“Open innovation and human sponges”
A positivistic, deductive, mono-method, correlational/observational, cross-sectional,
primary data-based study in the Dutch financial services industry.
Student: Mandy Carola Honingh (10545484)
Thesis supervisor: Bernardo Silveira Barbosa Correia Lima, MSc
Strategy Track
MSc. in Business Administration
Amsterdam Business School
1
STATEMENT OF ORIGINALITY
This thesis is written by MScBA Strategy student Mandy Carola Honingh who declares to take full responsibility for the contents of this thesis.
I declare that the text and the work presented in this thesis is original and that no sources other than those mentioned in text and references have been used in creating it. The University of Amsterdam’s Faculty of Economics and Business is responsible solely for
the supervision of completion of the thesis, not for its contents.
2
TABLE OF CONTENTS
Abstract.……….. 4
Introduction.……… 5
Literature Review……… 8
Individual Absorptive Capacity.………... 8
Individual Inbound Open Innovation Practices and Individual Absorptive Capacity….... 11
Moderating Role of Stressful Events.………. 14
Literature Gap.……… 15
Research Question.……….. 18
Hypotheses……… 19
Individual Inbound Open Innovation Practices as Antecedent to Individual PACAP and Individual RACAP………...19
The Moderating Role of Stressful Events on Individual Inbound Open Innovation Practices as Antecedent to Individual PACAP.………..22
Conceptual Model.……….. 23
Method……….. 24
Setting and Data Collection……… 24
3
Analysis and Results………. 30
Results………. 33
Discussion and Conclusion………... 36
Implications………. 36
Contributions………... 37
Limitations and Future Research Directions………... 39
References………. 43
Appendices……… 56
A Survey.………... 56
B Harman’s Single Factor Score………... 74
C Variables and Corresponding Items……….. 75
D Factor Loadings of the PCA After Oblimin Rotation………... 77
1. Individual ACAP four factor solution………... 77
2. Individual ACAP two factor solution……… 79
3. Individual Inbound Open Innovation Practices………. 81
4
ABSTRACT
Research on absorptive capacity is well established, yet our understanding of individual level absorptive capacity (i.e., the individual activities of recognizing, assimilating, transforming, and exploiting new external knowledge) and its antecedents remains rather unclear. Exploring how individual inbound open innovation practices (i.e., the opening up to, and establishment of relationships with external agents with the aim to access their competences and knowledge) affect individual potential (i.e., recognizing and assimilating new external knowledge) and realized absorptive capacity (i.e., transforming and exploiting new external knowledge), this study identifies significant effects for both components of absorptive capacity. In an
individual-level study of 110 employees in the Dutch financial services industry, results indicate that individual inbound open innovation practices enhance both individual potential and realized absorptive capacity. Furthermore, the results show no significant effect of the hypothetical moderator stressful events at the workplace. Through a richer explanation and empirical assessment, this study contributes to greater clarity of and emerging academic thinking on micro-foundations of absorptive capacity.
Key words: Organizational learning; individual absorptive capacity; individual inbound open innovation practices; micro-foundations; stressful events
5
INTRODUCTION
In order to achieve superior performance in today’s dynamic environment, firms are obliged to generate new knowledge and innovate their products and services. Crucial for firms in generating new knowledge internally, is employees’ ability to absorb knowledge from
external sources (Distel, 2017). Innovation, on the other hand, is becoming increasingly more open (Chesbrough, Vanhaverbeke & West, 2014). Van de Vrande, De Jong, Vanhaverbeke and De Rochemont (2009, p. 434) speak of “a trend towards increased popularity and dissemination of open innovation”. This makes Absorptive Capacity (ACAP) and open
innovation interesting topics for both theory and practice, since relatively little is known about their manifestation at an individual level of analysis. Besides a better understanding of these concepts, this study also provides useful suggestions for managers in the challenge of achieving superior performance.
The conceptualization of ACAP as organizational capability, has gained wide acceptance in strategy and organization literature (e.g., Lane, Koka, & Pathak, 2006; Todorova & Durisin, 2007). It is known for being an important antecedent for innovation performance (Cohen & Levinthal, 1990; Zahra & George, 2002). Despite the popularity of the concept and its wide use in academic contributions, prior empirical work has relatively neglected the role of
individuals in analyzing absorptive capacity (Lowik, Kraaijenbrink & Groen, 2012; Volberda, Foss & Lyles, 2010). This is highly problematic because it diminishes the role of employees as key assets of a firm and as a fundamental locus of knowledge (Felin & Hesterly, 2007; Lane et al., 2006). Underlying mechanisms of ACAP are still quite ambiguous (Lane et al., 2006; Volberda et al., 2010) and only few empirical studies have examined ACAP from an individual level of analysis (Lowik et al., 2012). This is striking, because the role of
individuals was already highlighted by Cohen and Levinthal in 1990. In their seminal paper, the authors stated the core assumption that “an organization’s absorptive capacity will depend
6 on the absorptive capacities of its individual members” (p. 131). 27 years later, individual ACAP has been defined as “an individual’s activities to recognize, assimilate, transform, and exploit new external knowledge” (Lowik, Kraaijenbrink & Groen, 2017, p. 1323). Although recent ACAP literature has highlighted the importance of antecedents of individual ACAP such as prior knowledge and experience (Cohen & Levinthal, 1990; Lane et al., 2006), cognitive models (Cohen & Levinthal, 1990; Zahra & George, 2002), and social networks (Todorova & Durisin, 2007), the effects of such antecedents have hardly been examined empirically. Also, Volberda, Foss and Lyles (2010, p. 943) state that “research on AC should be explicit about what kind of knowledge is being absorbed”. So an ongoing conversation on individual ACAP can be identified, but knowledge about this topic is still incomplete.
Scholars therefore encourage more research on individual ACAP (Lane et al., 2006; Volberda et al., 2010) and to examine existing and new antecedents of ACAP, including their relative importance (Todorova & Durisin, 2007).
This study seeks to answer the call for more empirical research on individual ACAP and its antecedents. It follows Volberda, Foss and Lyles’ (2010) advice to be specific about what knowledge is being absorbed, by focusing on individual inbound open innovation practices. Open innovation practices are increasingly present in (Chesbrough, 2006) and considered to be relevant for small, medium, and multinational enterprises (Van de Vrande et al., 2009). Especially inbound open innovation practices are most often used (Enkel, Gassmann & Chesbrough, 2009; Lichtenthaler, 2011). Inbound open innovation practices encompass knowledge diversity and external network diversity, which both have been found to
significantly increase individual ACAP (Lowik et al., 2017). Therefore, this study examines individual inbound open innovation practices as a source of external knowledge for individual ACAP. Since activation triggers are expected to moderate the impact of knowledge sources on Potential Absorptive Capacity (PACAP) development (Zahra & George, 2002), a
7 moderating role of stressful events at the workplace will also be examined. This study is guided by and aims at answering the following research question: “What is the relation between individual inbound open innovation practices and individual absorptive capacity and how do stressful events influence this relation?”.
