Hostname: page-component-76d6cb85b7-6jg5l Total loading time: 0 Render date: 2026-07-18T15:16:29.551Z Has data issue: false hasContentIssue false
  • English
  • Français

The Entrepreneurial University's Impact on Regional Socioeconomic Development: The “Alumni Policymaker” Mechanism

Published online by Cambridge University Press:  12 May 2023

Robyn Klingler-Vidra Open the ORCID record for Robyn Klingler-Vidra [Opens in a new window]  and
Adam William Chalmers Open the ORCID record for Adam William Chalmers [Opens in a new window]
Robyn Klingler-Vidra*
Affiliation:
King's College London, London, UK
Adam William Chalmers
Affiliation:
University of Edinburgh, Edinburgh, UK
*
Corresponding author: Robyn Klingler-Vidra; Email: Robyn_Klingler.Vidra@kcl.ac.uk
Rights & Permissions [Opens in a new window]

Abstract

Research has examined the impact of the “entrepreneurial university” on regional socioeconomic development by focusing on the entrepreneurial intentions and performance of alumni, staff, and students. The study of impact, to date, has focused on direct and short-term mechanisms, such as alumni's entrepreneurial activities, faculty spin-outs, and active public engagement with policy agendas. Our point of departure is in conceptualizing and empirically testing a longer-term and more systemic mechanism. We theorize and empirically test how the entrepreneurial university imprints on its graduates, some of whom take on leadership positions in innovation policymaking years later. We test this relationship by employing a text-as-data approach to examine the extent to which innovation policy leaders speak about startup-centric innovation, comparing the media coverage of entrepreneurial university alumni relative to their peers. Our original dataset comprises the 485 individuals who held senior innovation policy positions in East Asia's eleven largest economies from 1998 to 2019, detailing their educational background and media coverage (10,816 documents). We conceptualize the “alumni policymaker” mechanism, which constitutes entrepreneurial university alumni shaping the future of national innovation policy by referring to startup-centric innovation three times more than their peers. Those who completed MBAs at entrepreneurial universities express an even greater preference for startup-centric innovation policy.

Keywords

economic developmententrepreneurial universitiesinnovationpolicytext-as-data analysis

Information

Type
Research Article
Information
Business and Politics , Volume 25 , Issue 3 , September 2023 , pp. 251 - 292
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
Copyright © The Author(s), 2023. Published by Cambridge University Press on behalf of V.K. Aggarwal

Introduction

The “entrepreneurial university”Footnote 1 fosters startup activity and innovative mindsets through its teaching and learning, research, service, spin-outs, and technology transfer activities.Footnote 2 To date, the impact of the entrepreneurial university been conceptualized and studied at the level of individual entrepreneurs (e.g., alumni, staff, or students founding high-profile startups) and institutions (e.g., incubator-university collaborations and technology transfer offices).Footnote 3 The prevailing understanding of the entrepreneurial university's impact on regional socioeconomic development focuses on direct and short-term mechanisms, such as alumni entrepreneurial activities, faculty spin-outs, and active public engagement with policy agendas.Footnote 4

Our point of departure is in conceptualizing and empirically testing a more systemic and long-term mechanism. We study how the entrepreneurial university imprints on its graduates, some of whom take on leadership positions in innovation policymaking years later. These entrepreneurial university alumni shape the way that innovation policy is conceived, with a preference for startups as essential drivers of innovation. We theorize an “alumni policymaker” mechanism, which constitutes a more systematic impact on regional socioeconomic development than the effects acknowledged to date. This adds evidence of a new knowledge spillover mechanism of the entrepreneurial university in delivering societal value, or its “third mission.”

Theoretically, we extend state-of-the-art research on the impact of the personal characteristics of leaders on policy preferences.Footnote 5 This body of scholarship finds that personal characteristics—such as the location of the university and the subject that policymakers study—shape policy preferences for years to come. We combine this approach with state-of-the-art scholarship on “imprinting,” in which formative experiences leave a long-lasting impact on preferences.Footnote 6 Imprinting is defined as exposure that “establishes persistent organizational routines and structures that sustain traditions, vested interests, and ideologies” and occurs when people are “most susceptible to influence” in their life.Footnote 7 We explore the extent to which studying at an entrepreneurial university—a formative experience early in one's adult life—could imprint persistent preferences for a startup-centric form of innovation. Such imprinting as a spillover mechanism of the entrepreneurial university reveals an additional social return to higher education, which poses an interesting avenue for further research given the ongoing public debate about the societal value of universities.

Our core research question is: to what extent does studying at an entrepreneurial university affect graduates’ long-term preference for startup-centric innovation? We study this “alumni policymaker” preference in the context of innovation policy and conceive of startup-centric innovation policies as a form of national innovation system (NIS) policy,Footnote 8 which focus on the starting and scaling up of high-growth, technology-centric entrepreneurs.Footnote 9 We define startup-centric innovation policy in accordance with Klingler-Vidra and Pacheco Pardo,Footnote 10 who specify that the term refers to policies that are “focused exclusively on the aim of creating more, and higher quality, startups in a bid to advance a Silicon Valley-styled innovation cluster.” As we explain in the Data and methods section, we operationalize the language of startup-centric innovation policy in terms of eight categories of instruments used, which include funding, taxation, education, and training.

We operationalize entrepreneurial university by identifying the cohort of universities that, according to Crunchbase,Footnote 11 boast the most graduates creating startups that raise venture capital funding of $1 million or more. This is consistent with other state-of-the-art studies on the entrepreneurial universityFootnote 12 in that we use established rankings to establish the set of entrepreneurial universities. We then assess the extent to which the alumni of these most active entrepreneurial universities speak about startup-centric innovation relative to their policy-leading peers who did not attend an entrepreneurial university. We do so using text-as-data methods to analyze our original dataset of education and media coverage for the 485 innovation policy leaders responsible for innovation policy between 1998 and 2019 for the largest eleven economies (in terms of gross domestic product in 2019) in the East Asian region: China, Hong Kong, Indonesia, Japan, Korea, Malaysia, the Philippines, Singapore, Taiwan, Thailand, and Vietnam. The dataset details each policy leader's university education (including university, country, and degree level, across undergraduate [UG], master's, MBA, and PhD programsFootnote 13) as well as news media coverage. We analyze 10,816 media documents to assess how frequently policy leaders reference startup-centric innovation in the media.

We find evidence that innovation policy leaders who are entrepreneurial university alumni are three times more likely to speak about startup-centric innovation policy than their peers. Those who completed MBAs at entrepreneurial universities express an even greater preference for startup-centric innovation policy. Evidence of this preference occurs years after their university studies, suggesting imprinting upon its policymaker alumni, leaving an indelible impact on how they conceive of innovation—with startups playing a key role.

Through our approach and findings, the article makes three contributions. First, the article contributes to research examining the relationship between the entrepreneurial university and regional economic activity,Footnote 14 potentially illuminating a new mechanism by which entrepreneurial universities deliver societal impact. We do so by delineating expectations about how studying for certain degree types, at certain universities, may imprint upon graduates, leading to more or less of a preference for startup-centric innovation, which is expressed when alumni take an innovation policy leadership role later in life.Footnote 15 This “policymaker alumni” mechanism constitutes a new empirical area (innovation policy) for research into the relationship between personal characteristics and policy preferences.

The second contribution is methodological, as we bring a text-as-data approachFootnote 16 further into to the mainstream study of the relationship between education and policy preferences.Footnote 17 We theorize and measure media coverage in the years during which each policy leader held their leadership role, which, while imperfect, is the closest indicator of the policymakers’ expressed preferences. This approach of measuring media coverage as evidence of preference is well established in studies of how personal characteristics shape the preferences of politiciansFootnote 18 and central bankers.Footnote 19 However, it has not been applied to the empirical area of innovation policy. To study this nexus, we developed a set of startup-centric innovation n-grams, based upon state-of-the-art literature on innovation policy. This approach offers an empirically grounded, specialized method for corpus-based dictionary generation in research using natural language processing.Footnote 20

Third, we offer an original dataset of the educational backgrounds of the leaders of innovation policy for East Asia's eleven largest economies over a twenty-year period (1998–2019). In addition to detailing who these policymakers are and their educational profiles, the dataset includes each policymaker's media engagement during the years they were in office.

This article is structured as follows. The following section reviews the literature on how the entrepreneurial university is hypothesized to shape the entrepreneurial intentions of its alumni, as well as scholarship on personal characteristics, imprinting, and policy preferences. Next, we present the data and methods, including data collection and coding and the text-as-data method on which the analysis is based. The next section presents an analysis of the extent to which entrepreneurial university graduates invoke coverage about startup-centric innovation policies relative to their peers who did not study at one of these universities over the 1998–2019 period. Then, we discuss the implications of our findings. The last section concludes by discussing the limitations of our study and future research.Footnote 21

Personal characteristics and the entrepreneurial university imprinting of startup-centric innovation preferences

Researchers have examined the impact of personal characteristics on a variety of policy and political decisions and outcomes. The approach is underpinned by the assumption that endowments of cultural, human, and social capital—such as family background and socioeconomic status,Footnote 22 university education,Footnote 23 and work experienceFootnote 24—affects actors’ policymaking preferences. The degree subject studied, as well as occupation, have been found to inform preferences.Footnote 25

Studying at certain universities impacts graduates’ long-term economic,Footnote 26 financial,Footnote 27 trade,Footnote 28 and legalFootnote 29 policymaking preferences. The causal mechanism underpinning this relationship is the shared, intensive training, which creates “strong professional identities and shared norms” that shape long-lasting views.Footnote 30 Seminal work on the role of university education in forming policymaking preferences, such as the “Chicago boys,” has revealed that socialization, rather than course curriculum alone, informs shared beliefs that later manifest in policymaking contexts.Footnote 31

