Professional Documents
Culture Documents
1 s2.0 S0048733321002055 Main
1 s2.0 S0048733321002055 Main
Research Policy
journal homepage: www.elsevier.com/locate/respol
A R T I C L E I N F O A B S T R A C T
Keywords: Extant research explores the role played by individuals and, in particular, founders in defining open innovation
Open innovation strategies at the firm level. We join this discussion by combining insights from imprinting literature that explores
Imprinting theory the enduring impact of a founder’s personal history, with inputs from literature that stresses the impact of past
Authoritarian regimes
experience on trust formation. We suggest that founders are less likely to engage in open innovation if their
R&D cooperation
Surveillance
experiences engender a generalized lack of trust. We use a unique database that includes East and West German
Organizational decision-making founders to identify regional differences in activities conducted by authoritarian regimes that could inhibit trust.
We find that founders who were exposed to high levels of secret police surveillance in the former socialist
German Democratic Republic (GDR) are less likely to engage in interfirm R&D cooperation. We contribute to the
literature on open innovation by exploring how a founder’s social, political, and cultural backgrounds influence
strategic decisions related to open innovation, and to recent imprinting literature by showing that variation in
oppressive enforcement practices in authoritarian regimes, such as surveillance activities, can leave an enduring
imprint. Our findings complement recent insights on ideological imprinting effects on young firms’ decision-
making.
1. Introduction and conceptual background industry and/or organizational level (e.g., de Faria et al. 2020, 2010,
Keupp and Gassmann 2009), some scholars have recently shifted their
“The net of spies was really broad. In nearly every institution, even in the attention toward the micro-foundations of open innovation (Bogers
churches […], there were many of them. I remember telling a leading Stasi et al., 2018), and explore the role that individuals and, in particular,
(secret police) officer, ’If you had sent an informant to me, I would surely founders play in leading an organization to adopt an open innovation
have recognized him.’ His-answer was, ‘We didn’t send anyone. We took strategy (Classen et al., 2012; Eftekhari and Bogers, 2015). Hahn et al.
those who were around you.’" Hubertus Knabe (former director of the Stasi (2019), for instance, provide evidence that the professional background
prison memorial) (Knabe, 2014). of founders, specifically the presence of scientists in the founder team,
In the last decades, closed innovation systems characterized by firms often leads to imprinting certain routines that encourage open innova
conducting most R&D activities internally, are transitioning to open tion practices.
innovation systems in which firms manage knowledge flows and search Even though existing research has increased our understanding of
for knowledge sources with external actors outside their organizational how founders’ human and social capital and their professional experi
boundaries (cf., Chesbrough and Bogers 2014, Chesbrough 2003, 2006). ence determine open innovation strategies, we know very little about if
This transition triggered researchers in innovation studies to explore the and how founders’ social, political, and cultural backgrounds also play a
open innovation phenomenon (cf., Dahlander and Gann, 2010, West role regarding open innovation decision making. We build on research
et al. 2014). They have mainly focused their attention to the identifi that links a founder’s characteristics to open innovation and focus on
cation of open innovation’s benefits and risks (cf., Dahlander and Gann how a founder’s socioeconomic environment affects the firm’s likeli
2010, Ebersberger et al. 2021, Srinivasan et al. 2021), and to the un hood to adopt open innovation strategies. More specifically, we inves
derstanding of why certain firms embrace open innovation and others do tigate how socialization in an authoritarian regime imprints a founder’s
not (Lee et al., 2010). While such discussions are often rooted in an later strategic decision-making at the firm level in a free-market
* Corresponding author.
E-mail addresses: m.wyrwich@rug.nl (M. Wyrwich), p.j.steinberg@rug.nl (P.J. Steinberg), f.noseleit@rug.nl (F. Noseleit), p.m.m.de.faria@rug.nl (P. de Faria).
https://doi.org/10.1016/j.respol.2021.104409
Received 9 February 2021; Received in revised form 11 October 2021; Accepted 18 October 2021
Available online 3 November 2021
0048-7333/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
M. Wyrwich et al. Research Policy 51 (2022) 104409
economy. tend to have lower levels of generalized trust (Bjørnskov, 2006; Rainer
As illustrated by the quote at the beginning of the introduction, and Siedler, 2009). Our conjecture that oppressive surveillance activ
authoritarian regimes actively limit individual freedom and the free ities conducted by authoritarian regimes negatively affect generalized
exchanges of ideas by implementing comprehensive surveillance activ trust levels is based on the work of Hardin (2006), who outlines that
ities. The purpose of these activities is to control and shape individuals’ trustworthiness gives rise to trust. Consequently, the oppressive
views about how society and the economy should be organized (Svolik, behavior of regimes negatively affects the general level of trustworthi
2012; Gerschewski, 2013). For citizens living under such regimes, these ness (Paldam and Svendsen, 2001), and lowers an individual’s gener
surveillance activities create low trust environments where individuals alized trust levels. As individuals’ trust levels typically remain constant
view each other with suspicion (Lichter et al., 2021). over time (Algan and Cahuc, 2010), this adverse effect has long-run
In the second half of the 20th century, several South American (e.g., consequences and lower trust levels remain even when an individual
Chile, Argentina, and Brazil), European (e.g., Spain, Czech Republic, and no longer lives under an oppressive regime (Bjørnskov, 2006; Rainer and
East Germany), Asian (e.g., South Korea), and African (e.g., South Af Siedler, 2009).
