The Limitations of the Use of IRR

Screen Shot 2015-04-05 at 10.53.08 PMWhen Yale University’s endowment reported venture capital returns of 92.7% per annum over the past two decades (Wall Street Journal), it raised some eyebrows in the industry. Yale Chief Investment Officer David Swensen explained that the figure was in fact the 20-year, dollar-weighted IRR of its venture capital portfolio, much different from its 20-year, time-weighted, compounded return of 32.3%.

Why such a wide variance?

In his letter to investors, David Swensen spoke of the limitations of IRR as an accurate reflection of performance, stating, “An anomalous period such as the Internet boom highlights this limitation of the use of IRRs, which is why we use several metrics to analyze the portfolio’s performance.” Yale’s 20-year IRR was inflated by deals completed shortly before the collapse of the technology bubble, leading to the staggering 92.7% return per year.

Dr. Oliver Gottschalg, who has devoted much of his research to the study of IRR as a PE performance benchmark, has found that IRR has inherent methodological biases that skew fund performance, as evidenced in Yale’s venture capital figures. Specifically, IRR disproportionately values a fund’s earlier returns, calculating performance as if all investments were made and realized in one lump sum. In other words, early big wins will inflate returns over the life of a portfolio by assuming those early returns were generated consistently throughout the life of the fund. Furthermore, IRR fails as an indicator of performance persistence and as a comparative benchmark with public markets.

Realizing the need for an accurate indicator of PE outperformance, Dr. Gottschalg developed the PERACS Alpha metric to correct for the biases of traditional private equity performance metrics, such as IRR.