Session Abstracts

***Draft agenda as of February 23, 2016 - Subject to Change***

Friday, February 26

9:00 a.m. – Session 1: Entrepreneurial Characteristics and Human Capital (Troost Room)

Chair: Martin Ganco, University of Wisconsin-Madison

Living the "American Dream" in Finland: The Social Mobility of Innovators by Philippe Aghion, Ufuk Akcigit, University of Chicago, Ari Hyytinen, and Otto Toivanen

This paper matches individual patenting data with individual revenue data in Finland over the period 1990-1999 to analyze how innovation affects an individual's wage income and the probability that she reaches a top income bracket. We compare the upward social mobility of inventors versus non-inventors, and we also analyze how factors such as innate ability, family situation, gender, education, parental education or parental income affect the probability for an inventor to make it to top income brackets.
Does Student Debt Discourage Entrepreneurship? by Margaret E. Blume-Kohout, Mount Holyoke College
This paper uses newly collected data from the 2013 National Survey of College Graduates to evaluate effects of higher education debt loads on self-employment and business ownership among college-educated U.S. resident workers. Previous research studied effects of student debt  accumulation on early-stage career outcomes among PhDs in science and engineering fields, finding that those who accrue debt over $40,000 are less likely to take R&D-focused jobs after graduation. However, PhDs who incurred zero student debt were actually less likely to become entrepreneurs than those who incurred graduate debt of up to $10,000. These previous empirical results suggest a possible nonlinear relationship between student debt and propensity towards entrepreneurship. Graduates who avoided debt entirely may exhibit greater financial risk-aversion overall, as compared to those who accept typically low-rate student loans. On the other hand, graduates encumbered with high debt loads may be less able to attract external funding for entrepreneurial ventures, or may feel greater need for certain income flows to cover their looming debt payments.
A Tale of Two Datasets: Business Survival in Administrative versus Survey Data by Tami Gurley-Calvez, University of Kansas Medical Center, Donald Bruce, John Deskins, and Brian Hill
Entrepreneurs, particularly those that start and operate new businesses, are thought to be key drivers of economic innovation and growth.  A large and growing literature examines small business survival and the factors associated with business growth. Researchers use a variety of datasets including administrative records and survey data, but little is known about the implications of each data source for reported estimates and the broader understanding of small business outcomes.  This analysis addresses two fundamental data differences and assesses the implications for understanding the existing literature.  First, we examine business survival using exits from a matched administrative and survey dataset.  This analysis suggests that results differ meaningfully across data sources and analysis method suggesting caution in data selection and project design.  Second, we use additional information available in the survey data to assess whether results differ when one considers business closure, not just exit for any reason.  We find that owner characteristics such as age and education are more strongly related to exits from survey data, but not exits from taxpayer data or measures of firm closure available in the survey data.  These results might suggest a closer relationship between owner characteristics and survey attrition not firm longevity.
Go or No-Go? Testing an Antecedent Model's Ability to Predict Individual Entrepreneurial Action by Elaine C. Rideout, North Carolina State University
The study reviewed the literature for the antecedents of entrepreneurial action, and attempted to sort out the antecedent mechanisms (including entrepreneurship education) and isolate their impacts. The study’s Person x Context “Bounded Rationality” E-Correspondance model was grounded in Shapero & Sokol’s (1982) entrepreneurial event model, but their “desirability” and “feasibility” constructs were augmented to include Lofquist & Dawis’ (1969) contextual correspondence conceptualization. The model theorized that in addition to the elements of entrepreneurial “desirability” (motivations/wants/needs) and “feasibility” (efficacy in abilities and skills), there must be correspondent goodness-of-fit for both in the workplace context. The model was tested empirically via quasi-experimental research (4,000 University alumni). Structural Equation Model (SEM) analysis showed that work personality and workplace fit both play a significant role in predicting entrepreneurial production. Two of most important individual differences were individual attitudes and needs for personal autonomy, and individual styles for handling failure, as well as to Hard and Soft Self Efficacy and the mindsets and skillsets mediated by these variables. In sum, personal predispositions and individual differences in outlook and perspective impact entrepreneurship, as do knowledge and skills preparation. These results are particularly salient for E-ed practitioners. When educating the prospective entrepreneur, both skillsets and mindsets matter when it comes to equipping students to act on their entrepreneurial proclivities.
Risky Business: Some Empirical Aspects of Household Risk Management in Entrepreneurship by Sarada, University of Wisconsin-Madison, and Oana Tocoian
This paper investigates how individuals at different points on the wealth distribution scale seek to minimize the risk inherent to entrepreneurship. We segment entrepreneurs into three groups: the self-employed, business investors and business owners. This allows us to study how risk manifests for those who are exposed to just labor income volatility from self-employment versus asset returns volatility from being a business investor versus the combined risk to business owners where labor and asset income are now positively correlated. Empirical patterns show significant differences in the portfolio holdings of these different types of entrepreneurs in ways that correspond to their ability to be diversified and to tolerate short term income shocks. We find that initial wealth is associated with both entry into and survival in entrepreneurship, especially for the very wealthy. This relationship is particularly salient for those who bear both asset and labor income risk. Entrepreneurship presents a big potential upside for middle class households that succeed - far greater than that afforded by wage work. At the same time, those who fail face greater downward wealth mobility.
Does Entrepreneurship Pay or Satisfy? by Briana Sell Stenard, Mercer University (@Bstenard)
Much research has been conducted studying entry into entrepreneurship and the corresponding rewards and consequences of entrepreneurship on wages. This literature finds mixed results regarding whether workers enter entrepreneurship to maximize their wages or to maximize their non-pecuniary benefits. I suggest that whether workers are able to increase their pecuniary or non-pecuniary benefits in entrepreneurship depends on their motivations for entering entrepreneurship in the first place. I use longitudinal SESTAT data on over 28,000 scientists and engineers to track employment changes and changes in work outcomes over time for the same individuals. I examine changes in pecuniary and non-pecuniary work outcomes when wage workers transition to entrepreneurship, compared to those who do not change employers, as well as to those who change employers but do not transition to self-employment. This allows me to gain a deeper look at the implications of different types of mobility on work outcomes, instead of just comparing entrepreneurs to wage workers. I find that in general those who transition to entrepreneurship experience less growth in their wages and more growth in their non-pecuniary benefits compared to those who move to a new employer in wage work or do not change employers, however this result does differ depending on the motivation for moving. Additionally, using job satisfaction as a proxy measure for non-pecuniary benefits, I find that while those who enter entrepreneurship from wage work do generally improve their job satisfaction, job satisfaction rises for all movers in general, not just for those who enter entrepreneurship.

9:00 a.m. – Session 2: Entrepreneurial Management & Strategy (Kansas City Room)

Chair: E.J. Reedy, Kauffman Foundation (@EdgeReedy)