rica) countries transitioned from authoritarian regimes with closed These enduring consequences also play a central role in the
economies to democracies with market economies. Since the contrast conceptualization of imprinting that considers historical factors
between systems is stark, these transitions led to economic and social (Marquis and Tilcsik, 2013; Simsek et al., 2015). To explore the effect of
challenges that attracted the attention of economics and management imprinting in the context of open innovation strategies, we focus on an
scholars (e.g., Aslund and Djankov 2014, Svolik 2012). These scholars open innovation strategy where trust is essential: the decision to engage
investigated the interaction between long-term imprinting caused by in R&D cooperation with other organizations. From all the different
authoritarian regimes’ efforts to control individual behavior, and an types of open innovation (for an overview, see Dahlander and Gann
individual’s behavior as an economic actor even after transitioning to 2010), inter-organizational cooperation is specifically dependent on
more open societies (Alesina and Fuchs-Schuendeln, 2007; Rainer and trust and cooperative behavior (Nooteboom et al., 1997; Faems et al.,
Siedler, 2009). 2008; Ertug et al., 2013). Due to the high degree of endogenous project
Marquis and Qiao (2020) focus on how exposure to authoritarian uncertainty and a high likelihood for incomplete contract protection,
regimes imprints young firms’ internationalization decisions, and their opportunistic behavior in R&D cooperation partnerships is difficult to
evidence begs for further research on authoritarian regimes’ imprint on avoid via hierarchical or markets mechanisms (Mayer and Nickerson,
founders’ firm-level strategic choices. We integrate the insights from the 2005; Ulset, 1996). Trust, therefore, plays an important role in the de
research on authoritarian regimes’ imprinting into the literature on cision to engage in R&D cooperation activities because it increases a
open innovation and argue that an authoritarian regime’s surveillance is firm’s willingness to accept vulnerability towards a third party (Ertug
a trust-destroying mechanism that negatively affects a founder’s later et al., 2013).
decision to participate in open innovation activities. Our assertion is While relational trust between cooperating partners is critical (e.g.,
based on imprinting theory, which emphasizes the lasting influence of Krishnan et al. 2006, Lee and Cavusgil 2006), a more generalized trust
historical experiences on individual and organizational behaviors and level within an organization is likely to affect its decision to initiate
outcomes (Stinchcombe, 1965; Marquis and Tilcsik, 2013; Simsek et al., cooperative activities with other organizations. For example, Ferrin and
2015; Ferriani et al., 2020). Our argument relies on two related ideas. Gillespie (2010) show that individuals from countries with high gener
First, that a founder’s individual experiences, skills, and personal alized trust levels will be more cooperative than individuals from
characteristics influence subsequent strategic choices (Burton and low-trust countries. Because trust occurs between individuals rather
Beckman, 2007; Hahn et al., 2019; Bryant, 2014; Roche et al., 2020). than organizations (Zaheer et al., 1998), the personal experiences of
Second, that an individual’s socioeconomic context shapes these char central actors can have a substantial influence on how an organization’s
acteristics (Tabellini, 2010). Bringing these two ideas together, we claim cooperative behavior evolves. Ertug et al. (2013) show that high levels
that a founder’s strategic choices are shaped by her socioeconomic of generalized trust within a society positively influence firms’ perceived
context (Waguespack et al., 2018). Indeed, empirical research shows trustworthiness of specific partners. Thus, we expect that erosion of
that growing up in a specific political regime can leave a long-lasting generalized trust levels among an organization’s founding members is
imprint on an individual (e.g., Alesina and Fuchs-Schuendeln 2007, likely to impede the adoption of open innovation and cooperative R&D
Lichter et al. 2021). Based on the evidence that not all individuals living strategies.
under an authoritarian regime have similar experiences (e.g., Lichter We address recent calls for a better understanding of the micro-
et al. 2021), we claim that the imprinting of individuals stemming from foundations of open innovation by combining the primary notion of
being exposed to the socioeconomic context of a particular authoritarian imprinting literature that prior history can have an enduring impact on
regime is not uniform. We further argue that imprinting stemming from strategic choices with the insights from literature of generalized trust
differences in exposure to oppressive regime enforcement practices is a that emphasizes the importance of past experiences on trust formation.
