Is the vaunted “illiquidity premium” partially an illusion?

illusion2The illiquidity of alternative investments often used to explain their risk-adjusted out performance (see previous coverage on this topic  here).  But what if some of that risk adjusted out performance was actually the result of so-called “return smoothing”?  Critics of hedge funds suggest that hedge fund managers have an incentive to mis-report returns in order to make themselves look good.   Are the critics right?  And if so, is the “illiquidity premium” really just a result of the mis-valuation of illiquid investments?

These are the questions tackled by a new paper written by Gavin Cassar of Wharton and Joseph Gerakos of the University of Chicago’s Booth School of Business.  Cassar and Gerakos use a database of hedge fund due diligence reports (the same one used in this study) to measure the serial correlation of hedge fund returns across funds with different valuation policies.

The study contains a table that you might find interesting if you ever have to conduct due diligence on a hedge fund.  Reproduced below, it shows the percentage of funds in each hedge fund strategy that use each of several different NAV calculation sources:

nav1

Note that with the exception of equity market neutral, long/short equity and global macro, most hedge fund classes do not use exchange values.  Fixed income, multi-strat and emerging markets funds tend to mark to model, funds of funds calculate NAVs themselves and most – especially convert arb – rely on dealers to provide pricing data.

If you think that smaller, less “institutional” hedge funds tend to do their own valuation, think again.  The table below shows that the more positions in the fund, the more likely the manager does their own valuations.  Ironically, these funds are apparently so “institutional,” they obviously feel comfortable performing this task internally.  (Funds that value only once a year seem to also prefer to do it internally).

nav2

So which pricing sources yield the lowest serial correlation in returns?  If you guess “exchange,” you’re right.  It takes a lot of effort to game the system if you’re relying on exchange prices to set your NAV – plus, there is no judgment involved.

The highest level of return smoothing was found in funds where the manager performs their own self-valuations or where they use a dealer’s price.  You’d be excused for wondering how pricing information from a third party (a dealer) could yield serial correlations.  The authors say that this finding suggests some managers “dealer shop” for the most favorable NAVs.

Another way to view the evidence of return smoothing is to look at the distribution of monthly returns from funds using each pricing source.  (Regular readers may remember a study, cited by Cassan and Gerakos, that found a dearth of “barely negative” returns).

Funds that rely on dealer quotes seem to have the amazing ability to rarely lose 10 bps, even though making 10 bps is one of the most likely outcomes.

nav3Not even self-valued funds displayed such a propensity to minimize tiny losses.  In fact, when managers actually valued their own funds, the return distribution looked a lot more normal.

nav4a

“Reputable” Service Providers

After the Madoff affair, it’s easier than ever to argue that a reputable accounting firm should lead to a more objective picture of a fund’s affairs.  But this study finds that the presence of reputable service providers doesn’t necessarily mean that returns are devoid of any smoothing.  Using Alpha Magazine’s rankings of auditors and administrators as a proxy for “reputable,” Cassan and Gerakos concluded:

“…we do not find lower levels of smoothing in the returns of funds that use more reputable service providers and that exclude the manager from setting and reporting NAV.”

So as long as the manager isn’t calculating NAVs in isolation, smoothing appears not to be as much of a problem.

Whither the Illiquidity Premium?

Institutional investors with which we have spoken tell us that sometimes you have to hold your nose and invest with a manger that does their own valuations – for the simple reason that their strategy is esoteric or highly illiquid.  Cassan and Gerakos conclude that when they do conduct their own self-valuations, they aren’t all a bunch of mis-reporting liars…

“…asset illiquidity is the major factor that drives the anomalous properties of self-reported hedge fund returns…”

Whew! Illiquidity lives.

Be Sociable, Share!

Leave A Reply

← ETF hedge funds: The Infinite Monkey Theorem When skittish hedge fund investors and lenders become a problem for everyone →