Payback time

The most frequently used measure of private equity fund performance is the internal rate of return (IRR). Unfortunately, interim IRRs depend significantly on unrealised value. An alternative is the payback period, which measures the time required to break even on an investment without taking unrealised value into account. While this method ignores the time value of money and cash flows occurring after the payback period, it can provide an early indication of whether an investment was worthwhile.

This article examines three questions relating to the payback period for private equity funds. First, what is the private equity failure rate, i.e. the percentage of funds that do not pay back? Second, what is the payback period? Third, what is the impact of diversification: do portfolios of private equity funds pay back faster than single funds?

Our research is based on the Thomson Venture Economics private equity performance modeling dataset, which comprises anonymised cash flow and performance data from Thomson Venture Economics' performance database. All data is as of December 2003.

THE RELATION BETWEEN IRR AND PAYBACK PERIOD
Investors often focus on IRRs when analysing the performance of private equity funds. However, the final IRR of a fund is only known once the fund is liquidated, and interim IRRs can be misleading. For example, while venture capital funds raised in the United States in 1999 reported a pooled average IRR of almost 30 percent at the end of 2000, three years later that pooled average IRR had shrunk to around 15 percent according to Thomson Venture Economics. The following analysis shows that the payback period on the other hand can indicate the quality of a fund's performance well before it is liquidated.

We analyse the relation between IRR and payback period for all liquidated funds in the dataset. This sample contains 509 funds, diversified across the US and Europe (EU), across buyout (including mezzanine and special situations) and venture capital vehicles.

Figure 1 shows the link between IRR and the payback period. It is intuitively clear that short payback periods indicate a strong portfolio or favourable exit conditions or both. They raise the prospect of additional distributions later in a fund's life and hence high long-term IRRs. Conversely, long payback periods typically indicate a weak portfolio and/or an unfavourable exit environment, usually leading to low IRRs unless the fund can generate substantial exits at a later stage. For example, Figure 1 shows that all funds with a final IRR of more than 60 percent have had a payback period of less than six years. In contrast, no fund with a payback period of more than 12 years shows an IRR of more than approximately 15 percent.

A linear regression of the logarithmic IRR versus the logarithmic payback period exhibits a negative slope and an R2 of 0.5. The regression indicates that the payback period is indeed an early indicator of final performance, although the final IRR can still vary widely (for example, the IRRs of funds that returned all the paid-in capital in six years range between 10 and 31 percent).

Table 1: Payback percentage and payback period of liquidated funds as of Q4 2003.

Payback US US EU EU
characteristic venture buyout venture buyout
Payback percentage 77% 80% 56% 75%
Average payback 8.1 6.8 8.7 7.1
period
Median payback 8.0 6.9 9.0 7.3
period

IS CAPITAL BEING RETURNED FASTER?
It is important to consider how the payback characteristics described above might develop over various vintage years. In particular, can one observe trends in either the payback percentage or the payback period? In other words, is the industry paying back faster than it used to?

In order to investigate the behavior of the payback characteristics over time, we analyse a larger sample of 1,057 funds in the dataset that were raised in the time period from 1983 to 1994. It should be noted that when we investigate payback characteristics of nonliquidated funds per vintage year, the results are naturally biased downwards: assuming that the funds in the dataset do not change, every fund that reaches breakeven after December 2003 (the date of our data sample) will increase both the payback percentage (more funds pay back) and the average payback period (payback periods of funds reaching breakeven for any given vintage year are longer than the current average). We choose 1994 as a cut-off point to reduce the bias introduced through funds that are still active but have not yet reached break-even.

Figure 2 shows that the payback percentage of US private equity funds has proven to be surprisingly stable across vintage years. Generally 75 to 80 percent of US buyout funds paid back all drawn capital with only the 1990, 1991 and 1994 vintages showing a failure rate over 40 percent. For US venture, one observes that approximately 70 to 75 percent of all funds paid back all drawn capital. These figures coincide with the average payback percentage of all liquidated funds (see Table 1).

Interestingly, while vintage 1993 and 1994 venture funds might still manage to pay back their drawn capital, many of these funds were apparently not able to benefit from the excellent exit environment of the late 1990s. A comparable study of European buyout and venture funds shows similar results.