Based on survey data from employees working in the Dutch financial services industry (e.g., banking, insurance, pension funds, stock exchange, and consulting (Dutch Authority for the Financial Markets, 2018)), this study contributes to existing literature in two ways. First, this study is among the few to have taken an individual level perspective on ACAP. Being the first examining individuals’ open innovation practices as antecedent to individual ACAP, this study extends ACAP theory to better understand how capabilities are created, why they might differ, and provides thorough understanding of this specific relation. Second, the analysis of open innovation from a practice perspective, with a specific focus on inbound open
innovation practices, in a business context, and as an antecedent to individual ACAP,
provides new insights in open innovation literature. Moreover, practitioners are enriched with suggestions on how to enhance employee ACAP, based on the results of this study.
The remainder of this thesis is organized as follows: First, the theoretical background will be described in the literature review, resulting in the literature gap and research question. Second, the hypotheses will be elucidated and outlined in the conceptual model. Third, the methodological set-up in terms of data collection, measurement and validation of constructs will be explicated. Fourth, the used analyses and subsequent empirical results will be presented. Fifth, in the discussion and conclusion section, implications, contributions, limitations and suggestions for future research will be provided. Lastly, a reference list and the appendices can be found in the back of this thesis.
8
LITERATURE REVIEW
In this section relevant theory will be presented. First, relevant previous research on individual ACAP will be reviewed. Second, relevant previous research on the relation
between individual inbound open innovation practices and individual ACAP will be reviewed. Third, a review of relevant previous research on the moderating role of stressful events will be provided. Fourth, all three variables will address the literature gap. This section will end with the research question of this study.
Individual Absorptive Capacity
Although Kedia and Bhagat (1988) were the first who coined the term “absorptive capacity”, it is generally accepted that Cohen and Levinthal (1990) were the first authors who defined this concept. ACAP is one of a firm's fundamental learning processes: “to recognize the value of new, external information, assimilate it, and apply it to commercial ends” (Cohen &
Levinthal, 1990, p. 128). In this reasoning, ACAP is the mechanism that makes external knowledge available to and useful within an organization. Key assumption in ACAP literature is that the place where the knowledge is recognized and acquired is distant from the place where it is transformed and exploited (Cohen & Levinthal, 1990).
In recent decades, ACAP is still one of the most important constructs to emerge in
organizational research (e.g., Lane et al., 2006; Volberda et al., 2010; Yao & Chang, 2017). It has been embedded in multiple different theories and antecedents (Lane et al., 2006; Volberda et al., 2010), measured in different empirical contexts, and from different levels of analysis (e.g., Cockburn & Henderson, 1998; Lane & Lubatkin, 1998; Liu & White, 1997; Lowik, Kraaijenbrink & Groen, 2016; Mowery & Oxley, 1995). An individual level of analysis however, has rarely been adopted (Lowik et al., 2012).
9 Currently, a debate within ACAP literature can be detected. Contradictory findings exist on the relative importance of individual (Matusik & Heeley, 2005) and organizational (Zhao & Anand, 2009) factors in developing organizational ACAP. Additionally, in the
micro-foundations literature, central point of discussion is the locus of knowledge creation. Opinions are divided on whether this locus of knowledge creation takes place at the individual level or at the collective level (Felin & Hesterly, 2007; Hodgson, 2012). Other research suggests a combination of the two (Distel, 2017; Nemanich, Keller, Vera & Chin, 2010), since some empirical results showed that individual and collective factors complement each other (Lowik et al., 2016).
This study follows a call for further empirical research on ACAP from an individual level of analysis (e.g., Lane et al., 2006; Lowik et al., 2012; Volberda et al., 2010). This is also called “individual ACAP” (Lowik et al., 2017) and its existence was already illustrated by Cohen and Levinthal in 1990. They argued that individual level ACAP is an important antecedent to organization level ACAP. The latter would be even supervenient on the former, since there would be no organization level ACAP without individual level ACAP. A firm’s ACAP is not resident in any single individual, but instead depends on the ACAP of its individual members (Volberda et al., 2010).
Lowik, Kraaijenbrink and Groen (2017, p. 1323) were one of the first authors who defined individual ACAP as “an individual’s activities to recognize, assimilate, transform, and exploit new external knowledge”. An individual’s recognition concerns searching for new
knowledge, locating it, and evaluating it as opportunity for beneficial use. It concerns the individual activities that are related to the notion of “entrepreneurial alertness”: Scanning and searching, associating and connecting, and evaluating and judging (Tang, Kacmar & Busenitz, 2012). Assimilation makes the individually recognized and acquired knowledge adapt to the
10 organizational context, by making it understandable and transferable to other organizational members. It concerns the process of interpretation, articulation, and codification to make newly acquired knowledge available to other individuals by incorporating it into the
organizational memory (Lowik et al., 2012). Recognition and assimilation together constitute individual Potential Absorptive Capacity, i.e., “individual PACAP” (Ojo & Raman, 2016; Rai & Prakash, 2016). In the transformation process, the newly assimilated knowledge is
combined and integrated with existing knowledge from other organizational members to create new ideas. It concerns the individual ACAP activity of generating new ideas in collaboration with other individuals. Lastly, exploitation at the individual level concerns the internalization of the new knowledge into one’s own work routines (Nonaka, 1994). It concerns the individual ACAP activity of applying new knowledge in own work routines (Lowik et al., 2012). Transformation and exploitation together constitute individual Realized Absorptive Capacity, i.e., “individual RACAP” (Ojo & Raman, 2016; Rai & Prakash, 2016).
Although heterogeneous effects of individuals on knowledge acquisition capabilities remain mainly unsubstantiated (Ojo & Raman, 2016), some research did find that individual ACAP can be increased through networking, motivation, education, and skills (Lenox & King, 2004; Minbaeva, Pedersen, Björkman, Fey & Park, 2003). Another way to augment individual ACAP is by overcoming hindrances to the identification and integration of new knowledge, for instance cognitive biases (Eggers & Kaplan, 2013).
Recently, a bifurcation of individual ACAP has taken place. Schweisfurth and Raasch (2018) made a distinction between “solution absorptive capacity” and “need absorptive capacity”. They argued that literature focused almost exclusively on ACAP in relation to technical solution knowledge, providing functionalities (Lane et al., 2006; Volberda et al., 2010), which they refer to as solution absorptive capacity. Relative silence was noticed on ACAP in
11 relation to need knowledge, which is about unmet needs arising in the use of a given product or service, which they refer to as need absorptive capacity. Both knowledge types are often located outside a firm and need to be absorbed in order for innovation to occur (Schweisfurth & Raasch, 2018). Narrowing ACAP to solution knowledge only would be problematic because this impedes exploration of how ACAP operates for other knowledge types in other contexts (Lane, Salk & Lyles, 2001). Solution absorptive capacity and need absorptive capacity would also fundamentally differ in terms of knowledge types. Although this study does not address the two distinct types of individual ACAP, this does highlight the role of particular knowledge sources in relation to individual ACAP. Recent research therefore suggested to explore how individual ACAP differs, depending on the sources of the knowledge to be absorbed (Schweisfurth & Raasch, 2018).