Causes of Extreme Outcomes in Entrepreneurship: Expectations, Endowments, Engagement, & Environments by Christopher Crawford, Ohio University (@DrRedApple)
What drives the emergence of firms that produce massive, creative destruction in the market? Whereas the study of high-impact entrepreneurial organizations is of primary interest in the domain, little is understood about the causal mechanisms underlying the initial creation and growth of such firms. Indeed, senior scholars suggest that entrepreneurial growth is random, and lament the field's lack of a comprehensive theory to explain and predict the extreme outcomes in the domaint firms like Facebook, Google, or Tesla Motors' outliers in the population that are quantitatively and qualitatively different from the "normal" population of ventures. This paper develops a framework to explain and predict the emergence of extreme outcomes (i.e., high-growth outliers) in entrepreneurship. Using a complexity science perspective, I proffer four antecedent meta-constructs which apply to all ventures, regardless of size: expectations, endowments, engagement, and environments. Continuing the work of Crawford, Aguinis, Lichtenstein, Davidsson, and McKelvey (2015), I analyze variables from three representative datasets at different states of firm emergence (N=11,000+) using nonparametric long-range correlations, semi-parametric maximum-likelihood techniques, and agent-based simulation modeling, I demonstrate that all constructs" and all consequent measures of entrepreneurial outcomes" are best characterized by highly skewed power law distributions. Here, I proffer that outlier inputs create an increased probability of successful interaction with potential stakeholders and are disproportionately likely to result in outlier outcomes over time. These findings suggests that 1) extreme inputs beget extreme outcomes, and 2) expectations for novel future outcomes are the primary driver of growth in the domain.
Anti-takeover Provisions, Corporate Governance, and Firm Performance by Wei Du and Shan He, Louisiana State University
In this paper, we take advantage of the unique circumstance of anti-takeover provision (ATP) adoption in corporate spin-offs to examine several important questions relating to the roles of ATPs in corporate governance, firm performance, and managerial characteristics. Our main findings can be summarized as the following. First, we find spun-off firms typically adopt more ATPs than their parents. Second, we find that the adoption of ATPs in the spun-off firms is mostly driven by “efficiency” but not “entrenchment” motivations. That is, the higher number of ATP adoptions are associated with increased likelihood of becoming a takeover target and high growth prospect of the firm. Spun-off firms that are to be managed by parent top management (who also designs the ATPs for spun-off firms) do not have higher ATPs than spun-off firms that are to be managed by division manager, parent management (none-top) member, or outsiders. Third, we find that firms with strong alternative governance in place (such as more productive market competition and higher block holder ownership) also tend to adopt higher number of ATPs. Fourth, we document a positive relationship between the ATP-based indices and subsequent stock and operating performances. The above findings highlight the endogenous nature of ATP adoption and call for a caveat on relying solely on the ATP-based indices as corporate governance quality measures.
Deciding When to Change: How Problems and Opportunities Trigger Change in Entrepreneurial Firm Strategy by Jacqueline (Jax) Kirtley, Boston University (@jxkirtley)
Entrepreneurial firms are characterized as inherently flexible, able to make strategic changes as needed. However, we have a limited understanding of what triggers strategic change in these firms. When and why do entrepreneurial firms decide to change their strategy? With a longitudinal field study of seven early stage entrepreneurial hardware firms in the energy and cleantech sector, I show how different types of triggers prompted 100 strategic decisions. Surprisingly, most firms retained their strategy even in the face of problems. When firms did make strategy changes, they did so in response to newly identified opportunities. I offer a framework that explains why some triggers are more likely to lead to strategic change than others and show how emerging opportunities affect the evolution of entrepreneurial strategy. This research contributes to our understanding of the triggers of strategic change by examining entrepreneurial firms early in their decision making process rather than retrospectively exploring changes after they are made. From this examination, I explain why and when entrepreneurial firms make strategic changes and when they do not. I also contribute to the literature on technological entrepreneurship by showing how the unfolding of new opportunities post-founding affects the evolution of entrepreneurial strategy. Finally, in contrast with the literature which tends to assume that strategic change is a binary decision, I identify more fine grained variations of the ways in which entrepreneurial firms retain their current strategy as well as how they change.
Discovering the Antecedents of Learning Inertia in the Corporate's New Product Development Process by Yinghong Susan Wei, Texas A&M International University
Entrepreneurship is viewed as the "parent of innovation." Without entrepreneurship, business and society would be neither dynamic nor adaptive but stagnant. Entrepreneurship leads firms to seek innovative and flexible means to exploit opportunities and achieve desired objectives. However, learning inertia is the one of biggest barriers for corporate entrepreneurship. What cause or create of learning inertia in the corporate company's innovation creation process is a very important and interesting research questions. This research conceptualizes three different types of learning inertia in new product development process: vision inertia, technology inertia and procedure inertia. This research investigates whether the firm's size, age, square terms of firm size and age, past firm performance, technology uncertainty, competition intensity, and perceived industry dominance are related to three different types of learning inertia. Using survey, this study collects data from 150 firms from a high-technology industrial zone in China with two respondents per firm. A structural equation modelling is used for data analysis. The findings suggest that past performance and perceived industry dominance are positively related to vision inertia. Perceived industry dominance is negatively related to technology inertia. It is very interesting that the firm age and technology inertia has inverted U shape relationship, which means middle aged firms have the highest technology inertia. In contrast, the firm size and technology inertia has U shape relationship, which indicates that medium size firms have the least technology inertia. Past performance and technology uncertainty are positively related to procedure inertia.
Exploring Heterogeneity of Boomer Entrepreneurs by Ting Zhang, University of Baltimore (@TingTingTime)
The proposed research categorizes boomer entrepreneurship in the context of a less physically-demanding US shared knowledge economy and explores the factors driving different subcategories of boomer entrepreneurship. The subcategories of boomer entrepreneurs include nascent versus existing boomer entrepreneurs, necessity versus opportunity entrepreneurs, incorporated versus unincorporated boomer entrepreneurs, and full-time versus part-time boomer entrepreneurs. This empirical study that rely on descriptive statistics, the utility maximization theory, and a series of binomial multilevel mixed-effects logistic regressions will test two series of hypotheses: (1) Boomer entrepreneurs are not a homogeneous group and (2) entrepreneurial propensity factors differ across different categories of boomer entrepreneurs.  2005-2014 data from the Current Population Survey and the Bureau of Labor Statistics are used.

1:00 p.m. – Session 3: Demographics in Entrepreneurship (Troost Room)

Chair: Alicia Robb, Kauffman Foundation (@aliciarobb)

Demographics of Inventors in the Historical United States by Michael Andrews, University of Iowa, Sarada, and Nicholas Ziebarth

We match the 1870-1940 Annual Report of the Commissioner of Patents to the corresponding U.S. Federal Population Censuses. This provides a rich set of demographic information for a comprehensive dataset of inventors. Using this dataset, we first document that patentees over this seventy year period are more likely to be older, white, male and to have migrated between states than is the population at large. While these patterns hold generally over time, they do so to varying degrees. In particular, we find an increasing share of women inventors while the black share remains constant. We also find that patenting levels tend to be geographically concentrated and present some basic correlates between patenting geography and local demographics. We further provide evidence on the effects of new schools on the patenting rate. In particular, we explore the effects of Historically Black Colleges and Universities (HBCUs) on the rate of black patenting and the knowledge spillovers onto white patenting rates.
Gender Bias in Equity Investments:  Evidence from a Randomized Field Experiment by Sofia Bapna, University of Minnesota (@BapnaSofia), and Gordon Burtch
A large body of prior research suggests that female founders seek and receive less capital for their businesses (Coleman and Robb 2009). However, previous research also suggests that women self-select into ventures in traditional sectors that have limited size and growth potential (Constantinidis, Cornet and Asandei 2006). Moreover, prior work has observed that female founders experience lower barriers to capital acquisition as the quality of their ventures increases (Guzman, Kacperczyk and Stern 2015). Taken together, these findings raise questions as to whether there indeed exists a gender bias in start-up investment, or whether women obtain less funding simply because of self-selection and other unobserved factors.  We therefore employ a randomized field experiment in the context of equity crowdfunding (in partnership with an equity crowdfunding platform) to identify and quantify gender bias in equity investment. Holding venture characteristics fixed, we provide information about the gender of female firm founders to only a random set of potential investors. Employing a randomized control trial in this manner enables us to determine the presence of a causal relationship and, moreover, whether the effect of this treatment varies by i) investor gender ii) investor experience iii) venture quality and iv) venture industry. The outcomes observed are interest in investing and actual investments.  Past work on gender bias has largely relied on observational / archival data and laboratory experiments, which can be subject to omitted variable, selection and actor observer bias. The field experiment methodology used this study helps overcome some of these limitations.
Unequal Disadvantage in Entrepreneurship by Jorge Guzman, Massachusetts Institute of Technology (@GuzmanJorge_), and Aleksandra (Olenka) Kacperczyk
Using data on all businesses registered in the state of California between 1988 and 2011, we examine gender differences in access to entrepreneurial resources. We propose and empirically show that female entrepreneurs are more likely than male entrepreneurs to pursue lower-quality ventures. We also find that female entrepreneurs are less likely than male entrepreneurs to obtain VC funding, even for comparable levels of entrepreneurial quality. Our findings further indicate that the gender disadvantage in entrepreneurship disappears for top-quality ventures, suggesting that gender-stereotyped perceptions are weaker for top-performing women. Consistent with the notion of statistical discrimination, we also show that gender disadvantage is amplified for lower-quality evaluators, presumably more likely to rely on stereotypes when assessing new-venture potential. More generally, the study emphasizes the need to take the distribution of entrepreneurial quality into consideration when examining the mechanisms behind gender gap in entrepreneurship.
Can equity crowdfunding alleviate gender differences in access to capital? by Mingfeng Lin, University of Arizona (@mlin2008), and Brian Zhang
The literature has long documented that when trying to obtain financial capital, women entrepreneurs face more challenges than men, even though there is some evidence that women founders are in fact more likely to be successful. Expanding access to capital for women entrepreneurs, therefore, has long been an important research topic. In recent years, crowdfunding, where individuals can appeal to and solicit funds on the internet, has been lauded as a potential new channel to change this dilemma. Some recent studies, notably Marom, Robb, and Sade (2015)and Greenberg and Mollick (2014), show that on reward-based crowdfunding websites such as Kickstarter, women are better able to raise funds than they could offline. No study, to date, has examined the gender differences in equity crowdfunding.   I use detailed transaction level data from one of the largest equity crowdfunding platforms in the UK to study gender imbalances in equity crowdfunding. In particular, gender information is obtained from administrative data for both investors and entrepreneurs. Preliminary results suggest that  in equity crowdfunding, women are still less likely to receive funding than men. There is also interesting gender differences on the investor side: While female investors are more likely to invest in female (than male) entrepreneurs, the funding amount is significantly lower when they invest in females, compared to when they invest in male entrepreneurs.
Residential Segregation, Discrimination, and African-American Theater Entry During Jim Crow by Ricard Gil and Justin Marion, University of California, Santa Cruz
We examine the role of residential segregation and racial discrimination in determining the entry of movie theaters serving African-American customers in the 1950s. These theaters provided an alternative to the segregated theaters of the Jim Crow era. Consistent with preference externalities in racial and ethnic enclaves, we find that a greater degree of residential segregation leads to more African-American theater entry. We support this conclusion using estimates from a Bresnahan and Reiss model of theater entry, which indicate that this effect is due to higher variable profits in residentially segregated cities rather than lower fixed costs of entry. The effect of racial bias among whites is found to be complex. Using several measures of racial discrimination, we conclude that bias leading to a taste for segregation leads to greater entry, while more generally racial bias results in fewer theaters.
Androgynous Names and Entrepreneurial Outcomes by Amy Nguyen-Chyung, University of Michigan
While recent research suggests that gender signals impact entrepreneurial success by affecting perceptions of entrepreneurial competence, quality or investment worthiness (Shane et al. 2015, Thébaud 2015, Woods et al. 2014), less work has examined whether it can be beneficial for individuals to mute gender signals. Since personal names often reveal one’s gender, it is not surprising that in some entrepreneurial settings where personal names are also often the business names – some women have chosen to adopt androgynous names. In this study, I ask whether doing so boosts entrepreneurial outcomes for women and analyze the phenomenon in the context of residential real estate brokerage, where it is common for individuals to use their personal names as their business names. Combining a 17-year panel dataset containing detailed productivity, entrepreneurial roles, legal names and business names and other individual characteristics including actual gender for 40,000 agents, many of whom are business owners with additional data on their perceived gender and ethnicity and that of their clients, I aim to expose outcome differences between businesses owned by women with feminine names versus those with androgynous names. To improve causal inference with the observational data, I will employ a coarsened exact matching (Iacus et al., 2011) approach. Preliminary findings show that those who intentionally androgenize their names have higher initial productivity, suggesting that selection can be a driving mechanism behind significant differences in entrepreneurial outcomes.
Increasing Women’s Representation in Entrepreneurial Activity: A Longitudinal Study of the Rising Tide Program by Renee Rottner, Kyle Lewis, Sarah Thébaud, University of California, Santa Barbara (@sthebaud1), Joseph Broschak, and Christine Beckman
A wealth of research demonstrates that, even when women have the same human and financial capital as their male counterparts, they are substantially less likely to start or to invest in new enterprises. Previous literature suggests that structural mechanisms, such as women’s position in networks of access and influence, as well as cultural mechanisms, such as stereotypes about women’s abilities to competently run or invest in new companies, may explain these disparities. In an effort to evaluate and to develop new theoretical insights into such processes, we utilize novel data collected through the Next Wave Initiative’s Rising Tide Program, a new program that has been designed to increase the supply of women angel investors. The talk will cover research goals and hypotheses, as well as preliminary data collection efforts.

