In the 1990s indexing in the quoted equity markets came of age in two senses. First, many institutional investors chose to move from active to passive management of their portfolios (ie from individual stock picking to pro-rata investment in a basket of stocks forming an index), and second, the performance of investment managers was assessed ever more frequently and intensively against an index-based benchmark.
Many public equity investors now rely on the comfort and cost advantages of indexed investing: by definition returns mirror the market, and passive management fees are far lower than active ones. Investors can also evaluate active public equity managers’ performance in the highly quantitative and generally intellectually defensible manner that comparison against indexed data allows.
Presently, private equity offers very limited, if any, means for investors to choose indexing over active management, and benchmarking managers’ performance versus an appropriate index is at best an exercise in approximation. In theory, given investors’ increasing appetite for indexed public equity investing, the private equity industry should attract more capital from institutions if it offered an alternative indexed product comparable with the public model in cost-effectiveness and reliability. Also, the industry’s reputation for transparency and accuracy would improve if a private equity manager’s past performance could be evaluated readily against a properly constructed benchmark.
How would a private equity index work? Probably a substantial fund-of-funds would have to invest in every private equity fund within a universe defined by factors such as vintage, sector and geography. Achieving an appropriately weighted basket would create a number of practical problems such as access to funds and liquidity.
Quoted private equity investment trusts and funds-of-funds, increasingly a feature of both the UK and Swiss markets, are poor proxies for a true index. Their constituents, even if many in number, do not reflect the entire private equity universe. Also, their returns compared to overall industry performance may be distorted by the use of leverage (in some UK examples) or via intentional over-commitment of available capital to mitigate the ‘J curve’ effect of delayed draw-downs typical of private equity investing (a strategy utilised by some Swiss vehicles).
Similarly, a single, quoted private equity manager on the scale of 3i (that has amongst the largest market share and the most diversified portfolios in Europe) might be viewed by many investors as a quasi-index. Here again, a diversified portfolio does not necessarily give an accurate cross-section of the market. Even if 3i captured 10% of deals in the market it would equate to a quoted manager selecting only 10 of the FTSE100 companies in which to invest (ie it would not be regarded as an index player by any measure). Also, 3i has deliberately restricted its scope with an evolving strategy which has variously embraced specific sectors and deal sizes.
More significantly, even if an appropriately weighted private equity industry basket could be achieved, it is unlikely there would be much investor appetite for this kind of ‘indexed fund-of-funds’ product. Private equity investors do not aspire to average returns. Unlike the efficient market of quoted equities where out-performance is difficult and uncorrelated to past performance, the inefficiencies of private equity allow high quality managers to generate superior returns. Using previous fund performance alone, industry studies have calculated that a new fund has a 70% chance of outperforming the median if the manager’s previous fund was in the top quartile. Combining past performance with an assessment of the “softer” factors of team and strategy can increase this already high likelihood.
Given these problems with indexing, which reflect the nature of the private equity asset class as currently constituted, it seems unlikely that investors seeking low risk, low fee, low maintenance exposure to the breadth of the private equity universe will be satisfied in the foreseeable future.
Private equity investors are instead more concerned with an index as a benchmarking tool. It is therefore necessary to consider the adequacy of the benchmarks currently used to assess private equity managers and the prospects for their improvement.
Unlike the quoted manager, the private equity investor has no screens to consult, nor any centralised sources of data on corporate earnings histories, major announcements, management changes and peer group or sector performance. Instead, since private equity is by definition private, the equivalent information is accessible only by current or prospective investors in individual partnerships, and peer group data is typically incomplete. Even the most active of investors in private equity can acquire directly only a small fraction of the industry’s privately-held, proprietary information, as they do not invest in all partnerships, in all places, in all seasons. Accessing the set of investment opportunities that might compose a benchmark is also difficult, even with perfect information on which funds are available for investment. Those offered by general partners with outstanding track records are often all but closed to new investors, and the private equity operations of banks and insurance companies generally rely on their parent, not third parties, to provide all or the majority of their capital.
At present the most used performance benchmarks, whether for the US or Europe, or for venture capital or buyout funds, rely on data compiled by Venture Economics. This is assembled from a database of net cash flows and quarterly valuations provided by some, but not all, private equity managers. While the Venture Economics data is the best on offer, it has severe limitations – intellectually, and by public equity index standards. There is selection bias – the sample of managers is constructed from those firms that choose to submit their data, leaving out general partners who (for reasons of privacy or, presumably, sub-par performance) do not provide information. It should be noted that British Venture Capital Association (BVCA) data is more comprehensive and defensible but covers the UK only. Common to all benchmark data is reliance on interim valuations of unrealised investments that are frequently calculated on different bases (no global industry standard for the marking-to-market of unsold positions exists) and can lack independence. Consequently, all current benchmark data should be analysed cautiously, and seen more as general guideposts than scientifically sound, conclusive results.
Improving the integrity of these benchmarks awaits a greater commitment by most, if not all, private equity firms to contribute their results to industry databases. In a world of largely private firms, clearly it is not easy for Venture Economics or others to capture all the data, and missing players cannot be compelled to contribute their results.
Yet in these times of capital constraint for private equity firms generally, perhaps general partners can be convinced that a greater commitment to accurate, consistent and comprehensive data (even if it were to produce lower benchmarks) might increase investor confidence in the industry and encourage investors to allocate larger sums to the asset class.
John Barber and Laurence Zage are a director and an associate, respectively, of Helix Associates in London

Topics