Tracking the IRR metric is not for the faint of heart.
The internal rate of return (IRR) metric is the most controversial investment performance tool in the North American financial industry. British economists John Maynard Keynes and Kenneth Ewart Boulding first introduced it in 1935 and, for the past 82 years, it has been criticized, modified and criticized again by academics in a gargantuan tome of research.
Despite its unresolved intellectual quandaries, the financial industry has turned the IRR metric into the most popular capital budgeting metric in the USA and the second most popular metric in the UK.
Why? Practitioners favour its intuitive appeal and comparable format over what's perceived to be technical deficiencies that create “immaterial distortions in relatively isolated circumstances."
We believe that within the private equity ecosystem, the IRR metric produces distorted results that rob multiple stakeholders (portfolio companies, general partners, limited partners, the public) at three key levels (project, deal and fund level).
Distortion at each level compounds the overall impact and leaves the PE community particularly vulnerable to the metric’s flaws.
The Project Level
PE firms invest in small, private, high-growth companies and research has shown that these are exactly the kinds of companies that most commonly use the IRR metric to make capital budgeting decisions. These companies usually have a stream of high value projects and limited capital. As a result, financial managers utilize the IRR metric to help them rank and select their projects.
Mature companies, on the other hand, more commonly use the net present value (NPV) metric when selecting projects because they generally have ample funds and a limited supply of high value projects. Using IRR to rank and choose between mutually exclusive projects often results in inflated return figures and an incorrect project hierarchy.
In one published example, managers at an industrial company approved several capital projects on the basis of an average IRR of 77%. When these projects were reviewed using a modified internal rate of return (MIRR), average returns fell from 77% to 16%.
Moreover, the #1 ranked project dropped to #10. The MIRR metric corrects only one basic problem inherent to the IRR metric; the fact that project cash flows are often discounted at an unattainably high rate.
The MIRR corrects this by using the company’s cost of capital as the project’s cash flow reinvestment rate, which is a much more accurate assumption. Unfortunately, financial managers are not applying this basic correction consistently and PE deal performance is suffering as a result.
Companies often fund projects that run for a considerable number of years. When managers assess these projects using the IRR, their assessment uses one reinvestment rate for all future cash flows, whether it is the IRR or the MIRR. In reality, though, a company cannot always reinvest future cash flows at the same rate throughout the entire life of the project.
In situations where the reinvestment rate changes over time, the IRR metric cannot be used accurately and, therefore, it cannot assess the value of the company’s project.
The IRR metric’s most frustrating project flaw may be the fact that it often provides managers with a project ranking that is not equivalent to the NPV metric’s ranking. We encounter this problem when we compare mutually exclusive projects that differ in size or scale, or when the timing of each project’s cash flows differs.
These are very common occurrences in capital budgeting and we are advised to trust the NPV ranking over the IRR ranking because it represents the direct value that a project is adding to shareholders.
There are many other project-related IRR flaws but these common ones encourage us to emphasize that all PE investors (general and limited) ensure that their portfolio companies know the problems associated with the IRR metric, because there is a very good chance that many do not.
The Deal Level
When we measure deal performance using the IRR metric, we are assuming that the cash flow generated by that deal can be reinvested at the IRR. As discussed previously, this is a very poor assumption.
A simpler and more accurate performance measure for PE deals is the annualized rate of return (ARR). This metric is calculated by using the deal’s total return multiple and duration. Its significance was recently highlighted in a study of 5,038 realized investments.
It was found that the IRR metric overstated deal performance by more than 10% vs. the ARR metric for 12.8% of the sample (644 deals). This means that if a limited partner is looking at a deal performance of 30% then there is a 1 in 8 chance that the performance is actually less than 27%. Worse than this, in 3.7% of the sample (186 deals), the distortion exceeded 25% of the stated IRR. Limited partners have the most to lose from the IRR metric’s flaws at this stage and should be wary when tracking IRR performance measurements.
The Fund Level
IRR distortions at the fund level are far greater than those at the deal level because of the impact of compounding and the irregular cash flows that we normally see at the fund level.
The IRR and ARR for 422 PE funds were recently compared and it was discovered that the fund IRR deviated by more than 10% for over 30% of the funds (over 126 funds) in the study. For 10% of the funds (42 funds), the IRR deviated by more than 50%.
Distorted fund level performance figures impact a wide range of PE stakeholders, including institutional investors and the general public.
Institutional funds use IRR to track fund performance figures in order to find high-performing general partners. The public’s perception of the PE industry is also at stake, as seen by the recent news coverage that Yale University’s venture capital portfolio received when it reported returns of 92.7% per annum over the past 20 years.
The endowment’s Chief Investment Officer, David Swensen, had to explain that this inflated performance number was an IRR figure and that their 20-year compounded return was actually only 32.3%. The CIO went on to explain the limitations of the IRR metric, which was inflated by deals completed shortly before the collapse of the technology bubble. Despite these distortions, the CFA Institute continues to advise the financial industry to report PE fund performance using the IRR metric in their Global Investment Performance Standards guide.
As discussed above, the IRR metric is riddled with flaws that the academic community has not been able to completely resolve in 82 years. Despite this, the North American financial industry continues to demand a metric that conveniently compares an investment’s rate of return.
The IRR metric is particularly damaging to the PE industry because its distortions occur at multiple levels within the PE ecosystem, compounding its impact.
As a result, PE stakeholders need to be extra vigilant when tracking and reviewing it. We encourage all PE investors to work with their portfolio companies and partners to avoid the aforementioned complications inherent to the metric.
 Carlo Alberto Magni, “Average Internal Rate of Return and Investment Decisions: A New Perspective,” The Engineering Economist (June 2010). Taylor & Francis Group.
 CFA Institute, “Global Investment Performance Standards (GIPS), 2010.” Pearson Learning Solutions.