Individual Inbound Open Innovation Practices and Individual Absorptive Capacity
The ability of absorbing new knowledge is particularly important in the context of innovation, which requires new combinations of previously unconnected pieces of knowledge. This points to the growing paradigm of “open innovation” (Chesbrough, 2006). Literature on open
innovation stressed that firms’ openness to external sources of knowledge is an important driver of innovation performance (e.g., Chesbrough, 2006; Laursen & Salter, 2006; West & Bogers, 2014). Firms source this external knowledge by engaging in open innovation
practices. Open innovation practices are the “inter-organizational exchange of knowledge and the inflow of external knowledge into an organization” (Lowik et al., 2017, p. 1320). This involves a variety of sources such as suppliers, customers, or universities and a variety of mechanisms such as scouting, sourcing, licensing, or collaborating (Bogers, Foss & Lyngsie, 2018). Literature stressed that firms do not have to create entirely new processes for managing open innovation but instead can integrate it within their existing organizational processes,
12 which partially explains its growing relevance and popularity (Lichtenthaler, 2011) at small, medium, and multinational enterprises (Van de Vrande et al., 2009).
Since open innovation emphasizes knowledge flows (Chesbrough, 2006), the direction of a knowledge flow is commonly used as a way to classify amongst open innovation practices (Gassmann & Enkel, 2004). This explains the two dimensions of “inbound open innovation practices” and “outbound open innovation practices”. Inbound open innovation practices involve the opening up to, and establishment of relationships with external agents with the aim to access their competences and knowledge, and thus improving a firm’s own innovation performance (Chesbrough et al., 2014). Outbound open innovation practices involve the establishment of relationships with external agents with the aim to commercially exploit innovation opportunities (Chesbrough et al., 2014). In this study, open innovation practices are examined as a source of external knowledge for individual ACAP, therefore it only focuses on individual inbound open innovation practices.
To find out how to improve the capabilities to better capture value from inbound open innovation practices, most studies have examined the open innovation practices at an (inter)organization level of analysis (Bogers et al., 2017). Of growing interest however is open innovation at an individual level of analysis (Bogers et al., 2018). Research on open innovation still seems to neglect “the human side” (Gassmann, Enkel & Chesbrough, 2010, p. 218) in a sense that the role individuals have in open innovation practices, remains unclear (West & Bogers, 2014). This is remarkable since individuals are the ones who bring open innovation into practice (Bogers et al., 2017) – or not. For instance, when an individual employee is misaligned with top management’s strategy to implement (inbound) open
innovation practices, managers might be confronted with significant challenges, which might even threaten the survival of a firm (Huston & Sakkab, 2006). So, the micro-foundation of
13 (inbound) open innovation is rooted in an individual’s intentional actions, experience, and preferences (Ahn, Minshall & Mortara, 2017).
Although some research has focused on individual open innovation performance (e.g., Dahlander, O'Mahony & Gann, 2016; Salter, Criscuolo & Ter Wal, 2014; Salter, Wal, Criscuolo & Alexy, 2015), individuals as innovators (e.g., Füller, Mühlbacher, Matzler & Jawecki, 2009; Randhawa, Wilden & Hohberger, 2016), and individual motivational factors to practice inbound open innovation in the first place (e.g., Chesbrough, 2006; Lichtenthaler, 2011), research on individuals’ involvement in knowledge flows across the boundary of the firm is currently lacking. This is striking since academic work on “gatekeepers”, i.e.,
individuals who connect external and internal sources of innovation (Allen, 1977; Tushman, 1977), is well established (Tushman & Katz, 1980). Even more since the roles that enable individuals to bring external knowledge into the firm are becoming more distributed across firms (Whelan, Teigland, Donnellan & Golden, 2010). Nowadays, individuals are the locus of knowledge creation and innovation (Felin & Hesterley, 2007) through knowledge sharing and integration.
Therefore, this study examines inbound open innovation as a source of external knowledge for individual ACAP and from an individual practice perspective. Examples of such practices are: Searching for new trends or technology via internet, trade organization information, in technical magazines, or by purchasing licenses, patents, or R&D work from others (De Araújo Burcharth, Knudsen & Søndergaard, 2014). In this way, this study captures the multifaceted nature of the concept by including practices that transcend the relationship approach (Laursen & Salter, 2006). Inbound open innovation practices encompass prior knowledge diversity by facilitating a variety of knowledge which results from education, licenses, and work
14 network diversity by facilitating contact with people outside an individual’s organization such as lead users and innovation brokers (Lowik et al., 2017). This is somewhat reflected in Cohen and Levinthal’s (1990) accumulation of past experiences and “outward-looking ACAP”, which regards to one’s touch points with external sources of knowledge.
In practicing inbound open innovation, one needs to develop and maintain capabilities to connect newly encountered, external knowledge with one’s existing knowledge base (Lowik et al., 2017). Central is the ability to tap into external sources of knowledge and consequently integrate it, as is facilitated by individual ACAP (Cohen & Levinthal, 1990). When
employees practice inbound open innovation, they hold diverse knowledge and external networks, which facilitates the recognition and assimilation of relevant external knowledge to create new recombinations (Allen, 1977; Cohen & Levinthal, 1990; Mednick, 1962; Salter et al., 2015). Therefore, employees’ knowledge bases can be important determinants of a firm’s ability to access and absorb external sources of knowledge (Larrañeta, Zahra & González, 2012).
Moderating Role of Stressful Events
The recognition, assimilation, transformation, and exploitation of external knowledge can be constrained or encouraged by several factors (Dyer & Singh, 1998; Simonin, 2004). This is also indicated in the model of Zahra and George (2002) in which they state that activation triggers are expected to moderate the impact of knowledge sources on ACAP development. Activation triggers encourage to respond to specific internal or external stimuli (Zahra & George, 2002). An internal trigger could for instance be a crisis, such as performance failure. Crises might have a negative connotation, but can also intensify efforts to achieve and learn new skills, and to develop new knowledge that increases ACAP (Kim, 1998). External triggers are events that may influence the future of the firm in which an individual employee
15 operates. Examples are technological shifts or changes in policy. Both internal and external triggers intensify efforts to seek external knowledge (Winter, 2000).
Since employees working in the financial services industry are susceptible to a high degree of job stress (Montgomery, Blodgett & Barnes, 1996), the internal activation triggers in this study are stressful events at the workplace. This regards to for instance being criticized for work, making a mistake which will have consequences, being left alone in a difficult situation, or being treated unfairly (Orth, Robins & Meier, 2009). Stressful events create a sense of crisis as an individual response to urge the effort and investment, necessary to acquire knowledge to overcome the stressful event. By way of comparison, Greve (2003) found that problemistic search (i.e., search that is stimulated by a problem, and is directed towards finding a solution) is triggered when people find that organizational performance is below their aspiration level. Problemistic search results in increased R&D when decision makers judge that upgrading their technology and product portfolio can solve the faced performance problems (Greve, 2003). Likewise, it is expected that when an individual experiences stressful events which are wide in scope and potential impact or are persistent, that person is likely to seek external knowledge (Winter, 2000) by engaging in inbound open innovation practices. As the intensity of the stressful events increases, one is likely to allocate additional resources to develop the ACAP of recognizing and assimilating (i.e., PACAP) by these practices generated knowledge. Therefore, it is expected that stressful events at the workplace will intensify efforts to develop individual PACAP. These efforts are made with the intention of exploiting, to overcome the stressful event or to improve one’s performance.
