Study finds many hedge funds simply hold back liquidity to power returns

With hedge fund redemption gates now being shut with increasing frequency, it has become vogue to question the ethics, if not the very legality, of “not giving investors their money back.”  But holding onto investors’ capital – assuming such a possibility is included in the fund’s O.M. – is actually neither bad nor good on its own.  In fact, this illiquidity might actually have a fair market price according to a Swiss study.

After all, when a bank takes a term deposit, investors don’t get their money back for several years.  Yet few people complain that these guaranteed investment instruments won’t give investors their money back.  Instead, they simply demand (and receive) a higher return from these illiquid instruments.

A hedge fund O.M. with an optional redemption gate clause is analogous to a hybrid between a term deposit and a more liquid demand deposit.  The price an investor should be willing to pay for that fund (or put another way, the return they should expect from it) should therefore reflect the possibility of the “demand deposit” essentially becoming a “term deposit”.

Although the likelihood of the manager shutting a redemption gate is a function of the size of investors’ redemption requests, the illiquidity of many hedge fund portfolios makes them more exposed to possible gate closures than more liquid traditional portfolios.

The question many investors are now asking themselves is “so how much of my hedge fund’s return was simply a fair compensation for that illiquidity?” Several AllAboutAlpha.com guest contributors have wondered the same thing (Ranjan Bhaduri, AlphaMetrix; Pierre Laroche, Innocap; Konstantine Danilov, Bank of America).

Now Rajna Gibson and Songtao Wang of the Swiss Finance Institute join them by exploring how much of hedge funds’ alpha is actually just a fair compensation for taking on this illiquidity risk.

In a paper released earlier this month, they find that a lot of hedge fund alpha is actually just an illiquidity premium paid by investors to tie-up their capital.

Observes the duo:

“…dissociating their market timing and selectivity skills from their pure risk bearing compensations remains quite a challenge for most investors and fund advisors.

“Typically, continuous-time arbitrage or equilibrium asset pricing models ignore liquidity since the cost and time required to transfer financial wealth into cash is assumed to be nil and since trading is often ruled out by most equilibrium asset pricing models. Yet, in practice, financial crises (such as in Asia or in Russia during the nineties), the debacle of the LTCM hedge fund or the most recent Subprime credit crisis suggest that at times of tight credit and market conditions, liquidity can decline and even temporarily dry out.”

They develop a “liquidity factor” and add it to two commonly-cited factor models used to describe hedge fund returns (the ubiquitous Carhart Four-Factor Model and a model developed by Hasanhodzic and Lo that uses 6 factors to explain hedge fund returns – see related post).

Then they compare the alphas of these strategies with the alphas produced after the liquidity factor is introduced into the equation.  What they discover may come as bad news to some hedge fund investors.  They conclude that most of the supposed alpha produced by Convertible Arbitrage, Event Driven, Emerging Markets, Fund of Funds, Global Macro and Long/Short Equity hedge funds are just a fair illiquidity premium.

However, introducing the new liquidity factor to the returns of Equity Market Neutral, Fixed Income Arbitrage, Managed Futures and Multi Strategy funds didn’t seem to change the resulting alpha.  In other words, these strategy alphas could not be explained as a plain old illiquidity premium.

So why do alphas from Equity Market Neutral, Fixed Income Arb, Managed Futures and Multi-Strat funds defy explanation?  The authors suggest that a lack of (long) equity exposure is the reason:

“…Fixed Income Arbitrage and Multi-Strategy are not equity-oriented, and thus it is possible that
a liquidity risk factor constructed using stock data does not significantly explain the performance of portfolio strategies formed with these hedge fund styles.”

The recent illiquidity of hedge fund portfolios has been blamed for their poor returns.  So does Gibson and Wang’s model simply reflect these extreme liquidity problems (and other financial crises such as LTCM and the tech bubble) or do the researchers’ conclusions hold up during normal times too?

To find this out, the authors removed these anomalies and re-ran the analysis.  The same result was observed – the illiquidity premium still explained most of the alpha in the hedge fund strategies identified above.  As the authors point out:

“…financial crises were not the main liquidity events that affected the performance of hedge fund portfolios in our sample. Frequent yet small liquidity shocks seem to matter as well and to be an important systematic risk factor affecting most hedge funds portfolios’ returns.”

“…many hedge funds act as liquidity providers and should be as such compensated for their exposure to liquidity risk both during and outside of financial crises…”

While not directly addressed by this study, redemption gates are a reflection of the relatively illiquid nature of underlying securities in many hedge fund portfolios.  As Gibson and Wang find out, that illiquid nature has a fair price that is – for some strategies in aggregate – enough to swamp what might otherwise have been identified as alpha.

The good news: locked-in hedge fund investors were already being compensated for that illiquidity.  The bad news: that compensation was a passive risk premium, not actual alpha.

Be Sociable, Share!

One Comment

  1. Peter Urbani
    November 27, 2008 at 2:13 am

    For more on the topic of the impact of (i)Liquidty on risk please see the discussion and associated papers and spreadsheets on the discussion forum for Carol Alexander’s new four volume set at

    http://www.carolalexander.org/phpBB/viewtopic.php?f=6&t=2

    This incudes both the normal (‘gaussian’) and Cornish Fisher (‘modified’) VaR and CVaR numbers for simple returns at different holding horizons. Whilst it does not directly answer the question of liquidity premium it does give you a better idea of what the drawdown at risk or liquidity VaR you might be exposed to over a longer time horizon, also for the non-normal case, than simply scaling by the square root f time rule.


Leave A Reply

← Hedge fund industry enters time-warp in January 1970, pops out virtually unchanged in 2008 Report: Second half of '08 just a warm-up for more "slashing" at asset managers →