A survey carried out on behalf of the UK’s Investment Property Forum (IPF) in 2017* found that the analytical, metric-of-choice for the private equity real estate industry in the UK was the internal rate of return, or IRR. In fact, 93% of respondents to the study’s survey cited the IRR in the top three of the analytical metrics used, with 66% ranking it as the primary measure used. Unfortunately, according to the very same study,

“…finance theory comes up with a clear preferred approach…discounted cash flow (DCF) analysis and the use of Net Present Value (NPV) to determine the financial benefits of projects”, stating that the “IRR is never a better investment criterion than the NPV”.

Not to worry, real estate is “different” after all. But maybe it is worth recapping some of the flaws of the IRR.

1 - The IRR assumes that you can re-invest cashflows paid out at the IRR rate

A relatively recent global real estate IRR index released by INREV / ANREV / NCREIF provides global fund performance information, one of the goals of enhancing transparency in the industry. Such is the transparency of real estate however, that the performance data are only available to members of those organisations. Luckily there is a snapshot available to non-members which allows a good guess at at least some of the headline data. This snapshot shows that the median fund launched in 2013 had delivered an IRR of around 18.4% (with an inter-quartile range of around 25%…gulp). This assumes then that it is feasible to reinvest the cashflows at the IRR, i.e. 18.4%. However, the index itself shows that 2013 was the peak of fund performance and that the average performance of funds of subsequent (and prior) vintages was lower than those in 2013. Reinvestment (in those real estate funds) at the average IRR of 18.4% would not have been possible.

Unless of course you could select better performing funds than on average. Just as we know that 95% of MBA students think they are above their class average, or that 94% of professors think they are above average, examples of overconfidence in one’s own abilities, it would be reasonable to assume that around 100% of those who select funds would say they are better at it than average. Why, after all, would you spend time selecting funds if you thought you were worse at it than average and what kind of private equity real estate marketing pitch has ever been successful by claiming a lack of selection skill?

2 - The IRR contains no measure of risk

Being free from any concept of risk, the IRR is simply a number. As Bollinger (2020)** points out, despite the finance industry’s love of the phrase ‘risk-adjusted returns’, those with fiduciary responsibility may want to ask themselves the question, when they weigh up risk and return using the IRR, whether they are actually fulfilling their fiduciary responsibility at all since, by using the IRR, risk is not assessed at all.

“PE managers making large fees often find themselves at the intersection of luck and financial ignorance on behalf of PE investors. Such PE investors evidently think that an 8% IRR is some sort of magical hurdle, and above that, they have no problem splitting the returns with the investment manager, regardless of how much risk is taken to clear the hurdle. This situation flies in the face of finance orthodoxy and legal fiduciary duty, but this is how the PE industry largely operates.”

3 - The NPV of the cashflows, discounted at the IRR, is always zero

If investment is a positive NPV activity (you expect to get back more money than you put in adjusting for the risk associated with each cashflow), then the IRR, by definition is of no use in determining whether an investment is being made, since the NPV is always zero. This is no accident, it is the definition of IRR to goal seek the rate which which delivers an NPV of zero. Those who can recall calculating IRRs without a spreadsheet will remember a tedious process of iteration using a calculator and a book of paper for all the calculations, guessing the IRR and performing numerous DCF calculations until the NPV was close enough to zero that further iterations would just consume even more valuable time.

Maybe if IRRs still needed to be calculated in this way, people might use NPV instead as you only need to make one set of calculations. But first, and with the time you save not making all those calculations to get to an IRR, you will need to think about an appropriate discount rate for the bundle of risks you are entertaining. That relies on some understanding of risk and, consequently, requires some knowledge of finance theory and is potentially very hard work, but it would end up helping you work out whether you are making an investment or not and fulfil fiduciary duty by considering risk: it is as if the ideal metric already exists.

All in all then, the IRR doesn’t seem like the ideal metric to be using: =IRR(investment:fiduciary,human).

* IPF (2017) “An Investigation of Hurdle Rates in the Real Estate Investment Process”

** Bollinger, M (2020) “IRR: Snake Oil by any other Name”, Journal of Investing