mechanism that inhibits an individual’s ability to develop generalized Our analysis makes two important contributions. First, building on
trust. recent research on imprinting mechanisms, we offer a more fine-grained
Trust can be defined as “a more or less well-grounded expectation analysis of the context and improve our understanding of the imprinting
about the preferences of other people” (Delhey and Newton, 2005, 311), environment (Simsek et al., 2015; Marquis and Qiao, 2020). Marquis
and is considered to be a central determinant of social cooperation (e.g., and Qiao (2020), for example, focus on the (uniform) ideological
Hardin 2006, Mayer et al. 1995). Higher levels of trust lead to a greater imprinting of authoritarian regimes. Ideological imprinting refers to the
willingness to accept the vulnerabilities of a third party (Ertug et al., process of imprinting a “perceptual framework of interrelated ideas and
2013). While there are different classifications of trust, we focus on beliefs” (Wang et al., 2019, p. 1113). In contrast to Marquis and Qiao
trusting strangers (Rotter, 1971; Dinesen and Bekkers, 2017), referred to (2020), we shed light on imprinting caused by an oppressive regime’s
as generalized or social trust (Kramer and Lewicki, 2010; Nannestad, enforcement practices. We empirically separate the effect of uniform
2008). It is not entirely clear when and/or how individuals form ideological imprinting from the effect of variation in exposure to an
generalized trust levels, however, recent research by Glanville and authoritarian regime’s oppressive actions. We exploit an empirical
Paxton (2007) supports the social learning perspective that suggests that setting that is particularly pertinent to understand the imprinting effects
generalized trust is formed by a variety of social interactions (Burns of authoritarian regimes’ surveillance practices on R&D cooperation. By
et al., 2003; Rotter 1971). Although generalized trust varies across so doing, we expand our understanding of the micro-foundations of open
cultures and countries (Ferrin and Gillespie, 2010; Huff and Kelley, innovation in new ventures by highlighting the role of a founder’s prior
2003), it has been shown that individuals living in a communist system experiences in strategic decisions. We expect that those founders
2
M. Wyrwich et al. Research Policy 51 (2022) 104409
experiencing more severe levels of surveillance and control will have 2.2. Sample
lower levels of trust that affect their decision to adopt open innovation
strategies. Second, we give a more nuanced perspective to the notion Our analysis uses data from the GAW-ADIAB survey (also used in
that trust plays a central role in open innovation and R&D collaboration previous research, e.g., Bublitz et al. 2018), a representative survey of
(Carson et al., 2003; Brockman et al., 2018). Focusing on generalized East and West German founders across six sample regions (for details,
trust, we complement research on how relational trust impacts the see Wyrwich et al. 2018). The sample is based on the addresses of new
performance of collaborations (Faems et al., 2008), by suggesting that firms in manufacturing and knowledge-intensive business services
generalized trust may be of crucial importance, specifically in the (KIBS) and is taken from the Establishment History Panel (BHP)
context of new ventures. collected by the German Federal Employment Agency’s Institute for
Employment Research. The BHP includes information on all German
2. Empirical approach firms with at least one employee. All new manufacturing firms in the
sample regions that appeared in the BHP between 2003 and 2007 for the
2.1. Study context first time1 and were still active in 2010 were contacted for the survey.
For KIBS, about 75–80% of the respective new start-ups were contacted.
We study founders in East and West Germany about three decades The overall response rate was 31%, which is more than twice as large as
after German reunification to test our prediction that firms founded by in the Global Entrepreneurship Monitor adult population survey, a
individuals who experienced high surveillance levels during the socialist standard dataset for the analysis of entrepreneurship (GEM, 2020). The
era are imprinted with low levels of trust and will be less likely to GAW-ADIAB sample of technology-oriented manufacturing and
cooperate with other firms. Before reunification, East Germany, the knowledge-intensive business service firms (based on the NACE sector
former German Democratic Republic (GDR), was part of the Soviet bloc classification) comprises 772 observations.2
and had a rigid authoritarian regime. In 1990, East Germany adopted In order to ensure that the focal founder was socialized in the GDR
the formal institutional market economy framework of West Germany. before German reunification and the dissolution of the Stasi we used two
The opportunities for and the relative pay-offs of entrepreneurship additional criteria to create our final sample. First, because the GDR did
changed dramatically as a result of the transition. Entrepreneurship not exist before 1949, we excluded firms whose focal founder was born
usually plays a minor role in non-democratic authoritarian regimes before 1949. Second, we follow Lichter et al. (2021) and other studies on
(Audretsch and Moog, 2020). The socialist institutions in the GDR were legacy effects of the GDR (Fuchs-Schuendeln and Masella, 2016) that
particularly hostile toward entrepreneurial activities, and running a utilized the same spying data, and considered only respondents born
privately owned venture in that environment was almost impossible before 1974, and therefore were at least 17 years old at the time of
(Fritsch et al., 2014). German reunification in 1990, to ensure a clean treatment effect of being
Like other authoritarian regimes, the GDR government had powerful exposed to Stasi surveillance activity. Thus, all respondents reached or
intelligence agencies and used undercover informants to surveil its cit passed already the age (i.e., 18 to 25 years) that is regarded as crucial for
izens. This created an environment of suspicion and long-lasting low the formation of beliefs in accordance to social-psychology literature
levels of trust, not only towards strangers but even towards friends and and confirmative recent empirical evidence (for an overview, see Giu
family members (Bjørnskov, 2006; Rainer and Siedler, 2009). After the liano and Spilimbergo 2014). After data cleaning, our final sample in
fall of the Berlin Wall in November 1989, the East German intelligence cludes 620 firms, with 214 firms involved in R&D activities. In a
agencies were dissolved in preparation for German reunification in robustness check, we consider founders born after 1973, and this full
October 1990. This historical reality means that East German founders sample comprises 714 firms with 251 firms engaging in R&D.
were exposed to an authoritarian regime, while West German founders
represent non-treated counterfactual observations.
One advantage of our data and the empirical set-up is that both allow 2.3. Variables
us to identify regional differences in surveillance levels and go beyond a
simple binary East-West comparison that would only capture a general 2.3.1. Dependent variable
“GDR/East German effect”. Because there were, in fact, regional varia Our dependent variable captures a firm’s involvement in interfirm
tions in GDR surveillance levels conducted by the East German Secret R&D cooperation activities. It is measured by a dichotomous variable and
Police (Stasi), we can test if these variations left varying imprints on the refers to any R&D cooperation since the start-up’s foundation. The
founders. survey question explicitly refers to R&D cooperation with other private
The Stasi developed a system of recruiting undercover informants to firms, just as in the standard EU community innovation survey (CIS,
spy on their friends, neighbors, and families, which implied: “the 2020).3
breakdown of the bonds of trust between officers and men, lawyers and cli
ents, doctors and patients, teachers and students, pastors and their commu 2.3.2. Variable of interest
nities, friends and neighbors, family members and even lovers” (Childs and Also in line with Lichter et al. (2021), we measure spying intensity as
Popplewell, 1996, p. 111). Lichter et al. (2021) find that East Germans the local share of the population working either as an official employee
who were socialized in regions with high levels of surveillance reveal a or undercover informant for the Stasi in the region (at NUTS3 level)
lower level of trust, even several years after reunification. We use the
same regional dataset to measure local spying intensity and analyze
whether growing up in spying-intense regions reduces the likelihood of
the respective East German founders to engage in R&D cooperation. If 1
Since the first appearance in the BHP indicates that the venture hired an
this is the case, this would be in line with our prediction that the employee liable to Social Insurance for the first time, the actual start-up deci
trust-inhibiting effects of surveillance within authoritarian regimes sion could have taken place several years earlier.