1:00 p.m. – Session 4: Entrepreneurial Finance (Kansas City Room)

Chair: Yael Hochberg, Rice University (@yaelhochberg)

The Economic Effects of Public Financing: Evidence from Municipal Bond Ratings Recalibration by Manuel Adelino, Duke University (@madelino), Igor Cunha, and Miguel Ferreira
We show that the supply of local public financing can have important effects on local economies. We identify these effects by exploiting exogenous variation on municipal bond ratings due to Moody's recalibration of its scale in 2010. We find that local governments increase expenditures and employment due to an expansion of their debt capacity following an upgrade. These increases in local government spending lead to multiplier effects on employment and income. The increase in employment flows through an increase in new firms, particularly in the non-tradable sector. Our findings suggest that debt-financed increases in government spending can improve economic conditions during recessions.
Trust, Social Structure and Venture Capital Investment by Sampsa Patrikki Samila, Waverly W. Ding, University of Maryland, and Ying Geng
A growing body of research has documented the importance of generalized trust in shaping economic outcomes. At the cross-country-border level, for example, Guiso et al. (2009) documented that generalized trust between countries can significantly affect the direction of international trade, financial portfolio choices, and foreign direct investment destinations. When it comes to economic decisions of individual firms, Bloom et al. (2012) showed that generalized trust influences patterns of (de)centralization of multinational firms and Bottazzi et al. (2012) established a causal impact of cross-country generalized trust on venture capital firms’ choice of investment targets. The gist of the finding from this body of research is that if the generalized trust of one country towards another one is low, the level of economic interest and ensuing co-operations between the two countries are likely to be low as well.

Our question is: if lower generalized trust has been shown to reduce chances of economic cooperation, what do economic actors, firms in particular, rely on as mechanisms to overcome a deficiency of generalized trust that may jeopardize the chances of economic cooperation? The revelation of these mechanisms is important to us in understanding patterns of cross-country economic cooperation as driven by firm-level choices. The perspective we offer to answer our proposed question is rooted in sociological theories that relate social structure to trust between actors. In particular, we propose three broad classes of social structural factors that have the potential of modifying the strength of generalized trust in economic cooperation: relational, structural and status-based.
The Prevalence and Drivers of Entrepreneurial Bootstrapping by Benjamin Hallen, University of Washington (@benhallen), Pai-Ling Yin, and Jason Davis
Extensive prior literature has studied the impact and positive benefits of entrepreneurial ventures raising equity funding from venture capitalists (VCs) and angel investors. Indeed, based on this literature, entrepreneurs are often advised “don’t go it alone” (Baum, Calabrese, and Silverman, 2000). Less clear, however, is the prevalence of bootstrapping – or not raising equity funding – by otherwise high-growth ventures. One challenge to studying bootstrapping has been obtaining reliable data identifying high-growth ventures that have not raised equity funding. We seek to address this gap using a novel dataset that captures all competitors in the mobile app ecosystem on the Apple iPhone and iPad ecosystem over a period of 7 years. Whereas prior literature has generally portrayed VC and angel funding as primarily a story of sorting whereby the highest-quality ventures are most likely to raise, we show that a substantial number of high-quality ventures never raise external equity funding. We then investigate the drivers of entrepreneurial bootstrapping using a novel mixed-method approach, where we interview pairs of matched ventures that exhibited similar performance and growth but where one raised and the other did not.
Reputation and Relational Contracting: Experimental Evidence on Contracting Frictions in West African Trade by Meredith Startz, Yale University, and Shelby Grossman
Businesses worldwide face contracting problems – how to ensure that you receive goods, services, or payment as promised? A large theoretical literature on self-enforcing contracts suggests that freely flowing information can substitute for formal enforcement, deterring opportunistic behavior by firms or individuals concerned about maintaining a good reputation. However, little is known about whether or how reputation alleviates contracting frictions in practice.   To provide empirical evidence on this theoretical mechanism, we plan to randomize the introduction of a reputation-based mobile money escrow service among used clothing wholesalers in Lagos, Nigeria, in early 2016. These traders typically travel abroad in person every time they restock in order to monitor suppliers and check their goods before making payment. The prevalence of this costly coping strategy suggests that the lack of effective contract enforcement is a potentially large source of friction in this market, and is not fully solved by the repeated nature of the transactions.  The escrow service facilitates arms’-length business transactions, motivating follow-through on informal agreements by tying behavior to public reputations. We investigate whether increased access to reputational information affects the price, quality, and quantity of goods traded, and whether it substitutes for costly monitoring via travel. Through a second ‘information only’ treatment, we also test whether reputation operates by identifying good and bad types among potential partners, or by providing a direct disincentive to behave badly in a given transaction.
Organizational Decision-Making and Information: Angel Investments by Venture Capital Partners by Andy Wu, University of Pennsylvania (@AndyWuAndyWu)
We study the role of information in organizational decision-making for the financing of entrepreneurial ventures. We formally model a committee of agents who vote to allocate resources to a project with unknown outcome. The agents are endowed with costless explicit information, and they can each acquire costly tacit information to improve their decision quality. Equilibrium outcomes suggest a theoretical tension for group decision-making between the benefits of information aggregation and a cost from the participation of uninformed agents, and this tension presents a boundary condition for when a group decision is superior to an individual decision.   We test our predictions in the setting of a particular phenomenon in venture capital: private angel investments by the partners outside of their employer, which represent investments passed on by the employer. Venture capital partners, acting independently with their personal funds, make investments into younger firms with less educated and younger founding teams than their employing VC firms, but these investments perform financially similarly or better on some metrics even when controlling for investment size, stage, and industry. Geographic distance and technological inexperience by the VC increase the probability the investment is taken up by a partner and not the VC. This work contributes to an emerging stream of literature on information aggregation in organizations and the established literature on resource allocation and incumbent spin-outs.
Information Cascade and Efficiency in Crowdfunding by Xin (Power) Xue, University of Virginia (@powergascan)
Using a unique high frequency fundraising dataset from LendingClub, a crowdfunding platform, I find strong empirical evidence of herding behavior among the investors. Herding appears more salient in riskier loans. Moreover, applying the historical loan data, I find strong positive correlation between herding and loan performance, indicating possible information cascade.   This paper further provides a counterfactual welfare analysis as if investors make decisions simultaneously rather than sequentially so that any possible information cascade is eliminated. I hope to estimate the value added/deducted to the investors by the information that induces herding, and derive policy implications on information a financial intermediary should/should not provide.
Top Management Turnover, Inventor Mobility, and Corporate Innovation in Venture-Backed Private Firms by Qianqian Yu, Boston College
Using a hand-collected dataset on top management turnover in venture capital (VC)-backed private firms, I analyze the effect of top management turnover on subsequent innovation in VC-backed private firms. I find that management turnover is associated with significantly more and higher quality innovation, as measured by the number of patents and the total number of citations following the turnover event. I conduct an instrumental variable analysis using a plausibly exogenous shock to the supply of new managers as an instrument, and establish a positive causal effect of management turnover on innovation. I show that adding new managers to the top management team plays a more important role than removing existing managers in generating more and higher quality innovation. I also show that the management turnovers in my VC-backed sample are likely to be investor led and that the effect of management turnover on innovation is stronger for firms with higher VC-investor power. Finally, I analyze possible mechanisms through which management turnover affects innovation, and establish that one mechanism is through new management teams hiring a larger number of inventors and scientists.