Existing scholarship on the personal characteristics of East Asian political leaders anecdotally suggests that education may affect preferences. China's think tanks, which play a central role in national policy advisory groups, are found to hire US-trained graduates who have brought in “the American model, particularly the ‘Chicago model.’”Footnote 32 Historical accounts of the backgrounds of the regions’ scientific leaders reveal that large numbers of them graduated from elite universities. This includes Tokyo University for Japan's Ministry of International Trade and Industry leaders,Footnote 33 Korea's prestigious “SKY” universities,Footnote 34 and leading technical universities in China, particularly Peking University and Tsinghua University.Footnote 35 Research has also shown a lasting impact on those who studied in the former Soviet bloc. Tsai, for instance, asserts that Chiang Ching-kuo's state-led industrialization policies in Taiwan were probably attributable to his training in the Soviet Union.Footnote 36

Drawing together these insights on the relationship between personal characteristics and policy preferences, we hypothesize that the subject studied and the university attended may have a lasting impact on graduates’ preferences. Cognate scholarship, in the context of how funding from a venture capital fund influences the growth of a startup, conceives of the notion of “imprinting,” in which the values of investors are transposed onto startups during their nascent stage.Footnote 37 Applying the same logic to the lives of policy leaders, we hypothesize that studying at an entrepreneurial university could constitute imprinting, in which the accumulation of particular cultural, human, and social capital derived from studying at the entrepreneurial university shapes the worldviews and values of graduates for years to come. Studies of imprinting in the startup–venture capital nexus draw on seminal imprinting scholarship.Footnote 38 Marquis and Tilcsik advance a general theory of imprinting, delineated in the context of individuals as “institutional conditions (e.g., an organization's culture) influence the norms, schemas, and skills that early-career individuals develop and carry with them in subsequent periods.”Footnote 39 This conceptualization informs our expectation that studying at an entrepreneurial university imprints a veneration of Schumpeterian modes of innovation, in which risk-taking innovators form new technology-oriented firms that drive society's economic growth.Footnote 40

This stems from the variety of entrepreneurship-related activities undertaken within the entrepreneurial university, to raise awareness, educate and train, and foster social networks. This collectively drives the accumulation of cultural, human, and social capital, which informs the generation of what Audretsch calls the “entrepreneurial capital” of the university.Footnote 41 To be more precise about the ways in which the institutional setting of the entrepreneurial university imprints upon graduates, we briefly develop our treatment of each form of capital.Footnote 42

First, the entrepreneurial university's cultural capital is conceived in terms of the promotion of shared values around entrepreneurship, including creativity, grit, mindset, perseverance, problem-solving, and risk-taking.Footnote 43 This valorization stems from activities such as the organization of clubs and events that foster dispositions in favor of entrepreneurship, as well as the universities’ production of (marketing) materials that reinforce these values.Footnote 44 Outside the classroom, entrepreneurship clubs invite successful founders and venture capitalists to give guest talks, raising awareness of, and veneration for, high-growth entrepreneurship.Footnote 45 Second, human capital comprises (formal) education and trainingFootnote 46 related to entrepreneurship.Footnote 47 In addition, research has shown that the American MBA curriculum has become increasingly centered on entrepreneurship.Footnote 48 MBA teaching that heavily employs the “case study method” cultivates entrepreneurial intentions and capabilities.Footnote 49 Third, social capital theory has been applied to study how being embedded in the entrepreneurial university shapes entrepreneurial intentions and performances.Footnote 50 Taking a Granovetterian tack,Footnote 51 in which social capital is primarily conceived of in terms of social networks, the activities of the entrepreneurial university foster “community engagement and networking/professional social skills”Footnote 52 through mentorship and coaching by “alumni entrepreneurs, experienced volunteers, and professors with prior academic entrepreneurship experience.”Footnote 53

Table 1 synthesizes these points about accumulated cultural, human, and social capital that may imprint on entrepreneurial university students, leaving alumni with a sustained preference for startup-centric forms of innovation.

Table 1.

Cultural, human, and social capital imprinting on entrepreneurial university students.

Based on the cultural, human, and social capital accumulation expectations of students embedded in the entrepreneurial university, as illustrated in Table 1, we hypothesize that studying at an entrepreneurial university, especially completing an MBA, is most likely to imprint startup-centric preferences upon alumni. Our central expectation is that policy leaders who are graduates of an entrepreneurial university, especially those who obtained an MBA, are especially likely to express a preference for startup-centric innovation. We hypothesize that they will speak about startups in their media engagement with greater frequency than their peers who did not attend one of these universities.Footnote 54

Data and methods

To comprehensively cover the range of senior leaders responsible for a country's innovation policy, we first identified the leading innovation policymaking organizations in the eleven largest economies in East Asia.Footnote 55 We identified flagship innovation policies in each country to determine the set of policymaking organizations. To capture high-profile innovation policies, we canvassed the Global Entrepreneurship Monitor's Startup Nations Atlas of Policies (SNAP) platform,Footnote 56 the Innovation Policy Platform (IPP), and policy studies produced by the Organisation for Economic Co-operation and Development, World Bank, European Union,Footnote 57 and national governments for the 2014–2019 period.Footnote 58 The appendix presents a full list of the agencies that we identified through this process.

We then identified the two most senior positions (e.g., director and deputy director) in each organization. This required searching government websites to specify which two titles were the most senior in each organization. This approach is consistent with studies of leaders’ personal characteristics in other empirical realms, such as that by Chwieroth, who mapped the two most senior positions; in that case, they were the finance minister and the head of the central bank.Footnote 59 The difference, for us, is that the nature of innovation policymaking meant we needed to identify the two top leadership posts across several organizations in each country. We mapped these two senior positions from the East Asian financial crisis in 1998 to 2019; 1998 is the starting point for this analysis, given the evidence that the crisis acted as a critical juncture, shaking up postwar economic development approaches and instigating interest in alternative models.Footnote 60

Next, we identified the individuals who held these posts from 1998 to 2019. To do this, we conducted in-depth desk research using social media (e.g., LinkedIn, Japan's Line, and China's Weibo), media coverage, and government websites to find the names of each position holder. The result of these efforts is a dataset of 485 innovation policy leaders. The appendix provides full details, and the Independent variable section explains how this dataset was compiled.

Dependent variable: Startup-centric innovation

Our dependent variable is policymakers’ preference for startup-centric innovation. There are two approaches to studying such preferences as a result of personal characteristics: one is to consider policy change or economic performance as the indicator of preference. Such studies link policy changeFootnote 61 and economic outcomesFootnote 62 to personal characteristics. In so doing, they assume that the policy change or economic outcome reflects the leader's preference. Others pursue a more causally proximate strategy for testing the relationship between personal characteristics and preferences expressed by studying the speeches and media data of leaders.Footnote 63

We take this second approach—focusing on speech as expressed in media as the best indicator of policy preference—for three reasons. First, innovation policies are often authored and led by multiple organizations,Footnote 64 and so it is difficult to reliably filter out the impact of any single policymaker. For example, China's 1999 “Opinions on Establishing a Venture Investment System” was coauthored by the Ministry of Science and Technology, the State Development Planning Commission, the State Economic and Trade Commission, People's Bank of China, the State Administration of Taxation, and the China Securities Regulatory Commission. While China's Ministry of Science and Technology was likely the lead author, the other entities and their leaders undoubtedly influenced the regulations, and so the resultant policy is unlikely to clearly reflect the preferences of the minister of science and technology alone.

Second, studying policy change or economic outcomes as the dependent variable presumes that those other determinants, such as the organizational structure, policymaking process, political environment, and broader economic context, matter relatively less. We, and others using the speech and media data approach, contend that what leaders say “is relatively unconstrained” in comparison to their policies.Footnote 65 Said differently, although media coverage may of course be constrained by the broader context, there is relatively more freedom for policy leaders to indicate their preferences in their speech than in the policies or initiatives they enact.

Third, and relatedly, we choose to study media data based on the observation that policy leaders may launch initiatives that have been years in the making, and, similarly, that policies may not be implemented until after they leave office. In Taiwan, for instance, some policy leaders are only in their position for a year; their preferences are unlikely to translate into policies within that (short) period.Footnote 66

For these reasons, our dependent variable, startup-centric innovation, is the frequency with which policy leaders communicate about startup-centric innovation policies in the media. In short, we contend that the way policy leaders talk about policy is a more robust measure if our goal is to understand and explain the relationship between personal characteristics and policy preferences. Therefore, we argue that media coverage of policymakers’ activities in regard to startup-centric innovation policies provides the best—although not a perfect—measure of the expressed preferences of individual policy leaders.

To collect the data for our dependent variable, we used Factiva,Footnote 67 a searchable index of global media and communications data. Following best practices established in text-as-data research,Footnote 68 we created n-grams specific to what we are testing, rather than using a large “off-the-shelf” dictionary, to aid the robustness of testing for the prevalence of startup-centric innovation. We generated this list of n-grams by situating startup-centric innovation policies within the state-of-the-art literature. Within wider industrial policy analysis, the literature describes three primary types of innovation policy: (1) invention- or R&D-focused, (2) national innovation system (NIS), and (3) transformational or mission-oriented.Footnote 69

Within these three types, scholars refer to startup-centric innovation policies as a specific variety of NIS policy—one that is focused on elements such as entrepreneurial skills, incubators and accelerators, unicorns, and venture capital.Footnote 70 Startup-centric innovation policies strive to create an enabling environment for startups across their life cycle—from antecedents, through founding conditions, to scaling up, and, ultimately, outcomes. We adopt a typology that distinguishes policies according to instrument type: (1) funding; (2) taxation; (3) regulation; (4) clusters, networks, and institutes; (5) attracting talent and investment; (6) stock market access; (7) technology infrastructure and public procurement; and (8) education and training.Footnote 71 The typology specifies the kinds of specific tools that make up these instrument types, including coaching, mentorship, coaching and training for founders, cultivation of entrepreneurial finance (equity funding—particularly angels and venture capital funds), and the provision of physical infrastructure (such as coworking spaces, subsidized office space, incubators).

