LP outperformance is more than dumb luck: study

Pension giant Massachusetts PRIM has outdistanced many of its peers in the race to achieve top-quartile private equity returns. Are the superior results just a fluke?

A working paper issued recently by the National Bureau of Economic Research suggests that its outperformance stems from skill and not just chance.

The four authors compared the returns achieved by investors with those that would have resulted had they picked funds at random. An analysis suggests “higher and lower skilled investors consistently outperform and underperform,” the authors wrote.

Similarly, a second analysis determined that the more skill demonstrated by an investor, the better it performs — specifically, an implied three-percentage-point increase in IRR for every standard-deviation advance in skill.

Such are the main results of the paper, “Measuring Institutional Investors’ Skill from Their Investments in Private Equity,” co-authored by Daniel Cavagnaro and Yingdi Wang, professors at the Mihaylo College of Business and Economics at California State University, Fullerton; and Berk Sensoy and Michael Weisbach, professors at the Fisher College of Business at Ohio State University.

The question of whether skill matters in determining PE returns for investors is an important one to pose and to study. Salaries and bonuses paid to investment officers — often among the highest paid at their organizations — take a big bite out of investment returns every year.

Imagine if investors could instead put their resource-intensive private equity portfolios on auto-pilot and still perform as well as before. They would save a lot of money that could be put to more productive use.

To be sure, it makes intuitive sense that investor skill would make a difference in PE returns. If ever there was an asset class built to supply peer-beating returns, it would be private equity, given the vast space between top- and bottom-performing firms. Pick nothing but bottom-quartile PE funds, for example, and you’ll wind up with returns that average 19 percentage points less than had you picked nothing but top-quartile funds, according to the paper.

But the authors still needed data and sophisticated analytical tools to demonstrate it. To conduct their analysis the four professors studied investments in 12,043 venture capital and buyout funds by 630 different limited partners — data supplied by Thomson Reuters and Standard & Poor’s Capital IQ.

From 1991 to 2006 each LP had to have made at least four commitments to be included in the sample pool. Using a type of random sampling called bootstrapping, the authors estimated the distribution of performance that would result were LPs to all have the same level of skill.

“The comparison with the bootstrapped distributions suggests that more LPs do consistently well (above median) or consistently poorly (below median) in their selection of private equity funds than what one would expect in the absence of differential skill,” the authors found.

The professors also generated statistics on the standard deviation of LP performance, a measure that shows just how many investors outperform their peers and by how much. No matter how they varied the samples studied — by time period, by fund type, by investor type — they found “more variation in performance [than] one would expect in the absence of differential skill.”

A further series of calculations involved the use of “Monte Carlo” techniques of repeated random sampling. This number-crunching crystallized the results into a ratio: that a three-percentage-point increase in IRR results from every one standard-deviation boost in LP skill for all private equity investments. For venture capital funds, famous for their wide variability in performance, a one-standard-deviation increase in skill results in an even bigger five-percentage-point increase in IRR.

But hold on a minute. Could it be that some investors are taking more risk with their PE investments than others, accounting for the differences?

The authors tested this hypothesis by pooling the LPs into groups, such as endowments and pensions, which would be expected to have similar tastes for risk. “Within each type, we also observe more variation in LP performance than would be expected if investors had no differential skill,” the authors wrote.

They also investigated whether limited fund access, a result of top-performing funds having too few slots to fill available demand, accounts for some LPs outperforming others. Restricting their analysis just to first-time funds, typically open to all, the authors again found that actual investors do either consistently well or poorly compared with how they would have done with a random selection of first-time funds.

“In summary,” the authors concluded, “our results are consistent with skill being an important factor in the performance of institutional investors in their private equity investments. Relative to their peers, some LPs perform consistently well, while some perform consistently poorly.”

Highly paid investment officers, so long as they beat their benchmarks, can breathe a sigh of relief.