Do Private Equity Funds Manipulate Reported Returns?

In our article, Do Private Equity Funds Manipulate Reported Returns? we examine the evidence on performance manipulation by buyout and venture funds. Our study is motivated by the potential incentive for general partners (GPs) of a fund to exaggerate performance to attract limited partners (LPs) to a follow-on fund. We consider if there is evidence consistent with funds manipulating their self-reported net asset values (NAVs) around the time commitments are raised for a next fund. We utilize data provided by Burgiss that includes cash flows and NAV reports for a sample of 2,071 buyout and venture funds. These data are sourced (and cross-verified) from over 200 institutional investors that represent approximately $750 billion in committed capital to private equity. We supplement these data with an independent database of private equity firms provided by StepStone. The StepStone database contains a nearly exhaustive record of institutional private equity fundraising between 1971 and 2016. This combination of data sources allows us to examine the relation between private equity (PE) performance reporting and fundraising success with a high degree of confidence in the data.

In our primary analysis, we conduct several tests to see if fundraising is related to abnormal performance. In addition to a simple event-study methodology, we estimate a probability model of fundraising success. Specifically, we model success as a function of returns (in excess of public equity markets) and cash distributions to investors while controlling for the variation in fundraising environment and fund characteristics. The results of this analysis suggest that exaggerated NAVs are associated with lower probability of raising a follow-on fund. When we examine performance of funds that are unsuccessful at raising a follow-on fund, we observe clear evidence of performance reversals toward the end of fund life that is consistent with investors seeing through attempts of performance manipulation. Furthermore, we show that conservative reporting and credible signaling (via distributing capital back to investors) have, on average, stronger effects on fundraising-odds than market-adjusted performance. The results are similar for both buyout and venture funds.

Our results are distinct from several other studies of private equity interim reporting and subsequent fundraising. Similar to these studies, we find abnormal returns for a typical fund are smaller after the time that fundraisings occur. However, unlike related papers, we show that the abnormal returns remain positive, on average, for successful fundraisers. We also show that our results and interpretations differ because other tests of NAV manipulation with only fund-level data are inherently misspecified as the measurement error in regressions is correlated with the history of cash flows and market returns. This makes such tests prone to spurious results as many explanatory variables of interest are correlated with the fund cash flows and market returns. We consequently show how the methods we utilize for constructing our variables are valid using a variety of statistical tests and simulated data.

While we confirm that relative performance tends to peak during fundraising, we go a step further in examining the response of LPs. Our evidence suggests that a ‘signal-jamming’ equilibrium where all funds exaggerate performance (as proposed by some other studies) is not an accurate characterization of the PE fundraising equilibrium. Rather, the true equilibrium likely features elements of costly signaling as well as conflicts of interests between the current fund’s investors and the follow-on fund’s investors (e.g., with respect to resources for monitoring and nurturing the portfolio investments). We note that the importance of these tensions has been recognized in practice as evidenced by the existence of covenants in fund partnership agreements that address these issues specifically. Overall, our results indicate that overstating interim returns has not been a winning strategy for general partners (GPs) on average. While current fund performance impacts the odds of a successful fundraising, aggressive NAV marks are associated with a lower probability of raising a next fund. Consequently, GPs who are performing well should have an incentive to be truthful, or even conservative, with their unrealized investment valuations.

We also examine how reported changes in NAVs depend on the performance of peer funds (i.e., those of similar vintage and strategy) and identify the determinants of the reporting bias over a fund’s life. We find evidence consistent with “peer-chasing” where top-ranked funds subsequently report lower interim returns and bottom-ranked funds subsequently report higher returns. We attempt to determine whether the adoption of new mark-to-market accounting standards in 2006-08 (FAS 157) has affected the quality of NAV reporting by private equity funds. We find some evidence that regulation has improved accuracy of reported NAVs, however the analysis is confounded by the 2008-09 financial crisis.

Taking the results together, we find that the data are consistent with an equilibrium where overstatement has different costs and benefits to different GPs. In this equilibrium, LPs do not assume the interim performance reports are unbiased. In contrast, LPs undertake costly due diligence and “punish” GPs for the appearance of overstated performance by not providing capital to subsequent funds. Correspondingly, top-performing GPs may try to safeguard their long-term reputation from subsequent bad luck by reporting conservative interim NAVs. They are more likely to do this when it does not jeopardize their high relative performance rank. For underperforming GPs, these long-term reputational concerns appear to be dominated by a short-term concern related to firm survival and possibly a lack of credible ways to signal that valuations are more conservative than those of similarly underperforming funds. Therefore, certain poorly performing funds are incentivized to boost interim NAVs in hopes of fooling LPs though this is not successful on average. We develop a simple theoretical framework that rationalizes our empirical results as well as those of the related papers.

The complete article is available here.