inhibit R&D cooperation. We include West German founders with strong 2
This comprises all high, medium-high, and medium-low technology-inten
similarities in culture and language that were not subject to an author sive manufacturing industries and all knowledge-intensive service (NACE Rev
itarian regime’s influence as a benchmark for cooperative behavior. 2, 2-digit level: 19-30, 33, 58-63, 69-75; Eurostat, 2018).
3
These founders experienced a local environment where spying activity For description of the dataset, see https://ec.europa.eu/eurostat/web/mi
was virtually non-existing. crodata/community-innovation-survey
3
M. Wyrwich et al. Research Policy 51 (2022) 104409
4
M. Wyrwich et al. Research Policy 51 (2022) 104409
5
Table 3
M. Wyrwich et al.
Correlation matrix.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]
start-up: Paid
employee
[15] Employment 0.028 0.024 0.067* − 0.058 0.125*** 0.018 0.063 0.058 − 0.023 − 0.006 0.047 0.071* − 0.13*** − 0.722*** 1
status before
start-up: Self-
employed
[16] Self-employed 0.042 − 0.07* − 0.088** 0.038 0.11*** − 0.073* 0.064 0.022 − 0.051 − 0.003 0.042 − 0.031 − 0.035 0.039 − 0.017 1
role models
before start-up
[17] Takeover after 0.045 0.032 0.051 − 0.009 − 0.032 0.09** 0.021 0.028 0.058 0.076 0.051 0.067* − 0.004 − 0.027 0.037 − 0.057 1
start-up
[18] Number of 0.071 − 0.02 − 0.017 0 − 0.034 0.148*** − 0.028 0 − 0.076* 0.006 0.069* − 0.07* − 0.065 − 0.038 0.103 0 0.028 1
founders
[19] Years in self- 0.043 0.035 0.153*** − 0.083** 0.113*** 0.042 0.094** 0.04 − 0.06 − 0.078 0.031 0.163 − 0.054 − 0.414*** 0.556*** − 0.016 0.032 0.054 1
employment
before start-up
[20] East-to-West − 0.075 0.133*** 0.042 0.112*** − 0.018 − 0.055 0.029 − 0.088** 0.016 0.091 − 0.049 0.008 0.006 0.035 − 0.048 − 0.062 0.023 − 0.057 − 0.059 1
Move
[21] West-to-East 0.043 − 0.292*** − 0.009 − 0.011 − 0.042 0.049 0.078* 0.159*** 0.086** 0.006 0.052 − 0.051 0.007 − 0.045 0.049 0.023 0.077* 0.091** 0.05 − 0.039 1
Move
Notes:.
***
p<0.01.
**
p<0.05.
*
p<0.10. + Since spying intensity is zero for West German respondents, the correlation for spying intensity refers to East German respondents only.
M. Wyrwich et al. Research Policy 51 (2022) 104409
7
M. Wyrwich et al. Research Policy 51 (2022) 104409
Table 4
The relationship between East German origin and R&D cooperation behavior.
1 2 3 4 5
Notes: Probit model with sample selection. First stage results are provided in Table B1. Robust standard errors in parentheses.
***
p<0.01.
**
p<0.05.
*
p<0.10. Average marginal effects reported (unconditioned on selection). Constant not reported for brevity. There are two missing values on the R&D employment
share. This explains the minor difference in case numbers across models.
imprint on founders that influence their firms’ strategic choices. Because the combination of different imprints influences strategic choices in new
barriers to R&D cooperation and the relative economic success of firms ventures. Our study does not investigate how a founder’s imprint is
and regions may be rooted in history and depend on the intensity of transferred to the organization, nor whether a founder’s imprint affects
specific historical events, policy-makers need to address “soft” barriers the strategic decisions of the firm after the founder leaves the organi
to R&D cooperation to promote firm- and regional-level development. zation. Because our study’s sample includes mostly smaller firms, it is
Our findings suggest that founders from authoritarian regimes who do difficult to generalize our findings to large firms. In large firms, strategic
not transition well to a market economy may impose imprinted biases on decisions are usually made by management teams, and in that context,
their ventures and ignore the full range of strategic options free-market relational trust may be more important than generalized trust.
economies offer. Policies that support awareness campaigns and trust- An interesting avenue for future research is investigating the inter
facilitating networking might help founders be more aware of and action between ideological imprinting (cf., Marquis and Qiao 2020) and
address personal biases that negatively affect firm-level decision- oppressive surveillance in empirical contexts where certain regions,
making. organizations or people were differently exposed to the ideology of the
Our research has limitations that future research could address. We authoritarian regime. It is also interesting to explore the interaction
were not able to explicitly consider the detailed composition of the between pre-existing local cultures and exposure to authoritarian re
founding team (e.g., Bryant 2014, Hahn et al. 2019). Future research gimes (for a general discussion in the context of Germany, see Becker
might look at the heterogeneity of founders in order to investigate how et al. 2020). Our findings might also encourage future research on how
8
M. Wyrwich et al. Research Policy 51 (2022) 104409
Fig. 1. The marginal effect of spying intensity on cooperation behavior (with 95% confidence interval, CI).
