AAA What is the performance of government-sponsored VCs?

What is the performance of government-sponsored VCs?

Government venturing is an increasingly important part of today’s global venture capital activities. According to the Global Government Venturing (GGV) database, in Q4-2014, there were 173 investment deals with a government-backed fund in the syndicate and 12 exits through sale or flotation. During the same quarter, governments disclosed $1.25bn of commitments to 20 VC funds raising in total more than $3bn. In addition, more than 40 funds were raised with government commitments with closed or set target allocation from budgets of $18.65bn.

Following the remarkable success of private and corporate venture capital (CVC) funds in accelerating innovation and creating jobs, a significant number of governments have sponsored the provision of finance to entrepreneurial ventures. The active engagement of governments in VC activities is an important innovation strategy that complements, rather than substitute, the range of tax and research and development policies for venture capital.

But what is the record of government support for venture capital? What is the impact of government-sponsored VCs (GVC) on the likelihood of success of new entrepreneurial ventures? A recent NBER working paper co-authored by J. A. Breander, Q. Du and T.F. Hellmann empirically addresses this question using international enterprise-level data, which includes 21,852 ventures located in 25 countries that received funding in the 2000-2008 period. The remarkable sample offers a substantial representation of US, European and East Asian markets along with Australia, Brazil, Canada, India and Israel.

The results show that “compared to a benchmark of ventures financed by private VCs, a small amount of GVC investment appears to be a good thing, but larger amounts of GVC decrease the likelihood of successful exit (initial public offerings and acquisitions).”  Moreover, the study examines whether there is a difference in terms of performance between government-owned and government-sponsored VCs. The key insight here is that the superior performance of minority GVCs applies only to government-supported VCs, that is, support outperforms ownership. As the authors observe, the results indicate that “some market discipline helps make government promotion of venture capital more effective.”

The following graphs highlight the key findings of GVC (government venture capital) at different intensity levels also vis-à-vis PVC (private venture capital), as well as GVC involvements by different global geographies and the GVC participation in the various industry sectors where PVC invests. The three performance measures are based on (1) the value of the enterprise, (2) when exits occur through initial public offering / merger and acquisitions, and (3) by the total venture capital investment received by the enterprise. These “value creation” measures are closely related to the economic value of the enterprise. In summary, these performance measures are of interest in part because they reflect “private” returns – returns to venture capitalists, other investors and entrepreneurs. In addition, these measures also reflect benefits to other parties such as customers, workers and other enterprises. In addition, successful enterprises help generate tax and other revenues for governments.

References

James A Brander, Qianqian Du, Thomas F Hellmann (Working Paper 16521, November 2010; http://www.nberorg/papers/w16521). “The Effects of Government-Sponsored Venture Capital Evidence.” National Bureau of Academic Research (USA)

Boris Battistini is a senior research fellow at the Swiss Federal Institute of Technology (ETH Zürich) and an associate at Metellus, a venture capital firm based in Zürich, London and San Diego. Email: boris.battistini@metellus.ch Twitter:  @bbattistini. Martin Haemmig is an adjunct professor at CeTIM at UniBW Munich and Leiden University. Email: martinhaemmig@cetim.org

Leave a comment

Your email address will not be published. Required fields are marked *