As already mentioned, the average payback period of the sample underestimates the final payback period for the vintages 1983 to 1994 (downward bias). Nevertheless, trends and cycles in the average payback period of private equity funds are clearly observable. Figure 3 illustrates what most investors realised when monitoring their portfolios; the 90s vintages paid back much faster than the 80s vintages.

In contrast to the relatively stable payback percentages described above, the average US venture payback period has fallen from above eight years for vintages in the mid-eighties to approximately five years for the 1993 vintage year, as can be seen from Figure 3. Given the relative lack of venture capital exits during 2000- 2003, we expect this trend to reverse for late 1990s vintage venture funds that did not benefit from the TMT bubble. Average payback periods for such funds should rise toward the long-term mean going forward.

For US buyout funds, the average payback period illustrates the impact of economic cycles on the average payback period. The difficult environment during the early 1990s recession lengthened the average payback period for late 1980s vintages. The payback period of the early 1990s vintages returned to its prior levels with the recovery of the economy. We expect late 1990s vintage buyout funds to exhibit payback periods similar to late 1980s vintages due to the difficult economic environment of 2000 to 2002.

Venture funds had longer average payback periods than buyout funds except for vintages 1990 and 1993. These exceptions occurred for different reasons. Vintage 1990 shows the lowest average IRR for all US buyout vintages between 1984 and 1994, whereas vintage 1993 shows a very strong venture performance.

THE BENEFITS OF DIVERSIFICATION
The analysis above describes the payback characteristics of individual private equity funds. But because investors usually diversify their private equity portfolios across many funds, it is important to know how the payback characteristics of broadly diversified portfolios behave. It is commonly known that, in diversified portfolios, successful funds usually more than offset underperforming funds (see also “Cash-flows in a rightskewed world”, published in the February 2004 edition of Private Equity International). Does this skewness alter the payback characteristics of a diversified private equity portfolio relative to a single investment?

We analyse the payback percentage for portfolios of funds, with each portfolio consisting of all funds in the same segment and vintage year (e.g. pooled cash flows for all vintage 1990 US buyout funds). While 20 percent or more of the individual private equity funds failed to pay back, Figure 4 shows that the diversified investor represented by our portfolios reached breakeven for each of the different US segments and vintage years analysed. The same observation holds true for all European buyout vintages between 1983 and 1994. European venture was the only segment where the portfolios analysed had approximately the same failure rate as individual funds (40 percent, corresponding to vintages 1985, 1991, 1993 and 1994). Hence, diversified investors significantly improve their chances of reaching break-even.

While diversifying the portfolio can increase the payback percentage, it does not significantly shorten the payback period. Comparing Figure 4 with Figure 3 shows that the pooled payback period for US buyout and US venture funds is up to a year longer than the average payback period for these two segments. Of course, this phenomenon can be explained by the fact that the pool of funds includes the 20 percent bad performances; for the calculation of the average payback period, these are not included. Nevertheless, calculating the median payback period for all funds (assuming an infinite payback period for the funds that do not break even) results in approximately the same payback period as for the pooled funds observed in Figure 4.

Hence, while diversified investors usually outperform the typical fund, our analysis shows that this outperformance only materialises after the funds have reached break-even.

CONCLUSIONS
Our analysis shows that the payback percentage is surprisingly stable over time at approximately 70 to 80 percent, highlighting the importance of thorough fund manager due diligence. Furthermore, our study stresses the long-term nature of the asset class: primary investors in private equity funds have to wait five to nine years until funds return their drawn capital, although investors may shorten the average payback period of their portfolios through purchases on the secondary market.

Finally, diversification across various funds significantly improves the probability of reaching break-even. However, although successful funds can compensate for underperformers, diversification does not significantly shorten the payback period; surprisingly, diversified investors don't begin to outperform the typical fund until six to eight years after the vintage year in question.

André Frei and Michael Studer are members of Partners Group's quantitative private equity management and research team. They have developed the latest generation of Partners Group's private equity commitment and cash flow models, which are used for commitment steering and cash flow management of Partners Group's private equity mandates and securitised private equity products. They can be contacted at andre.frei@partnersgroup.net (+ 41 (0)41 768 85 79) and michael.studer@partnersgroup. net (+41 (0)41 768 85 45).

This material is provided for educational purposes only. In the event any of the assumptions used in this material do not prove to be true, results are likely to vary substantially from those shown herein. Opinions expressed are current opinions as of the date appearing in this material only.