Applying techniques used in state-of-the-art natural language processing research,Footnote 72 we identified thirty terms, or “n-grams”—that is, a set of words (grams) that are most referenced in the aforementioned scholarship on startup-centric innovation policies. The resulting set of n-grams included the following terms: accelerat*, angel, cluster, coworking, coach, cohort, early-stage, early-stage finance, ecosystem, entrepreneur, entrepreneurial finance, equity funding, founder, hi-tech, incubat*, mentor, network, new firms, risk-taking, Silicon Valley, startup, startup ecosystem, survival, tax breaks, technolog*, training, unicorn, venture, and venture capital.

We used these n-grams in our search strings in Factiva media data for each of the 485 innovation policy leaders. Each policy leader had a unique search string, including their name, our n-grams,Footnote 73 and a search time corresponding to the years during which they were in their post—from the beginning of the year (e.g., 1 January) they took office until the end (31 December) of their last year in post. For policy leaders currently in their role, we used 31 May 2020 as the end date. The result was 10,816 individual media documents.

It should be noted that our search strings were limited to English-language media coverage in each country. This approach is consistent with existing studies.Footnote 74 We acknowledge that this could exclude relevant domestic media sources from our analysis. Nevertheless, by focusing on English-language media, we ensure a level of consistency and reliability in our search strings and n-grams that would otherwise be lost or complicated in translation (especially across eight national languages). An analysis of the news media sources used in our analysis supports this claim. In most cases, the media sources used in our analysis are among the English news sources with the highest circulation in each country (or the leading news source) and, often, the English-language version of the highest-circulation local language newspaper. A full overview of our assessment of news sources can be found in Table A4 in the appendix.

Next, we employed a text-as-data approach using the R package tidyr Footnote 75 to analyze the documents and assess the extent to which the policymakers referred to startup-centric innovation initiatives in the media. For the purposes of our study, a mention of any of our n-grams constitutes a reference to startup-centric innovation policy, our dependent variable. We test these n-grams against our entire news media corpus using a dictionary approach. In short, we examine the term frequency of our n-grams relative to each document and the total corpus of documents per policymaker. This is commonly referred to as a term frequency-inverse document frequency (TFIDF) score.Footnote 76 In other words, TFIDF assesses the importance of terms relative to all other terms in each document within an entire corpus of documents.Footnote 77 Because we have data for the entire duration of each policymakers’ tenure as well as n-grams, we calculate our dependent variable, startup-centric innovation, as the sum of all TFIDF scores per policymaker. The higher the score, the more the policy leader mentions startup-centric innovation policies.Footnote 78

Independent variable

Our independent variable is based on the university education of the 485 innovation policy leaders in East Asia for the period 1998–2019. Our main interest lies in determining whether policy leaders obtained degrees from a leading entrepreneurial university. To this end, we operationalized our main independent variable, being an alumni of an entrepreneurial university, using the combined 2017 and 2018 startup activity rankings from Crunchbase. This is a list of universities whose current students or recent graduates founded the most startups that had raised venture capital funding of $1 million or more in the academic years in which the rankings were conducted. Combining data for these two periods (2017 and 2018) resulted in a list of thirty-three entrepreneurial universities.Footnote 79 Our reliance on an existing ranking is consistent with the approach of Guerrero and colleagues; in their study focused on the United Kingdom, they used the Russell Group set of twenty universities to determine the country's (most) entrepreneurial universities.Footnote 80 Rather than relying on university rankings, we use the Crunchbase list to identify the universities with graduates who have created the most venture-capital-backed startups. In doing so, we strive to disentangle the entrepreneurial nature of the university from its elite status. In comparing our list with the thirty-three top-ranked US News & World Report national universities in 2022,Footnote 81 we find that fifteen of the thirty-three (45.5 percent) elite-ranked universities are not included in our Crunchbase-generated list. For instance, the University of Chicago, Johns Hopkins University, and Vanderbilt University are all highly ranked by US News & World Report, but they are not included in our entrepreneurial university list.Footnote 82

An alternative means of identifying a cohort of entrepreneurial universities would be survey results, such as the Global University Entrepreneurial Spirit Students (GUESS) Survey, which reflects spirit across the university community.Footnote 83 However, the GUESS captures entrepreneurship in a wide sense, including family firm takeover intentions, rather than the technologically oriented startup-centric orientation that we are interested in assessing. Thus, we opted for the Crunchbase ranking, which is more focused on high-growth startups and not only universities’ wider rankings.

We contend that attendance at one of these entrepreneurial universities may shape graduates’ likelihood of speaking of startup-centric innovation activities later in their career as policymakers. In entrepreneurial university settings, such as the MBA program at Stanford University, students would have been immersed in an environment in which their peers were more likely to be taking entrepreneurship-themed courses, discussing entrepreneurial ideas, planning to build their businesses, or planning to work for a startup after graduation. It logically follows that this accumulated cultural, human, and social capital increases the propensity for graduates to form a preference for startup-centric modes of innovation. A potential limitation is that rankings have only recently begun to identify these universities, and the rankings are a US-centric list. Most of our policy leaders attended these universities before such identification began. However, the prevalence of startup activity on these campuses has developed over time. Thus, we contend that it is reasonable that the recent Crunchbase rankings reflect a long-standing ethos—one that would have been accumulating when our policy leaders attended these universities.

Entrepreneurial university is measured as a binary variable, taking the value of 1 if a policy leader obtained any degree from our list of universities and 0 otherwise. Based on our data, about 28 percent of our policy leaders have at least one entrepreneurial university degree. To examine how an entrepreneurial university experience differs from obtaining degrees elsewhere, we include two further binary variables: whether a policy leader studied abroad at an international university (31 percent) or at a domestic university (82 percent). It should be noted that these categories overlap because most of our policy leaders have multiple degrees at different levels. We also add a series of control variables reflecting the level (UG, master's, MBA, and PhD) at which they studied.Footnote 84 Because individuals can have more than one of the same kind of degree (e.g., two master's degrees), we code degree level as a series of individual dummy variables.

Finally, we include the following control variables in our analysis: the gender (1 = female; 0 = male) of policy leaders, the agency where they work(ed), and the country in which the agency is located. Lastly, to capture unobserved time variance, we include a variable, Year, based on the year when the policy leader joined the agency. Table 2 presents summary statistics for all variables.

Table 2.

Summary statistics.

Note: There are 508 observations because some of the 485 individuals identified as innovation policy leaders held more than one leadership role over the period studied.

Analysis

Before presenting our regression results, we first reveal where East Asia's innovation policy leaders studied. Figure 1 presents (1) the twenty most-attended universities on the left-hand side and (2) the twenty most-attended universities, excluding domestic universities, on the right-hand side. Entrepreneurial universities are indicated by dark navy bars. We can see that, when considering all 311 universities in our dataset, policy leaders overwhelmingly attend their country's high-ranking national universities, especially to complete a bachelor's degree.Footnote 85 A number of entrepreneurial universities also appear to be some of the most-attended universities, with Stanford (2.46 percent) and Harvard (2.03 percent) Universities leading the pack and the University of California, Cornell University, University of Pennsylvania, and University of Illinois following. Results are far more striking, however, when we look only at policy leaders who studied abroad, as illustrated on the right-hand side of Figure 1.

Figure 1.

Percentage of degrees obtained at the twenty most-attended universities. Entrepreneurial universities are indicated by dark blue bars. All other universities indicated by light blue bars.

Entrepreneurial universities dominate the top of this list, as illustrated in Figure 1. Nearly 28 percent of the degrees obtained at nondomestic universities came from one of the entrepreneurial universities, making these universities a popular destination for the policy leaders.

To answer our central question about the extent to which the place one studies, and the degree that is completed, affects leaders’ preference for startup-centric innovation, we estimate a series two-stage least squares (2SLS) regression models with additional dummies controlling for agency and time. 2SLS regression is a widely used approach for addressing endogeneity.Footnote 86 In our study, there is some concern about endogeneity between our dependent variable, startup-centric innovation, and one main explanatory variable, entrepreneurial university. Our argument is that attending an entrepreneurial university has a positive impact on the how policymakers talk about startup policy. There is, at the same time, a reasonable possibility that individuals with a propensity for using language around startup-centric innovation in their normal speech may be more inclined to select and attend an entrepreneurial university, and they may, in turn, continue to use startup-leaning language in their speech acts as policymakers. 2SLS addresses concerns about endogeneity by including an instrumental variable in the regression equation. The instrumental variable is selected because it is independent from the dependent variable but related to the explanatory variables. Our instrumental variable is country dummy. We argue that country dummies may shape where individuals in our dataset attend university but are unlikely to shape individual speech acts as their pertain to startup-centric language. Finally, there is, unsurprisingly, a perfect negative correlation between international university and domestic university. Therefore, we present the results of two regression models in which these additional explanatory variables are included separately. The results are presented in Table 3.

Table 3.

Two-stage least squares regression analysis of startup-centric innovation and education

Notes: Regression coefficients with standard errors in parentheses. * p < .05; ** p < .01; *** p < .001.

Our regression results provide considerable evidence supporting our central expectation. As shown in Models 1 and 2, obtaining at least one degree at an entrepreneurial university is strongly and positively correlated with reference to startup-centric innovation policy in the media. Specifically, the regression coefficients suggest that the mean difference between policymakers with an entrepreneurial university degree and those without is about 0.37 (Model 1) to 0.35 (Model 2). We also see no statistically significant differences for individuals obtaining degrees either from international or domestic universities. The effect appears to be unique to those with a degree from an entrepreneurial university. Plotting marginal effects (as we do in Figure 2) for this variable as well as international university and domestic university helps put this in context.

Figure 2.

Marginal effects of education on startup-centric innovation references.

