Dutton, 1986; Kraatz & Zajac, 2001). An example from this perspective is given by Lewin & Volberda (1999) in that the degree of innovation can be determined by the amount of organizational slack and whether this slack is used for innovation. In contrast, motives for a negative relationship between size and innovation appear to be based on the population ecology theory (Aldrich & Auster, 1986; Haveman, 1993; Kelly & Amburgey, 1991). Regarding this theory, one can reason that large organizations seem to show an inability to innovate due to a variety of inertial pressures (Lewin & Volberda, 1999).

(7)

7

From a methodological perspective, the lack of understanding among scholars related to the size-innovation relationship can be explained by that researchers do not appear to use a common metric for both organizational size and innovation. In their review study, Josefy et al. (2015) make a contribution to the discussion about what organizational size actually is. The authors propose several definitions of organizational size related to different theories within the field of strategic management (e.g. theory of the firm, transaction costs economics, resource-based view, knowledge-based view, and stakeholder theory; see Appendix 1 for a comprehensive overview). Josefy et al. (2015) extend the work of Kimberly (1976) who argued that both the way size is conceptualized as well as the measurement used might have an effect on the relationship between size and other organizational outcomes. Overall, Josefy et al. (2015) argued that organizational size can be measured ideally as revenue, amount of resources / assets, number of employees, or capacity of an organization.

According to the meta-analyses of Damanpour (1992) and Camisón-Zornoza et al. (2004), it seems that there is a lack of a common measurement for innovation as well. These studies demonstrate that innovations are measured along several dimensions, for example technical versus administrative innovations, product versus process innovations, or radical versus incremental innovations. In addition, there are contradictory results due to the level of

analysis, which can be divided into industry, organization or subunits. The stage of innovation (generation vs adaptation) and the scope of innovation (one vs multiple innovations) appear to be causing varied findings as well. Altogether, the mixed results can be explained by

measurements from one specific underlying theory (e.g. behavioral theory of the firm, and population ecology). Additionally, the use of different measurements of organizational size and innovation explained the mixed results as well. Yet, the focus of this paper will be on the theoretical contradiction, which will be discussed in the next section.

(8)

8

1.2 Theoretical gap

Current literature demonstrates a weakness regarding mediating roles within the size-innovation relationship for organizational slack (for positive relationships) as well as structural inertia (for negative relationships). Both are mentioned -as single forces- within a considerable amount of research as a possible explanation for the relationship between size and innovation (see Table 1). However, to my knowledge, their potential mediating effects regarding the size-innovation relationship are not yet scrutinized. By incorporating organizational slack and structural inertia as mediator, this study aims to deepen the insights for either the positive (in the case of organizational slack) or the negative relationship (in the case of structural inertia). As such, this study aims to strengthen the cause-effect explanation between organizational size and innovation.

Related to the mediating effects of organizational slack and structural inertia, this study also aims to combine these forces in order to explore a curvilinear relationship between organizational size and innovation. This may lead to an examination of both a positive and a negative relationship between size and innovation. Within current literature there are a few studies that demonstrate a curvilinear relationship. First, Kelm, Narayanan & Pinches (1995) found a curvilinear relationship between size and innovation. However, this study argued only from the population ecologist perspective. The positive relationship is explained through economies of scale and specialization, while the negative relationship emphasizes commitment to a firm’s existing technology and an increase in formalization (Kelm et al., 1995). Second, Nohria & Gulati (1996) conducted research on organizational slack and innovation and they found a curvilinear relationship between these variables. However, this research did not incorporate organizational size as an independent variable (Nohria & Gulati, 1996). In addition, the emphasis is only on the behavioral theory of the firm, whereby the positive relation can be explained through experimentation and slack search, whereas the negative relation occurs as a

(9)

9

result of inefficient use of resources and the likeliness of managerial self-interest (Nohria & Gulati, 1996). Lastly, Leiblein & Madsen (2009) found a curvilinear relationship between size and innovation as well. However, this study argues from the perspective of the theory of the firm in combination with the resource based view (Leiblein & Madsen, 2009). More specifically, the authors emphasize operating experience. As such, a positive relationship can be explained due to an increase in available operating experience, whereas a negative relationship is declared through a decrease in operating experience (Leiblein & Madsen, 2009).

Overall, one can argue that the current literature lacks the integration of organizational slack and structural inertia separately (as mediating effects) as well as in combination. Therefore, this paper aims to answer the following question:

How do organizational slack and structural inertia, both separately and in combination, influence the relation between organizational size and innovation of firms?

1.3 Contributions

This paper purposes to make two theoretical contributions to the current literature. First, by incorporating either organizational slack (which is rooted in the behavioral theory of the firm) or structural inertia (which is rooted in the population ecology) as mediating variables in order to discover whether the relation between organizational size and innovation can be declared by these constructs. As such, the potential role of organizational slack and structural inertia mentioned in current studies is empirically tested and this may increase the understanding of both concepts on the size-innovation relationship. Second, this paper contributes to the existing literature through the integration of both organizational slack and structural inertia. As such, an inverted U-shape relationship between organizational size and innovation can be demonstrated. Josefy et al. (2015) suggested this curvilinear relationship as a path of future research in their study of the effect of organizational size on a considerable number of organizational outcomes, including innovation.

(10)

10

A practical contribution is made in that managers can get deeper insights regarding the effect of organizational slack and structural inertia on innovation. More specifically, managers may get an explanation why particular organizations might be able to be more innovative in comparison with others or why some firms may face difficulties regarding innovative activities.

1.4 Thesis structure

The structure of this paper will be as follows. After the introduction the current literature regarding innovation, organizational slack, and structural inertia will be reviewed. After that, the conceptual model will be described, followed by the methodology section. Then, the results will be presented. Lastly, the discussion section will contain a summary of this paper and elaborates further on the contributions and limitations of this study.

(11)

11

2. Literature Review

This section contains a review of the current size-innovation literature. First, the domain regarding innovation will be presented followed by an overview of the ongoing debate concerning the size-innovation relationship. Subsequently, the influence of the behavioral theory of the firm and the population ecology perspective on the size-innovation relationship will be discussed. Lastly, the theoretical framework of this research will be explained.