3:15 p.m. – Session 5: Innovation, Technology, and Knowledge-Transfer (Troost Room)

Chair: Ethan Mollick, University of Pennsylvania (@emollick)

The Unequal Gains from Product Innovations by Xavier Jaravel, Harvard Business School (@Xjaravel)
This paper shows that product innovations disproportionately benefit high-income households due to increasing inequality and the endogenous response of supply to market size. Using detailed product-level data covering 40% of total expenditures on goods, the paper presents two new stylized facts: (1) from 2004 to 2013, high-income households have consistently spent a higher share of their income on new products; and (2) quality-adjusted inflation has been lower for these households, relative to more modest households. Using changes in market size driven by demographic trends exogenous to innovation patterns, the paper provides causal evidence that a shock to the relative demand for goods can (1) affect the direction of product innovations, and (2) lead to a decrease in the relative price of the good for which demand became relatively larger (the strong equilibrium bias hypothesis). Finally, the paper examines which theories of directed innovation and entrepreneurship are consistent with the evidence.
Frictions in the Flow of Academic Knowledge to Industry: Evidence from Simultaneous Discoveries by Michaël Bikard and Matt Marx, Massachusetts Institute of Technology (@marxmatt)
Scientific discoveries in academia can spur innovation and growth, but only if that knowledge flows to relevant industry actors. One possible friction in the flow of academic knowledge to industry could be that many academic scientists are geographically isolated from firms performing R&D in relevant fields. But research at isolated institutions may be less applied, or simply of lower quality, confounding inference. We address the unobserved-quality problem by analyzing simultaneous discoveries where multiple researchers in different locations report the same finding in separate papers. We find that “twin” papers reporting a simultaneous discovery are 10-23% less likely to be referenced as prior art in firm-owned patents when not within commuting distance of a significant concentration of R&D activity in relevant fields. No effect is found for references from university-owned patents. Our results suggest that discoveries at isolated institutions may become orphaned, suggesting both implications for the science of science policy as well as firms’ commercialization strategies.
Markets for Technology and the Technological Positioning of Entrepreneurial Startups in Nascent Industries by Mahka Moeen, University of North Carolina at Chapel Hill, and Seth Carnahan (@sethcarnahan)
This paper studies how dominant firms affect the technological positioning of other players in the industry.  Prior research suggests that competitors typically tend to position themselves away from dominant firms to avoid product market competition.  By contrast, we suggest that entrepreneurial startups, due to their lack of commercialization assets, may view a dominant firm as a potential partner in the market for technology.  Therefore, startup firms may position themselves in close proximity to a dominant firm within the technology landscape, in order to increase their attractiveness as a partner in the market for technology.  Using the population of field experiments within plant biotechnology during 1987-2010, we examine whether entry by a dominant firm into a technical domain (i.e., a new crop-genetic trait) spurs follow-on investments by entrepreneurial startups while discouraging follow-on investments by established firms. We further examine the moderating effect of the direction of scientific progress, the commercial applicability of a domain, and the density of the dominant firm’s alliance portfolio. Our contributions are twofold. First, to the entrepreneurship literature, we suggest that a startup’s opportunity set for participating in the market for technology is not exogenous.  Rather, the startup may engage in endogenous, strategic behavior and direct its technological investments towards opportunities in the market for technology.  Second, to the strategic positioning literature, we suggest that a firm’s position in the technology landscape may be driven by the prospects of value capture through market for technology, rather than the previously-identified mechanism of product market rivalry reduction.
Does Public Data Encourage Entrepreneurship? Evidence from the Role of NASA Satellite Maps in the Gold Exploration Industry by Abhishek Nagaraj, Massachusetts Institute of Technology (@abhishekn)
A number of governments have launched "open data" policies to boost private-sector performance and support entrepreneurship, yet the economic effectiveness of such programs remains unexamined. In this paper, I study the impact of public investments in satellite maps from the NASA Landsat program on the gold exploration industry. Using random variations in the timing of mapping in different parts of the world (from technical failures in data collection and cloud-cover in imagery), I estimate the value of open data on the discovery of new gold deposits and entrepreneurship in the mining industry. I find that mapped regions are almost twice as likely to report the discovery of new gold deposits as compared to unmapped regions, and open data also boosts discoveries from entrepreneurial firms in this industry. However, the positive impact of open data programs on entrepreneurship is largely dependent on the quality of local institutional conditions like property rights and labor laws. Back-of-the-envelope estimates suggest that the satellite mapping program led to the discovery of over $10 billion worth of new gold discoveries in the United States alone. Ultimately the results provide evidence of how public investments in information can be used to boost private sector performance and entrepreneurship, and also highlight conditions and mechanisms through which public sector investments could affect entrepreneurship.
Bridging the Academic-Practitioner Gap in an Online Community: Evidence from the US Shellfish Industry Listserv Email List by Bogdan Prokopovych, University of Massachusetts Amherst, and Staffan Furusten
One of the challenges of the academic-practitioner gap identified by researchers is the difficulty in communication between experts from academia and industry practitioners. Institutional literature has viewed experts as institutional agents who create cultural-cognitive and normative categories for their organizational fields. How do experts who subscribe to different institutional logics and represent differing organizational fields bridge this academic-practitioner gap and share information? We draw on the selective attention and institutional theories to explore the mechanisms of information exchange between the communities of practice and academic researchers. We use dyad-level data from a listserv of shellfish industry used by both practitioners and academic experts. We explore how the logics of science and industry manifest themselves in the vocabulary of the message content, and use it to predict which message generates more attention among both communities.
Competitive Poaching of Customers in the Mobile Service Market by Xiahua Wei, University of Washington, and Kevin Zhu
In technology-intensive service markets, firms spend substantially on acquiring customers. The race for customers has intensified due to rapid technological change, reduced switching costs, and increased competition. Does such investment really pay off, or is it simply the cost of competing? How would new technologies be leveraged to retain the right kinds of customers when they become so easy to be lured away? This paper studies two strategies of customer poaching: paying customers to switch (acquisition spending) versus attracting customers by new technology. We analyze how these two strategies affect the three stages of customer relationship management – acquisition, retention, and revenue generation – in different ways.  We evaluate these issues using a firm-level panel dataset from the global mobile service industry. We find that acquisition spending helps to attract new customers. However, it does not improve customer retention or revenue generation. Rivals’ competitive actions substantially undercut the effectiveness of the focal firm’s acquisition effort. Using new technologies alleviates this adverse effect, which suggests the strategic role of technology in the race for customers. Hence, it becomes important that firms synchronize their marketing strategies and technology strategies to acquire/retain the most valuable customers in technology-intensive markets.