Literature Gap
The construct of Absorptive Capacity (ACAP) has often been researched in various contexts. However, most attention has been given to the outcomes of ACAP, thereby relatively
16 neglecting its antecedents. Volberda, Foss and Lyles (2010) suggest that researchers should build on prior work to accumulate the knowledge about ACAP as a dependent variable. This study responds to this call by considering individual ACAP as the dependent variable.
Moreover, ACAP is mostly studied on a firm-, business unit-, or team-level of analysis (Jansen, Van Den Bosch & Volberda, 2005). This resulted in the current situation in which individual ACAP is largely under-researched (e.g., Lane et al., 2006; Volberda et al., 2010; Rai & Prakash, 2016). Research has only more recently begun to address the sources and nature of ACAP by investigating its micro-level foundations (Ter Wal, Criscuolo & Salter, 2017). Our understanding of knowledge recognition, assimilation, transformation, and exploitation on lower levels of analysis is currently at best limited (Lowik et al., 2012). Further examination of individuals’ abilities to integrate acquired knowledge into an existing knowledge base is needed (Lowik et al., 2017). Adopting a closer look at the underlying tensions might reveal contributions to successful individual ACAP management (Ojo & Raman, 2016). This study responds to this call by considering ACAP on an individual level of analysis, and thereby stressing the relative importance of individuals in the current debate within ACAP literature (Matusik & Heeley, 2005).
Literature has also stressed the importance of antecedents of individual ACAP, but effects of these antecedents have hardly been empirically examined (Lowik et al., 2012). “Effects of individual differences on the associated dimensions of [individual] ACAP have been overlooked, while data have mostly originated from single respondent or proxy construct” (Ojo & Raman, 2016, p. 724-725). Investigating existing and new antecedents of ACAP, including their relative importance, is desired (Todorova & Durisin, 2007). To the author’s knowledge, individual inbound open innovation practices as a specific antecedent has never been empirically tested in relation to individual ACAP. Also, Volberda, Foss and Lyles (2010,
17 p. 943) state that “research on AC should be explicit about what kind of knowledge is being absorbed”. This survey-based study responds to this call by considering individual inbound open innovation practices as a specific source of external knowledge, being a new antecedent to individual ACAP.
Parallel to a call for individual level ACAP studies (Volberda et al., 2010), a call for individual-level open innovation studies can be detected (Lowik et al., 2017). Thus far, research on open innovation seems confined to innovation-specific journals, focusing on few and selected open innovation issues. In this way, little influence is exerted on the wider business community (Randhawa et al., 2016). A clearer understanding of characteristics and practices of open innovation is needed (Lichtenthaler, 2011). Individual inbound open innovation practices are often complex processes which need to be further systematized (Lichtenthaler, 2011). Interactions of these processes deserve particular attention
(Lichtenthaler & Lichtenthaler, 2009). Randhawa, Wilden and Hoberger (2016, p. 750) call for “a more comprehensive understanding of open innovation by including diverse
perspectives” of, for instance, “users, networks, and communities”. Future studies should increase our understanding of how individuals can be leveraged and managed as sources of innovation (Bogers, Afuah & Bastian, 2010). This study responds to this call by considering inbound open innovation from an individual, practice, and business perspective.
Studies in open innovation need to be sufficiently grounded in prior research on both open innovation and related fields. Lichtenthaler (2011, p. 87) mentioned that “there is a lot to learn from earlier research into related topics” such as “absorptive capacity”. Some attention has been given to ACAP as being instrumental in integrating externally sourced knowledge (West & Bogers, 2014), but few studies provide a theory-grounded explanation of the connection between open innovation and absorptive capacity, and how they cut across at individual levels
18 of analysis (Bogers et al., 2017). Further research should provide understanding of how open innovation practices and interactions of individuals create ACAP (Felin, Foss, Heimeriks & Madsen, 2012). Both open innovation and ACAP, have important complementarities that need much more attention by strategy scholars (Chen, Lin & Chang, 2009; Spithoven, Clarysse & Knockaert, 2011). This study responds to these calls by relating inbound open innovation practices as an antecedent to individual ACAP, and thereby combining both fields of literature.
Lastly, it has been stressed to consider potential internal and external contingency factors in relation to open innovation practices (Lichtenthaler, 2011). In this study, stressful events may be such critical contingency factors which affect the relation between individual inbound open innovation practices and individual PACAP. By looking more into this, boundaries of open innovation practices and ACAP might be identified (Lichtenthaler, 2011). This study responds to this call by considering stressful events as potential moderators in the relation between inbound open innovation practices and individual PACAP.
Research Question
In this study three key variables can be distinguished: Individual inbound open innovation practices as the independent variable, individual ACAP as the dependent variable and stressful events as the moderating variable. By relating the three in the setting written above, this study seeks to explore the relation between individual inbound open innovation practices and individual ACAP, moderated by stressful events. Hence, this leads to the following research question: “What is the relation between individual inbound open innovation practices and individual absorptive capacity and how do stressful events influence this relation?”
19
HYPOTHESES
In this section the hypotheses will be introduced and presented. First, individual inbound open innovation practices will be described as antecedent to both individual PACAP and RACAP. Second, the moderating role of stressful events will be articulated. This section will end with the conceptual model which outlines all the hypotheses.
Individual Inbound Open Innovation Practices as Antecedent to Individual PACAP and
Individual RACAP
This study posits individual inbound open innovation practices as an antecedent for individual ACAP. Key assumption in ACAP literature is that “an organization’s ACAP will depend on the absorptive capacities of its individual members” (Cohen & Levinthal, 1990, p. 131). This points to the assumption that there is a distinction between individual ACAP and
organizational ACAP. Organizational ACAP is widely considered as a dynamic capability (Zahra & George, 2002). Based on this, the micro-foundations theory suggests that
organizational ACAP consists out of routines (Feldman & Pentland, 2003) at the collective level, which are influenced by individuals’ activities at the individual level (Abell, Felin & Foss, 2008).
Congruent with micro-foundational research, this study follows Lowik et al.’s (2012) and Ojo, Raman, Chong, and Chong’s (2014) conceptualizations of individual ACAP as the activities of individuals to recognize, assimilate, transform, and exploit new external knowledge. The activities of recognizing and assimilating new external knowledge is conceptualized as
Potential ACAP (PACAP), whereas the activities of transforming and exploiting new external knowledge is conceptualized as Realized ACAP (RACAP). These conceptualizations differ from most extant literature, in which individual ACAP is conceptualized as a set of
20 Individual inbound open innovation practices involve the opening up to, and establishment of relationships with external agents with the aim to access their competences and knowledge (Chesbrough et al., 2014). They serve as a source of external knowledge for individuals. These practices are assumed to affect all four activities of individual ACAP.
First, individual inbound open innovation practices determine individuals’ ability to recognize and assimilate new, external knowledge. Recognizing the value of this new, external
knowledge is more likely when a connection can be made to knowledge one already possesses (Cohen & Levinthal, 1990). The more one practices inbound open innovation, the more likely that new associations with existing knowledge can be made, therefore enhancing the
recognition of new, external knowledge’s potential value. This, in turn, enhances the
absorption of the newly recognized, external knowledge (Cohen & Levinthal, 1990; Zahra & George, 2002).