the socioeconomic context of immigrant founders (cf. Kerr and Kerr Appendix A
2020) influences their willingness to participate in open innovation
activities. Table A1:
Our data has only limited information about the scope of R&D ac Table A2:
tivities and the technology-related contexts in which firms operate, an
issue that future research might address. Additional research into the
role of R&D contracts that might mitigate trust-related obstacles for Table A1
cooperation is also called for. Future research may also want to explore Summary statistics: Full sample.
how exogenous measures affecting local trust levels link to endogenous
Mean S.D. Min Max
trust formation at the individual level. Here, a better understanding of
the interrelation between socioeconomic imprinting and the emergence Engaging in R&D activity (Dummy, Y = 1) 0.351 0.478 0 1
GDR origin (Dummy, Y = 1) 0.491 0.5 0 1
of generalized trust and relational trust could be an especially promising
Spying intensity (in%) 0.185 0.209 0 0.742
avenue for future research in the context of studying collaborative West 0 0 0 0
behavior. East 0.377 0.127 0.166 0.742
Age 45.582 7.763 25 61
Women (Dummy, Y = 1) 0.144 0.351 0 1
Funding
Start-up year 2001 4.314 1990 2007
Start-up size 2.099 2.41 1 20
This research did not receive any specific grant from funding Industry 5 1.941 2 7
agencies in the public, commercial, or not-for-profit sectors. Regional degree of agglomeration 3 1.833 1 7
Level of Education 4.824 1.693 1 7
Industry experience 9.27 8.011 0 39
CRediT authorship contribution statement
Employment status before start-up: Other 0.127 0.334 0 1
(Dummy, Y = 1)
Michael Wyrwich: Conceptualization, Methodology, Writing – Employment status before start-up: Paid 0.736 0.441 0 1
original draft, Writing – review & editing, Formal analysis. Philip J. employee (Dummy, Y = 1)
Steinberg: Conceptualization, Methodology, Writing – original draft, Employment status before start-up: Self- 0.137 0.344 0 1
employed (Dummy, Y = 1)
Writing – review & editing. Florian Noseleit: Conceptualization, Self-employed role models before start-up 0.601 0.49 0 1
Methodology, Writing – original draft, Writing – review & editing. (Dummy, Y = 1)
Pedro de Faria: Conceptualization, Methodology, Writing – original Takeover after start-up (Dummy, Y = 1) 0.145 0.353 0 1
draft, Writing – review & editing. Number of founders 1.55 0.894 1 5
Years in self-employment before start-up 2.807 5.905 0 32
East-to-West Move (Dummy, Y = 1) 0.014 0.118 0 1
Declaration of Competing Interest West-to-East Move (Dummy, Y = 1) 0.085 0.28 0 1
Openness (Dummy, Y = 1) 0.443 0.497 0 1
None.
Notes: Interfirm R&D cooperation, share of R&D-employees, and use of state-of-
the-art-technology was only asked if the company was involved in R&D
activities.
9
M. Wyrwich et al.
Table A2
Correlation Matrix: Full sample.
[1] [2] [3] [4] [5] [6] [7] [8] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20]
[14] Self-employed role models before start-up 0.042 − 0.046 − 0.123*** 0.041 0.118*** − 0.055 0.048 0.024 0.038 − 0.061 0.011 − 0.009 0.001 1
[15] Takeover after start-up − 0.029 0.023 0.084** − 0.011 − 0.043 0.082** 0.011 0.035 0.052 0.072* − 0.027 − 0.005 0.032 − 0.053 1
[16] Number of founders 0.131*** − 0.019 − 0.056 − 0.012 − 0.023 0.134*** − 0.025 − 0.017 0.065* − 0.081** − 0.028 − 0.071* 0.119*** 0.002 0.013 1
[17] Years in self-employment before start-up 0.073* 0.006 0.166*** − 0.097*** 0.092** 0.043 0.09** 0.031 0.04 0.172*** − 0.064* − 0.372*** 0.54*** − 0.011 0.034 0.049 1
[18] East-to-West Move − 0.013 0.121*** 0.037 0.121*** − 0.016 − 0.054 0.034 − 0.062* − 0.058 0.005 − 0.01 0.017 − 0.013 − 0.049 0.018 − 0.047 − 0.055 1
[19] West-to-East Move − 0.057 − 0.3*** 0.039 − 0.025 − 0.048 0.05 0.064* 0.152*** 0.055 − 0.031 − 0.011 − 0.021 0.038 0.024 0.073* 0.087** 0.051 − 0.036
[20] Openness 0.218*** − 0.019 0.068* − 0.041 − 0.033 − 0.033 0.045 − 0.053 − 0.008 0.055 − 0.053 0.011 0.038 0.044 − 0.048 0.022 0.023 0.038
[21] Spying intensity [in%]+ 0.035 0.011 − 0.027 − 0.013 0.04 − 0.089* 0.404*** − 0.073 − 0.017 0.009 0.009 − 0.02 0.057 − 0.07 − 0.129** 0.007 − 0.118** .
Notes:.
***
p<0.01.
**
p<0.05.
*
p<0.10. + Since spying intensity is zero for West German respondents, the correlation for spying intensity refers to East German respondents only. Interfirm R&D cooperation, share of R&D-employees, and use of state-
of-the-art-technology was only asked if the company was involved in R&D activities.
Appendix B dummy are both not significantly related to engaging in R&D activities.
The models presented in Table 4 do not include an inverse mills ratio.
