According to standard finance theory, the only risk investors care about is systemic, or market, risk, since they can neutralize idiosyncratic risk through diversification. Venture capitalists (VCs) negotiate deals with entrepreneurs on behalf of investors, so in an ideal world, VCs wouldn’t care about idiosyncratic risk either.
The problem, according to Professors Matthew Rhodes-Kropf and Charles Jones, is that VCs hold large equity stakes in the projects they fund but, unlike investors, can’t diversify away idiosyncratic risk. Moreover, the sequence in which VCs strike deals — first with investors, then with entrepreneurs — means that the terms of the investor-VC contract will influence the price at which VCs are willing to invest in entrepreneurs’ projects. The stage is set for a classic principal-agent problem: the VCs, who have previously signed a contract with investors based on expected portfolio returns, are pursuing their own agenda when they negotiate with entrepreneurs.
Absent the principal-agent problem, VCs would set prices that just compensate investors for market risk; all excess returns would accrue to the entrepreneur, who holds all of the market power in this scenario. But Rhodes-Kropf and Jones posit that the three-way interaction among investors, VCs and entrepreneurs changes the pricing structure of the private equity market. If the total risk of a particular project exceeds the expected portfolio risk, the entrepreneur must compensate the VC for the idiosyncratic risk, although the extra risk is irrelevant from the investors’ viewpoint. Thus, VCs’ returns, net of fees, should be correlated with total risk, not just market risk.
Rhodes-Kropf and Jones tested this hypothesis by examining data on VC and buyout fund returns from 1980 through 1999. They found that funds with higher realized risk do indeed have higher average net returns, as predicted by the researchers’ model. The results of their analysis suggest that the excess returns that accrue to VCs and investors could amount to as much as 7 percent of the total amount invested. So in 1999, a year in which the private equity industry held $175 billion in capital, entrepreneurs lost more than $12 billion as a result of the principal-agent problem. Another drawback from the entrepreneurs’ perspective is that the pricing of idiosyncratic risk can sometimes lead VCs to turn down otherwise profitable investment opportunities.
The principal-agent problem means that VCs with a lower-than-average risk aversion can earn excess returns. VCs can further maximize their returns by investing in a larger number of projects than other VCs. Because VCs play an active role in managing the projects they invest in, they can invest in only a few projects at any given time, which makes them vulnerable to idiosyncratic risk. In the Rhodes-Kropf–Jones model, VCs should compete, not to find better projects — since competition among investors ensures that excess returns go to the entrepreneur — but rather to find more projects.
From the investor’s point of view, the principal-agent problem has two distinct benefits. First, investors share in any excess returns that VCs gain by pricing idiosyncratic risk into their contracts with entrepreneurs. Second, VCs’ susceptibility to idiosyncratic risk makes them work harder at managing the projects they invest in, since the VCs have a large personal stake in each project. This incentive issue is one of the reasons that private equity is and should remain a fragmented industry. If private equity firms were more concentrated — and therefore more diversified — individual VCs would be less motivated to ensure the success of the companies in their portfolios.
Matthew Rhodes-Kropf and Charles Jones are associate professors of finance and economics at Columbia Business School. This research project was partially funded by a fellowship from the Eugene M. Lang Center for Entrepreneurship.