Additionally, individual inbound open innovation practices influence one’s locus of search. People tend to practice open inbound innovation in areas they are familiar with and where they have been successful earlier (Zahra & George, 2002). For instance, when one has worked with lead users and with success, it is more likely one will do this again. Another example is when one has purchased licenses, patents, or know-how which brought success, it is more likely one will do this again. Individuals who are experts in a specialized knowledge field tend to search in-depth for new knowledge, while generalists tend to search more broadly (Laursen & Salter, 2006). Individuals who practice open innovation are inclined to search more broadly and are, for this reason, more likely to recognize new opportunities.
Individual inbound open innovation practices also facilitate contact with people outside of an employee’s own organization, such as with lead users, innovation brokers, or trade
organizations (De Araújo Burcharth et al., 2014). The more diversity in the people with whom one practices inbound open innovation, the more likely one is to be exposed to potential new,
21 external knowledge. This positively affects the recognition of new, external knowledge
(Cohen & Levinthal, 1990; Tushman, 1997).
In short, individual inbound open innovation practices affect both recognition and
assimilation activities, which together constitute individual PACAP. Thus, individuals who practice much inbound open innovation, will exhibit a higher level of individual PACAP. This suggests the following hypothesis:
Hypothesis 1a: Individual inbound open innovation practices will be positively related to individual recognition and assimilation of new external knowledge (that is, to individual potential absorptive capacity).
Second, individual inbound open innovation practices determine individuals’ ability to
transform and exploit new, external knowledge. Individual inbound open innovation practices are considered to facilitate knowledge transfer and learning, which are required for
transformation and exploitation activities (Fiske & Taylor, 2016). These practices can enhance innovation and creativity because new insights emerge from combinations of encountered knowledge. Also, diversity of individual inbound open innovation practices facilitate the foundation of a shared knowledge base, which is required for knowledge integration (Grant, 1996). Individuals who practice much inbound open innovation have a broader domain-specific repertoire and can better understand individuals from another domain or expertise (Madhavan & Grover, 1998).
Additionally, during transformation and exploitation activities, individuals integrate the knowledge from individual inbound open innovation practices. The acquired external knowledge in these activities is “easily available and readily applicable (which does not concern the new external knowledge that is acquired through recognition activities)” (Lowik et al., 2017, p. 1324). The more one practices inbound open innovation, the more diverse an
22 individual’s external network becomes. This simplifies finding the required knowledge and makes it more likely that one will engage in transformation and exploitation activities (Lowik et al., 2017).
In short, individual inbound open innovation practices affect both transformation and exploitation activities, which together constitute individual RACAP. Thus, individuals who practice much inbound open innovation, will exhibit a higher level of individual RACAP. This suggests the following hypothesis:
Hypothesis 1b: Individual inbound open innovation practices will be positively related to individual transformation and exploitation of new external knowledge (that is, to individual realized absorptive capacity).
The Moderating Role of Stressful Events on Individual Inbound Open Innovation
Practices as Antecedent to Individual PACAP
The impact of antecedents on PACAP has been found to be moderated by activation triggers (Zahra & George, 2002). Discontinuous or non-linear learning takes place in the presence of such activation triggers (Kim, 1998). Especially when activation triggers are large in scale and impact or when they are persistent, seeking external knowledge is likely. Increasing intensity of a trigger results in increasing investment in building PACAP (Zahra & George, 2002). Activation triggers can be internal or external in nature (Zahra & George, 2002). Given the susceptibility to high degrees of job stress of employees working in the financial services industry (Montgomery et al., 1996) and the individual level of analysis, the internal triggers in this study are stressful events at the workplace. This regards to, for instance, being criticized for work, making a mistake which will have consequences, being left alone in a difficult situation, or being treated unfairly (Orth et al., 2009).
23 When one experiences stressful events at the workplace, existing (cognitive) frames might become broken and what at first made sense no longer does so (Fosfuri & Tribó, 2006). Stressful events trigger the feeling of a need for a different type of knowledge which one currently does not possess. This feeling, also known as problemistic search (Greve, 2003), encourages search activities (Cyert & March, 1963) and stimulates learning (Winter, 2000). Stressful events can therefore make it more attractive for individuals to expend resources in developing the capabilities of recognizing and assimilating externally generated information (e.g., PACAP). They are hypothesized to intensify the effort of searching external knowledge, and to favor the assimilation of that knowledge (Fosfuri & Tribó, 2006). Hence, stressful events, if experienced, will increase the impact of individual inbound open innovation practices on individual PACAP. This suggests the following hypothesis:
Hypothesis 2: The positive relationship between individual inbound open innovation practices and individual potential absorptive capacity (PACAP) is moderated by stressful events, so that this relationship is stronger for higher levels of stressful events.
Conceptual Model
The hypothesized relationships are illustrated in the conceptual model of figure 1.
24
METHOD
In this section the methodological set-up of this study will be explained. First, the overall research setting and the process of data collection will be discussed. This section will end with elaboration on the used measurement constructs and their validation.
Setting and Data Collection
Since it was not possible to have a sampling frame, a non-probability convenience sample has been used. The empirical research was conducted at firms operating in the Dutch financial services industry (e.g., banking, insurance, pension funds, stock exchange and consulting (Dutch Authority for the Financial Markets, 2018)). This entails primarily the private sector, but also the Dutch Authority for the Financial Markets (AFM). Most recent figures reveal that in 2016, the Dutch financial sector had total assets of 10,446,142 million euros (Central Bureau for Statistics, 2018a). Key requirement was that respondents needed to work in the same industry in order to control for contextual and environmental aspects (Lowik et al., 2016). Since this study examines innovation and knowledge absorption, the (Dutch) financial services industry is considered to be an adequate sample, because this particular industry regards knowledge use and innovation as sources of competitive advantage (Zaheer & Mosakowski, 1997).
After a pilot-test among five potential respondents, some adaptations in question formulations were made and complex concepts were clarified with explanatory sentences. Afterwards, the online Qualtrics survey (see Appendix A) was administered in April 2018 to the employees of firms operating in the Dutch financial services industry. Reviewed survey invitations were sent via e-mail and LinkedIn, with a cover letter from the researchers. Participation was voluntary and the survey was completely anonymous, because it was believed that this would
25 achieve a higher response rate and alleviate potential confidentiality concerns. After one reminder a total of 197 questionnaires were returned. Our final sample was reduced to 110 respondents, due to incomplete responses.
Respondents were capable of reading English and not required to sell products and/or services directly to clients, since staff and support functions were also involved. A percentage of 63.6% of the respondents was male, 29.1% belonged to the age category of 40-49 years old, and 48.2% attained higher professional education (HBO) as their highest level of education. Roughly three-quarters of the respondents (73.6%) worked in a firm with more than 250 employees. With regard to organizational departments, 21.8% was working for
accounting/finance, 12.7% for marketing/communication, and 14.5% for IT support. The average size of the teams was 22.51 (SD = 48.38) full-time employees. The respondents had a mean company tenure of 10.89 years (SD = 10.06) and a mean function tenure of 5.15 years (SD = 5.76).
A Harman's one factor test was performed on items included in the regression model to examine whether common method bias augmented relationships. It showed multiple factors, and the first factor did not account for the majority of the variance (see Table 4 of Appendix B). So there was little concern about potential problems associated with common method bias (Podsakoff & Organ, 1986).