2.1 Innovation: Defining the domain

Within the current literature, innovation appears to be studied from a variety of perspectives such as administrative versus technical innovations (Daft, 1978; Kimberly & Evanisko, 1981), radical versus incremental innovations (Dewar & Dutton, 1986; Ettlie et al., 1984; Nord & Tucker, 1987), and the initiation versus the implementation phase of an innovation (Marino, 1982; Zmud, 1982). In order to avoid reasoning from a particular perspective, this study uses the following overarching definition of innovation; ‘’the adoption of an internally generated or purchased device, system, policy, program, process, product or service that is new to the adopting organization’’, provided by Damanpour (1991, p. 556) and based on the work of Daft (1982) and Damanpour and Evan (1984). This definition incorporates several perspectives, for example different parts of an organization, several parts of an innovation operation and the several types of innovations (Damanpour, 1992). Additionally, the adoption of an innovation is proposed to contribute either to an organizations’ performance or effectiveness (Damanpour, 1992).

Taking this definition of innovation in to a broader perspective, one can argue that innovation may be an answer to both internal as well as external changes within the environment of an organization. As such, innovation can be seen as a form of strategic renewal, which is, following Volberda et al. (2001, p. 160), defined in this research as ‘’the activities a firm undertakes to alter its path dependence’’. According to the study of Volberda et al. (2001),

(12)

12

strategic renewal can be either selective or adaptive. Within a selective renewal, organizations are constrained by insufficient resources as well as structural inertia (Volberda et al., 2001). Organizations that strategically renew themselves from an adaptive perspective might be able to learn to act differently (as opposed to competitors) and therefore explore new / different competencies (Volberda et al., 2001).

Building further on this, innovations that are developed or purchased in order to respond to internal / external changes may be part of strategic change. Following Rajagopalan & Spreitzer (1997, p. 49), strategic change is defined as follows within this study: ‘’a difference in the form, quality, or state over time in an organizations’ alignment with its external environment.’’, whereby an organizations’ alignment is defined as ‘’the fundamental pattern of present and planned resource deployments and environmental interactions that indicates how the organization will achieve its objectives’’ (Hofer & Schendel, 1978, p. 25; in Rajagopalan & Spreitzer, 1997, p. 49). Based on the above mentioned definition of strategic change, one can state that both the selective (deterministic) and the adaptive (voluntaristic) perspectives appear here as well (Müller & Kunisch, 2017) (see Table 1).

Table 1 Single lens perspectives; size-innovation relationship

Authors Perspective Main antecedent Outcome

Ginsberg & Buchholtz (1990) Population Ecology (deterministic) Formalization, bureaucratization, complex structures

Lower conversion time (time between a particular

event within an environment and the response of an organization Kelly & Amburgey

(1991) Population Ecology (deterministic) Formalization, bureaucratization, complex structures Lower probability of a change within the core of an

organization Haveman (1993) Population Ecology

(deterministic)

Political constraints, rigidity, bureaucratization

Slower response towards dynamics within the

environment of an organization Barker & Duhaime

(1997)

Behavioral Theory of the Firm (voluntaristic)

Financial slack resources Greater extent of changes of an organizations’ strategy Kraatz & Zajac

(2001)

Human resources, financial assets

Greater extent of strategic change

Dawley, Hoffman & Lamont (2002)

Absorbed slack, unabsorbed slack

More adequate responses towards environmental dynamics after a bankruptcy

(13)

13

As a result of external changes, organizations can be constrained to follow their external environment and thus react from a selective perspective. In turn, the presentation and planning of resource deployments may give organizations the ability to experiment with new competencies and therefore transform its current competitive landscape.

Elaborating further on the nature of strategic renewal, Volberda et al. (2001) argue that the basis for selective and adaptive strategic renewal can be found within several underlying theories. Herein, population ecology, institutional theory, evolutionary theory, and resource-based theory are mainly selective, whereas the dynamic capability theory, behavioral theory of the firm, learning theories, and strategic choice theories are primarily adaptive (see Table 2).

With regard to this study, especially the behavioral theory of the firm and population ecology seem interesting because these theories have received much attention within current literature about innovation (e.g. Dewar & Dutton, 1986; Ginsberg & Buchholtz, 1990; Graves & Langowitz, 1993; Müller & Kunisch, 2017; Nord & Tucker, 1987; Rajagopalan & Spreitzer, 1997).

Table 1: Theories on Journeys of Strategic Renewal (Volberda et al., 2001, p. 162)

Mainly Selection Journeys Mainly Adaptation Journeys - ‘’Population Ecology: Renewal journeys are

based on and limited to accumulation of structural and procedural baggage through retention processes (Aldrich & Pfeffer, 1976;

Hannan & Freeman, 1977, 1984)’’

- ‘’Dynamic capability theory: Renewal journeys

are promoted by firms’ latent abilities to renew, augment, and adapt its core competence over time

(Teece, Pisano, & Shuen, 1997)’’ - ‘’Institutional theory: Renewal journeys result

from coercive, normative, and mimetic isomorphism. Renewal is achieved through maintaining congruence with shifting industry norms and shared logics (DiMaggio & Powell,

1983; Greenwood & Hinings, 1996)’’

- ‘’Behavioral theory of the firm: Renewal

journeys are determined primarily by the availability and control of organization slack and

by the strategic intent to allocate slack to innovation (Cyert & March, 1963)’’

- ‘’Evolutionary theory: Renewal journeys are

based on proliferation of routines and reinforce incremental improvements (Nelson &

Winter, 1982)’’

- ‘’Learning theories: Renewal journeys as a

process of alignment of firm and environment based on unique skills for learning, unlearning, or

relearning (Argyris & Schön, 1997; Huber, 1991)’’

- ‘’Resource-based theory: Renewal journeys

are converging trajectories of exploitation of unique core competencies (Penrose, 1959;

Wernerfelt, 1984)’’

- ‘’Strategic choice theories: Renewal journeys as a

dynamic process subject to managerial action and environmental forces (Child, 1972; Miles & Snow,

1978)’’

(14)

