The revised GIPS for Asset Owners guidance statement hasn’t yet been finalized, but we can pretty much expect that the time-weighted composite return requirement will remain.
At a recent NYSSA (New York Society of Security Analysts) GIPS program, that I had the pleasure to speak at, I asked what might be considered an unfair or trick question:
What question does the GIPS for Asset Owners composite return answer?
Why unfair? Because there is no answer. That is, there is no question that the returns provide an answer to.
Mixing private and public equities, plus much more …
There is quite a bit of evidence that asset owners frequently include private equities in their portfolios: as much as 20% or more. In addition, they’ll have public equities, bonds, real estate, infrastructure, hedge funds, etc. Quite a plethora of asset classes.
How are the returns on private equities calculated? Using the Internal Rate of Return. That is, money-weighting, or a money-weighted rate of return (MWRR).
And public equities? Here we expect to find time-weighting, or time-weighted rates of return (TWRR).
The GIPS for Asset Owners guidance requires composite returns to be time-weighted.
Time-weighting eliminates or reduces the effect of cash flows, while money-weighting takes them into consideration.
If GIPS required asset owners to report the IRR, then this would answer the question “how did the plan do?” But a TWRR of a composite that holds both private equities (whose returns should be measured using IRR) and public (whose returns are measured using TWRR, to let us know how the manager(s) did, and IRR to tell us how the client or fund did) will give us an answer to an unknown question; well, actually, a non-existent question.
What should the GIPS for Asset Owners guidance do?
GIPS for Asset Owners recommends that managers also report the IRR: this is a great thing. I’d prefer that it was a requirement, but a recommendation is a good start. This answer the question “how did the fund do?” Or, “how did ‘we’ do?,” where “we” is the asset owner.
But what if we want to know how the collection of money managers did?
It is my humble opinion that the return should be an asset-weighted combination of MWRR and TWRR. This isn’t the first time I’ve recommended this, but I can’t recall if I’ve ever written it down before. Nor, do I recall if I’ve actually done the math.
Years ago (make that, decades (two) ago), this topic came up with one of our clients. And, at that time we concluded that an asset-weighted return would be the best approach to judge the collection of managers. However, I didn’t write it down then, and, as just noted, unsure I’ve written it down since. But, write down I am, starting here!
Oh, and “what should the GIPS for Asset Owners guidance do?” Require an asset-weighted composite return, which blends the TWRR and MWRR together.
Questions to ask
In general, returns provide answers to two questions:
- How did the asset owner do?
- How did the manager(s) do?
More than five decades ago, Peter Dietz concluded that the proper approach to answering question #2 is to eliminate or reduce the impact of cash flows. He was thinking of only cases where the client controls the cash flows. This idea was immediately adopted by the Bank Administration Institute (BAI: who brought us the term “time-weighting”) and the Investment Council Association of America (ICAA: now the Investment Advisers Association (IAA)).
Peter (as well as many others) recognized that to answer the first question, the IRR (MWRR) was the best approach. Sadly, this view somehow got lost over the intervening decades, but has had a resurgence of late.
As noted above, if we want to answer the first question, the composite return needs to be derived using the IRR. And while some might think the TWRR will answer the second question, they would be, of course, wrong, if the composite/fund includes assets (e.g., private equities) for which MWRR is used to judge the managers (who control the flows). Thus, the TWRR employment doesn’t work.
What to do? Again, use an asset-weighted approach, asset-weighting the underlying assets by the returns used.
More to come on this topic!
Now that I’ve gotten started, I can’t (and won’t) stop: in the June newsletter, I will expand upon this topic. And, expect to see an article (hopefully in The Journal of Performance Measurement(r)) later this year or in early 2018.