Measurement and Validation of Constructs
This study used existing scales from previous research to ensure validated instruments. Prior to running the analyses, reliability and validity of the scales have been checked. All items of the measurements can be found in Appendix C.
26 Individual absorptive capacity. To measure individual Absorptive Capacity (ACAP), a scale based on existing items from Lowik, Kraaijenbrink and Groen (2016) was used. Respondents were asked to answer the items by a seven-point strongly disagree/strongly agree scale. Individual ACAP consists of Potential Absorptive Capacity (PACAP) and Realized
Absorptive Capacity (RACAP). Individual PACAP consists of recognition and assimilation of new external knowledge. Four items assessed the intensity and direction of efforts expended in knowledge recognition. In addition, three items measured assimilation and gauged the extent to which individuals were able to analyze and understand new external knowledge. Individual RACAP consists of transformation and exploitation of new external knowledge. Four items measured transformation and assessed the extent to which individuals were able to facilitate recognizing opportunities and consequences of new external knowledge. In addition, three items tapped into the extent to which individuals were able to exploit new knowledge. The scale gauged the ability of individuals to incorporate new external knowledge.
In ACAP theory, evidence exists for both a four factor solution (recognition, assimilation, transformation, and exploitation) (Lowik et al., 2016) and a two factor solution (PACAP and RACAP) (Zahra & George, 2002). Therefore, an exploratory factor analysis (EFA) via a principal component analysis (PCA) was conducted on the scales. The Kaiser–Meyer–Olkin measure verified the sampling adequacy for the analysis, KMO = .813. Bartlett’s test of sphericity χ² (78) = 403.188, p <.001, indicated that correlations between items were sufficiently large for PCA. An initial analysis was run to obtain eigenvalues for each
component in the data. Four components had eigenvalues over Kaiser’s criterion of 1, but the scree plot revealed a levelling off after the second factor. Also, the factor loadings after rotation roughly pointed to two factors. The first item of recognition “I am always actively looking for new knowledge for my work” was excluded, which slightly improved the bifurcation of two factors. Excluding more items was not possible, since this would result in
27 less than three items (Yong & Pearce, 2013) measuring the underlying subdimension of recognition. This factor would then not be able to measure individual PACAP anymore, because of inferior measurement of recognition and excessive focus on measurement of assimilation. Besides, this is undesirable since it is appreciated to keep as many of the original items as possible, and to avoid modifying the scales. Although a four factor solution was also produced (see Appendix D1), a two factor solution proved to be considerably congruent with the two factors and therefore superior (see Appendix D2). Thus, two factors were retained and rotated with an Oblimin with Kaiser normalization rotation and in combination explained 45.43% of the variance. The items that cluster on the same factors suggest that factor 1 represents realized absorptive capacity, and factor 2 potential absorptive capacity. As the results suggest, the item ‘I intentionally search for knowledge in many different domains to look “outside the box”’ of individual PACAP shows high cross-loadings on the factor of individual RACAP as well. This could be due to the content of the item, since the words “in many different domains” might be interpreted as both externally search for knowledge (as congruent with individual PACAP) and internal transformation (as congruent with individual RACAP). Though it should be remarked that future research should test again whether the items really fit the scale. Both scales of individual PACAP (α = .71) and individual RACAP (α = .79) have proven to be reliable.
Individual inbound open innovation practices. To measure individual inbound open innovation practices, employees were asked to indicate which inbound open innovation practices they used within the past three years. The final measure was generated from the “inbound” dimension of the “open innovation practices” measure from De Araújo Burcharth, Knudsen and Søndergaard (2014). The initial selection of these inbound practices was based on a literature review (Bahemia & Squire, 2010; Chesbrough & Garman, 2009; Van de
28 Vrande et al., 2009) and resulted in an extensive rather than compressed list of practices, captured in eight items. The items were measured on a three-point never/often scale.
A PCA was conducted on the scales. The Kaiser–Meyer–Olkin measure verified the sampling adequacy for the analysis, KMO = .740. Bartlett’s test of sphericity χ² (28) = 203.222, p <.001, indicated that correlations between items were sufficiently large for PCA. An initial analysis was run to obtain eigenvalues for each component in the data. Two components had eigenvalues over Kaiser’s criterion of 1 and in combination explained 53.68% of the variance. In agreement with Kaiser's criterion, examination of the scree plot revealed a levelling off after the second factor. Two factors were rotated with an Oblimin with Kaiser normalization rotation. Appendix D3 shows the table of the factor loadings after rotation.
However, only two items loaded on factor 2. In general, to keep the factor it should have at least three items to be retained (Yong & Pearce, 2013). “Models with factors that have only two indicators are more prone to estimation problems, especially when the sample size is small” (Kline, 2005, p. 172). This implies that even though this second factor was identified, it would probably not be very stable because only two items are loading on it. Thus, the factor structure identified was not optimal because, contrary to the expectations that a single factor will be extracted, two items demonstrated high factor loading on another factor. However, since only two items loaded on this second factor, it was decided to not consider it as a
separate factor and still consider one factor solution as the most appropriate. This single factor explained 37.94% of the variance. Since the scale is based on a literature review (De Araújo Burcharth et al., 2014) and more often used (Oltra, Flor & Alfaro, 2018), it was decided to not discard these two items. Though it should be remarked that future research should test again whether the items really fit the scale. In this study, the scale has proven to be reliable (α = .75).
29 Stressful events. To measure stressful events, a scale based on existing items from Orth, Robins and Meier (2009) was used. Five items measured the extent to which employees experienced stressful events at their workplace within the past three years. Items were measured on a five-point never/very often scale.
A PCA was conducted on the scales. The Kaiser–Meyer–Olkin measure verified the sampling adequacy for the analysis, KMO = .690. Bartlett’s test of sphericity χ² (10) = 72.478, p <.001, indicated that correlations between items were sufficiently large for PCA. An initial analysis was run to obtain eigenvalues for each component in the data. One component had
eigenvalues over Kaiser’s criterion of 1 and explained 42.28% of the variance. In agreement with Kaiser's criterion, examination of the scree plot revealed a levelling off after the first factor. Thus, one factor was retained and rotated with an Oblimin with Kaiser normalization rotation. Appendix D4 shows the table of the factor loadings after rotation. The items that cluster on the same factor suggest that factor 1 represents Stressful Events. The scale was reasonably reliable (α = .64), which exceeded the reliability of the scale in its original setting (α = .61). However, in this paper it was noted that “coefficient alpha is not an appropriate indicator of reliability for event scales” (Orth et al., 2009, p. 314).
Control variables. Besides gender (0 = male, 1 = female) (Lowik, Kraaijenbrink & Groen, 2012), age (clustered in years), company tenure (in years), and supervisory role (0 = no, 1 = yes) (Lowik et al., 2017) there were two other variables to control for. Firm size is expected to impact the extent to which firms - and therefore their employees - engage in open practices positively (Huizingh, 2011). Furthermore, larger firms may have more resources to let their employees recognize and assimilate new external knowledge (Jansen et al., 2005). Its measure was constructed as the number of employees, which is acknowledged as an appropriate
measure of overall firm size in a given industry (Audia & Greve, 2006). This resulted in three categories: Less than 50 employees, less than 250 employees, and 250 or more employees
30 (Central Bureau for Statistics, 2018b). Job rotation is assumed to enhance external knowledge recognition, assimilation, and transformation (Jansen et al., 2005). It augments one’s contacts, capacity for establishing new associations (Cohen & Levinthal, 1990), and awareness of employees' knowledge and skills in other functional areas (Campion, Cheraskin & Stevens, 1994). Whether a respondent had engaged in job rotation in the past three years was measured by a dichotomy (0 = no, 1 = yes).
