The returns to private equity, measured both as an entire asset class and at the individual fund level, remain somewhat of a mystery for interested investors. Anecdotal stories abound of venture capital investments generating returns over a thousand percent. Yet systematic, accurate and timely measures of private equity performance are not available because of the absence of observable market prices. Venture Economics publishes quarterly aggregate level IRRs for venture capital and buyout firms, such as the median IRR, but does not provide specific fund data. In addition, the funds' IRRs are self-reported and subject to measurement error for the net asset value of unrealized investments.
Further complicating matters is the challenge of devising a statistical measure that reflects the true returns. IRR is the industry standard, but can often give a misleading picture of the actual performance (for much more detail the Encyclopedia Premium Section: Beyond IRR). Ideally, the measure should track the cash flows into and out of a fund, starting from the takedowns of committed capital through to the disbursements after an exit event, while taking into account the time value of money and the level of fund risk.
A number of recent academic studies provide comprehensive and thorough examinations of the returns to private equity. Using extensive databases, including data supplied by Venture Economics, the studies have been largely consistent in their findings. The findings confirm some commonly held beliefs about the return patterns, but also contradict the widely held assumption that the private equity asset class has been a source of abnormally high returns when compared with public equity.
Rates of Return
Returns to private equity investment are reported at the individual fund level, which can then be aggregated to make a broad statement about the private equity asset class. The returns reported in the subsequent studies are based on the cash flows received by the limited partners, net of the management fees and carried interest. Thus the returns measure the actual payoff to limited partners from private equity, which they would most care about, and not the total return generated from private investments.
Individual private equity funds differ in the type of investments they make, varying according to style, region, industry, and stages. For the purpose of measuring returns all funds are categorized into two groups, venture capital and buyout funds. Two separate studies examined venture capital returns using fund data supplied by Venture Economics. Steven Kaplan and Antoinette Schoar analyzed 580 venture capital funds started from 1980 to 1994 that have realized all relevant cash flows.  The average annualized IRR for the funds was 17 percent. This return measure is based on the timing of actual cash takedowns and disbursements, avoiding the problem of the GP subjectively valuing unrealized investments. Similarly, Charles Jones and Matthew Rhodes-Kropf analyzed the returns to 866 funds that were started between 1980 and 1999.  The annualized IRR for this sample was 19.25 percent. The obvious caveat to this reported IRR is that the latter funds have not been fully liquidated, potentially biasing the average IRR lower. A study by Alexander Ljungqvist and Matthew Richardson reported IRRs for 19 funds started between 1981 to 1993 based on explicit cash flow data provided by a limited partner.  The average annual fund IRR for this small sample was 14 percent.
Two aditional observations about the returns to venture capital funds are worth stating. First, there is considerable variance across the returns to individual funds. Kaplan and Schoar found that a quarter of all funds had IRRs of 3 percent of less, while another quarter had IRRs in excess of 22 percent. Second, after making adjustments for fund risk, the payoff to the LPs was not significantly different from the returns generated by the public equity markets. Technically speaking, the fund alphas were not statistically different from zero.
Buyout fund returns were also examined in the three aforementioned studies. Kaplan and Schoar found that the average annual IRR for 166 buyout funds was 19 percent. Jones and Rhodes-Kropf reported an average IRR of 9.67 percent for the 379 buyout funds examined. Finally, the Ljungqvist and Richardson sample of 54 buyout funds had an average IRR of 21.8 percent. Once again, the fund returns exhibited considerable volatility. However, whereas venture fund returns tended to track the public market indexes somewhat closely, buyout fund returns had little correlation with the market indexes.
An important caveat to interpreting the returns to buyout funds is making the proper adjustments for risk. A simple comparison of the fund returns to market indexes like the S&P 500 or Nasdaq suggests that the returns have not been abnormally positive. Such a comparison can be quite misleading. On the one hand, the individual portfolio companies that comprise the funds generally produce consistent, low volatility cash flows necessary for a successful buyout. As a result, these funds have betas that are usually less than one, implying that the risk-adjusted alphas are positive.
The bigger concern when examining buyout returns is failing to account for the effects of leverage. The risk of an investment, and thus the required return, increases in the amount of leverage used to finance the purchase of the asset. Buyout firms finance the acquisition of their portfolio companies by borrowing heavily. Debt-to-total capital ratios are frequently in the 70 to 80 percent range, far above the 30-35 percent for the public market as a whole. After adjusting for the additional leverage-induced risk, the returns to buyout funds appear far less attractive. For example, the 21.8 percent annual IRR found by Ljungqvist and Richardson was about 5 percent higher than the returns to the S&P 500 or Nasdaq. However, a quick calculation demonstrates that if the companies were three-quarters financed with risk-free debt and one-quarter equity, the risk-adjusted alphas would be negative. The debt is obviously not risk-free, but the point remains that risk-adjusted buyout fund returns have not been good.
