Cost of Experimentation and the Evolution of Venture Capital

Michael Ewens is Associate Professor of Finance and Entrepreneurship at the California Institute of Technology; Ramana Nanda is the Sarofim-Rock Professor of Business Administration at Harvard Business School; and Matthew Rhodes-Kropf is Visiting Associate Professor of Finance at the MIT Sloan School of Management. This post is based on their recent article, forthcoming in the Journal of Financial Economics.

Related research from the Program on Corporate Governance includes Carrots & Sticks: How VCs Induce Entrepreneurial Teams to Sell Startups, by Jesse Fried and Brian Broughman (discussed on the Forum here), and Agency Costs of Venture Capitalist Control in Startups by Jesse Fried and Mira Ganor.

The introduction of cloud computing services in the mid 2000s was a fundamental technological shift that has also had an impact on the financing landscape for Internet and web-based startups. A key benefit of cloud computing for such startups is the ability to “rent” hardware space in small increments and scale up as demand grows, instead of making large upfront investments when the outcome of the venture is still uncertain. Entrepreneurs and investors can therefore learn about the viability of startups with substantially less funding, lowering the cost of financing initial “experiments” that can help investors learn about the potential of new ventures before committing further capital.

In our article, Cost of Experimentation and the Evolution of Venture Capital, we show that subsequent to the introduction of cloud computing services by Amazon Web Services (AWS) in 2006, startups founded in sectors benefiting most from the introduction of AWS raised much less capital in their first round of VC financing. On average, initial funding fell 20%, but quantile regressions demonstrate that this fall was even larger for many parts of the funding-size distribution. While startups in treated sectors raised less capital in their initial two years, total capital raised by firms in treated sectors that survived three or more years was unchanged. This is consistent with the notion that the primary effect that AWS had was on the initial fundraising by startups rather than on the cost of running the business at scale—effectively allowing startups to shift their large investments to later stages when uncertainty had been resolved.

This fall in the cost of starting businesses also impacted the way in which VCs managed their portfolios. We show that in sectors impacted by the technological shock, VCs responded by providing a little funding and limited governance to an increased number of startups, which they were more likely to abandon after the initial round of funding. The number of initial investments made per year by VCs in treated sectors nearly doubled from the pre- to the post-period, without a commensurate increase in follow-on investments. In addition, VCs making initial investments in treated sectors were less likely to take a board seat following the technological shock. The investment approach of supplying a small initial amount in a large number of startups, providing less governance, and offering follow-on funding less frequently, is colloquially referred to as one where investors “spray and pray.”

Importantly, however, we also document how this falling cost of experimentation allowed a set of entrepreneurs who would not have been financed in the past to receive early-stage financing. Many of these were ventures with a low probability of success, but with a high return if successful (what we refer to as “long shot bets”). We show theoretically how “long shot bets” have high failure rates but experience large increases in valuation if they receive another round of funding. This is because long-shot bets have a small chance of a huge payoff, so the relatively infrequent times where the initial experiment generates positive information will lead investors to update their posterior beliefs about the startup’s expected value much more—and hence lead to higher step-ups in value from the first to the second round of funding.

Our empirical results comparing firms in sectors benefiting from AWS to those in unaffected sectors over the period 2001 to 2010 document that relative to the pre-2006 period, VCs increased their investments in startups run by younger, less experienced founding teams. These firms were more likely to fail post-2006, but conditional on another round of funding, had nearly 20% higher step-ups in value across rounds than equivalent startups in untreated sectors. This finding is consistent with the notion that on balance, there was an increased financing of long-shot bets by VCs. We perform a number of other tests to check for alternative explanations.

An interesting implication of our results is that the value-added role of VCs may have fallen at precisely the point in the startup’s lifecycle that value-add is needed the most. Moreover, younger and more-inexperienced founders, who have a technology that holds a lot of promise but are not experienced at running a firm, arguably need mentorship and governance even more than those run by more experienced entrepreneurs. Yet VCs’ need to “spray and pray” to make the investment profitable implies that these startups only get financed with limited mentorship and governance. This finding helps to explain the rise of new financial intermediaries such as accelerators (which have emerged in the last decade) that provide scalable, lower cost forms of mentorship to inexperienced founding teams. These are a natural response to the gap in value add created by the evolution of VCs’ investment behavior in early rounds to a more passive spray and pray investment approach.

The full article is available here.

Both comments and trackbacks are currently closed.