ANALYSIS AND RESULTS
In this section the descriptive statistics, correlations, and results of this study will be reported. The results will be clarified by Table 1, 2, and 3.
Table 1 presents descriptive statistics and correlations for the study variables. The bivariate relationships were examined using Pearson correlation coefficient. Considering the
correlations between the independent and the dependent variables, individual inbound open innovation practices shows a medium, positive correlation with both individual PACAP (r = 0.28) and individual RACAP (r = 0.33). The two dependent variables of individual PACAP and individual RACAP show a strong, positive correlation (r = 0.54)Considering the correlation between the dependent variable and moderator, individual PACAP shows a medium, negative correlation with stressful events (r = -0.35). Considering the control
variables, gender shows a small, negative correlation with individual inbound open innovation practices (r = -0.21). Company tenure shows a small, positive correlation with individual PACAP (r = 0.20), whereas job rotation shows a small, negative correlation with individual PACAP (r = -0.19).
31 The hypotheses 1a and 1b were tested via multiple hierarchical regressions in SPSS. The specific conditional effect of hypothesis 2 was tested by use of Process (Hayes, 2012). To examine the issue of multicollinearity, the variance inflation factors (VIFs) in each of the regression equations was calculated. The maximum VIF within the models was 2.453, which was well below the rule-of-thumb cut-off of 10 (Neter, Wasserman & Kutner, 1990).
32 Table 1
Means, Standard Deviations, and Correlations
Variable M SD 1 2 3 4 5 6 7 8 9 10 11
1. Potential Absorptive Capacity 5.56 0.66 (.71)
2. Realized Absorptive Capacity 5.00 0.85 .54** (.79)
3. Individual Inbound Open Innovation Practices 1.73 0.40 .28** .33** (.75) 4. Stressful Events 2.05 0.54 -.35** -.10 -.05 (.64) 5. Gender .36 .48 -.17 -.12 -.21* -.03 6. Age 30-39 .19 .40 -.03 -.05 .07 .21* .16 7. Age 40-49 .29 .46 .12 .16 -.06 -.09 -.11 -.31 ** 8. Age ≥ 50 .20 .40 .16 .04 .17 -.09 -.14 -.24* -.32** 9. Company Tenure 10.89 10.06 .20 * .07 -.07 -.14 -.09 -.15 .10 .55** 10. Supervisory Role .31 .46 .05 .19 * .12 .16 -.30** -.13 .22* .16 .20* 11. Firm Size .74 .44 .04 -.01 -.15 -.16 -.02 .03 .11 -.01 .22 * -.18 12. Job Rotation .28 .45 -.19* -.01 -.11 .27** .03 .06 .04 -.21* -.09 .11 -.04 * Correlation is significant at p <.05 (two-tailed).
** Correlation is significant at p <.01 (two-tailed).
33
Results
The results of the hierarchical regression analyses for individual inbound open innovation practices and individual PACAP are reported in Table 2, and individual RACAP in Table 3.
Standardized coefficients and standard errors were reported. Prior to the creation of the interaction term in Model 3 (Table 2), the independent variable and moderator were standardized by creating Z-variables in SPSS, in order to avoid multicollinearity issues (Aiken & West, 1991). The baseline models (Models 1 and 4) contain control variables. Models 2 and 5 introduce individual inbound open innovation practices as an antecedent of individual PACAP and RACAP, and Model 3 examines the moderating effect of stressful events on individual PACAP.
As expected, Model 2 shows that individual inbound open innovation practices had a positive and significant effect on individual potential absorptive capacity (β = 0.221, p = 0.022). Hypothesis 1a, which posits a positive relationship between individual inbound open innovation practices and individual PACAP, was supported. Regarding the effect of individual inbound open innovation practices on individual realized absorptive capacity, Model 5 shows that the coefficient was positive and significant (β = 0.349, p = 0.001). Hypothesis 1b, which posits a positive relationship between individual inbound open innovation practices and individual RACAP, was supported. As shown in Model 3, the interaction between individual inbound open innovation practices and stressful events did increase individual potential absorptive capacity (β = 0.009, p = 0.887). Although the coefficient was slightly positive, it was not significant. Thus, hypothesis 2, which posits a moderating effect of stressful events on the positive relationship between individual inbound open innovation practices and individual PACAP, was not supported.
34 Table 2
Results of Hierarchical Regression Analyses: Effects of Antecedents on Individual Potential Absorptive Capacity Individual PACAP
Model 1 Model 2 Model 3
Variables Control SE Main SE Interaction SE
Gender -0.15 0.14 -0.12 0.13 -0.16 0.13 Age 30-39 0.12 0.18 0.14 0.18 0.23 0.20 Age 40-49 0.20 0.18 0.16 0.17 0.23 0.17 Age ≥ 50 0.15 0.23 0.08 0.22 0.14 0.25 Company Tenure 0.11 0.01 0.12 0.01 0.01 0.01 Supervisory Role -0.05 0.15 -0.01 0.14 -0.01 0.15 Firm Size -0.02 0.15 -0.03 0.14 -0.05 0.15 Job Rotation -0.16 0.14 -0.07 0.14 -0.10 0.15
Individual Inbound Open Innovation Practices 0.22* 0.16 0.15* 0.07
Stressful events -0.32* 0.12 -0.21* 0.07
Individual Inbound Open Innovation Practices x Stressful Events 0.01 0.06
Model F 1.64 3.38* 4.02**
R² .12 .26 .26
ΔR² .14 .00
Note: Standardized regression coefficients are reported. * p <.05 (two-tailed), ** p <.01 (two-tailed).
35 Table 3
Results of Hierarchical Regression Analyses: Effects of Antecedents on Individual Realized Absorptive Capacity Individual RACAP
Model 4 Model 5
Variables Control SE Main SE
Gender -0.06 0.18 0.01 0.18 Age 30-39 0.04 0.24 -0.03 0.23 Age 40-49 0.17 0.24 0.12 0.23 Age ≥ 50 0.08 0.31 -0.04 0.30 Company Tenure -0.02 0.01 0.07 0.01 Supervisory Role 0.13 0.20 0.12 0.19 Firm Size -0.00 0.20 0.04 0.19 Job Rotation -0.02 0.19 0.01 0.18
Individual Inbound Open Innovation Practices 0.35** 0.21
Model F 0.79 2.13*
R² .06 .16
ΔR² .10
Note: Standardized regression coefficients are reported.
36
DISCUSSION AND CONCLUSION
In this section, the significance of the study findings will be discussed. First, the implications section will highlight the major findings. Second, the theoretical and practical contributions will be expressed. This section will end with the limitations of this study and suggestions for future research.
