The Hazards of Using IRR to Measure Performance: The Case of Private Equity
23 Pages Posted: 27 Mar 2008 Last revised: 23 Sep 2009
Most of the investments in asset classes such as real estate and private equity (include buyout, mezzanine and venture capital) are made via private partnerships. Measuring performance in these partnerships is important for investment allocation decision as well as for compensation. The main performance measure that is used in practice is the Internal Rate of Return (IRR). However, it is known that the effective Rate of Return (RoR) experienced by investors differs from IRR. This difference means that the incentives of the asset managers partly differ from the objective of the investors.
This article makes two main contributions to the literature on performance evaluation. First, it shows the problems that arise when IRR is used as a performance measure in the context of private partnership investments. It shows that in addition to the well-known pitfalls, IRR leads to a number of issues. First, it provides severely distorted incentives for the timing of cash flows and grouping of funds. Second, it biases upward volatility estimates. Third, at least for venture capital and buyout investments, simple average performance measures are significantly upward biased. Fourth, in a situation where "kick-backs" can happen, the use of IRR provides incentives to alter significantly cash flow amounts. The second contribution is that it describes in details a solution. While it is known that using Modified IRR (labeled MIRR), or, equivalently, Net Present Value, (labeled NPV) largely tackles the well-known pitfalls of IRR, its practical implementation in a private partnership context is not obvious. I show how MIRR can be implemented at the investment level and at the fund level in order to not only tackle the well-known pitfalls of IRR but also provide the right incentives to the fund managers.
Suggested Citation: Suggested Citation