Research Projects
The Foundations of Advice: A Quantitative Exploration
Files:[Main Paper][Appendices][Supplemental Material]
ABSTRACT: Advice is a key input to early firm development, but our understanding of it remains quite limited. Using a sample of 7,914 mentoring decisions by seed-stage investors participating in a mentoring program, I develop a framework to systematically analyze advice and offer among the first empirical insights into its nature and provision. The characteristic element of mentors' advice is to do less and learn more, nudging entrepreneurs to balance business objectives that entail costly to reverse commitments with objectives focused on discovering and evaluating options. Although mentors are quite aligned in this message, I find significant differences between angel investors and venture capitalists in the provision of hands-on mentoring to help founders achieve their objectives. Angels are more likely than VCs to help founders design and execute product market experiments, while VCs provide more support on business analysis and planning tasks. I find evidence consistent with the hypothesis that experimentation is a skill developed via learning-by-doing, and angels have a skill advantage in that domain due to having more operational experience. These and other findings reported offer several new research opportunities in economics, entrepreneurial strategy, and finance.
Information Frictions and Employee Sorting Between Startups
w/ Kevin Bryan and Mitchell Hoffman
Additional Files: [Main Paper][AEA RCT Registry][Appendices][RCT Screenshots][Supplemental Material]
ABSTRACT: Would workers apply to better firms if they were more informed about firm quality? Collaborating with 26 science-based startups, we create a custom job board and invite business school alumni to apply. The job board randomizes across applicants to show coarse expert ratings of all startups' science and/or business model quality. Making this information visible strongly reallocates applications toward better firms. This reallocation holds even when restricting to high-quality workers. The treatments operate in part by shifting worker beliefs about firms' right-tail outcomes. Despite these benefits, workers make post-treatment bets indicating highly overoptimistic beliefs about startup success, suggesting a problem of broader informational deficits.
Does Mentorship Drive Startup Performance? Yes, But Only for High Learners
w/ Ajay Agrawal and Avi Goldfarb
(manuscript ready for conference presentation)
ABSTRACT: Mentorship is a staple component of private sector accelerators designed to maximize equity value and also of public sector initiatives created to support economic development. This paper examines whether, how, and when mentorship enhances startup performance. We show that mentorship drives startup performance. To address endogeneity concerns due to mentor selection, we exploit randomness in the availability of mentors to spend time with startups due to personal scheduling conflicts. We then show that one channel through which mentorship operates is founders learning how to set priorities for their companies. We conduct this empirical study using a novel panel of 289 high-technology startups participating in a global eight-month program for seed-stage companies.
How does Industry Affiliation of Academic Scientists Affect the Rate and Direction of Research?
(manuscript ready for conference presentation)
ABSTRACT: The implications of academic collaborations with industry have long been the nexus of contentious debate. The prevailing concern is that industry causes research to lose its fundamental depth and become commercially driven. This paper presents evidence that these concerns are misplaced for areas of research that have commercial value--that is, the specific domains for which these concerns are raised. The empirical analysis uses large-scale, manually improved bibliometric data from artificial intelligence research. For identification, I use the unexpected and significant success of the neural network techniques revealed at the ImageNet 2012 benchmark competition, which sharply increased the industry's demand for AI scientists, but more so for scientists with higher expertise in the breakthrough field. While industry affiliation significantly increases the usefulness of research, it does not diminish its novelty. In addition, industry affiliation raises both the publication and the quality of science produced by academic scientists. Results are consistent with the explanation that, at least in the medium term, academics in short supply can negotiate higher academic freedom while utilizing commercial resources for their research.
Media Mentions: The Economist (Million Dollar Babies and Battle of the Brains)
Learning vs. Doing: The Effect of Business Uncertainty on Entrepreneurial Activities
(preparing manuscript for conference presentations)
ABSTRACT: Over the past decade, public and private startup mentorship programs have proliferated. Yet the empirical investigation of this phenomenon is scant. I examine advice in the context of change in startup activities. Resource-constrained entrepreneurs trade off prioritizing between learning about and evaluating their options versus implementing them. In a setting where mentorship advice regulates this trade-off, I show that, relative to mentors, entrepreneurs under-prioritize simple search and planning activities—a form of entrepreneurial learning that is broadly termed “analysis.” Mentors’ call for more learning through analysis is precisely at the expense of de-emphasizing the implementation of ideas in the short term. I show that this result is driven by mentors’ perceived uncertainty of the startup’s quality, where perceived uncertainty is proxied from mentors’ expectation error dispersion and sentiment variation.
Activity Sequencing in Startups
(preparing manuscript for conference presentations)
ABSTRACT: In this paper, I investigate the sequence of startup activities over time to understand the mechanisms underlying the prioritization of activities in startups. I develop a novel typology of startup activities using a database of 371 early-stage, science-based startups. I show that entrepreneurs, particularly first-time founders, under-prioritize learning. Using Latent Markov Models I show that the sequence of activities in early-stage startups from learning to implementation of ideas and acquisition of resources increases startups’ success in accessing capital.
Other Work in Progress
The Effect of Noisy Learning on Startup Performance, with Joshua Gans, Erin Scott, Scott Stern (data collection and empirical analysis)
Database Development
Database, Methodological Tools, and Research Opportunities: Creative Destruction Lab and Early-Stage Technology Ventures
Perhaps the most significant obstacle to conducting entrepreneurship research is the lack of high-quality systematic data on startups that allow us to test critical hypotheses about how startups form, grow, and succeed. To address this challenge, I have hand-collected detailed data from several hundred high-technology startups participating in Creative Destruction Lab (CDL), a global entrepreneurship program. This database is, to my knowledge, the largest of its kind, offering granular longitudinal visibility on the process of firm development in many high-growth startups across various technological domains, ranging from quantum computing to space. The relational database developed consists of nearly 20 tables that cover information ranging from the background of founders and their mentors to details about changes in the product and target market of each startup, as well as the type and amount of financial and non-financial resources they received along the way.
To help accelerate entrepreneurship research, these data are made available to 20 scholars across eight institutions. As of 2023, 15 research projects using these data are underway, of which the following two have been published:
2021: Per Davidsson, Denis Gregoire, Maike Lex; Venture Idea Assessment (VIA): Development of a needed concept, measure, and research agenda; Journal of Business Venturing
2022: Álvaro Parra, Ralph A. Winter; Early-stage venture financing; Journal of Corporate Finance
Creative Destruction Lab is an entrepreneurship program for early-stage science-based start-ups founded in 2012 by Professor Ajay Agrawal at the Rotman School of Management. My work to compile its data has been generously supported by CDL, the Federal Government of Canada (Strategic Innovation Fund), and many faculty and staff of the University of Toronto’s Rotman School of Management. As of June 2020, eighteen scholars at the University of Toronto, Harvard University, the University of Chicago, HEC Montreal, HEC Paris, Dalhousie University, the University of British Columbia, and the University of Calgary have been approved and given access to use these data for fifteen research projects.
The Rate and Direction of Academic Research in Artificial Intelligence
Ajay Agrawal and I collected data on a decade of conference proceedings of major AI conferences to understand how the state of the labor market for AI scientists and the distribution of scientific productivity has changed since the 2012 ImageNet Competition.
Findings from these data were presented to world leaders from government, industry, and the scientific community at the 2015 Future of Life Institute conference. For an overview on this conference see this Washington Post article.