14

2.2 The size-innovation relationship

According to several conducted meta-analyses (e.g. Camisón-Zornoza et al., 2004; Damanpour, 1992; Josefy et al., 2015), the direction and the strength of the size-innovation relationship still is an ongoing debate within the current literature (see Table 3)

Table 3: Overview ongoing debate within size-innovation relationship

Study Outcome

Size-Innovation relationship

Independent Variable

Dependent Variable Method

Kimberly & Evanisko (1981) Positive relationship Number of employees and log number of employees

Technical / administrative innovations

Database Research & Survey

Ettlie et al. (1984) Positive relation Log of number of year-round employees

Adoption of radical / incremental product or process innovations

Mail survey & interviews Dewar & Dutton (1986) Positive relation Log of the number of

employees

Radical / incremental innovation

Interviews & Database Graves & Langowitz (1993) Negative relation Number of employees R&D intensity Database Research Haveman (1993) Inverted U-shape Scale of operations &

total assets

Investments in new products / client markets

Database Research Hadjimanolis (2000) No significant relation Number of employees Innovation activities

(new products and markets)

Case Study

Ahuja and Katila (2001) Positive relation Log of the number of employees

Patents Database Research Leiblein & Madsen (2009) Inverted U-shape Log of revenue New adopted process

technologies

Database Research

By reviewing the current studies concerning the size-innovation relationship, it seems that the underlying theory of a particular research determines whether the relationship is positive or negative. The positive relationship seems to be based on the behavioral theory of the firm, whereas the negative relationship tends to be based on the population ecology theory.

2.3 The size-innovation relationship from a behavioral theory of the firm perspective

Cyert and March (1963) were one of the first authors who tried to develop an overarching theory regarding behavior and decision making within organizations. They wrote their book in order to open the black box of firms (Argote & Greve, 2007). By opening this black box they

(15)

15

tried to find an explanation regarding the internal systems within organizations, especially focusing on behavior and decision making patterns (Argote & Greve, 2007). To point out the impact and the value of a behavioral theory of the firm, Cyert and March (1963) started their work by giving four commitments. These commitments were as follows: ‘’focusing on a small number of key economic decisions, develop process-oriented models of the firm, link models of the firm as closely as possible to empirical observations and develop a theory with generality beyond the specific firms studied’’ (Cyert & March, 1963, p. 2). With these commitments and previous research (e.g. March & Simon, 1958; Simon, 1957) regarding behavior and decision making in mind, Cyert & March (1963) elaborated further on concepts like ‘’bounded rationality, problemistic search, dominant coalition, standard operating systems, and slack search’’ (Argote & Greve, 2007, p. 339).

Regarding the relationship between organizational size and innovation, in particular slack search can be seen as interesting part of the behavioral theory of the firm, because it may be a driver of innovation. Within the current literature organizational slack is defined in many ways. Originally, Cyert and March (1963, p. 36) defined it as ‘’the disparity between the resources available to the organization and the payments required to maintain the coalition.’’ Regarding the maintenance of coalition, Cyert and March (1963) argued that organizations consist of several subgroups (e.g. sales, production and finance) that all want to defend their own interests or goals. Yet, this may lead to conflicts between groups, which might jeopardize the internal organization. However, due to the availability of slack, it seems possible to solve these goal conflicts and therefore bring the members of an organization back to a unity instead of loosely coupled groups (Cyert & March, 1963; in Nohria & Gulati, 1996). The reasoning behind this is that the presence of sufficient slack may allow organizations to distribute choice opportunities to all of its members (Moch & Pondy, 1977). This function of slack is mentioned as conflict resolution by Bourgeois (1981).

(16)

16

For the purpose of this study the definition of Geiger & Cashen (2002, p. 69) is used, which is stated as follows: ‘’the resources in or available to an organization that are in excess of the minimum necessary to produce a given level of organizational output.’’ This definition is comprehensive in that both current and potential slack resources are incorporated, which may give organizational slack a multidimensional character (Geiger & Cashen, 2002). As such, both internally generated (e.g. people or profits) and externally generated (e.g. debt financing) organizational slack seems incorporated within the provided definition by Geiger & Cashen (2002). Both might influence the amount of organizational slack and therefore the innovativeness of an organization. Additionally, this definition goes beyond keeping together the coalition by incorporating workflow buffers as another function of organizational slack (Bourgeois, 1981).

Elaborating further on current and potential resources of an organization, one can state that there are different types of organizational slack. Organizational slack can be specified as absorbed or unabsorbed (Bourgeois & Singh, 1983; Singh, 1986). Absorbed slack consists of resources that are already incorporated within the operations of an organization, for example skilled employees, overhead costs and assembled inventory (Bourgeois & Singh, 1983; Sharfman, Wolf, Chase, & Tansik, 1988). Unabsorbed slack, in turn, entails resources which tend to be much more redeployable, such as for instance cash, credit lines, and raw material inventory (Bourgeois & Singh, 1983; Sharfman et al., 1988; Singh, 1986). In line with the distinction between absorbed and unabsorbed slack is the degree of discretion of certain slack resources (Sharfman et al, 1988). Higher discretion resources might not be restricted to particular situations within an organization and seem therefore more freely deployable. Slack resources with a lesser level of discretion tend to be, in general, ‘’fixed’’ to particular situations and thus more difficult to use in a variety of situations (Sharfman et al., 1988). As such, it can be assumed that high discretion slack resources are comparable to unabsorbed resources,

(17)

17

whereas low discretion slack resources seem similar to absorbed resources. In addition, current studies regarding slack tend to make a specification in terms of the ease of recovery or the extent of future employability (Sharfman et al., 1988), Thereby, cash resources might be -in general- easy to recover, whereas resources (for example skilled labor or processed inventory) are much more difficult to recover. Again one can argue that there is an overlap with unabsorbed (high discretion) and absorbed (low discretion) resources in that unabsorbed resources are easy to recover in comparison to absorbed ones (Nohria & Gulati, 1996). This study focuses on unabsorbed, high discrete, and easy recoverable slack resources, because such resources seem in general more applicable in order to stimulate innovation within an organization as compared to absorbed, ‘’fixed’’ slack resources (Nohria & Gulati, 1996).