3:15 p.m. – Session 6: Regional Dynamics and Entrepreneurial Ecosystems (Kansas City Room)

Chair: Yasuyuki (Yas) Motoyama, Kauffman Foundation

Geographic Locations and Entrepreneurship: Evidence from Randomized Housing by Laura Doering, University of Toronto, and Chirstopher C. Liu
Research on entrepreneurship emphasizes the importance of geographic locations. However, much of this literature focuses on regions and cities, overlooking the potentially important impact of neighborhood configurations. We refer to these configurations as micro-geographies and theorize that proximity to local markets can affect the rate of entrepreneurship. To test this notion, we use data from a public housing complex in Colombia where residents are randomly assigned to housing. We consider whether residents who live on the ground floor of a multi-story apartment building are more likely to become entrepreneurs than those assigned to upper floors. We also test the effect of residents’ proximity to the local market on entrepreneurial earnings. For the entrepreneurs in this context, operating a small business on the ground floor versus an upper floor means the difference between living above or below the poverty line. Overall, this study extends the literature on entrepreneurial location choice to highlight the important effect of neighborhood configurations.
Innovation Districts as an Urban Economic Development Strategy by Joshua Drucker, University of Illinois at Chicago, and Amanda Kass
Innovation districts are becoming the latest craze in urban economic development, promoted by the Brookings Institution’s Bruce Katz and bolstered by apparent early successes in Barcelona and Boston. Roughly expressed, innovation districts are designated areas in which several kinds of economic actors employ a variety of strategies to generate and support innovation and entrepreneurial activity. Although the idea is becoming broader and “fuzzier” as its application spreads, the central goal of an innovation district policy is to provide a practical mechanism for focusing and targeting economic development efforts on entrepreneurship and innovation. At the 2014 KES conference, I described my initial impressions and opinions of innovation districts, gained through teaching a studio course that culminated with a research project for World Business Chicago. I have been conducting my own research since. My first approach has been to consider precisely the characteristics that ought to define an innovation district, and to evaluate the strategy’s economic development potential by gauging  how well the characteristics align with theoretical and empirical understandings of how innovation and entrepreneurship operate. My second approach is to conduct case studies, exploratory in nature because most innovation districts have not been in place long enough to assess empirical outcomes. I will present one or portions of both of these approaches, depending on my progress and the fit with the session theme. (I suspect that I will synopsize the findings from the first approach and present an outline of the research design for the second approach seeking feedback.)
Small Establishments and City Growth: Evidence from Industrial Labor Demand by Rebecca Diamond and Naomi Hausman, Hebrew University, Jerusalem (@NaomiHausman)
Entrepreneurs are often thought to be vital contributors to economic growth, bringing new ideas to life. The emphasis in growth theory on ideas as being as essential to economic progress means that the entrepreneurs that realize these ideas are crucial. Certainly, actualizing these ideas drives growth globally, but it can be especially valuable in cities, where innovators can interact in close proximity, building quickly on each others' ideas, and generate cascading growth in their locality. Substantial evidence suggests the presence of numerous small establishments in a city predicts its growth, but do the small establishments generate this growth? We present new evidence on the role of entrepreneurs in city growth using the natural experiment that occurs when a city has a concentration of small establishments in an industry that experiences a national growth shock.  Does this city grow more than a comparison city that has a different establishment size distribution in the growth-shocked industry? Using highly detailed and comprehensive data from the Census' Longitudinal Business Database (LBD), 1982-2012, we show that the smallest establishments cause the most growth, and we proceed to unpack the mechanisms through which these effects may operate.  Small establishments appear to be particularly operative in industries that are in flux, experiencing rapid growth or decline, and they create the greatest employment growth in establishments larger than them. Our results on decadal growth effects complement previous evidence on the long-run growth effects of a city's culture of entrepreneurship and provide insight into the life cycles of cities.
Flooded Cities by Adriana Kocornik-Mina, Alterra - Wageningen UR, Thomas K.J. McDermott, Guy Michaels, and Ferdinand Rauch
Does economic activity in cities relocate away from areas, which are at high risk of recurring shocks? We examine this question in the context of floods, which are among the costliest and most common natural disasters. Over the past thirty years floods worldwide killed over 500,000 people, displaced over 650,000,000 people and caused over $500 billion in damages. This paper analyses the effect of large scale floods, which displaced at least 100,000 people each, in over 1800 cities in 40 countries, from 2003-2008. We conduct our analysis using spatially detailed inundation maps and night lights data. We find that low-lying areas are about 3-4 times more likely to be hit by large floods, and yet they concentrate more economic activity than other urban areas. When cities are hit by large floods, the low elevation areas also sustain more damage, but like the rest of the flooded cities they recover rapidly, and there is no evidence that economic activity relocates away from these risky areas. Our findings have important policy implications for aid, development and urban planning when the costs of residing in low-lying areas are partly borne by those paying for flood defenses and for reconstruction.
Repeat Winners (or Repeat Offenders): When Does Repeat Certification Help Ventures Acquire Resources? by Lauren Lanahan, University of Oregon, and Daniel Armanios
New firms often have trouble acquiring resources because there is significant uncertainty (Aldrich & Fiol, 1994). Institutional theory argues that if a new firm can demonstrate their legitimacy and value (Suchman, 1995), they can help alleviate liabilities that inhibit access to resources (Hallen, 2008; Zott & Huy, 2007). Firms can acquire legitimacy through external certification. This occurs when a reputable institution publicly attests to a firm's quality (Rao, 1994) or behavior ( Sine, David, & Mitsuhashi, 2007), and thus confers legitimacy for the firm.  Institutions also gain reputation by evaluating the firm quality based on its demonstrated record and commercial promise (Podolny, 2001; Stuart, Hoang, & Hybels, 1999), and by its perceived expertise in making such assessments (King, Lenox, & Terlaak, 2005).   The overall implication is that certification from reputable institutions is useful for new firms hoping to assuage concerns regarding uncertainty due to novelty. The prior literature, however, does not look at the dynamics of certification. This literature overlooks repeat certification "“ the occurrence of multiple forms of certification over time. Repeat certification could be a story of the "rich getting richer," demonstrating the quality of the firm (Chen, Hambrick, & Pollock, 2008). However, it could lead to questions of whether the firm is actually fulfilling the potential, or whether the certifying body is indeed equipped to make such assessments (Cox-Pahnke, Katila, & Eisenhardt, 2015). As such, existing theory would predict conflicting outcomes. Therefore, we find it important to ask: when does repeat certification benefit new firms?
Plant Closure, Organizational Demise and Community Entrepreneurship by Henry Renski, University of Massachusetts Amherst, and Ian Walsh
The closure of a major employer can have devastating impacts on a community. Lost paychecks wreak havoc on the lives of layoff victims, producing serious public policy consequences that are emblemized by boarded-up factories and crowded unemployment offices. Communities often find themselves often straddled between shrinking tax revenues and increased demand for services. However, some scholars point to potentially constructive outcomes of organizational demise. When organizations close, new entrepreneurial ventures rise up in its wake (Walsh & Bartunek, 2011). This study seeks to better understand community entrepreneurial responses, by documenting changes in the level and composition of new firm creation following the closure of a major employer. Scholars argue that the effects of organizational decline on innovation may be positive or negative depending on environmental, organizational, and individual-level variables (Mone, McKinley, & Barker, 1998). We seek to explain this variability--asking why some communities exhibit more resilience in the face of organizational demise through entrepreneurial activity than others? Our approach uses both quantitative and qualitative research methods. We focus on rural and small urban counties, which are likely to be more sensitive to the impacts of plant closures. We employ a quasi-experimental framework to distinguish the impact of large plant closures on new firm formation from other competing causes and estimate a panel regression model to measure the degree of community entrepreneurial response. We complement this with an in-depth investigation of a small subset of afflicted communities: examining newspaper reports, published documents, and interviews.
Place and Space: Communication System Development and Competitive Differentiation Among Startups by Heather Haveman and Chris Rider, Georgetown University (@ProfChrisRider)
As communication systems develop, organizational foundings become less sensitive to local context and more sensitive to distant competitors (i.e., space matters less). We propose that this expanding spatial scope of competition causes new organizations to increasingly differentiate their offerings from those of established organizations by identifying with their place of production (i.e., place matters more). We test this argument using data on the development of the first modern communication system, the US post office, and foundings of organizations that depended on it for distribution, magazine-publishing ventures. We find that as the postal system developed, new magazines were increasingly likely to be founded with names that explicitly identified their place of production; this effect was accentuated for organizations that were situated far from established competitors. We also find that this effect was due to magazines founded farther from established ones being more likely to adopt localistic (e.g., state or municipal) identities than universalistic (e.g., national) ones. We conclude that as new organizations face increasing competition from distant incumbents, place of production becomes an increasingly common basis for competitive differentiation.

