ABSTRACT
We review statistical methods used to estimate the impact of crowding out of private venture capital (VC) by government VC. We review three types of failures that have plagued the VC literature and resulted in policy implications that are precisely the opposite of what the data actually indicate. The first failure involves the mistaken use of measures that give rise to country rankings where the best VC markets in the world are countries like Austria and Hungary, and the worst VC market in the world is the U.K. The second and more recent failure involves the use of data that do not predate the creation of government VC. The third type of failure involves not accounting for the nonrandom matching between entrepreneurs and government VC programs. We show that statistical inference in recent work that makes this latter mistake can give rise to remarkably incorrect conclusions; including, for example, a bizarre and clearly false inference that a market with more than 89% investment by government funds exhibits no evidence of displacement of private funds. In view of these issues, we offer suggestions for future research and raise some new questions that could guide policymakers in the future.
Acknowledgments
We owe thanks to Richard Harrison and Colin Mason for many excellent comments and suggestions.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
2. By analogy, it is akin to making an absurd argument that socialism does not crowd out capitalism by only looking at data that postdate the introduction of socialism and observing little variability in and scant levels of capitalism.
3. Some inferences might be properly drawn with data that only postdate the program on topics other than domestic crowding out, but those topics are not among the issues raised in Dahaj, Cozzarin, and Talebi (Citation2018).