Regarding the part ‘’in excess of’’ within the chosen definition of organizational slack, slack can be seen in surplus from an input as well as an output perspective (Nohria & Gulati, 1996). From the input perspective slack resources contains an excess of employees, capacity or redundant capital expenses compared to the minimum required level. Regarding the output perspective, slack resources may exist in excess of the current way of doing business. As such, new opportunities can be exploited, possibly resulting in the growth of margins or revenues because of having more customers or high-tech innovations (Nohria & Gulati, 1996). Besides the availability of slack resources within an organization, it seems also crucial to deploy these resources in order to take advantage of it. Considering the ways in which slack resources can be deployed, one can argue that they can be used as response to weak performance (Kamin & Ronen, 1978), shocks within the environment of an organization (Meyer, 1982) or for experimenting (Greve, 2003; Levinthal & March, 1981). Altogether, organizational slack has an adaptive or voluntaristic character, whereby managers of an organization can influence their strategic decisions as well as their environment or structure (Müller & Kunisch, 2017). In other

(18)

18

words, managers may have the capability to shape and learn about an organizations’ environment (Miles & Snow, 1978).

However, organizational slack may also constrain organizations due to the use of slack resources in a value-destroying and inefficient way (Jensen & Meckling, 1976; Leibenstein, 1969). Arguments regarding the negative influence of slack resources on innovation find its origin in the scholar of organizational economists (Nohria & Gulati, 1996; Vanacker, Collewaert, & Zahra, 2017). In general, these economists (e.g. Leibenstein, 1969; Williamson, 1963) agree with Cyert and March (1963) that organizations consist of several coalitions including its competing interests. However, these conflicting interests are a consequence of the principal-agent relationship and therefore agents within an organization might follow their own interests rather than the organizations’ interest (which is also called the agency problem) (Eisenhardt, 1989). An obvious principal-agent relationship within an organization is the relation between middle and top managers. According to organizational economists, structured incentives can solve the principal-agent problem more effective than organizational slack (Nohria & Gulati, 1996). Therefore, organizational slack can provide companies with unnecessary costs and thus value-destroying inefficiency (Nohria & Gulati, 1996). In addition, organizations that have a considerable amount of slack may invest this into uncertain (and unrelated) innovation projects (Jensen, 1993). As such, slack may not generate the intended innovation

2.4 The size-innovation relationship from a population ecology perspective

Hannan & Freeman (1977, 1984) can be seen as one of the first authors that tried to establish the theory of population ecology of organizations. They wrote these papers to offer a different view concerning the relation between an organization and its environment and thereby they extend the work of Burns and Stalker (1961). Until that moment, the adaptive view, whereby an organization creates its own environment, dominates the literature

(19)

19

regarding the organization-environment relation (Hannan & Freeman, 1977, 1984). However, Hannan & Freeman (1977) argued that there are some inertial pressures at play that can influence the structure of an organization and therefore the power of an organization over its environment. When organizations face a strong level of these pressures, their adaptability or flexibility can diminish in comparison to organizations with a weak level of inertial pressures (Hannan & Freeman, 1977, 1984). As a result of strong pressures, organizations might act reactive towards the actions within an environment rather than establishing the environment by themselves (Aldrich & Pfeffer, 1976; Hannan & Freeman, 1977, 1984).

Elaborating further on the inertial pressures on organizational structure one can divide them into internal (particularly structural elements) and external (environmental) pressures. Concerning internal pressures, investments of an organization into nontransferable assets (e.g. plants, specialized personnel, and equipment) can generate sunk costs that can constrain adaptation (Hannan & Freeman, 1977, 1984). Second, information flows may pressure the adaptability of an organization. As a result of several hierarchical levels, leaders of an organization may not obtain complete information regarding the organizations’ activities (Hannan & Freeman, 1977, 1984). Third, political constraints may play a considerable role when the structure of an organization is transformed (Hannan & Freeman, 1977, 1984). Therefore, resources might be reallocated across business units and this may create conflicts within an organization. As such, some subunits can resist a restructuring and this may lead to short-term costs. Because of these costs, leaders may decide to not alter the structure of the organization (Downs, 1967; Hannan & Freeman, 1977, 1984). Lastly, an organization may face inertial pressures as a result of its own history (Hannan & Freeman, 1977, 1984). Hereby, standardization of procedures and task and the allocation of authority might play an essential role. Such activities may increase resistance to change as well as constrain alternative

(20)

20

Considering the external pressures, legal and fiscal barriers may limit the entry and exit decisions of an organization and thus its ability to adapt (Hannan & Freeman, 1977, 1984). Second, external information flows (similar to internal information flows) pressure

organizations to change due to the high costs of obtaining crucial information of a relevant environment (Hannan & Freeman, 1977, 1984). Finally, legitimacy violates adaptation as well as problems with generating a collective rationality (Hannan & Freeman, 1977, 1984). This study focuses particularly on the internal inertial pressures.

With these pressures in mind, the question raises why these inertial pressures exist. According to Hannan & Freeman (1984) formal organizations tend to have the ability to act reliable and seem rationally accountable for their actions. Regarding the reliability,

organizations want to deliver its products or services on time and at a particular quality level (Hannan & Freeman, 1984; Kelly & Amburgey, 1991). An organizations’ accountability refers to the specific use of resources and particular decisions / rules behind organizational outcomes. (Hannan & Freeman, 1984; Kelly & Amburgey, 1991). To accomplish these abilities, formal organizations can be structured around hierarchical levels and formal, standardized procedures that are repeatable and steady over time (Hannan & Freeman, 1984; Nelson & Winter, 1982).