Saturday, February 27

8:45 a.m. – Session 7: Industry Studies (Troost Room)

Chair: Michael J. Fern, University of North Carolina at Chapel Hill (@mjfern)

Temporary Ventures: Entrepreneurship and the Creative Industries by Daphne A. Demetry, University of Oxford, and Antonio Paco Giuliani
Scholarship on entrepreneurs has largely understudied cultural production as an industry worthy of investigation. Yet, entrepreneurship in the realm of creative work – for example, art, fashion, and advertising – is a rapidly growing economic sector. For instance, in 2014 one in eleven jobs in the UK were classified as part of the creative economy. As entrepreneurship in this ‘symbolic intensive’ field expands, the question remains of what modes of start-up organizing are most effective? In this paper, we explore one particularly fruitful organizational form creative entrepreneurs use to balance their need for creativity and economic survival: temporary project-based organizations. We propose that temporary projects are an effective model for ventures in the creative realm because they: (1) provide spaces of negotiation with key stakeholders encouraging the co-creation of meaning, (2) are characterized by fluid boundaries that enable the growth of key professional and social networks, (3) require lower initial resource, allowing entrepreneurs to focus on innovation and creativity rather than economic choices, and lastly, (4) they boost career capital by enabling entrepreneurs to gradually build up their craft/skills.
Regulatory Discretion and New Venture Outcomes: An Examination of the U.S. Hydroelectric Power Sector, 1970-2014 by Jake Grandy, University of Southern California (@jakegrandy)
While research on institutions and entrepreneurship has examined the role of policy formation by legislators on new ventures, scholars have yet to examine the effect of policy implementation by regulatory agencies on ventures startup and operation. The dominant logic stemming from the management literature generally assumes that regulators have the freedom to shift policy in response to the firms' attempts at influence; however research in other fields acknowledges that regulators don't always have the ability to shift policy due to firms' influence, but instead are constrained by legislative bodies that monitor and can overturn regulatory decisions. I attempt to reconcile these positions by exploring how the degree of regulatory discretion granted to regulatory agencies by legislators affects ventures. Empirically, I examine the licensing of new hydroelectric facilities in the United States from 1970 to 2014. I suggest that the discretion given to regulatory officials to interpret and implement policy creates opportunities for both entrepreneurs and key external stakeholders to influence regulatory decision-making. Preliminary results suggest that as discretion increases, so does the likelihood of venture entry. However, external stakeholders who oppose entry are able to mitigate this effect when regulators have a moderate, but not large amount of discretion. I discuss the implications of this study for the study of entrepreneurship and non-market strategy.
Impact of Market Contestation on Entrepreneurial Collective Action in the Emergent U.S. Wood Pellet Industry by Shon Hiatt, University of Southern California, and Sangchan Park
While collective action seldom arises without free riding, this problem has not been fully examined in the emergence of new markets where entrepreneurial collective action is critical for gaining and defending legitimacy. We address this issue by exploring how entrepreneurial ventures’ common efforts to promote a collective identity unfold in response to social movement contestation of the market. We propose that opposition activism has a bifurcated impact on verbal and nonverbal dimensions of identity claim-making by causing entrepreneurs to perceive differently the costs between the two. Using the emergent U.S. wood pellet market as an empirical context, we show that environmental activists’ targeting of forest-product manufacturers is associated with increased similarity to what pellet firms claim to be (i.e., renewable energy producers), but decreased similarity to what they actually do. The study has implications for new market legitimation, social-movement, and sustainability research.
Product Identity and Competitive Positioning in the Nascent Synthesizer Industry by Callen Anthony, Andrew J. Nelson, University of Oregon (@nelson_a_j), and Mary Tripsas
It is well established that firms make a series of positioning choices that shape how they compete within an industry. However, much of this work has examined competition within established industries where performance attributes are well-understood. By contrast, we know little about how firms position their products within nascent industries, which are often characterized by extreme uncertainty about what the product even is. We address this gap through an inductive study of the emergence of the music synthesizer, drawing upon a unique data set of four leading firms’ complete product offerings and advertisements from 1976 to 1986. We discover that conventional dimensions of competitive positioning, such as features and price, do not capture important distinctions in how firms framed their products. Rather, firms interpreted the synthesizer as having primarily one of two distinct meanings: a new instrument that enables a technologically-sophisticated musician/synthesist to create and/or play new “synth” sounds, or a substitute for acoustic instruments that allows a musician with limited technical knowledge to play realistic emulations of existing instruments. We use the phrase product identity to capture these fundamental differences in meaning and, based on our data, we find that product identity addresses four fundamental questions: What does the product do? How is it used? Who uses it? And what does it look like? Our study illustrates that product identity is an important aspect of positioning in nascent industries.
Why Some Markets Refuse Innovation by Entrepreneurial Firms, When they Say they Love it: Evidence from Uber’s Transportation Network Service by Yongwook (Yong) Paik, Washington University in St. Louis, and Sukhun Kang
Scholars widely believe that entrepreneurship and innovation is beneficial to society and economic growth. Yet, often times, there exists strenuous resistance to such innovation that entrepreneurial firms introduce to the marketplace, in the forms of regulation, litigation, or even protests attempting to ban the commercialization of the innovation despite strong demand from consumers. Such resistance also displays substantial heterogeneity across markets. In this study, we investigate the political determinants of commercializing innovation and argue that the more politically entrenched a market is, the more resistance commercializing innovation will face. This is because political leaders in politically entrenched markets are less likely to respond to public demand and more likely to respond to demand by incumbent businesses. We measure the degree of political entrenchment in a market using the turnover rate of the political leader’s party affiliation and the proportion of tax revenue contributed by local businesses. In addition, using hand-collected data from the online sharing economy, we use Uber’s operations in 161 cities across the fifty U.S. states and some of the resistances it faced when launching its innovative mobile transportation network service in various U.S. markets to test our hypothesis. We find that our data broadly support our hypotheses. This study contributes to the entrepreneurship, innovation, and strategy literature by highlighting the significance of political determinants and heterogeneity of institutional environment across markets in successfully commercializing innovation.
Price Discrimination Across Internet-Connected Devices: The Case of the Hotel Industry by Vincenzo Palermo, University of Toronto (@Vi_Palermo), Dan Breznitz, and Avi Goldfarb
Shopping and buying on internet connected mobile devices has become more common over the recent years: it is predicted that there will be more than two billion smartphone users in the world by 2016. The fast growth of mobile ad spending has triggered an increasing body of empirical research on the topic. While marketers keep increasing their investment in mobile internet services, little research is done on the behavioral differences across users that adopt a mobile device and those that prefer a personal computer (PC). We aim to fill the gap in the existing literature, by providing two different insights (1) on how firms can capitalize from discerning between mobile and PC users by adopting a price discrimination strategy and (2) on how price can also be affected by the timing of the users’ searches. Our data contains daily information on hotel bookings that occur both on mobile and PC devices for 1.8 million unique users between April 07, 2013 and July 31, 2014. The adoption of tracking technologies can be an innovative strategy for large corporations and start-up that offer online purchases. The ability to discern among users offers the opportunity to rapidly increase profitability by reducing search costs and offer a better (more tailored) product.
Careers As An Industry Structure Problem: Specialization in Training and the New Ports of Entry by Tiantian Yang, Duke University (@Yang_Tiantian), and Matthew Bidwell
Just as firms differ in their capabilities to provide goods and services, we argue that they also differ in their capabilities to effectively socialize workers into careers. Drawing on theories of industry structure and vertical disintegration, we theorize that differences in firms’ abilities to engage in such socialization will shape careers, with workers disproportionately starting careers in a few firms that specialize in training and socializing new workers, firms that we call “ports of entry”. Using longitudinal data from Sweden that follow employees from 1993–-2012, our results show three important findings: (1) entry into the labor market post-graduation does, indeed, take place through a highly concentrated set of firms; (2) firms that hire a large number of new graduates are better able to train them and offer better opportunities for future employability; (3) workers who begin their careers in those organizations subsequently earn more over the course of their careers than workers who don’t receive similar training. Our research thus makes important contributions to career mobility and firm strategy.