Implications
The concept of ACAP has received substantial attention in academic research (Jansen et al., 2005), yet our understanding of individual ACAP and of its antecedents remains rather unclear. Although prior research suggested “to be explicit about what kind of knowledge is being absorbed” (Volberda et al., 2010, p. 943), empirical examinations at an individual level have failed to appear. Moreover, a potentially moderating role of stressful events on the effectiveness of individual PACAP has never been empirically examined. Focusing on the Dutch financial services sector, the objective of this study was to explore the relation between individual inbound open innovation practices and individual absorptive capacity, and the potential influence of stressful events in this relation.
This study underscores previous assertions that external knowledge sources significantly influence individual ACAP (Zahra & George, 2002). Prior knowledge diversity and external network diversity have been studied recently and are supposed to positively influence
individual ACAP (Lowik et al., 2012; Lowik et al., 2017). However, the relationship between specific external knowledge sources and individual ACAP remains rather unclear. The findings of this study demonstrate that individual inbound open innovation practices
positively and significantly influence both individual PACAP and individual RACAP. Hence, on the one hand, individual inbound open innovation practices increase recognition and
37 assimilation of new, external knowledge. On the other, these practices increase transformation and exploitation of new, external knowledge.
Prior research has argued that internal activation triggers moderate the impact of knowledge sources on ACAP development (Zahra & George, 2002, p. 193). Contrary to this study’s prediction, individual inbound open innovation practices in stressful workplace-related circumstances do not significantly increase the recognition and assimilation of new external knowledge (i.e., individual PACAP). Although a slightly positive effect was detected, this effect was not significant. This could be due to the used convenience sample of this study. One possible explanation could be that the sample has been too small, resulting in too little statistical power to significantly approve the prediction. Another possible explanation could be that the sample captured too little variance in the measurement of stressful events due to, for instance, socially desirable answering. This would explain the slightly positive, but non-significant effect. To further investigate the potentially moderating role of stressful events at the workplace, future research is encouraged to realize a larger sample, with more variance in scores on experienced workplace-related stressful events. A larger sample would also be beneficial for explaining influences of background characteristics on the outcomes. In this study, none of the control variables showed a significant influence in the regression analyses.
Contributions
This study contributes to research on ACAP and combinative capabilities in several ways. First, by examining individuals’ inbound open innovation practices and their influences on individual ACAP, combined with the key assumption that individual ACAP is a micro-foundation for organization-level ACAP (Cohen & Levinthal, 1990), this study extends ACAP theory to better understand its relation to open innovation practices as a new antecedent. This micro-foundational lens is useful in understanding the underlying
38 mechanisms of individuals’ abilities to absorb new knowledge (Ter Wal et al., 2017). It
enables to better understand how absorptive capabilities are created and why they might differ (Ojo & Raman, 2016). Since external knowledge is needed for the execution of ACAP
(Cohen & Levinthal, 1990; Zahra & George, 2002), differences in external knowledge explain why some ACAP is more successful than others. This study is among the few to have taken an individual level perspective, and is the first to show that individuals’ external knowledge absorptive capabilities are increased by individual inbound open innovation practices. In this way, it empirically examined the effects of this new antecedent (Lowik et al., 2012; Todorova & Durisin, 2007), it underscores the importance of knowledge sources (Zahra & George, 2002) for individual ACAP (Lowik et al., 2017), and is explicit about what knowledge is being absorbed (Volberda et al., 2010). Also, it stresses the relative importance of individuals in the debate within ACAP literature, by highlighting the locus of knowledge creation at the individual level (Matusik & Heeley, 2005).
Second, this study contributes to the relatively undeveloped individual open innovation theory. It provided the new insights that individuals’ inbound open innovation practices increase the ability of recognizing and assimilating external knowledge (PACAP), and the ability of transforming and exploiting external knowledge (RACAP). By examining inbound open innovation from a micro (Ahn et al., 2017) and practice perspective (De Araújo
Burcharth et al., 2014; Lichtenthaler, 2011), with a specific focus on practices (Enkel et al., 2009; Lichtenthaler, 2011), and as a new antecedent to individual ACAP (Lowik et al., 2012; Todorova & Durisin, 2007) in a business context (Randhawa et al., 2016), this study provides further understanding of open innovation literature. Particularly, it tapped into “the human side” of open innovation (Gassmann et al., 2010), by focusing on the role of individuals in
39 inbound open innovation practices and knowledge flows across the boundaries of a firm (Lichtenthaler, 2011).
From a practical perspective, also some contributions can be noted. Because this study proved individual inbound open innovation practices as antecedent to individual ACAP, managers might acknowledge their relevance and encounter suggestions on how to enhance employee ACAP. The findings might help managers to design knowledge management practices to advance employees’ absorptive capacities (Lowik et al., 2017). For instance; stimulating the use of inbound open innovation practices by investing in training or education, providing access to sources such as technical magazines, or setting up exchange programs with lead users. With regard to a firm’s hiring processes, managers might also assess the level of inbound open innovation practices one potential employee has shown in previous jobs, to get a sense of one’s individual ACAP. Staffing employees with high individual ACAP implies that these employees have broad knowledge and experience and a diverse personal network to access new knowledge (Lowik et al., 2012).
Limitations and Future Research Directions
Several limitations, which provide meaningful directions for future research, virtue discussion. First, this quantitative study relies on a correlational design, which impedes us from inferring causal relationships (Bono & McNamara, 2011). Together with the cross sectional set-up, this results in low internal validity. ACAP is typically viewed as a construct that captures the evolution of learning and utilization of new knowledge that accumulates over time (Cohen & Levinthal, 1990). It could also take three years before the positive effect of open innovation practices materialize (Lichtenthaler & Lichtenthaler, 2009). The cross-sectional survey used cannot relate individual absorption efforts to specific instances of knowledge transfer and absorption. This study cannot observe how external knowledge
40 absorption unfolds over time, which leaves the developmental and path-dependent character of the construct quite untouched. Longitudinal research is desired for studying individual ACAP and its dynamics.
Second, Although Harman's one factor test proved sufficient, data is collected by only one means – an online survey – which can cause common method bias (Podsakoff, MacKenzie & Podsakoff, 2012). A deeper understanding of the way in which individual inbound open innovation practices influence individual ACAP can be reached by qualitative research, for example by conducting interviews (Harrison, 2013). Moreover, this study relies on self-reporting on own capabilities. This is a highly subjective method and may stimulate social desirable answering (Meade & Craig, 2012). However, self-reported data sources are inevitable because objective measures of individual inbound open innovation practices and individual ACAP hardly exist. Though, future research is highly encouraged to incorporate objective data from other sources if possible.
Third, a major limitation of this study can be found in the used measurements. Unfortunately, the measurement of individual ACAP (Lowik et al., 2016) is relatively new and has not been published in any peer-reviewed journal yet. This leaves the measurement non-previously validated which threatens its reliability, although previous reliability tests have proven sufficient. Originally, individual ACAP was operationalized as a second-order construct consisting of four individual dimensions (recognition, assimilation, transformation, and exploitation). Previous Kendall’s tau examinations showed moderate to strong correlations between the four dimensions (Lowik et al., 2012). In addition, previous confirmatory factor analyses showed low distinctiveness of the four dimensions. So, although the four dimensions of individual ACAP are conceptually distinctive, they are too strongly interrelated to show sufficient discriminant validity. Also, with regard to the measurement of individual inbound