Elaborating further on hierarchical organizations, every activity is localized within subunits. Therefore specific commands, information and resources are used only in these organizational silos (Simon, 1962). As such, changes within subunits may not influence other subunits (Hannan & Freeman, 1984). A possible explanation for structuring along hierarchies is that stable subunits seem to be able to resist possible shocks within the environment of the organizations and therefore provide organizations with the assurance that their production will be completed without interruptions (Šiljak, 1975; Simon, 1962). However, due to

(21)

21

may arise (Hannan, Polos, & Carroll, 2002), for example due to an increase in geographic or product diversification (Josefy et al., 2015). As such, complex organizations might face significant costs beyond its administrative costs (Josefy et al., 2015). First, complexity can lead to disagreement among an organizations’ top management team concerning strategic issues, which seems to be particularly driven by a lack of coordination and integration among top executives (Iaquinto & Fredrickson, 1997). As such, due to the siloed organizational subunits, executives may not be able to make unanimous strategic decisions. This ineffective way of decision making may potentially make an organization more vulnerable compared to, for example, competitors’ actions regarding innovation (Josefy et al., 2015). As such, these complex organizations might suffer to change rapidly within highly fast-changing

environments. Second, complexity seems to demand more information processing capabilities of the top management team (Henderson & Fredrickson, 1996). This may be difficult when an organization is organized around hierarchical levels, because information will be restricted to a particular subunit.

Regarding the execution of formal and standardized procedures, bureaucratization can occur, whereby the influence of managers on decision making changed into the application of institutionalized rules (Chen & Hambrick, 1995; Nelson & Winter, 1982). The elements of bureaucracy are ‘’differentiation, specialization, administration and routinization’’ (Sørensen, 2007, p. 389). Bureaucracy can facilitate organizations with structures in order to manage its employees effectively (Haveman, 1993; Sutton & Dobbin, 1996), as well as enable

organizations to standardize its decision making process (Baker & Cullen, 1993). As a result of formalized processes, particular responsibilities (e.g. operational decisions or the

positioning of a business unit) can be delegated towards lower management levels within an organization (Josefy et al., 2015). As such, the managers of these units seem to receive a specific amount of resources, which they are accountable for. In addition, they might be

(22)

22

obliged to report the financial results to the top management. That way, top executives

focuses on administrative oversight rather instead of regulating all subunits separately (Josefy et al., 2015). However, due to this administrative oversight, top executives might be, to a greater extent, focus on variations in performances (particularly short term) among

organizational units instead of searching for new opportunities (particularly long term) within the environment of an organization (Josefy et al., 2015). That way, senior executives tend to act in a reactive rather than an active way, which might constrain these organizations to respond adequately towards environmental changes (Josefy et al., 2015).

2.5 Conceptual model

One can state that organizational size has a positive influence on the degree of innovation of an organization. Larger firms may, in general, have access to more diverse and complex facilities compared to smaller firms, for example research capabilities, knowledgeable workers, experience with regard to product or process development, and marketing / sales competencies (Haunschild & Beckman, 1998; Nord & Tucker, 1987; Sirmon et al., 2010). In addition, larger firms seem to possess more financial resources that can be used to fund innovation projects. These larger firms can exert their facilities in order to enhance their innovation. Smaller firms, in turn, may not have these advantages. A possible explanation for this is that smaller firms may not have access to financial resources in order to obtain technical or human resources. Lastly, as organizations grow in size, they can become less vulnerable for constraints related to resource allocation, for instance resources allocated towards exploitation or exploration (Lin, Yang, & Demirkan, 2007). In other words, larger firms seem to have the ability to exploit more resources in order to realize innovation. Hence,

Hypothesis 1: There is a positive relationship between organizational size and innovation.

As opposed to smaller firms, larger firms might be able to possess more slack resources, which can be a result of its greater amount of financial and physical capacity (Sharfman et al.,

(23)

23

1988). This capacity may give these large firms a considerable amount of excess resources (organizational slack) compared to smaller organizations. Regarding the financial capacity, larger firms might be able to hold more cash and financial instruments. Therefore they can attain a higher amount of unabsorbed slack (Greve, 2003). A possible explanation for this is that larger firms can, in comparison to smaller firms, accumulate (financial) resources beyond the minimum level that is required in order to run an organization. This seems a result of a greater amount of input or output volume that these large companies can generate (Damanpour, 1992). In addition, large, diversified firms might be able to obtain economies of scale (Barney, 2002 in Josefy et al., 2015). As such, larger firms, may receive higher margins, which in turn can positively influence their financial capacity and thus their amount of organizational slack. Smaller firms may not have, in general, the opportunity to achieve either high efficiency advantages or economies of scope. Hence,

Hypothesis 2: There is a positive relationship between organizational size and organizational slack.

Organizational slack may enhance innovation for two reasons. Firstly, Organizational slack may lead to a reduction of controls within organizations and it may provide companies with a fund that they can use in times of uncertainty (Nohria & Gulati, 1996). Secondly, organizational slack seems to offer companies the opportunity to conduct innovative projects. That way, slack resources seem to provide organizations with protection regarding the possible uncertain outcomes of such projects. As such, an experimentation culture might be established (Bourgeois, 1981). This culture can allow organizations to try new strategies (e.g. new products or market) (Hambrick & Snow, 1977) and may be a driver of innovation. Additionally, slack search can enhance innovation as well (Greve, 2003). Slack search may lead to the execution of innovation projects in which high potential, but uncertain inventions might appear (Levinthal & March, 1981). Hereby, the role of slack might be that these resources may influence

(24)

24

decisions whether to continue an innovation project or not (Greve, 2003). In general, the possession of more slack resources can lead to a reduced amount of performance monitoring (Greve, 2003). Performance monitoring may occur when firms might not have the experience to determine whether innovation projects will result in an improvement of their performance (Lounamaa & March, 1987). As such, more organizational slack might positively influence innovation. Hence,

Hypothesis 3: There is positive relationship between organizational slack and innovation.

Altogether, one can state that the degree of organizational slack can explain the positive relationship between organizational size and innovation. As a result of a considerable amount of slack resources, larger firms can afford to hire more knowledgeable, professional workers, which may give these organizations an advantage over smaller firms with regard to technical competencies. Technical competencies might be essential in order to conduct innovative projects. In addition, these technical employees might be able to collaborate with other knowledgeable, professional workers and therefore they seem to have the opportunity to develop their capabilities even further, for example through the accessibility of new information (Haunschild & Beckman, 1998). As a result of synergy and the existence of knowledge pools, smaller firms can fall behind regarding innovation compared to larger firms. Furthermore, larger firms seem to invest more in innovation due to the availability of extra unabsorbed slack resources. As such, these firms might sell more products, due to greater marketing and sales efforts. Therefore, larger firms can earn back research and development costs earlier relative to smaller firms (Cohen & Klepper, 1996). Lastly, as a result of the availability of unabsorbed slack resources, larger firms might bear potential losses related to innovation as well as decreasing the risk of failure that may be related to experimentation (Haveman, 1993; Hitt et al., 1990). Therefore,

(25)

25

Hypothesis 4: The degree of organizational slack mediates the positive relationship between organizational size and innovation.