8:15 a.m. – Session 8: The Rise of Programmatic Research in Entrepreneurship (Kansas City Room)

Chair: Amisha Miller, Kauffman Foundation (@AmishaMiller)

Behavioral Biases in Hiring: Evidence from a Field Experiment in Machine Learning by Bo Cowgill, Columbia University (@bocowgill)
How much do human behavioral biases distort hiring decisions? I share results from a large natural experiment and from a separate large scale field experiment. In the natural experiment, screeners are randomly assigned to job applicants. I show evidence that outcomes for applicants are highly dependent on the screener, and particularly the characteristics of the applicant-screener match. Next, I evaluate a large scale field experiment attempting to manage these behavioral biases. I find large differences (about 20%) in the numbers of candidates who proceed to interviews between treatment and control. Treatment candidates are ultimately 7% more likely to be extended an offer and accept the offer than control candidates -- this difference is driven not only by differences in interviewing rates but also in performance during the interviews.

The results demonstrate the ability of modern algorithms to automate, improve and displace white-collar labor. Hiring is a particularly interesting area for this result, as prior studies have emphasized the nuanced judgments of character, interpersonal skills, career trajectory and other intangible characteristics in screening workers. Several results suggest that the effect is driven by a decrease in homophily. The effect is particularly strong for non-traditional candidates with atypical academic backgrounds. These results are also surprising particularly because the machine was trained using human data. My results also suggest that the machine's performance improvement is the result of aggregating individual screeners' tastes, which may cancel out idiosyncratic tastes that would otherwise affect human screeners' decisions.
On Becoming an Entrepreneur: Unpacking the Role of Shared Work Space in Entrepreneurial Identity Development by Eliana Crosina, Boston College (@ecrosina)
Over the past decade, the number of “entrepreneurial hubs” of various kinds - coworking facilities, incubators, and accelerators - has grown exponentially. This raises questions about their role in supporting entrepreneurial activity, broadly, and entrepreneurial identity development, more specifically. The latter focus is important as entrepreneurial identity has been found to influence entrepreneurs’ decisions and their actions, and it constitutes “…a central aspect of how entrepreneurs secure the resources they require for new venture success…” (Navis & Glynn, 2011: 495). Through an inductive, longitudinal field study of entrepreneurs working in a shared facility for startups (“Phoenix Nest”), I explore the role of space (physical and social) in entrepreneurial identity work. My observations (510 hours) and semi-structured interviews (N=92), suggest that nascent entrepreneurs struggle to co-create an occupational and business identity: that is, to develop a sense of “who they are” while at the same time launching their organizations. Phoenix Nest constitutes an environment that mitigates these challenges by legitimizing their activities and offering a forum for identity experimentation. Identity work, then, allows some to: (a) become more resilient in the pursuit of their projects while others (b) stall or (c) abandon their entrepreneurial paths altogether. In my paper, I explore how entrepreneurs’ different utilization of Phoenix Nest physical space, and a range of other practices, leads them down one of these paths.
Student Experiences in University Incubators at State Schools by Daniel Davis, University of California, San Diego
Of the roughly 1,250 venture incubators in the US, some researchers claim that as many as one in three are now sponsored by a college or university. This is a relatively new phenomena, with the vast majority of these campus programs springing up in the last decade. This project surveys student participants in campus affiliated business incubators across the state of California, and also ethnographically follows student teams at a California State University campus and at a University of California campus. Research questions center on student experiences, academic impacts, and program strengths; as well as differences by race, class, gender, and institutional type.
Network Dynamics and Idea-Based Performance Advantages: Experimental Results from an Entrepreneurship Boot Camp by Rem Koning, Stanford University (@orgRem)
Do networks plentiful in ideas provide early stage startups with performance advantages? On the one hand, network positions that provide access to a multitude of ideas are thought to increase team performance. On the other hand, research on network formation argues that such positional advantages should be fleeting as entrepreneurs both strategically compete for the most valuable network positions and form relationships with others who have similar characteristics and abilities. I embed a field experiment in a three-week-long startup boot camp to test if networks that are plentiful in ideas lead to performance advantages. Using detailed data from the boot camp's custom-designed learning management platform, I find support for the first hypothesis. Teams with networks more plentiful in ideas receive better peer evaluations and more crowdfunding page views. I find little evidence that entrepreneurs actively build networks to others who could have provided a greater quantity of information and ideas. Instead, entrepreneurs seek feedback from those they have collaborated with in the past or who share similar ascriptive characteristics. These findings provide first-order evidence for the importance of knowledge spillovers within boot camps, incubators, and accelerators. Furthermore, the findings provide a potential explanation for the durability of idea and information-based network advantages.
Learning from Experimentation by Sandy Yu, University of California, Berkeley (@sandyyu_), and Alicia Robb (@aliciarobb)
How and what do founders learn from feedback? The existing literature on early-stage financing tends to focus on the determinants of entrepreneurial success. However, the majority of entrepreneurs fail in their new ventures, and there is a question of whether any value is created during this process. In this paper, we use a proprietary database of applicants to accelerators and pitch competitions to analyze how and what entrepreneurs learn from various feedback mechanisms. Early and frequent feedback may help start-ups fail faster, but they may also fill a unique role in motivating experimentation and learning.

10:30 a.m. – Session 9: Entrepreneurship and Public Policy (Troost Room)

Chair: Wie Yusuf, Old Dominion University (@wie_rockstar)