The models in Table B1 show that openness to new ideas and This is because of the binary nature of the dependent variable (interfirm
experimentation is positively linked to the likelihood of engaging in R&D cooperation). In a Probit model with sample selection the entire
R&D. Hence, the data suggests that this is a valid instrument in the two- model is estimated via maximum likelihood without inverse mills ratio
stage model with our sample selection. Some of the control variables, correction. This is in contrast to a standard Heckman model where the
such as gender and education, also determine selection into R&D ac dependent variable is continuous. As a consistency check, we ran a
tivities. Nevertheless, sample selection does not seem to be a severe issue standard two-step Heckman model and calculate the inverse mills ratio
as the correlation coefficient (rho) between the error term of the selec despite the fact that our dependent variable is not continuous. The re
tion model and the second-stage model is relatively low and insignifi sults are provided in Table B2. The results for our main variables of
cant, which is indicated by the values for athrho, the inverse hyperbolic interest remain robust while the inverse mills ratio is insignificant,
tangent of rho. Interestingly, spying intensity and the GDR origin implying that sample selection is not a severe issue.
Table B1
Selection stage results: Probit with sample selection.
Dep var: Engaging in R&D activities (Dummy, Y = 1) 1 2 3 4 5
Notes: First stage results from Probit model with sample selection. Robust standard errors in parentheses.
***
p<0.01.
**
p<0.05.
*
p<0.10. Average marginal effects reported. Constant not reported for brevity. There are two missing values on the R&D employment share. This explains the minor
difference in case numbers across models. Atrho measures the correlation between error terms in the first and second stage of the selection model. Based on a Wald test,
the null hypothesis that this correlation is zero cannot be rejected across model specifications. Hence, selection is likely not problematic.
11
M. Wyrwich et al. Research Policy 51 (2022) 104409
Table B2
Second stage results: Standard Heckman twostep model.
Dep Var: Interfirm R&D Cooperation (Dummy, Y = 1) 1 2 3 4 5
References Chesbrough, H.W., 2006. Open Business Models: How To Thrive In The New Innovation
Landscape. Harvard Business Press, Boston.
Chesbrough, H.W., Bogers, M., 2014. Explicating Open Innovation: clarifying an
Agarwal, R., Croson, R., Mahoney, J.T., 2010. The role of incentives and communication
Emerging Paradigm for Understanding Innovation. In: Chesbrough, H.,
in strategic alliances: an experimental investigation. Strateg. Manag. J. 31 (4),
Vanhaverbeke, W., West, J. (Eds.), New Frontiers in Open Innovation. Oxford
413–437.
University Press, pp. 3–28.
Alesina, A., Fuchs-Schuendeln, N., 2007. Good bye Lenin (or not?): the effect of
Childs, D., Popplewell, R., 1996. The Stasi: The East German Intelligence and Security
communism on people’s preferences. Am. Econ. Rev. 97, 1507–1528.
Service. New York University Press, New York, NY.
Algan, Y., Cahuc, P., 2010. Inherited trust and growth. Am. Econ. Rev. 100 (5),
CIS (2020). Community Innovation Survey, https://ec.europa.eu/eurostat/web/mi
2060–2092.
crodata/community-innovation-survey.
Amore, M.D., Murtinu, S., 2019. Tobit models in strategy research: critical issues and
Classen, N., Van Gils, A., Bammens, Y., Carree, M., 2012. Accessing resources from
applications. Glob. Strategy J. Online 1–25.
innovation partners: the search breadth of family SMEs. J. Small Bus. Manag. 50,
Aslund, A., Djankov, S., 2014. The Great Rebirth: Lessons from the Victory of Capitalism
191–215.
over Communism. The Great Rebirth: Lessons from the Victory of Capitalism over
Corneo, G., Gruner, H.P., 2002. Individual preferences for political redistribution.
Communism. Peterson Institute for International Economics.
J. Public Econ. 83 (1), 83–107.
Audretsch, D.B., Moog, P., 2020. Democracy and entrepreneurship. Entrep. Theory
Dahlander, L., Gann, D.M., 2010. How open is innovation? Res. Policy 39, 699–709.
Pract., 1042258720943307
De Faria, P., Lima, F., Santos, R., 2010. Cooperation in innovation activities: the
Bauernschuster, S., Falck, O., Gold, R., Heblich, S., 2012. The shadows of the socialist
importance of partners. Res. Policy 39, 1082–1092.
past: lack of self-reliance hinders entrepreneurship. Eur. J. Polit. Econ. 28 (4),
De Faria, P., Noseleit, F., Los, B., 2020. The influence of internal barriers on open
485–497.
innovation. Ind. Innov. 27, 205–209.
Becker, S.O., Mergele, L., Woessmann, L., 2020. The separation and reunification of
Delhey, J., Newton, K., 2005. Predicting cross-national levels of social trust: global
Germany: rethinking a natural experiment interpretation of the enduring effects of
pattern or nordic exceptionalism? Eur. Sociol. Rev. 21 (4), 311–327.
communism. J. Econ. Perspect. 34 (2), 143–171.
Dinesen, P.T., Bekkers, R., 2017. The foundations of individuals’ generalized social trust:
Bjørnskov, C., 2006. The multiple facets of social capital. Eur. J. Polit. Econ. 22 (1),
a review. In: Van Lange, P.A.M., Rockenbach, B., Yamagishi, T. (Eds.), Trust in Social
22–40.
Dilemmas. Oxford University Press, New York, NY, pp. 77–100.
Bogers, M., Foss, N.J., Lyngsie, J., 2018. The “human side” of open innovation: the role of
Ebersberger, B., Galia, F., Laursen, K., Salter, A., 2021. Inbound open innovation and
employee diversity in firm-level openness. Res. Policy 47 (1), 218–231.
innovation performance: a robustness study. Res. Policy 50, 104271.