+ +

+

Figure 1: Conceptual model regarding the mediating effect of organizational slack

Compared to smaller firms, larger firms may face considerable inertial pressures as a result of growing complexity and bureaucracy (Child, 1972; Josefy et al., 2015). As organizations grow in size, more employees, strategic business units and decision making might appear. Therefore, in order to keep the organization manageable, larger organizations can be structured along hierarchical levels (Hannan et al., 2002) as a result of product or geographic differentiation. However, Mintzberg (1979) argues that innovation requires collaboration between different parts of an organization that seems to be difficult for larger firms to establish as a result of divisionalization. In general, collaboration between organizational parts can be achieved more easily in smaller organizations as compared to larger organizations (Haveman, 1993; Nord & Tucker, 1987). In addition, hierarchy might cause organizational silos, which in turn may lead to a diversity of opinions within an organization (Iaquinto & Fredrickson, 1997). This diversity of opinions may result into disagreement among senior executives concerning strategic decisions. Consequently, this disagreement can result in a political conflict, which is one of the inertial pressures according to Hannan & Freeman (1977, 1984). In addition, organizational subunits might constrain the flow of information within an organization (Henderson & Fredrickson, 1996). Again, this can enhance the inertial pressures of larger

Organizational Size

Innovation Organizational

(26)

26

organizations. Furthermore, in order to regulate the developments related to a growth in firm size, larger organizations may compose rules and regulations, which in turn can result in bureaucratization (Chen & Hambrick, 1995; Nelson & Winter, 1982). Lastly, in order to hold a larger organization competitive, economies of scale can be pursued. To accomplish this, large investments seem to be made into fixed assets as well as into the hiring of specialized personnel (Josefy et al., 2015). However, fixed assets and specialized personnel can generate sunk costs and therefore inertial pressures may appear (Hannan & Freeman, 1977, 1984; Nickerson & Silverman, 2003). Altogether, larger firms, compared to smaller organizations, might adjust their way of organizing in order to keep the organization manageable by incorporating hierarchy and formalized processes, which in turn may enhance inertial pressures. Hence,

Hypothesis 5: There is a positive relationship between organizational size and structural inertia.

Structural inertia may inhibit innovation for several reasons. First, the disagreement among top executives caused by structuring along hierarchy and formalized processes may lead to slow and ineffective decision making (Iaquinto & Fredrickson, 1997). As such, organizations with high inertial pressures seem not be able to respond quickly to developments within their environments (Josefy et al., 2015). Additionally, top executives might serve as supervisors within an organization due to the decentralization of activities towards lower level managers (Josefy et al., 2015). However, these executives seem to focus primarily on the current (financial) situation and therefore they might act reactive. In order to be innovative, executives may anticipate on changes within the environment of an organization. Therefore, as a result of their reactive attitude, they may not be able to respond adequately to market opportunities. Furthermore, hierarchy can create a considerable distance between executives and operational staff (Dougherty & Hardy, 1996). As such, executives may not be able to obtain the right information in order to make their decisions, which in turn may result in ineffective decisions

(27)

27

(Henderson & Fredrickson, 1996). In sum, structural inertia may generate rigidity or inflexibility with regard to changes within the environment of organizations and thus can inhibit the rate of innovation (Delacroix & Swaminathan, 1991; Haveman, 1993). Firms with a lesser extent of inertial pressures might have the ability to respond more adequate to its environment due to the absence of control and coordination mechanisms (e.g. hierarchy and standardized processes). Hence,

Hypothesis 6: There is a negative relationship between structural inertia and innovation.

Altogether, one can argue that the negative relationship between organizational size and innovation can be explained through the effect of structural inertia. As a result of an increase in firm size, the amount of hierarchical levels tends to grow (Hannan et al., 2002). This may give a rise in complexity, because these firms may have to hire more employees and might involve in executing different (sometimes incoherent) strategic business units. As such, the increasing amount of employees and strategic business units can demand more information processing from executives (Henderson & Fredrickson, 1996), which in turn may lead to structural distance between the operational part of the organization (e.g. R&D personnel ) and the senior executives (Dougherty & Hardy, 1996). This structural distance seems to lead to conflicting interests between top executives and operational personnel (agency problem) (Vanacker et al., 2017). Top executives may therefore follow their own interests rather than the interest of their organization, which in turn seems to result in using organizational resources in a value-destroying manner (Nohria & Gulati, 1996). For example, the usage of slack resources into unrelated innovation projects (Jensen, 1993). Additionally, due to an increase in hierarchical levels, disagreement among senior executives tends to grow, which in turn inhibits an adequately response towards fast-changing environments (Iaquinto & Fredrickson, 1997). Lastly, the standardization of processes (as consequence of an increase in size) might enhance decentralized decision making by lower level managers, which gives top executives the

(28)

28

opportunity to have an overview of an organization. However, this overview seems to have a reactive character as it focuses only on fluctuation within performances of business units (Josefy et al., 2015). This may result in ineffective responses towards environments as well. Hence,

Hypothesis 7: The degree of structural inertia mediates the negative relationship between structural inertia and innovation.

+ -

+

Figure 2: Conceptual model regarding the mediating effect of structural inertia

By combining both organizational slack and structural inertia one can make arguments for an inverted U-shape relationship between organizational size and innovation. When organizations are relatively small, they may need all their resources in order to survive. However, as firms increase in size, they can obtain more resources than they actually need in order to survive, for example better research capabilities, experience related to products and markets, knowledgeable workers, and sales / marketing competencies (Haunschild & Beckman, 1998; Nord & Tucker, 1987; Sirmon et al., 2010). As such, organizations can increase further in size, which in turn may have a positive influence on their slack resources.