Should they stay or should they go? The migration patterns of university-licensed startups by David Benson, Brigham Young University (@dbenso01), Paul Godfrey, and Gov Allen
Can technology transfers from university-sponsored research fuel local and regional economic development? Prior research generally assumes that successful technology licensees operate locally, near the university. We question that assumption and investigate the following question: (1) How often startups locate away from the university? And (2) Are those that leave more successful than those that stay? We test this question on a sample of 2,051 firms from top 50 U. S. research universities. We find that roughly 30% of startups locate more than 60 miles from the university, but that these startups are far more likely to raise venture capital, be acquired or go public than those that stay local. The exceptions to this are startups from universities already located in a cluster. We identify implications of our findings for theory development, public policy, and managerial decision-making.
Firms, Productivity, and Earnings Inequality in Brazil from 1988-2012 by Christian Moser, Princeton University, Niklas Engbom, and Jorge Alvarez
This paper highlights the importance of firms and productivity in understanding the evolution of earnings inequality among Brazilian workers. Brazil experienced a dramatic decline in earnings inequality over the last two decades, with the variance of log earnings falling by 26 log points. Using a large matched employer-employee data set, we fit models with log additive worker and firm fixed effects within six overlapping subperiods in order to understand the sources of this decline. We find that compression in firm-specific pay accounts for 45 percent of the decline in the variance of log earnings between 1996 and 2012 and compression in worker fixed effects for 28 percent, with changes in worker observables, covariances and the residual explaining the remainder. We show that the compression in the firm component is not a result of compression in underlying measures of firm performance and compression in the worker component is only to some extent driven by compression in measures of worker ability. Rather, we argue that a significant fraction of the decline in inequality resulted from changes to the way firm and worker characteristics are reflected in pay. Models explaining what caused the rapid decline in inequality in Brazil over this time period need to be consistent with this fact.
Taxes, Regulations of Businesses and Evolution of Income Inequality in the U.S. by Sebastian Dyrda, University of Toronto, and Benjamin Pugsley
The share of business receipts of pass-through entities (S corporations, partnerships and sole proprietorships) in the total business receipts in the U.S. increased from 13 percent in 1980 to 37 percent in 2012. This significant change in the structure of the legal forms of organization of the U.S. businesses has been accompanied by increase in the top shares of individual income and change in their composition. The top 1 percent income share has increased from 8.2 percent to 18.9 percent in this period. At the same time the share of entrepreneurial income in the top 1 percent has risen from 7.9 percent to 29.0 percent, contributing significantly to the increase. In this paper we document, using The Longitudinal Business Database, new facts about the shift in the organization forms and provide its decomposition into the entry and switching margin. We argue the change in the organization forms has been largely driven by the tax reforms and regulations, which made running a business as a pass-through entity more attractive. To validate this claim we write down heterogeneous firms, general equilibrium model with endogenous choice of the business organization form and simulate it with the time varying taxes and regulations of businesses disciplined by the data. Given the model replicates the key facts from the data about the shift in the organization forms we use it to quantify the impact of the changes in taxes and regulations of businesses on the increase in the top incomes shares in the U.S.
The Law and Economics of Prospective Patent Remedies by Erik Hovenkamp, Northwestern University
The law and economics of property disputes offers a number of insights on the comparative efficiency of prospective remedies for ongoing harms. But these contributions have not been meaningfully embraced by patent law, nor have they been adapted to address all of the distinct institutional and competitive concerns that arise in intellectual property disputes. Accordingly, this paper makes two principal arguments. The first is that prospective patent remedies do not adequately encourage the efficient allocation of patent rights. For example, in many instances prospective damages are enhanced to account for \"willfulness,\" but such factors have no bearing on the efficient terms of trade. Thus the courts risk undermining transactional efficiency by including them in the prospective component of the remedy. Second, the law and economics of property disputes reaches some very different conclusions when the parties happen to be competing firms, as is often the case in patent disputes. The Coase theorem says that private parties who can bargain will allocate property rights efficiently, meaning it will maximize their joint welfare. But competitors\' joint welfare may be maximized by suppressing competition, which is socially inefficient. Thus, for example, even if the parties could have reached mutually-beneficial licensing terms ex ante, a patentee may refuse to bargain around an injunction ex post, because exclusion provides immense anticompetitive value, and may raise joint profits in a concentrated market. Thus, choosing the right remedy is even more important in patent disputes, as ex post bargaining may not achieve a socially efficient result.
When Are Economically Distressed Areas a Source of Competitive Advantage for Technology Entrepreneurs? Evidence from HUBZones and the Small Business Technology Transfer (STTR) Program, 1999-2012 by Amol M. Joshi, Oregon State University, Todd M. Inouye, and Jeffrey A. Robinson
By law, U.S. Federal government agencies are required to allocate a percentage of their budgets to procurement from certified small businesses located in economically distressed areas specially designated as Historically Underutilized Business Zones (HUBZones). These areas tend to be disproportionately composed of minority and socioeconomically disadvantaged groups. We examine the extent to which technology entrepreneurs pursuing federal funding from the STTR Program benefit from locating and conducting research in a HUBZone. Our empirical analysis of 6991 STTR Phase I awards reveals that research contracts conducted by firms in a HUBZone have 169% higher odds of obtaining Phase II follow-on funding compared to research contracts conducted by non-HUBZone firms. For Department of Defense (DOD) sponsored research contracts, this effect is further amplified. When a HUBZone firm conducts DOD research in STTR Phase I, it has 288% higher odds of obtaining Phase II follow-on funding in comparison to DOD research conducted by firms not in a HUBZone. Because firm owners may exercise managerial discretion in choosing where to locate their research and development (R&D) activities and in identifying which sources of R&D funding to pursue, our study implies that deciding to conduct DOD-sponsored R&D in a HUBZone may maximize the firm's chances for obtaining the follow-on funding needed to commercialize new technologies. Since HUBZone certification is dependent on place, rather than race, our findings imply that minority and non-minority firm owners alike may find it advantageous to conduct R&D in a HUBZone, especially when pursuing DOD funding via the STTR Program.
Too Costly To Convince: How Do Entrepreneurs Market Breakthrough Innovation Through Partnership? by Yujin Kim, University of California, Berkeley
This research examines the strategic conditions that drive entrepreneurial innovators to pursue novel innovation rather than innovation that is closer to existing technologies. To an increasing extent, startups commercialize innovation in a cooperative setup. Because radical breakthrough innovation is more difficult to communicate than its incremental counterpart, entrepreneurial innovators may avoid breakthrough innovation for which the cost of developing credible information is exceedingly high. In the context of the Orphan Drug Act (ODA), this study uses a difference-in-difference approach to measure whether entrepreneurs are more likely to bring novel innovations to the market when the policy change unexpectedly lowers the cost of obtaining information that will convince investors through a small market test. Using a new measure of the novelty of innovation and a detailed panel dataset of therapeutic molecules, this empirical study finds that biotech startups bring more breakthrough drugs to markets affected by the ODA. This research also finds that in ODA-affected areas, entrepreneurs hold novel projects longer before contracting with partners and aim to generate more revenue streams from pursuing novel innovation. Taken together, the results of this study suggest that the cost of convincing investors prevents entrepreneurs from marketing novel innovation and that a public policy can moderate inefficiency in the “market for ideas” by decreasing communication costs.

10:30 a.m. – Session 10: The Importance of Networks in Entrepreneurship (Kansas City Room)

Chair: Jason Greenberg, New York University

The Influence of Pre-Founding Inventor Collaborations and Post-Founding Managerial Initiatives on Venture Knowledge Quality by Warren Boeker, Sandip Basu, Baruch College, City University of New York, Michael Howard, and Arvin Sahaym
Some of the most enduring questions in the field of strategic management are concerned with the extent to which founding attributes of firms can be a source of lasing competitive advantage. For technologically-intensive firms, the successful development of new knowledge has important implications for survival and growth. Research has not examined if founders’ prior collaborative relationships can influence their firms’ knowledge development even in the absence of active ongoing collaboration. Further, there is little examination of whether certain important managerial initiatives such as forming alliances and hiring skilled employees can compensate for the founders’ lack of ties to talented collaborators. We address these issues by employing a knowledge recombination perspective in which knowledge creation is conceptualized as an outcome of the combination and synthesis of underlying knowledge elements. We test our theory on a sample of biotechnology ventures founded from 1990–2000, tracking the impact of the knowledge they create in their first ten years after founding. While we find evidence that pre-founding inventor collaborations of venture founders have continuing advantages for ventures’ knowledge development, our findings also show that such starting advantages can be partially compensated for by post-founding initiatives. Our study represents a preliminary attempt to integrate the contrasting perspectives that argue for the persistent benefits of favorable founding conditions and the efficacy of managerial agency in eroding such initial advantages. We close by discussing the implications of our study for research on inventor collaboration, external search, and new venture strategies.
Boundary Spanning in Converging Industries by Nachiket Bhawe, North Carolina State University, Shaker Zahra, and Don Neubaum
This study examines the performance of new products launched by boundary spanners in the video game industry. To understand this relationship, we use industry convergence as a research context where we can differentiate industry boundaries and identify inter-industry and intra-industry boundary spanning. While few studies systematically explore within and inter industry boundary spanning performance, using data from the converging video game industry, we find that boundary spanning both intra and inter industry enhances performance. The level of product complexity reduces the effect of intra-industry boundary spanning and increases the effect of inter-industry boundary spanning. Consistent with our theory, we also find that over time as industry boundaries erode over time the advantage of product complexity declines.
Entrepreneurial Adaptation and Social Networks: Evidence from a Randomized Experiment on a MOOC Platform by Charles Eesley, Stanford University (@eesley), and Lynn Wu
We examine the performance of early-stage entrepreneurs before and after randomly showing them different approaches to the strategic process. In isolation, a planning strategic process is more effective than a more adaptive approach. Contrary to the finding that entrepreneurs often change their business model and strategic direction frequently, we find that instructing entrepreneurs to have a strong, persistent vision for their startup often results in better performance in the early stages. In particular, the results show that adding a diverse network tie alone is less effective than combining a diverse tie with a specific strategic approach. In contrast to prior work that shows that entrepreneurs often begin their ventures with a cohesive, closed network high in trust then transition later to a more diverse network, we find that early stage ventures appear to be better off with more diverse social ties in the beginning, particularly if a more adaptive approach is adopted for the venture's strategy. The results suggest that when formulating a strategic approach there is an important interaction between social networks and the strategy formulation process that must be taken into consideration.
Is Bigger Better? Lab Productivity and Lab Size by Christopher C. Liu, University of Toronto
Although laboratories are ubiquitous features of science, which and what types of scientific labs should receive government funding remains a topic of ongoing and often contentious debate. Recently, there have been prominent assertions that labs have grown too large to be productive. However, these assessments have not taken into consideration the laboratory environment within which productive work is embedded. Here, I stress that labs should be assessed on their conversion of human capital to scientific outcomes, rather than a singular focus on increasing paper output. Using data from the MIT Department of Biology, the productivity measure developed in this paper illustrates a negative correlation between increases in lab size and lab productivity. Lastly, I show that the use of my productivity metric would have shifted over 20% of the funds allocated to MIT biology. Thus, if adopted by granting agencies such as the NIH, this metric could easily reshape the funding landscape on the order of billions of dollars per annum.