Why do small firms often fail to adopt new profitable opportunities, even in the absence of informational frictions, fixed costs, or misaligned incentives? We explore three potential mechanisms: present bias, memory, and trust in other firms. In partnership with a financial technology (FinTech) company in Mexico, we randomly offer firms that are already users of the payment technology the opportunity to be charged a lower merchant fee for each payment they receive from customers. The median value of the fee reduction is 3% of profits. We randomly vary the size of the fee reduction, whether the firms face a deadline to accept the offer, whether they receive a reminder, and whether we tell them in advance that they will receive a reminder. While deadlines do not affect take-up, reminders increase take-up of the lower fee by 18%, and anticipated reminders by an additional 7%. The results point to limited memory in firms, but not present bias. Additional survey data suggests trust as the mechanism behind the significant additional effect of the anticipated reminder. Upon receiving an anticipated reminder from the FinTech company, firms value the offer more and accept it even if they generally distrust advertised offers.

How should governments procure innovation? One choice facing policymakers is whether to tightly specify the innovations they seek (a “Conventional” approach) or to allow firms to suggest ideas (an “Open” approach). We study a natural experiment in the U.S. Air Force Small Business Innovation Research (SBIR) program where Open and Conventional competitions were held simultaneously. We compare them using within-competition regression discontinuity designs on administrative data. Open awards increase desired outcomes; they lead to more adoption of new technologies, measured by (non-SBIR) defense contracts, and more commercial innovation, measured by VC funding and patenting. In contrast, Conventional awards have no effects on these outcomes but do create lock-in through increasing the chances of winning a future SBIR award. The Open program succeeded in its aim of attracting new types of firms, but we demonstrate that openness has a differential impact beyond inducing selection: (i) comparing specific and non-specific Conventional topics; (ii) examining firms that applied to both Open and Conventional programs; and (iii) comparing Open with two other reform programs that attracted similar types of firms to Open but used specific topics. Overall, the results point to benefits from open approaches to innovation procurement.

We study the participation of nascent firms in open source communities and its implications forattracting funding. To do so, we exploit rich data on 160,065 US startups linking information fromCrunchbase to firms’ GitHub accounts. Estimating a within-startup model saturated with fixedeffects, we show that startups accelerate their activities on the platform as they approach their firstfinancing round. The intensity of their involvement on GitHub declines in the twelve months after.Startups intensify those activities that rely on external technology sources above and beyond thetechnologies they themselves control. Exploiting a shock that reduced the relative cost of internalcollaborations, we provide evidence that startups’ decision to integrate external sources of knowledgein their production function hinges on the relative cost vis-`a-vis internal collaboration. Applyingmachine learning to classify GitHub projects, we further unveil that the most prevalent among theseexternal activities are related to software development, data analytics, and integration. Our resultsindicate that VCs and renowned investors are the most responsive to these activities.

We use cross-state business registrations to track the geographic movement of startups with high growth potential. In their first five years, 6.6% percent of these startups move across state borders. Though startup births are concentrated geographically, hubs like Silicon Valley and Boston on net lose startups to entrepreneurial migration. A revealed preference approach nonparametrically identifies the average utility of cities to migrant founders. University towns and startup hubs have low relative utility. This pattern is due neither to vertical sorting nor industrial specialization. The higher-quality startups move to lower-tax, business-friendly cities, while less growth-oriented startups move to low-tax, high-amenity cities.

We study the sources of cross-sectional variation in the performance of first-time venture capital (VC) fund partners. We find that, relative to partners with startup experience, partners with VC experience are at least 20 percent more likely to invest in successful deals or start a follow-on fund. We investigate three potential mechanisms for this finding. Our tests do not support the hypotheses that partners with VC experience make riskier investments or have better deal-selection skills. Consistent with a network effect, we show that the higher success rate for partners with VC experience primarily comes from joining successful syndicates, not from leading successful deals. Our results suggest that a background in venture capital is an important channel for the success of first-time venture funds.