However, it may be possible that, at a certain point, organizations might possess too many slack resources that their senior executives are not able to use these resources most effectively, which in turn might be a consequence of their pursuit towards self-interest (Jensen, 1993; Nohria & Gulati, 1996). Additionally, in order to control and coordinate larger firms,

Organizational Size

Innovation Structural

(29)

29

hierarchy and standardized processes tend to be necessary, which in turn may cause inertial pressures for an organization (Haveman, 1993). As such, hierarchical levels and standardized processes may constrain the adaptability of organizations towards their environment and therefore might inhibit the degree of innovation of organizations. Theoretically, as a firm reaches a particular size, the advantages of organizational slack resources may be outweighed by the disadvantages of structural inertia. On the contrary, as a firm stays small, it can take advantage of slack resources in order to use them for innovation, while it also take advantage of its flexible structure. Hence,

Hypothesis 8: There is an inverted U-shape relationship between organizational size and innovation.

∩

Figure 3: Conceptual model regarding the curvilinear size-innovation relationship

Organizational Size

(30)

30

3. Methodology

This chapter contains the methodology that will be used for this study. First, the research design will be explained. Subsequently, the sample and operationalization of the dependent, independent, mediating and control variables will be described. Lastly, the reliability and validity will be assessed followed by the justification of the models that will be used in order to analyze the results.

3.1 Research Design

This research is conducted from a deductive perspective, whereby the influence of existing theories (behavioral theory of the firm and population ecology) on the relationship between size and innovation is empirically tested. As such, the theory concerning these concepts may be further developed (Saunders, Lewis, & Thornhill, 2016). These empirical tests have a quantitative character, whereby a database with several numerical organizational and financial data is used. Lastly, the time horizon of this paper is cross-sectional as the research is conducted at a fixed time (Saunders et al., 2016).

3.2 Sampling Strategy

The sample of this research consists of publicly listed US manufacturing firms (SIC code 2000 – 3999) covering the years 2000 to 2016. Particularly these firms are selected because this is an industry in where innovations are essential in order to stay competitive. On the one hand, manufacturing firms might strive to most efficient processes whereby exploitative (process) innovations might appear. On the other hand, with a view on the long term, these firms have to invent new products, which might have an explorative character. Furthermore, this sample is chosen because it enables to compare this research with previous studies in where manufacturing firms are used. (e.g. Dewar & Dutton, 1986; Leiblein & Madsen, 2009).

(31)

31

3.3 Data collection

In this study secondary data is used, which is collected through the CRSP – Compustat Merged Database, accessible via Wharton Research Data Services (WRDS). This database consists of all Standard & Poors’ 500 companies listed on the New York Stock Exchange or NASDAQ and encompasses loads of data with regard to annual and quarterly fundamentals as well as daily and monthly security figures and historical segments. The advantage of the merged database is that CRSP gives access to market and corporate data of companies (e.g. stock & bond prices), whereas Compustat offers fundamental data (e.g. sales, number of employees, assets). First, the separate figures are gathered from the CRSP – Compustat Merged database and subsequently the several ratios are calculated. Initially, the dataset contains 12.467 firm-year observations. Following the data-preparation method of Kim & Bettis (2014) and Villalonga (2004), firm-year observations (1) with missing data concerning key variables, and (2) with an R&D intensity (R&D expenditures dived by net sales) higher than 1 are excluded. As a result, the final sample consists of 6858 firm-year observations.

3.4 Measures

Innovation. Within this study, the dependent variable is innovation. The proxy that is used

is research & development expenditures dived by the amount of sales of a company (Net Sales). Subsequently, this ratio is divided by the average R&D industry intensity.

𝐼𝑛𝑛𝑜𝑣𝑎𝑡𝑖𝑜𝑛 = (

𝑅𝑒𝑠𝑒𝑎𝑟𝑐ℎ & 𝐷𝑒𝑣𝑒𝑙𝑜𝑝𝑚𝑒𝑛𝑡 𝐸𝑥𝑝𝑒𝑛𝑠𝑒𝑠

𝑁𝑒𝑡 𝑠𝑎𝑙𝑒𝑠 )

𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑅&𝐷 𝑖𝑛𝑑𝑢𝑠𝑡𝑟𝑦 𝑖𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦

This way of measuring innovation is consistent with a considerable amount of studies (e.g. Camisón-Zornoza et al., 2004; Hitt, Hoskisson, Ireland, & Harrison, 1991; Kim & Bettis, 2014). By using this measure, it is possible to have a relative measure of innovation that can reduce possible scale advantages of firms that are of larger size. Additionally, such a broad way of measuring innovation makes it possible to include all sorts of innovation (e.g. process & product

(32)

32

innovation). Furthermore, by correcting for the R&D industry intensity it can be possible to eliminate industry effects, because some industries might be more R&D intensive than others.

Organizational size. The independent variable for this research is organizational size,

which is proxied as the number of employees of organizations. Josefy et al. (2015) provided a list of ideal measurements which contains revenue, amount of resources / assets, number of employees, or capacity of an organization. Often the way of measuring is dependent on the underlying theory chosen for the firm size measurement (see appendix 1). Overall, the number of employees is mentioned as a most robust and direct measurement (Josefy et al., 2015), detached from any particular theory or framework. The number of employees will be logged in order to better capture its real effect on innovation (Dewar & Dutton, 1986; Ettlie et al., 1984).

𝑂𝑟𝑔𝑎𝑛𝑖𝑧𝑎𝑡𝑖𝑜𝑛𝑎𝑙 𝑆𝑖𝑧𝑒 = log(𝐹𝑖𝑟𝑚 𝑆𝑖𝑧𝑒 𝐸𝑚𝑝𝑙𝑜𝑦𝑒𝑒𝑠)

The use of this proxy is consistent with a considerable amount of prior studies (e.g. Dewar & Dutton, 1986; Ettlie et al., 1984; Kimberly, 1976).