Figure 2 indicates that the predicted value for startup-centric innovation for those with an entrepreneurial university degree is about 0.56 and, for those without, it is only 0.19. This is a sizeable increase of roughly three orders of magnitude or 194 percent. Critically, these differences are much smaller when we examine those with degrees from international universities (middle figure), where the mean difference is –0.07. Similarly, those with degrees from domestic universities show a negligible negative mean difference of –0.08. Results for both international and domestic universities are not statistically significant. An example of a policy leader's media coverage helps show the causal mechanism in action. Tony Tan, a senior minister in Singapore and MIT graduate, provides an example. He was quoted in the media stating, “While it is not possible to replicate Silicon Valley in Singapore, there are many lessons we can learn from Silicon Valley and elsewhere that can be adapted to our local environment.”Footnote 87 Tan, like the other entrepreneurial university alumni, used Silicon Valley as a reference point when discussing innovation more than his peers who did not graduate from entrepreneurial universities.

Turning back to Table 3, our regression analysis also revealed that having an entrepreneurial university MBA is positively correlated with referring to startup-centric innovation, and this is consistent across both models. The mean difference between those with an MBA and those without is about 0.04 to 0.06. To investigate this further, we estimated new regression models with interaction terms between our main independent variables (entrepreneurial university, international university, and domestic university) and an MBA degree specifically. In this case, we used ordinary least squares (OLS) regression to ease the interpretation of the interaction effects.Footnote 88 Results for the interaction terms are presented in Figure 3 as a series of coefficient plots with 95 percent confidence intervals. A full regression table is available in the appendix.

Figure 3.

Multilevel OLS regression of startup-centric innovation and educational backgrounds with interaction effects.

The interaction terms show no significant differences between policy leaders who obtain international degrees and those who hold domestic degrees. However, there is a sizeable positive and statistically significant effect for those with entrepreneurial university MBAs. These results suggest a mean difference between those receiving an MBA from an entrepreneurial university and those who hold an MBA from anywhere else, of 0.12 (p < .001). Thus, we conclude that while the MBA degree is important for speaking about startup-centric innovation, it is far more important when it is obtained at an entrepreneurial university. This supports our expectation that individuals with entrepreneurial university MBAs exhibit the greatest preference for startup-centric innovation as particularly desirable when they later take on policy leader positions.

Discussion: Entrepreneurial university alumni prefer startup-centric innovation

The results of our study reveal that 28 percent of East Asia's innovation policy leaders obtained degrees from an entrepreneurial university. Analyzing the media coverage of the full set of East Asian policy leaders, we find evidence that the entrepreneurial university alumni speak more frequently about startup-centric innovation than their peers who did not study at these entrepreneurial universities. Graduates of MBA programs at these universities go on to express startup-centric innovation policy preferences the most. This finding augurs well for a more systemic (e.g., national innovation policymaking) and longer-term (e.g., years later, when taking on policy leadership roles) mechanism for the entrepreneurial university to impact regional socioeconomic development.

This finding offers a new “alumni policymaker” mechanism by which innovation policy is shaped—namely, through graduates of entrepreneurial universities who later lead national innovation policymaking. Existing research on the entrepreneurial university has focused on more direct and shorter-term mechanisms that drive regional socioeconomic development, such as university-government interactions.Footnote 89 Research has also shown that returnee entrepreneurs can motivate innovation policy toward high-growth startups,Footnote 90 and that study visits undertaken while policymakers are in office act as mechanisms for the proliferation of startup-centric innovation policies.Footnote 91 Our findings suggest that the 28 percent of East Asia's innovation policy leaders who are alumni of the most active entrepreneurial universities in the United States form an additional mechanism for shaping regional socioeconomic development. While their peers who did not attend these entrepreneurial universities are almost certainly aware of the role of Schumpeterian entrepreneurship in innovation, our findings suggest that these entrepreneurial university alumni speak more often than their peers about partnerships or programs that they initiate with accelerators, investors, and startups.

Our empirical setting focused on the transmission of preferences through East Asia's policy leaders studying at entrepreneurial universities in the United States. It is worth noting that local universities in East Asia are also increasingly entrepreneurial.Footnote 92 We found, however, that graduates of domestic universities do not refer to startup-centric innovation to the same extent as those who studied—especially an MBA—at one of the United States’ most entrepreneurial universities. The potential for this US-based education to shape national policy directions raises important questions about the desirability of such an orientation. We do not contend that an innovation policy preference for or against this startup-centric innovation policy type is desirable; we simply suggest that the potential for innovation leaders’ academic background informing their preferences warrants policymaking organizations to give due consideration to this aspect of individuals background, and, especially, the diversity of educational backgrounds across policy leadership teams.

Conclusion

This article offers a novel investigation of the impact of leaders’ personal characteristics, especially the potential imprinting on alumni of entrepreneurial universities, on their innovation policy preferences. It strives to invigorate a new line of research into the relationship between entrepreneurial universities and national innovation policy, suggesting that policy leaders’ university education imprints long-lasting valorization of startup-centric innovation. In so doing, the imprinting may act as a spillover mechanism, thus extending the effects of the entrepreneurial university beyond direct collaboration between government and university, and beyond individual-level entrepreneurial pursuits of alumni, staff and students. Similar to the mechanisms driving the influence of education at particular universities on policy, we offer initial evidence that policy leaders’ studying at entrepreneurial universities could be contributing to the rise of startup-centric innovation policy across East Asia. We hope that this finding will invigorate further studies of this relationship between the entrepreneurial university and policy preference, in accordance with research in other domains that has shown that leaders’ personal characteristics, such as family status,Footnote 93 education and occupational background,Footnote 94 and migration experienceFootnote 95 affect policy preferences.

Three important issues emerge when interpreting these results, which could serve as avenues for research that takes this line of inquiry further. First, media coverage has limitations in that it includes both what policymakers say and what is said about them. This is not unique to our study, and other researchers have explained that it offers benefits in the form of capturing issues of reverse causality: how the media frames coverage of policymakers based on their background characteristics.Footnote 96 By analyzing media data, while we predominantly analyze the text of speeches and quotes made by policy leaders, we also capture the media's framing. This weakness is at least partially mitigated by the fact that we studied the same national media outlets for entrepreneurial university alumni and other policymakers, so the same narrative and sociotechnical systems are present for policy leaders operating in the same country at similar periods. Thus, while imperfect, we conclude that media coverage is a robust way of testing for policy preferences, as what policy leaders speak about is closer to their preferences than the policies implemented during their tenure.

The second limitation has to do with the set of entrepreneurial universities studied. The very nature of the Crunchbase rankings lends a US focus to our operationalization of the entrepreneurial university. We tried to mitigate this effect by differentiating international and domestic universities in the sample, so that we were testing for this set of entrepreneurial universities, and not transnational education (especially in the United States) more broadly. However, it would be ideal to have the most entrepreneurial universities identified in a more global way. Third, we note that there is an endogeneity issue here, in that individuals who choose to study certain programs, at certain universities, may have a proclivity toward embracing startups. We performed a 2SLS regression to help test for endogeneity but note that there may still be individual characteristics that are otherwise shaping innovation policy preferences.

Despite these limitations, we hope that our findings encourage scholars to take this line of research forward. First, in this initial test, we offer a new avenue for studying the impact of the entrepreneurial university—through the potential imprinting of cultural, human, and social capital in favor of high-growth startups—on public policymaking. Future research can extend this by interrogating what, precisely, about the entrepreneurial university experience shapes long-term preferences in this way. For instance, interviews or surveys with policy leaders could help unpack which aspects of the experience most affected their veneration of startup-centric innovation. Second, it also offers a novel approach to explaining the rising prevalence of startups in innovation policymaking. Our findings offer evidence that “alumni policymakers” from entrepreneurial universities may themselves, as a result of imprinting, act as a spillover mechanism for shaping regional socioeconomic development in favor of startup-centric modes of innovation. Additional research is needed to further unpack how the entrepreneurial university is shaping the future of innovation, and how innovation policymaking organizations consider the human resources implications of the finding.

Appendix

Table A1.

Innovation agency name (and abbreviation) by country.

Table A2.

List of policy leaders.

Table A3.

Example search string.

Table A4.

News media sources.

Table A5.

Entrepreneurial university list in comparison to US News & World Report 2022 rankings.

Table A6.

Multilevel OLS regression (robustness test against main regression models, Table 3 in the main text).

Table A7.

Multilevel OLS regression with interaction terms.

Footnotes

1 Etzkowitz (Reference Etzkowitz1983); Muscio and Ramaciotti (Reference Muscio and Ramaciotti2019); Sánchez (Reference Sánchez2013).

2 Binks, Starkey, and Mahon (Reference Binks, Starkey and Mahon2006).

3 Guerrero, Urbano, and Gajón (Reference Guerrero, Urbano and Gajón2020); Audretsch (Reference Audretsch2014).

4 Etzkowitz and Zhou (Reference Etzkowitz and Zhou2021); Klofsten et al. (Reference Klofsten, Fayolle, Guerrero, Mian, Urbano and Wright2019). We note that there are cognate terms and research streams to the entrepreneurial university. These include “academic entrepreneurship” (Shane Reference Shane2004; Wong Reference Wong2011; Geothner and Wyrwich Reference Geothner and Wyrwich2020), “professorial entrepreneurship” (Kenney and Goe Reference Kenney and Goe2004), “university entrepreneurship” (Rothaermel, Agung, and Jiang Reference Rothaermel, Agung and Jiang2007), and, to a lesser extent, research on the triple helix model (Etzkowitz Reference Etzkowitz2008).

5 Chalmers et al. (Reference Chalmers, Klingler-Vidra, Puglisi and Remke2021); Mercier (Reference Mercier2016); Hayo and Neueier (Reference Hayo and Neumeier2014); Besley, Montalvo, and Reynal-Querol (Reference Besley, Montalvo and Reynal-Querol2011); Dreher et al. (Reference Dreher, Lamla, Lein and Somogyi2009); Chwieroth (Reference Chwieroth2007); Gohlman and Vaubel (Reference Gohlmann and Vaubel2007).