Brambor, T., Clark, W.R., Golder, M., 2006. Understanding interaction models:
Eftekhari, N., Bogers, M., 2015. Open for entrepreneurship: how open innovation can
improving empirical analyses. Polit. Anal. 14, 63–82.
foster new venture creation. Creat. Innov. Manag. 24, 574–584.
Brockman, P., Khurana, I.K., Zhong, R.I., 2018. Societal trust and open innovation. Res.
Ertug, G., Cuypers, I.R., Noorderhaven, N.G., Bensaou, B.M., 2013. Trust between
Policy 47 (10), 2048–2065.
international joint venture partners: effects of home countries. J. Int. Bus. Stud. 44
Bryant, P.T., 2014. Imprinting by design: the microfoundations of entrepreneurial
(3), 263–282.
adaptation. Entrep. Theory Pract. 38 (5), 1081–1102.
Eurostat, 2018. Glossary: high-tech, https://ec.europa.eu/eurostat/statistics-explain
Bublitz, E., Nielsen, K., Noseleit, F., Timmermans, B., 2018. Entrepreneurship, Human
ed/index.php?title=Glossary:High-tech.
Capital, and Labor Demand: a Story of Signaling and Matching. Ind. Corp. Change
Faems, D., Janssens, M., Madhok, A., Looy, B.V., 2008. Toward an integrative
27, 269–287.
perspective on alli-ance governance: connecting contract design, trust dynamics, and
Burns, N., Kinder, D., Rahn, W., 2003. Social trust and democratic politics. In:
contract application. Acad. Manag. J. 51 (6), 1053–1078.
Proceedings of the Paper prepared for the Annual Meeting of the Mid-West Political
Ferrin, D.L., Gillespie, N., 2010. Trust differences across national-societal cultures: much
Science Association. Chicago Illinois. Mid-West Political Science Association,
to do, or much ado about nothing? In: Saunders, M.N.K., Skinner, D., Dietz, G.,
Chicago. April 2003.
Gillespie, N., Lewicki, R.J. (Eds.), Organizational trust: A cultural Perspective.
Burton, M.D., Beckman, C.M., 2007. Leaving a legacy: position imprints and successor
Cambridge University Press, Cambridge, UK, pp. 42–86.
turnover in young firms. Am. Sociol. Rev. 72 (2), 239–266.
Ferriani, S., Lazerson, M.H., Lorenzoni, G., 2020. Anchor entrepreneurship and industry
Carson, S.J., Madhok, A., Varman, R., John, G., 2003. Information processing moderators
catalysis: the rise of the Italian Biomedical Valley. Res. Policy 49 (8), 104045.
of the effectiveness of trust-based governance in interfirm R&D collaboration. Organ.
Fuchs-Schuendeln, N., Masella, P., 2016. Long-lasting effects of socialist education. Rev.
Sci. 14 (1), 45–56.
Econ. Stat. 98, 428–441.
Certo, S.T., Busenbark, J.R., Woo, H.S., Semadeni, M., 2016. Sample Selection Bias and
Fritsch, M., Bublitz, E., Sorgner, A., Wyrwich, M., 2014. How much of a socialist legacy?
Heckman models in strategic management research. Strateg. Manag. J. 37,
The Re-emergence of entrepreneurship in the East German transformation to a
2639–2657.
market economy. Small Bus. Econ. 43, 427–446.
Chesbrough, H.W., 2003. Open Innovation: The New Imperative for Creating and
Profiting from Technology. Harvard Business Press, Boston.
12
M. Wyrwich et al. Research Policy 51 (2022) 104409
Garcia, D., Lester, N., Cloninger, K.M., Cloninger, R.C., 2017. Temperament and Mayer, K.J., Nickerson, J.A., 2005. Antecedents and performance implications of
Character Inventory (TCI). In: Zeigler-Hill, V., Shackelford, T. (Eds.), Encyclopedia contracting for knowledge workers: evidence from information technology services.
of Personality and Individual Differences. Springer, Cham. Organ. Sci. 16 (3), 225–242.
GEM, 2020. German National Report. (downloaded on 11 October 2021 from https Mayntz, R., 1998. Socialist academies of sciences: the enforced orientation of basic
://gemconsortium.org/news/key-findings-from-the-2019%2F2020-german-nation research at user needs. Res. Policy 27 (8), 781–791.
al-report). Nannestad, P., 2008. What Have We Learned About Generalized Trust, If Anything?
Gerschewski, J., 2013. The three pillars of stability: legitimation, repression, and co- Annu. Rev. Polit. Sci. 11, 413–436.
optation in autocratic regimes. Democratization 20 (1), 13–38. Nooteboom, B., Berger, H., Noorderhaven, N.G., 1997. Effects of trust and governance on
Giuliano, P., Spilimbergo, A., 2014. Growing up in a recession. Rev. Econ. Stud. 81, relational risk. Acad. Manag. J. 40 (2), 308–338.
