Alternative investments, including private equity, hedge funds, and real estate, compose a meaningful portion of most institutional investor portfolios. There is currently a lack of available data for private equity funds because they are exempt from disclosure requirements. The author overcomes this transparency issue by creating a hand-collected dataset of private equity fund–level cash flows from over 500 Swiss pension funds. In contrast, many commercial private equity databases suffer from several biases, including survivorship bias and self-selection bias, because of self-reported data; thus, the author’s dataset can be considered more reliable.

Using his unique dataset and such performance measures as the internal rate of return (IRR) and the public market equivalent (PME), the author compares returns of private equity funds with those of publicly traded benchmarks. Based on cash outflows, cash inflows, and the total return of the S&P 500 Index, the PME is a performance measure used to assess whether a fund has outperformed the S&P 500 (public market benchmark). The author provides a cash-flow-based performance analysis of Swiss pension funds’ private equity fund investments; the results allow practitioners (such as pension fund managers) to benchmark their own fund selection decisions to the industry. The enhanced transparency of the private equity asset class should also lead to improved investment decisions.

Calculating returns on the basis of each fund’s historical cash flows, the author finds that the cross-sectional mean IRR is 9% for all direct private equity funds. In addition, the mean IRR is calculated as 18% for buyout funds, –4% for venture funds, and 12% for funds of funds. As measured by the PME, the private equity funds outperformed the broader public stock market by between 9% and 19% over their lifetimes.

The author uses a confidential dataset of private equity fund–level cash flows from 531 Swiss pension funds, representing $3.8 billion (45%) of the total $8.4 billion in private equity investments as measured by net asset value (NAV) as of 31 December 2012. The author’s dataset was hand-collected from Swiss pension funds and from providers of private equity funds of funds based in Switzerland. The author met with all data providers for which he concluded that survivorship bias and self-selection bias were unlikely.

First, he determines cash investments and cash inflows that a pension fund would have had if it had invested in each of the private equity funds during the 20-year research period (January 1993–December 2012). In the event that a fund was fully liquidated before the end of the research period, all the cash flows are known and thus no subjective estimates are required. If the fund was not liquidated by December 2012, a subjective residual NAV estimate is used as the final cash inflow. To minimize the potential bias of the NAV estimate, the author does not include any private equity fund whose NAV exceeds 30% of the total cash invested in the fund.

From the cash flows, the author calculates each private equity fund’s performance using three return measures. The first measure is the IRR of the stream of cash invested and cash received. The second measure is the total value to paid-in (TVPI), which is simply total cash distributions received (including the final NAV) divided by total cash invested. The third measure is the PME, which is the ratio of the cumulative discounted value of cash received divided by the discounted value of cash invested (using the total return on a market index as the discount rate).

From these results, the author concludes that private equity funds generally outperform market indexes of similar risk. The private equity funds’ IRRs are higher than those of the indexes over similar periods and their PME ratios are higher than 1, using a range of returns on market indexes as discount rates.