6 See Cooiman (Reference Cooiman2022).

7 Alakent, Goktan, and Khoury (Reference Alakent, Sinan Goktan and Khoury2020, 4).

8 Schot and Steinmueller (Reference Schot and Steinmueller2018); Edler and Fagerberg (Reference Edler and Fagerberg2017); Linden (Reference Linden2004).

9 See Siegel et al. (Reference Siegel, Wessner, Binks and Lockett2003); Audretsch et al. (Reference Audretsch, Colombelli, Grilli, Minola and Rasmussen2020).

10 We employ the taxonomy delineated in Klingler-Vidra and Pacheco Pardo (Reference Klingler-Vidra and Pacheco Pardo2022, 4) and Pacheco Pardo and Klingler-Vidra (Reference Pacheco Pardo and Klingler-Vidra2019).

11 The Crunchbase data can be accessed at https://www.crunchbase.com/ .

12 Notably, our approach is similar to that of Guerrero, Cunningham, and Urbano (Reference Guerrero, Cunningham and Urbano2015), in that we use rankings rather than survey data.

13 Here, UG includes bachelor of arts and bachelor of science degrees. Master's includes master of arts and master of science. MBA is a distinct category, as graduates can (and often do) obtain both a master's degree and an MBA; therefore, we separate the two types of degrees. Finally, we capture data on PhD degrees.

14 Arroyabe, Schumann, and Arranz (Reference Arroyabe, Schumann and Arranz2022); Forliano, De Barnardi, and Yahiaoui (Reference Forliano, De Barnardi and Yahiaoui2021); Klofsten et al. (Reference Klofsten, Fayolle, Guerrero, Mian, Urbano and Wright2019).

15 We conceive of startup-centric innovation policy as a form of national innovation system policy focused on the creation and growth of early-stage, high-growth firms, consistent with Audretsch et al. (Reference Audretsch, Colombelli, Grilli, Minola and Rasmussen2020), Klingler-Vidra and Wade (Reference Klingler-Vidra and Wade2020), and Breznitz (Reference Breznitz2006).

16 Panagis (Reference Panagis, Olsson, Atikcan and Morin2021); Gentzkow, Kelley, and Taddy (Reference Gentzkow, Kelley and Taddy2019); Riffe et al. (Reference Riffe, Lacy, Watson and Fico2019); Loughran and McDonald (Reference Loughran and McDonald2016).

17 Prüfer and Prüfer (Reference Prüfer and Prüfer2020); Gentzkow, Kelley, and Taddy (Reference Gentzkow, Kelley and Taddy2019); Johnson, Arel-Bundock, and Portniaguine (Reference Johnson, Arel-Bundock and Portniaguine2019); Riffe et al. (Reference Riffe, Lacy, Watson and Fico2019); Loughran and McDonald (Reference Loughran and McDonald2016); Benoit and Herzog (Reference Benoit, Herzog, Bachner, Ginsberg and Hill2017).

18 See Schwarz, Traber, and Benoit (Reference Schwarz, Traber and Benoit2017); Benoit and Herzog (Reference Benoit, Herzog, Bachner, Ginsberg and Hill2017).

19 Johnson, Arel-Bundock, and Portniaguine (Reference Johnson, Arel-Bundock and Portniaguine2019).

20 See Rice and Zorn (Reference Rice and Zorn2021).

21 Norris et al. (2022).

22 Hayo and Neumeier (2016).

23 Besley, Montalvo, and Reynal-Querol (Reference Besley, Montalvo and Reynal-Querol2011).

24 Chalmers et al. (Reference Chalmers, Klingler-Vidra, Puglisi and Remke2021); Gohlmann and Vaubel (Reference Gohlmann and Vaubel2007).

25 Dreher et al. (Reference Dreher, Lamla, Lein and Somogyi2009).

26 Chwieroth (Reference Chwieroth2007).

27 Chwieroth (Reference Chwieroth2015).

28 Weymouth and Macpherson (Reference Weymouth and Muir Macpherson2012).

29 Dezalay and Garth (Reference Dezalay and Garth2010).

30 Weymouth and Macpherson (Reference Weymouth and Muir Macpherson2012).

31 Becker (Reference Becker1997).

32 Li, (Reference Li2016, 15–16).

33 Callon (Reference Callon1995).

34 B. Lee, “How the ‘SKY’ Universities Dominate,” Korea JoongAng Daily, 27 January 2003.

35 Cao (Reference Cao2004).

36 Tsai (Reference Tsai1999, 77).

37 Alakent, Goktan, and Khoury (Reference Alakent, Sinan Goktan and Khoury2020); Cooiman (Reference Cooiman2022).

38 Stinchcombe (Reference Stinchcombe and March1965); Mezias (Reference Mezias1990).

39 Marquis and Tilcsik (Reference Marquis and Tilcsik2013, 58).

40 Arroyabe, Schumann, and Arranz (Reference Arroyabe, Schumann and Arranz2022).

41 Audretsch (Reference Audretsch2014).

42 Bourdieu (Reference Bourdieu and Richardson1986); Coleman (Reference Coleman1988); Becker (Reference Becker1993).

43 Hulen and Tumunbayarova (Reference Hulen and Tumunbayarova2020).

44 Geothner and Wyrwich (Reference Geothner and Wyrwich2020).

45 Pittaway et al. (Reference Pittaway, Rodriguez-Falcon, Aiyegbayo and King2011).

46 Becker (Reference Becker1993, 17).

47 Uslu et al. (Reference Uslu, Calikoglu, Nevra Seggie and Seggie2019).

48 Binks, Starkey, and Mahon (Reference Binks, Starkey and Mahon2006).

49 R. Jack, “Why Harvard's Case Studies Are Under Fire,” Financial Times, 29 October 2018.

50 Salamzadeh, Sangosanya, and Salamzadeh (Reference Salamzadeh, Sangosanya and Salamzadeh2022); Redondo and Camarero (Reference Redondo and Camarero2019); Fengqiao and Dan (Reference Fengqiao and Dan2015).

51 Granovetter (Reference Granovetter1973).

52 Martínez-Martínez and Ventura (Reference Martínez-Martínez and Ventura2020, 10).

53 Klofsten et al. (Reference Klofsten, Fayolle, Guerrero, Mian, Urbano and Wright2019, 153).

54 Bloch et al. (Reference Bloch, Mitterle, Paradeise and Peter2017).

55 The countries with the largest gross domestic product in East Asia are China, Hong Kong, Indonesia, Japan, Korea, Malaysia, the Philippines, Singapore, Taiwan, Thailand, and Vietnam.

56 For more on the Global Enterprise Monitor's SNAP platform, visit https://www.genglobal.org/startup-nations/snap#:~:text=The%20Startup%20Nations%20Atlas%20of,public%2Dsector%2Dsupported%20programs. Audretsch et al. (Reference Audretsch, Colombelli, Grilli, Minola and Rasmussen2020) also use the SNAP database.

57 The European Union funds the China Innovation Funding project: http://chinainnovationfunding.eu/china-innovation-policies/.

58 OECD (2016); OECD and World Bank (2014); Ambashi (Reference Ambashi2018).

59 See Chwieroth (Reference Chwieroth2007).

60 Pempel (Reference Pempel2021); Callon (Reference Callon1995).

61 Mercier (Reference Mercier2016); Chwieroth (Reference Chwieroth2007, Reference Chwieroth2015).

62 Hayo and Neumeier (Reference Hayo and Neumeier2014); Besley, Montalvo, and Reynal-Querol (Reference Besley, Montalvo and Reynal-Querol2011).

63 Johnson, Arel-Bundock, and Portniaguine (Reference Johnson, Arel-Bundock and Portniaguine2019); Schwarz, Traber, and Benoit (Reference Schwarz, Traber and Benoit2017); Benoit and Herzog (Reference Benoit, Herzog, Bachner, Ginsberg and Hill2017).

64 Breznitz, Ornston, and Stamford (Reference Breznitz, Ornston and Stamford2018).

65 Schwarz, Traber, and Benoit (Reference Schwarz, Traber and Benoit2017).

66 Jones and Olken (Reference Jones and Olken2005) showed this with respect to tenure in office.

67 Available at https://professional.dowjones.com/factiva/.

68 Loughran and McDonald (Reference Loughran and McDonald2016).

69 See research that delineates innovation policy types, notably, Schot and Steinmueller (Reference Schot and Steinmueller2018); Edler and Fagerberg (Reference Edler and Fagerberg2017).

70 Autio and Rannikko (Reference Autio and Rannikko2016); Audretsch et al. (Reference Audretsch, Colombelli, Grilli, Minola and Rasmussen2020).

71 Klingler-Vidra and Pacheco Pardo (Reference Klingler-Vidra and Pacheco Pardo2022, 5); Pacheco Pardo and Klingler-Vidra (Reference Pacheco Pardo and Klingler-Vidra2019).

72 See Loughran and McDonald (Reference Loughran and McDonald2016).

73 A subset of n-grams was used in our initial search because of limits on search string length in Factiva. Ultimately, we intentionally employed terms that would ensure that our initial media search cast our net wider to avoid unwittingly omitting relevant documents.

74 See Massey and Chang (Reference Massey and Chang2002) as an example of another East Asian regional study that uses English-language sources in its media analysis.

