We are glad our paper (with Virginia Sarria Allende and Gabriela Robiolo) went out in Venture Capital : A International Entrepreneurial Finance Journal.
A working paper version is available here. Here are some brief comments on the ideas, the econometrics, and data approach:
In the paper we study the “matching” between investors and startups in the entrepreneurial finance market. Broadly speaking, we are concerned with the question of who will invest in who, and in the role played by (social, professional) networks in the explanation. Specifically, we show evidence on a simple idea: due to information related frictions in the entrepreneurial finance market, being closer in the network of connections actually matters for matching. Being closer increases not only the attractiveness of a prospective match, but also makes observable attributes more attractive.
But “being closer” has a particular interpretation here. Our measured networked connections, are not the typical social (or follower) style of connections. We recognize a link if there is information that you have worked, invested, mentored, etc. a common startup or organization in the past. So we could say that these are really costly (or “signally meaningful”, in Spence’s sense) connections.
For founders (or prospective investors alike) the implication is to work out connections in order to approach desirable matches. If you are not “that” attractive (say, you are not the most experienced investor out there), you could still work on reducing your network distance to you investors, investing in real work in the entrepreneurial ecosystem, since common knowledge can serve in reducing informational assymetries.
Our data approach in reconstructing the massive network of connections in California consisted on recollecting data from Angellist, which is widely known in the entrerepeneurial ecosystem as the most important site for investing in startups, among many other market and social-related features. We gathered data on the history of roles individuals took in the entrepreneurial setting. Doing it efficiently and securely was an interesting project per se (we discussed the project here).
A added comment for those with an interest on econometrics. It turns out that estimating the drivers of a match was not trivial at all. Say for instance you are interested in estimating the specific probability between a match between startup i and investor j. You would only observe realized matches, but those other potential matches entail, in principle, all other possible combinations. And as you know, the number of pairs of possible combinations scale up really fast. The approach we follow in the paper is to sample among these counterfactual potential matches.
Another difficutly is that observations in the estimated model would neither be independent, since a realized match would lower the probability of some other potential matches. In dealing with these issues we followed a very ingenious, game theory based-approach, proposed by Jeremy Fox, that exploits the pairwise stability matching condition. As an intuition, this condition says that , if you to take two pair of realized matches, and switch them, the value of such a switch should be lower than the actual match. Eventually this condition is used in finding out which are the matching value function parameters that would maximize the number correct pair of matches (instead of their switching alternatives).
Another interesting insight is related to how to specify matching drivers. The insight here comes from matching theory: when estimating the driver of a match what actually matters is not, say, the individual characteristic of a startup (which is fixed among all possible matching partners) but how these characteritics interact with those of the counterpart. I understand this insight goes back, at least to the examination of marriage matches of Gary Becker in 1973. So, what drives a matches are located in the space of interactions…