Organizational slack. Organizational slack is the first mediator variable. It might be

difficult to measure this with one generic proxy because it appears in many forms within an organization. Current literature has made a lot of efforts to determine how slack may be measured in a most comprehensive way (Bourgeois, 1981; Singh, 1986). However, it remains difficult to generate a widespread proxy, especially because slack may exist, for example, as knowledge, facilities or human resources as well as financial buffers (Nohria & Gulati, 1996). Therefore, with an eye on the research method chosen for this paper, organizational slack is proxied as the cash & short investments.

(33)

33

This proxy is adapted from the study of Kim & Bettis (2014) and is closely related to the definition that is used in this study. It is important to mention that this proxy of organizational slack only covers the cash stock of an organization. However, a change exists that a company does have a lot of absorbed slack resources, but that it lacks financial slack resources. In order to address the above stated notion, a financial proxy for the dependent variable is chosen as well. Therefore, only the influence of financial buffers on the degree of R&D intensity of an organization is included in this study.

Structural inertia. Structural inertia is the second mediator variable in this research.

Because this study uses secondary data through a database, it is complex to find an

appropriate proxy in order to determine structural inertia. Previous studies measured structural inertia only through a survey or interviews or they used variables which are not applicable within the research setting of this study (e.g. Ginsberg & Buchholtz, 1990; Haveman, 1993; Kelly & Amburgey, 1991). In order to find a proxy that can be applied in this research setting, the inertial pressures provided by the articles of Hannan & Freeman (1977, 1984) are

revisited. Especially the internal inertial pressure with regard to the possession of or

investments in fixed assets like plants, equipment and specialized personnel seems interesting with an eye on the chosen research design. These kind of assets may cause sunk costs for organizations and therefore, for example, inflexibility (Hannan & Freeman, 1977). These sunk costs can be quantified in terms of financial figures. Therefore, structural inertia is proxied as the ratio between total current assets and fixed assets (total property, plant, equipment).

𝑆𝑡𝑟𝑢𝑐𝑡𝑢𝑟𝑎𝑙 𝑖𝑛𝑡𝑒𝑟𝑡𝑖𝑎 = 𝑇𝑜𝑡𝑎𝑙 𝐶𝑢𝑟𝑟𝑒𝑛𝑡 𝐴𝑠𝑠𝑒𝑡𝑠

𝑇𝑜𝑡𝑎𝑙 𝑃𝑟𝑜𝑝𝑒𝑟𝑡𝑦, 𝑃𝑙𝑎𝑛𝑡 𝑎𝑛𝑑 𝐸𝑞𝑢𝑖𝑝𝑚𝑒𝑛𝑡

Again, by using this proxy it is possible to compare the influence of the two financial proxies (structural inertia and innovation) with each other. However, caution is needed, because structural inertia is only measured related to one of the eight inertial pressures.

(34)

34

Therefore, it may happen that one organization is classified as structural inert based on fixed assets, whereas another organization, which may be very bureaucratic but do not possess a lot of fixed assets, is not mentioned as inflexible. This issue can be tackled by mentioning that all pressures are somewhat intertwined to each other, meaning that larger firms with a great amount of fixed assets may be, in general, diversified along product or geographic dimensions and therefore they might need a bureaucratic approach in order to keep these firms

controllable (Hannan et al., 2002; Haveman, 1993).

Past performance. This research controls for past performance, because it tends to

influence the relationship between size and innovation. Firms with a strong performance may be able to bear potential losses of innovation, whereas this can be more difficult for firms with weak firm performance. (Hitt et al., 1990). Therefore, this study incorporates lagged return on assets (t-1) as a proxy for past performance covering the years 1999 to 2015. This way of measuring is consistent with the studies of Chen & Miller (2007) and Vanacker et al. (2017).

𝑃𝑎𝑠𝑡 𝑝𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒 (𝑡 − 1) = 𝑁𝑒𝑡 𝐼𝑛𝑐𝑜𝑚𝑒 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠

Time. In order to control for a time-effect, the years 2000 to 2016 are incorporated as

dummy variables within this study. It may be possible that organizations adjust their

innovation expenditures as a result of an economic recession for example. In addition, it may happen that organizations plan some large R&D expenditures a few years ahead, which can result in fluctuations within their innovation intensity. Lastly, an explosive growth in demand caused by market forces in a particular year may lead to an increase in innovation in order to keep up with the market situation.

Industry profitability. To incorporate a possible industry effect, this study controls for

industry profitability. The industry profitability serves as a proxy to the average return on assets of firms that are active in the same industry (based on the four-digit SIC code). As

(35)

35

such, it is possible to compare the profitability of an industry with the profitability of single firms. This way of measuring industry profitability is consistent with the study of Vanacker et al. (2017).

3.5 Reliability / validity

Regarding potential reliability issues, the database used in this research (Wharton CRSP – Compustat Merged Database) is accessible through the Wharton University of Pennsylvania and includes company data from Standard & Poor’s. Additionally, this methodology is used in prior studies as well (e.g. Kim & Bettis, 2014; Villalonga, 2004). In order to use the gathered data in a reliable manner, the data-analyzing method of this research is based on the studies empirical studies of Kim & Bettis (2014) and Villalonga (2004). These studies examined the role of intangible and cash resources on competitive strategy and firm performance and are published in the Strategic Management Journal and the Journal of Economic Behavior & Organization.

To conduct a valid research, the internal, construct, and external validity are taken into consideration as well. First, to guarantee the internal validity, several control variables are added (time, past performance and industry profitability). These control variables are incorporated in order to examine possible influences of them on the dependent variable. As such, the results of this paper might be more accurate. Second, the construct validity is assured by using or adapting proxies that has been used in a considerable amount of prior studies (e.g. Dewar & Dutton, 1986; Hitt et al., 1991; Kim & Bettis, 2014; Kimberly, 1976; Vanacker et al., 2017). Lastly, regarding the external validity, the use of a sample that covers the whole publicly listed US manufacturing population may mean that the results are

generalizable to all publicly listed US manufacturing firms, but not to manufacturing firms outside the US, firms active in other industries (e.g. agriculture, retail, or services) or not publicly listed firms.

Firm size innovation : the influencing effects of organizational slack and structural inertia

Firm Size and Innovation: The influencing effects of

Organizational Slack and Structural Inertia

Statement of originality

Table of contents

Abstract

1. Introduction

2. Literature Review

3. Methodology