75 The tidyr package can be found at https://cran.r-project.org/web/packages/tidyr/index.html.

76 Gentzkow, Kelley, and Tabby (Reference Gentzkow, Kelley and Taddy2019, 538).

77 TFIDF (t, d) = TF(t, d) * IDF(t), where t = “term” and d = “document.”

78 Haggard and Zheng (Reference Haggard and Zheng2013) and Keller and Pauly (Reference Keller and Pauly2009).

79 The following thirty-three universities, and in some cases just their business schools, are the entrepreneurial universities identified: Berkeley-Haas; Brown; Carnegie Mellon; Columbia; Columbia Business School; Cornell; Dartmouth; Duke; Harvard; Kellogg (Northwestern University); MIT; MIT-Sloan; Northwestern; NYU; NYU-Stern; Penn State; Princeton; Stanford; UC Berkeley; UCLA; UCLA-Anderson; University of Colorado; University of Illinois; University of Michigan; University of Pennsylvania; University of Southern California (USC); University of Virginia; University of Washington; UC San Diego; USC-Marshall; UW Madison; Wharton; Yale.

80 Guerrero, Cunningham, and Urbano (Reference Guerrero, Cunningham and Urbano2015, 752–53).

81 The US News & World Report 2022 rankings are available at https://www.usnews.com/best-colleges/rankings/national-universities.

82 The full list of the top thirty-three ranked US News & World Report universities in 2022, in comparison to our Crunchbase-determined set of entrepreneurial universities, is included in Table A5 in the appendix.

83 Lechuga Sancho et al. (Reference Sancho, Paula, Ramos-Rodríguez and Vega2021) and Meek and Gianiodis (Reference Meek and Gianiodis2020) used this survey, for example.

84 This coding of university studies at four distinct levels is consistent with the approach taken in Klingler-Vidra et al. (Reference Klingler-Vidra, Tran and Chalmers2021, Reference Klingler-Vidra, Hai, Liu and Chalmers2022).

85 QS World Rankings 2020 name the top university for each country (with global ranking indicated), as follows: China (Tsinghua University, #16), Hong Kong (University of Hong Kong, #25), Indonesia (Universitas Indonesia, #296), Japan (University of Tokyo, #22), Korea (Seoul National University, #37), Malaysia (Universiti Malaya, #70), Philippines (University of the Philippines Manila, #365), Singapore (National University of Singapore, #11 ), Taiwan (National Taiwan University, #69), Thailand (Mahidol University, #316), Vietnam (Vietnam National University, #801-1000).

86 See Terza, Basu, and Rathouz (Reference Terza, Basu and Rathouz2009).

87 Ramchandani (Reference Ramchandani2007).

88 2SLS regression links the instrumental variable to our explanatory variable, making the interpretation of interaction effects on that same explanatory variable rather difficult. Therefore, we instead reran our regression models using OLS but with fixed effects for agency and year. Furthermore, we assessed consistency in the results of our 2SLS models and OLS models. The results are presented in the appendix.

89 Etzkowitz (Reference Etzkowitz2008).

90 Kenney, Breznitz, and Murphree (Reference Kenney, Breznitz and Murphree2013).

91 Klingler-Vidra (Reference Klingler-Vidra2018).

92 Park and Jeong (Reference Park and Jeong2015); Mok (Reference Mok2005, Reference Mok2013).

93 Hayo and Neumeier (Reference Hayo and Neumeier2014).

94 Dreher et al. (Reference Dreher, Lamla, Lein and Somogyi2009); Chwieroth (Reference Chwieroth2007, Reference Chwieroth2015); Gohlmann and Vaubel (Reference Gohlmann and Vaubel2007).

95 Mercier (Reference Mercier2016).

96 See Riffe et al. (Reference Riffe, Lacy, Watson and Fico2019); Gentzkow and Shapiro (Reference Gentzkow and Shapiro2010).

Notes: 485 individuals, some of which have an asterisk after the first mention, to indicate that they appear more than once because they held more than one top innovation policy position over time.

Notes: Standard errors in parentheses. * p < .10; ** p < .05; *** p < .01.

Notes: Standard errors in parentheses. * p < .10; ** p < .05; *** p < .01.

