We assess the heterogeneous impact of economic downturns on individuals’ decisions to bring high-technology ideas to the market in the form of new ventures. We thereby examine how worsening labor market conditions influence individuals’ opportunity costs of starting new ventures, the resulting composition of the entrepreneurial pool, and startup performance outcomes. Using a rich dataset of startup founders in the biotechnology and medical device sectors, we find that an increase in the unemployment rate is associated with a substantial rise in the share of entrepreneurs who are most sensitive to worsening labor market conditions. Additionally, we find that startups founded by these entrepreneurs display lower financial and innovative performance than startups founded by entrepreneurs who are relatively insensitive to business cycles. Finally, we provide suggestive evidence that individuals’ heterogeneous response to worsening labor market conditions is a relevant factor in explaining the negative relationship between unemployment and startup performance outcomes.

What role do differences in founders' occupational backgrounds play in new venture performance? Analyzing a novel dataset of 2,998 founders creating 1,723 innovative startups in biomedicine, we find that the likelihood and hazard of achieving a liquidity event are lower for academic than for non-academic startups. However, academic startups produce as many patents and receive as much funding as non-academic startups, suggesting that the observed differences in achieving a liquidity event are not driven by differential invention performance. Exploiting heterogeneity among academic startups, we also find that differences between professor and student startups do not explain academic startups' comparatively low performance on the exit market vis-à-vis nonacademic startups. Yet, startups founded by superstar professors perform similarly to non-academic startups on the exit market for new ventures, and better than startups founded by highly productive professors but without external certification.

In this paper, we analyze how the physical layout of cities affects innovation by influencing the organization of knowledge exchange. We exploit a novel data set covering all Census Block Groups in the contiguous United States with information on innovation outcomes, street infrastructure, as well as population and workforce characteristics. To deal with concerns of omitted variable bias, we apply commuting zone fixed effects and construct instruments based on historic city planning. The results suggest that variation in street network density may explain regional innovation differentials beyond the traditional location externalities found in the literature.

Entrepreneurs learn from a variety of sources. One particularly important channel is learning from fellow entrepreneurs. In this paper, we examine the influence of physical proximity on technology adoption decisions and subsequent startup performance outcomes at one of the largest technology co-working hubs in the United States. To deal with endogenous geographic clustering, we rely on the random assignment of office space to the hub's 251 startups. Using floor plans to measure geographic distances, we find that proximity greatly influences the likelihood of adopting an upstream technology also used by a peer firm. This effect quickly decays with distance where startups that are more than 20 meters apart no longer influence each other. Our results suggest that other social, informational, and competition-based dimensions alter the effect of distance and that peer interaction is a viable mechanism driving our results.