The article titled IRR as performance measure comes under fire describes a report that criticizes IRR as a measure of private equity returns.
This sort of criticism is nothing new. It is true that there is an unrealistic assumption of compound returns underlying the PE IRR calculation, this limitation is widely known by the practitioners.
The real question is: so, what are the implications? It is not common to compare PE returns to say publicly traded equity, or bonds using IRR. The metric is more often used to compare performance of one PE fund to another, and as these IRRs are calculated under the same assumptions, the numbers, while not entirely accurate, are comparable. They provide a practicable tool to compare investment returns across the asset class.
Another problem, which is not mentioned here, is the risk of having multiple mathematically valid IRRs when making the calculation. This can happen when several cash flows are negative, positive, and then negative again. There are ways to address this problem.
In summary, IRR in PE is indeed imperfect, but to state that it is used for "manipulation" is probably a stretch.
■
This sort of criticism is nothing new. It is true that there is an unrealistic assumption of compound returns underlying the PE IRR calculation, this limitation is widely known by the practitioners.
The real question is: so, what are the implications? It is not common to compare PE returns to say publicly traded equity, or bonds using IRR. The metric is more often used to compare performance of one PE fund to another, and as these IRRs are calculated under the same assumptions, the numbers, while not entirely accurate, are comparable. They provide a practicable tool to compare investment returns across the asset class.
Another problem, which is not mentioned here, is the risk of having multiple mathematically valid IRRs when making the calculation. This can happen when several cash flows are negative, positive, and then negative again. There are ways to address this problem.
In summary, IRR in PE is indeed imperfect, but to state that it is used for "manipulation" is probably a stretch.
■
Comments