References

Alakent, Ekin, Sinan Goktan, M., and Khoury, Theodore A.. 2020. “Is Venture Capital Socially Responsible? Exploring the Imprinting Effect of VC Funding on CSR Practices.” Journal of Business Venturing 35 (3): Article 106005. https://doi.org/10.1016/j.jbusvent.2020.106005.CrossRefGoogle Scholar
Ambashi, Masahito, ed. 2018. Innovation Policy in ASEAN. Jakarta, Indonesia: Economic Research Institute for ASEAN and East Asia. Accessed April 13, 2023, https://www.eria.org/publications/innovation-policy-in-asean/.Google Scholar
Arroyabe, Marta F., Schumann, Martin, and Arranz, Carlos F. A.. 2022. “Mapping the Entrepreneurial University Literature: A Text Mining Approach.” Studies in Higher Education 47 (5): 955–63. https://doi.org/10.1080/03075079.2022.2055318.CrossRefGoogle Scholar
Audretsch, David B. 2014. “From the Entrepreneurial University to the University for the Entrepreneurial Society.” Journal of Technology Transfer 39: 313–21. https://doi.org/10.1007/s10961-012-9288-1.CrossRefGoogle Scholar
Audretsch, David, Colombelli, Alessandra, Grilli, Luca, Minola, Tommaso, and Rasmussen, Einar. 2020. “Innovative Start-ups and Policy Initiatives.” Research Policy 49 (10): Article 104027. https://doi.org/10.1016/j.respol.2020.104027.CrossRefGoogle Scholar
Autio, E., and Rannikko, H.. 2016. “Retaining winners: Can policy boost high-growth entrepreneurship?.” Research Policy 45 (1): 42–55. doi:10.1016/j.respol.2015.06.002.CrossRefGoogle Scholar
Becker, Gary S. 1993. Human Capital: A Theoretical and Empirical Analysis with Special References to Education. Chicago: Chicago University Press.CrossRefGoogle Scholar
Becker, Gary S. 1997. “What Latin America Owes to the ‘Chicago Boys.’” Hoover Digest, 30 October. Accessed April 13, 2023, https://www.hoover.org/research/what-latin-america-owes-chicago-boys.Google Scholar
Benoit, Kenneth, and Herzog, Alexander. 2017. “Text Analysis: Estimating Policy Preference From Written and Spoken Word.” In Analytics, Policy and Governance, edited by Bachner, Jennifer, Ginsberg, Benjamin, and Hill, Kathryn Wagner, 137–59. New Haven, CT: Yale University Press.Google Scholar
Besley, Timothy, Montalvo, Jose G., and Reynal-Querol, Marta. 2011. “Do Educated Leaders Matter?” Journal of Economics 121: F205–27. https://doi.org/10.1111/j.1468-0297.2011.02448.x.CrossRefGoogle Scholar
Binks, Martin, Starkey, Ken, and Mahon, Christopher L.. 2006. “Entrepreneurship Education and the Business School.” Technology Analysis & Strategic Management 18 (1): 1–18. https://doi.org/10.1080/09537320500520411.CrossRefGoogle Scholar
Bloch, Roland, Mitterle, Alexander, Paradeise, Catherine, and Peter, Tobias. 2017. Universities and the Production of Elites: Discourses, Policies, and Strategies of Excellence and Stratification in Higher Education. Basingstoke: Palgrave Macmillan.Google Scholar
Bourdieu, Pierre. 1986. “The Forms of Capital.” In Handbook of Theory and Research for the Sociology of Education, edited by Richardson, John, 241–58. Westport, CT: Greenwood Press.Google Scholar
Breznitz, Dan. 2006. “Innovation-Based Industrial Policy in Emerging Economies? The Case of Israel's IT Industry.” Business and Politics 8 (3): 1–35. https://doi.org/10.2202/1469-3569.1155.CrossRefGoogle Scholar
Breznitz, Dan, Ornston, Darius, and Stamford, Steven. 2018. “Mission Critical: The Ends, Means and Design of Innovation Agencies.” Industrial and Corporate Change 5 (1): 883–96. https://doi.org/10.1093/icc/dty027.CrossRefGoogle Scholar
Callon, Scott. 1995. Divided Sun: MITI and the Breakdown of Japanese High-Tech Industrial Policy, 1975–1993. Stanford, CA: Stanford University Press.CrossRefGoogle Scholar
Cao, Cong. 2004. China's Scientific Elite. Abingdon: RoutledgeCurzon.CrossRefGoogle Scholar
Chalmers, Adam William, Klingler-Vidra, Robin, Puglisi, Alfio, and Remke, Lisa. 2021. “In and Out of Revolving Doors in European Union Financial Regulatory Authorities.” Regulation and Governance 16 (4): 1233–49. https://doi.org/10.1111/rego.12424.CrossRefGoogle Scholar
Chwieroth, Jeffrey. 2007. “Neoliberal Economists and Capital Account Liberalization in Emerging Markets.” International Organization 61: 443–63. https://doi.org/10.1017/S0020818307070154.CrossRefGoogle Scholar
Chwieroth, Jeffrey. 2015. “Professional Ties that Bind: How Normative Orientations Shape IMF Conditionality.” Review of International Political Economy 22 (5): 757–87. https://doi.org/10.1080/09692290.2014.898214.CrossRefGoogle Scholar
Coleman, James S. 1988. “Social Capital in the Creation of Human Capital.” American Journal of Sociology 94: S95–120. https://doi.org/10.1086/228943.CrossRefGoogle Scholar
Cooiman, Franziska. 2022. “Imprinting the Economy: The Structural Power of Venture Capital.” Environment and Planning A: Economy and Space. https://doi.org/10.1177/0308518X221136559.CrossRefGoogle Scholar
Dezalay, Yves, and Garth, Bryant G.. 2010. Asian Legal Revivals: Lawyers in the Shadow of Empire. Chicago: University of Chicago Press.CrossRefGoogle Scholar
Dreher, Axel, Lamla, Michael J., Lein, Sarah M., and Somogyi, Frank. 2009. “The Impact of Political Leaders’ Profession and Education on Reforms.” Journal of Comparative Economics 37: 169–93. https://doi.org/10.1016/j.jce.2008.08.005.CrossRefGoogle Scholar
Edler, Jakob, and Fagerberg, Jan. 2017. “Innovation Policy: What, Why and How.” Oxford Review of Economic Policy 13 (1): 2–23. https://doi.org/10.1093/oxrep/grx001.CrossRefGoogle Scholar
Etzkowitz, Henry. 1983. “Entrepreneurial Scientists and Entrepreneurial Universities in American Academic Science.” Minerva 21: 198–233.CrossRefGoogle ScholarPubMed
Etzkowitz, Henry. 2008. The Triple Helix: University-Industry-Government Innovation in Action. New York: Routledge.CrossRefGoogle Scholar
Etzkowitz, Henry, and Zhou, Chunyan. 2021. “Licensing Life: The Evolution of Stanford University's Technology Transfer Practice.” Technological Forecasting and Social Change 168: 120764. https://doi.org/10.1016/j.techfore.2021.120764.CrossRefGoogle Scholar
Fengqiao, Yan, and Dan, Mao. 2015. “The Impact of Social Capital on the Employment of College Graduates.” Chinese Education and Society 48 (1): 59–75. https://doi.org/10.1080/10611932.2015.1014708.CrossRefGoogle Scholar
Forliano, Canio, De Barnardi, Paola, and Yahiaoui, Dorra. 2021. “Entrepreneurial Universities: A Bibliometric Analysis within the Business and Management Domains.” Technological Forecasting and Social Change 165: 120522. https://doi.org/10.1016/j.techfore.2020.120522.CrossRefGoogle Scholar
Gentzkow, Matthew, and Shapiro, Jesse M.. 2010. “What Drives Media Slant? Evidence from U.S. Daily Newspapers.” Econometrica 78 (1): 35–71. https://doi.org/10.3982/ECTA7195.Google Scholar
Gentzkow, Matthew, Kelley, Bryan, and Taddy, Matt. 2019. “Text as Data.” Journal of Economics Literature 57 (3): 535–74. https://doi.org/10.1257/jel.20181020.CrossRefGoogle Scholar
Geothner, Maximilian, and Wyrwich, Michael. 2020. “Cross-Faculty Proximity and Academic Entrepreneurship: The Role of Business Schools.” Journal of Technology Transfer 45: 1016–62. https://doi.org/10.1007/s10961-019-09725-0.CrossRefGoogle Scholar
Gianiodis, Peter T., Markman, Gideon D., and Panagopoulos, Andreas. 2016. “Entrepreneurial Universities and Overt Opportunism.” Small Business Economics 47: 609–31. https://doi.org/10.1007/s11187-016-9753-6.CrossRefGoogle Scholar
Gohlmann, Silja, and Vaubel, Roland. 2007. “The Educational and Occupational Background of Central Bankers and Its Effect on Inflation: An Empirical Analysis.” European Economic Review 51: 925–41. https://doi.org/10.1016/j.euroecorev.2006.05.001.CrossRefGoogle Scholar
Granovetter, Mark S. 1973. “The Strength of Weak Ties.” American Journal of Sociology 78 (6): 1360–80.CrossRefGoogle Scholar
Guerrero, Maribel, Cunningham, James A., and Urbano, David. 2015. “Economic Impact of Entrepreneurial Universities’ Activities: An Exploratory Study of the United Kingdom.” Research Policy 44 (3): 748–64. https://doi.org/10.1016/j.respol.2014.10.008.CrossRefGoogle Scholar
Guerrero, Maribel, Urbano, David, and Gajón, Edouardo. 2020. “Entrepreneurial University Ecosystems and Graduates’ Career Patterns: Do Entrepreneurship Education Programmes and University Business Incubators Matter?” Journal of Management Development 39 (5): 753–75. https://doi.org/10.1108/JMD-10-2019-0439.CrossRefGoogle Scholar
Haggard, Stephan, and Zheng, Yu. 2013. “Institutional Innovation and Investment in Taiwan: The Micro-foundations of the Developmental State.” Business and Politics 15 (4): 435–66. https://doi.org/10.1515/bap-2012-0010.CrossRefGoogle Scholar
Hayo, Bernd, and Neumeier, Florian. 2014. “Political Leaders’ Socioeconomic Background and Fiscal Performance in Germany.” European Journal of Political Economy 34: 184–205. https://doi.org/10.1016/j.ejpoleco.2014.01.009.CrossRefGoogle Scholar
Hulen, Peter, and Tumunbayarova, Zhargal. 2020. “Building Students’ Entrepreneurial Mindsets: Results from an Intervention at a Russian University.” International Journal of Management in Education 18 (2): Article 100380. https://doi.org/10.1016/j.ijme.2020.100380.Google Scholar
Johnson, Juliet, Arel-Bundock, Vincent, and Portniaguine, Vladislav. 2019. “Adding Rooms onto a House We Love: Central Banking after the Global Financial Crisis.” Public Administration 97 (3): 546–60. https://doi.org/10.1111/padm.12567.CrossRefGoogle Scholar
Jones, Benjamin F., and Olken, Benjamin A.. 2005. “Do Leaders Matter? National Leadership and Growth Since World War II.” Quarterly Journal of Economics 120 (3): 835–64. https://doi.org/10.1093/qje/120.3.835.Google Scholar
Keller, William W., and Pauly, Louis W.. 2009. “Innovation in the Indian Semiconductor Industry: The Challenge of Sectoral Deepening.” Business and Politics 11 (2): 1–21. https://doi.org/10.2202/1469-3569.1270.CrossRefGoogle Scholar
Kenney, Martin, and Goe, W. Richard. 2004. “The Role of Social Embeddedness in Professorial Entrepreneurship: A Comparison of Electrical Engineering and Computer Science at UC Berkeley and Stanford.” Research Policy 33 (5): 691–707. https://doi.org/10.1016/j.respol.2003.11.001.CrossRefGoogle Scholar
Kenney, Martin, Breznitz, Dan, and Murphree, Michael. 2013. “Coming Back Home after the Sun Rises: Returnee Entrepreneurs and Growth of High-tech Industries.” Research Policy 42 (2): 391–407. https://doi.org/10.1016/j.respol.2012.08.001.CrossRefGoogle Scholar
Klingler-Vidra, Robyn. 2018. The Venture Capital State: The Silicon Valley Model in East Asia. Ithaca, NY: Cornell University Press.CrossRefGoogle Scholar
Klingler-Vidra, Robyn, and Pacheco Pardo, Ramon. 2022. “Bridging Developmental State and Entrepreneurial State Theory: A Typology of Startup Policies’ Incumbent Firm Benefits.” Journal of Asian Public Policy. https://doi.org/10.1080/17516234.2022.2132584.CrossRefGoogle Scholar
Klingler-Vidra, Robyn, and Wade, Robert H.. 2020. “Science and Technology Policies and the Middle Income Trap: Lessons from Vietnam.” Journal of Development Studies 56 (4): 717–31. https://doi.org/10.1080/00220388.2019.1595598.CrossRefGoogle Scholar