The general conclusion to be drawn from these studies is that the returns to private equity have been little different from the returns to the public market, and perhaps even lower on a risk-adjusted basis. This surprising assessment runs counter to the conventional wisdom that private equity has been a source of attractive returns relative to the stock market. One possible explanation for this finding is that reported returns measure the payoff to limited partners. The gross returns to private equity, before the 2 percent management fee and 20 percent carried interest is paid to the general partner, should be about 25 percent higher. This premium reflects the services provided by the GP to the portfolio companies and the poor liquidity of the portfolio. Since the LPs hold large diversified portfolios, with private equity constituting only a small fraction of the total, have less need to pay for a liquidity premium and provide few direct services to the portfolio companies, their risk-adjusted returns should not differ significantly from the returns generated by the market.
Further analysis of the fund data has revealed some interesting additional insights into the pattern of returns. Some venture capital funds specialize in seed and start-up financing, others in late stage or mezzanine financing. The different stages are associated with varying levels of risk and should generate distinct payoffs. Sanjiv Das, Murali Jagannathan, and Atulya Sarin examined data on over 52,000 early, expansion, and late stage financing rounds from 1980 to 2000 supplied by VentureXpert.  They compared the value of the investment at the time of the financing with the value at an IPO or acquisition exit. Financing rounds without a successful exit were assumed to be worthless.
They estimated that if it took an average of four years to exit an early stage investment, the expected annual return is about 50 percent. A 3-year cycle from expansion stage investment to exit yields an expected annual return of 26 percent. Finally, if the late stage investment occurs one year prior to the exit, a 12 percent return is expected. These returns are gross of all fees to the GP and overstate the actual returns to the LPs. Nonetheless, the return pattern is consistent with the risky early stage investments earning the highest return, whereas the payoff to late stage investment is little different from the expected return to the S&P 500.
The flow of capital into new funds also has a significant effect on the returns to private equity. Portfolio company values are positively related to the amount of money flowing into funds. This leads to "too much money chasing too few deals". Paul Gompers and Josh Lerner estimated that a doubling of the amount capital inflows into funds was related to a 7-21 percent increase in the valuation of portfolio companies.  This mechanically lowers the returns to private equity by about 15 percent, assuming an unchanged exit value. Ljungqvist and Richardson similarly found that the more money raised by all funds in a given fund's vintage year, the worse the fund's subsequent performance. Kaplan and Schoar reported that older VC funds were largely unaffected by the number of new funds started in a given year, but the returns to new funds were adversely impacted. Buyout returns across all fund ages were affected by the number of new funds. This is consistent with the sensitivity of buyout returns to market timing and herding effects.
The flow of capital into new funds started by new partnerships is highly cyclical. Not surprisingly, Kaplan and Schoar found that both funds and partnerships were more likely to be started when the industry had performed well. These partnerships were less likely to raise a follow-on fund, suggesting that these funds had performed very poorly. The increase of capital flowing into private equity during these flush markets does not go to the top performing funds, but rather to the new funds that produce poor returns. In fact, Kaplan and Schoar found that the decline in industry performance was due primarily to the new funds; the funds of established partnerships had only a small impact on total performance.
The higher returns to the funds of established partnerships reflect persistence in their performance. Kaplan and Schoar documented that partnerships that had a successful fund were more likely to have success with a new fund, and vice-versa for partnerships with a poorly performing fund. In particular, they estimated that a 1 percentage point increase in the current fund IRR was associated with a 54 basis point higher IRR in a subsequent fund. The persistence in the performance is partly attributable to the successful partnerships limiting the size of their new funds, allowing them to select only the best possible investments. Other factors include access to better investments at more favorable financing terms for the higher reputation funds.
The persistence in performance by good fund managers stands in contrast with the mutual fund industry, which exhibits relatively low persistence in managerial performance across time. The belief among LPs that the most important criteria for selecting a fund for possible investment are the reputation and skill of the general partner is supported by the data. If good managers are those who have generated positive excess returns in the past, then this is a pretty good predictor that they will generate high returns in the future. Thus, when GPs tout their high IRRs from past funds, the result should not be discounted as a matter of self-promotion.
 Steve Kaplan and Antoinette Schoar, 2003, "Private Equity Performance: Returns, Persistence and Capital Flows," University of Chicago Graduate School of Business working paper.
 Charles Jones and Matthew Rhodes-Kropf, 2003, "The Price of Diversifiable Risk in Venture Capital and Private Equity," Columbia University Graduate School of Business working paper.
 Alexander Ljungqvist and Matthew Richardson, 2003, "The Cash Flow, Return and Risk Characteristics of Private Equity," NBER working paper 9454.
 Sanjiv Das, Murali Jagannathan, and Atulya Sarin, 2002, "The Private Equity Discount: An Empirical Examination of the Exit of Venture Backed Companies," Santa Clara University working paper.
 Paul Gompers and Josh Lerner, 2000, "Money Chasing Deals? The Impact of Fund Inflows on Private Equity Valuations," Journal of Financial Economics 55, 239-279.