787–817. O’Sullivan, D., Zolotoy, L., Fan, Q., 2021. CEO early-life disaster experience and
Glanville, J.L., Paxton, P., 2007. How do we learn to trust? A confirmatory tetrad corporate social performance. Strateg. Manag. J. 42, 2137–2161 forthcoming.
analysis of the sources of generalized trust. Soc. Psychol. Q. 70 (3), 230–242. Paldam, M., Svendsen, G.T., 2001. Missing social capital and the transition in Eastern
Grilli, L., Jensen, P.H., Murtinu, S., Park, H.D., 2020. A close look at the contingencies of Europe. J. Inst. Innov. Dev. Trans. 5, 21–34.
founders’ effect on venture performance. Ind. Corp. Change 29 (4), 997–1020. Rainer, H., Siedler, T., 2009. Does democracy foster trust? J. Comp. Econ. 37 (2),
Hahn, D., Minola, T., Eddleston, K.A., 2019. How do scientists contribute to the 251–269.
performance of innovative start-ups? An imprinting perspective on open innovation. Roche, M.P., Conti, A., Rothaermel, F.T., 2020. Different founders, different venture
J. Manag. Stud. 56, 895–928. outcomes: a comparative analysis of academic and non-academic startups. Res.
Hardin, R., 2006. Trust. Trust. Wiley, Cambridge, UK. Policy 49 (10), 104062.
Huff, L., Kelley, L., 2003. Levels of organizational trust in individualist versus collectivist Rothstein, B., Stolle, D., 2002. How political institutions create and destroy social capital:
societies: a seven-nation study. Organ. Sci. 14 (1), 81–90. an institutional theory of generalized trust. In: Proceedings of the 98th meeting of
Kalnins, A., 2018. Multicollinearity: how common factors cause Type 1 errors in the American Political Science Association in. Boston, MA.
multivariate regression. Strateg. Manag. J. 39, 2362–2385. Rotter, J.B., 1971. Generalized expectancies for interpersonal trust. Am. Psychol. 26 (5),
Kerr, S.P., Kerr, W., 2020. Immigrant entrepreneurship in America: evidence from the 443.
survey of business owners 2007 & 2012. Res Policy 49 (3), 103918. Simsek, Z., Fox, B.C., Heavey, C., 2015. What’s past is prologue” A framework, review,
Keupp, M.M., Gassmann, O., 2009. Determinants and archetype users of open and future directions for organizational research on imprinting. J. Manag. 41 (1),
innovation. R&D Manag. 39, 331–341. 288–317.
Knabe, H., 2014. The dark secret of a surveillance state. In: Proceedings of the TED Srinivasan, R., Choo, A., Narayanan, S., Sarkar, S., Tenhiälä, A., 2021. Knowledge
Conferences (downloaded on 12 January 2021 from. https://www.ted.com/talks/hu sources, innovation objectives, and their impact on innovation performance: quasi-
bertus_knabe_the_dark_secrets_of_a_surveillance_state/transcript. replication of Leiponen and Helfat, 2010. Strateg. Manag. J. 42, 2104–2136
Kramer, R.M., Lewicki, R.J., 2010. Repairing and enhancing trust: approaches to forthcoming.
reducing organizational trust deficits. Acad. Manag. Ann. 4, 245–277. Stinchcombe, A.L., 1965. Social structure and organizations. Handb. Organ. 7, 142–193.
Larsson, R., Bengtsson, L., Henriksson, K., Sparks, J., 1998. The interorganizational Svolik, M.W., 2012. The Politics of Authoritarian Rule. Cambridge University Press, New
learning dilemma: collective knowledge development in strategic alliances. Organ. York, NY.
sci. 9 (3), 285–305. Ulset, S., 1996. R&D outsourcing and contractual governance: an empirical study of
Krishnan, R., Martin, X., Noorderhaven, N.G., 2006. When does trust matter to alliance commercial R&D projects. J. Econ. Behav. Organ. 30 (1), 63–82.
performance? Acad. Manag. J. 49 (5), 894–917. Tabellini, G., 2010. Culture and institutions: economic development in the regions of
Lee, Y., Cavusgil, S.T., 2006. Enhancing alliance performance: the effects of contractual- Europe. J. Eur. Econ. Assoc. 8 (4), 677–716.
based versus relational-based governance. J. Bus. Res. 59 (8), 896–905. Waguespack, D.M., Dunford, E.T., Birnir, J.K., 2018. Cultural imprinting, institutions,
Lee, S., Park, G., Yoon, B., Park, J., 2010. Open innovation in SMEs – An intermediated and the organization of new firms. Strategy Sci. 3 (2), 426–438.
network model. Res. Policy 39, 290–300. Wang, D., Du, F., Marquis, C., 2019. Defending Mao’s dream: how politicians‘ ideological
Lichter, A., Loeffler, M., Siegloch, S., 2021. The Long-Term Costs of Government imprinting affects firms‘ political appointment in China. Acad. Manag. J. 62,
Surveillance: insights from Stasi Spying in East Germany. J. Eur. Econ. Assoc. 19, 1111–1136.
741–789. West, J., Salter, A., Vanhaverbeke, W., Chesbrough, H., 2014. Open innovation: the next
Marquis, C., Qiao, K., 2020. Waking from Mao’s dream: communist ideological decade. Res. Policy 43, 805–811.
imprinting and the internationalization of entrepreneurial ventures in China. Adm. Wyrwich, M., Fritsch, M., Bublitz, E., Sorgner, A., 2018. GAW survey data linked with
Sci. Q. 65 (3), 795–830. administrative data of the IAB (GAW-ADIAB): a documentation and review of
Marquis, C., Tilcsik, A., 2013. Imprinting: Toward a multilevel Theory, 7. Academy of empirical evidence and research opportunities. JENA Econ. Res. Paper #2018-017.
Management Annals, pp. 195–245. Zaheer, A., Mcevily, B., Perrone, V., 1998. Does Trust Matter? Exploring the Effects of
Mayer, R.C., Davis, J.H., Schoorman, D., 1995. An Integrative Model of Organizational Interorganizational and Interpersonal Trust on Performance. Strateg. Manag. J. 9
Trust. Acad. Manag. Rev. 20, 709–734. (2), 141–159.
13