Klingler-Vidra, Robyn, Tran, Ba Linh, and Chalmers, Adam William. 2021. “Transnational Experience and High-Performing Entrepreneurs in Emerging Economies: Evidence from Vietnam.” Technology in Society 66: 101605. https://doi.org/10.1016/j.techsoc.2021.101605.CrossRefGoogle Scholar
Klingler-Vidra, Robyn, Hai, Steven Jiawei, Liu, Ye, and Chalmers, Adam William. 2022. “Is the Jack Ma Trajectory Unique? Assessing the Place-Based Hypothesis on Entrepreneurial Success.” Journal of Small Business & Entrepreneurship 34 (4): 419–42. https://doi.org/10.1080/08276331.2021.197.CrossRefGoogle Scholar
Klofsten, Magnus, Fayolle, Alain, Guerrero, Maribel, Mian, Sarfraz, Urbano, David, and Wright, Mike. 2019. “The Entrepreneurial University as Driver for Economic Growth and Social Change—Key Strategic Challenges.” Technological Forecasting and Social Change 141: 149–58. https://doi.org/10.1016/j.techfore.2018.12.004.CrossRefGoogle Scholar
Sancho, Lechuga, Paula, María, Ramos-Rodríguez, Antonio Rafail, and Vega, María Angeles Frende. 2021. “Is a Favorable Entrepreneurial Climate Enough to Become an Entrepreneurial University? An International Study with GUESSS Data.” International Journal of Management Education 19 (3): Article 100536. https://doi.org/10.1016/j.ijme.2021.100536.Google Scholar
Li, Cheng. 2016. “China's New Think Tanks: Where Officials, Entrepreneurs, and Scholars Interact.” China Leadership Monitor 29: 1–21.Google Scholar
Linden, Greg. 2004. “China Standard Time: A Study in Strategic Industrial Policy.” Business and Politics 6 (3): 1–26. https://doi.org/10.2202/1469-3569.1069.CrossRefGoogle Scholar
Loughran, Tim, and McDonald, Bill. 2016. “Textual Analysis in Accounting and Finance: A Survey.” Journal of Accounting Research 54 (4): 1187–1230. https://doi.org/10.1111/1475-679X.12123.CrossRefGoogle Scholar
Marquis, Christopher, and Tilcsik, András. 2013. “Imprinting: Toward a Multilevel Theory.” Academy of Management Annals 7 (1): 195–245. https://doi.org/10.1080/19416520.2013.766076.CrossRefGoogle Scholar
Martínez-Martínez, Sofía Louise, and Ventura, Rafael. 2020. “Entrepreneurial Profiles at the University: A Competence Approach.” Frontiers in Psychology 11: Article 612796. https://doi.org/10.3389/fpsyg.2020.612796.CrossRefGoogle ScholarPubMed
Massey, Brian L., and Chang, Li-jing Arthur. 2002. “Locating Asian Values in Asian Journalism: A Content Analysis of Web Newspapers.” Journal of Communications 52 (4): 987–1003. https://doi.org/10.1111/j.1460-2466.2002.tb02585.x.CrossRefGoogle Scholar
Meek, W. R., and Gianiodis, P.. 2020. “The Death and Rebirth of the Entrepreneurial University Model.” Academy of Management Perspectives 37 (1). https://doi.org/10.5465/amp.2020.0180.Google Scholar
Mercier, Marion. 2016. “The Return of the Prodigy Son: Do Return Migrants Make Better Leaders?” Journal of Development Economics 122: 76–91. https://doi.org/10.1016/j.jdeveco.2016.04.005.CrossRefGoogle Scholar
Mezias, Stephen J. 1990. “An Institutional Model of Organizational Practice: Financial Reporting at the Fortune 200.” Administrative Science Quarterly 35 (3): 431–57. https://doi.org/10.2307/2393312.CrossRefGoogle Scholar
Mok, Ka Ho. 2005. “Fostering Entrepreneurship: Changing Role of Government and Higher Education Governance in Hong Kong.” Research Policy 34 (4): 537–54. https://doi.org/10.1016/j.respol.2005.03.003.CrossRefGoogle Scholar
Mok, Ka Ho. 2013. “The Quest for an Entrepreneurial University in East Asia: Impact on Academics and Administrators in Higher Education.” Asia Pacific Education Review 14: 11–22. https://doi.org/10.1007/s12564-013-9249-x.CrossRefGoogle Scholar
Muscio, Alessandro, and Ramaciotti, Laura. 2019. “How Does Academia Influence PhD Entrepreneurship? New Insights on the Entrepreneurial University.” Technovation 82–83: 16–24. https://doi.org/10.1016/j.technovation.2019.02.003.CrossRefGoogle Scholar
Norris, William J., Vaughan-Naron, Katie, Kashyap, Neha, and Rodgers, Joseph Balmain. 2022. “Inoculating the University R&D Enterprise: How RISC Can Strengthen Post-COVID-19 Research Integrity and Global Supply Chains.” Business and Politics 24 (2): 113–32. https://doi.org/10.1017/bap.2022.2.CrossRefGoogle Scholar
OECD (Organisation for Economic Co-operation and Development). 2016. OECD Reviews of Innovation Policy: Malaysia 2016. Paris OECD Publishing,.Google Scholar
OECD (Organisation for Economic Co-operation and Development) and World Bank. 2014. Science, Technology and Innovation in Viet Nam. Paris: OECD.Google Scholar
Pacheco Pardo, Ramon, and Klingler-Vidra, Robyn. 2019. “The Entrepreneurial Developmental State: What Is the Perceived Impact of South Korea's Creative Economy Action Plan on Entrepreneurial Activity?” Asian Studies Review 43 (2): 313–31. https://doi.org/10.1080/10357823.2019.1589418.CrossRefGoogle Scholar
Panagis, Yannis. 2021. “Automated Text Analysis: The Application of Automatic Text Processing in the Social Sciences.” In Research Methods in the Social Sciences: An A-Z of Key Concepts, edited by Olsson, Christian, Atikcan, Ece Özlem, and Morin, Jean-Frédéric, 14–16. Oxford: Oxford University Press.Google Scholar
Park, Mun-su, and Jeong, Seung Ouk. 2015. “Entrepreneurial University for Science and Technology: Cases of KAIST and POSTECH.” Journal of Science and Technology Policies and Management 4 (4): 89–105.Google Scholar
Pempel, T. J. 2021. A Regime of Regimes: Prosperity and Plunder in the Asia-Pacific. Ithaca, NY: Cornell University Press.Google Scholar
Pittaway, Luke, Rodriguez-Falcon, Elena, Aiyegbayo, Olaojo, and King, Amanda. 2011. “The Role of Entrepreneurship Clubs and Societies in Entrepreneurial Learning.” International Small Business Journal 29 (1): 37–57. https://doi.org/10.1177/0266242610369876.CrossRefGoogle Scholar
Prüfer, Jens, and Prüfer, Patricia. 2020. “Data Science for Entrepreneurship Research: Studying Demand Dynamics for Entrepreneurial Skills in the Netherlands.” Small Business Economics 55: 651–72. https://doi.org/10.1007/s11187-019-00208-y.CrossRefGoogle Scholar
Ramchandani, N. 2007. Promote local high-tech start-ups: Tony Tan. The Business Times. Singapore, November 28.Google Scholar
Redondo, María, and Camarero, Carmen. 2019. “Social Capital in University Business Incubators: Dimensions, Antecedents and Outcomes.” International Entrepreneurship Management Journal 15: 599–624. https://doi.org/10.1007/s11365-018-0494-7.CrossRefGoogle Scholar
Rice, Douglas R., and Zorn, Christopher. 2021. “Corpus-Based Dictionaries for Sentiment Analysis of Specialized Vocabularies.” Political Science Research and Methods 9: 20–35. https://doi.org/10.1017/psrm.2019.10.CrossRefGoogle Scholar
Riffe, Daniel, Lacy, Stephen, Watson, Brendan, and Fico, Frederick. 2019. Analyzing Media Messages: Using Quantitative Content Analysis in Research. London: Routledge.CrossRefGoogle Scholar
Rothaermel, Frank T., Agung, Shanti, and Jiang, Lin. 2007. “University Entrepreneurship: A Taxonomy of the Literature.” Industrial and Corporate Change 16 (4): 691–791. https://doi.org/10.1093/icc/dtm023.CrossRefGoogle Scholar
Salamzadeh, Yashar, Sangosanya, Taofeek Adeyemi, and Salamzadeh, Aidin. 2022. “Entrepreneurial Universities and Social Capital: The Moderating Role of Entrepreneurial Intention in the Malaysian Context.” International Journal of Management Education 20: Article 100609. https://doi.org/10.1016/j.ijme.2022.100609.Google Scholar
Sánchez, José C. 2013. “The Impact of an Entrepreneurship Education Program on Entrepreneurial Competences and Intention.” Journal of Small Business Management 51 (3): 447–65. https://doi.org/10.1111/jsbm.12025.CrossRefGoogle Scholar
Schot, Johan, and Steinmueller, W. Edward. 2018. “Three Frames for Innovation Policy: R&D, Systems of Innovation and Transformative Change.” Research Policy 47 (9): 1554–67. https://doi.org/10.1016/j.respol.2018.08.011.CrossRefGoogle Scholar
Schwarz, D., Traber, D., and Benoit, K.. 2017. “Estimating Intra-Party Preferences: Comparing Speeches to Votes.” Political Science Research and Methods 5 (2): 379–96. doi:10.1017/psrm.2015.77.CrossRefGoogle Scholar
Shane, Scott. 2004. “Encouraging University Entrepreneurship? The Effect of the Bayh-Dole on University Patenting in the United States.” Journal of Business Venturing 19: 127–51. https://doi.org/10.1016/S0883-9026(02)00114-3.CrossRefGoogle Scholar
Siegel, Donald S., Wessner, Charles, Binks, Martin, and Lockett, Andy. 2003. “Policies Promoting Innovation in Small Firms: Evidence from the U.S. and U.K.” Small Business Economics 20 (2): 121–27. https://doi.org/10.1023/A:1022271914246.CrossRefGoogle Scholar
Stinchcombe, Arthur L. 1965. “Social Structure and Organizations.” In The Handbook of Organizations, edited by March, James G., 142–93. Chicago: Rand McNally & Co.Google Scholar
Terza, Joseph V., Basu, Anirban, and Rathouz, Paul J.. 2009. “Two-Stage Residual Inclusion Estimation: Addressing Endogeneity in Health Econometric Modeling.” Journal of Health Economics 27 (3): 531–43.CrossRefGoogle Scholar
Tsai, Pan-Long. 1999. “Explaining Taiwan's Economic Miracle: Are the Revisionists Right?” Agenda: A Journal of Policy Analysis and Reform 6 (1): 69–82.Google Scholar
Uslu, Baris, Calikoglu, Alper, Nevra Seggie, F., and Seggie, Steven H.. 2019. “The Entrepreneurial University and Academic Discourses: The Meta-synthesis of Higher Education Articles.” Higher Education Quarterly 73 (3): 285–311. https://doi.org/10.1111/hequ.12198.CrossRefGoogle Scholar
Weymouth, Stephen, and Muir Macpherson, J.. 2012. “The Social Construction of Policy Reform: Economists and Trade Liberalization around the World.” International Interactions 38 (5): 670–702. https://doi.org/10.1080/03050629.2012.726185.CrossRefGoogle Scholar
Wong, Poh-Kam. 2011. Academic Entrepreneurship in Asia: The Role and Impact of Universities in National Innovation Systems. Cheltenham: Edward Elgar.CrossRefGoogle Scholar
Figure 0

Table 1. Cultural, human, and social capital imprinting on entrepreneurial university students.

Figure 1

Table 2. Summary statistics.

Figure 2

Figure 1. Percentage of degrees obtained at the twenty most-attended universities. Entrepreneurial universities are indicated by dark blue bars. All other universities indicated by light blue bars.

Figure 3

Table 3. Two-stage least squares regression analysis of startup-centric innovation and education

Figure 4

Figure 2. Marginal effects of education on startup-centric innovation references.

Figure 5

Figure 3. Multilevel OLS regression of startup-centric innovation and educational backgrounds with interaction effects.

Figure 6

Table A1. Innovation agency name (and abbreviation) by country.

Figure 7

Table A2. List of policy leaders.

Figure 8

Table A3. Example search string.

Figure 9

Table A4. News media sources.

Figure 10

Table A5. Entrepreneurial university list in comparison to US News & World Report 2022 rankings.

Figure 11

Table A6. Multilevel OLS regression (robustness test against main regression models, Table 3 in the main text).

Figure 12

Table A7. Multilevel OLS regression with interaction terms.