The performance of hedge funds in different VIX “states”

There is a common belief among many that hedge funds thrive in times of market chaos.  Hedge funds provide the market with much needed liquidity during times of crisis.  Whether you call this market making, or contrarian investing, hedge funds have the flexibility to trade with desperate sellers (“fear”) on one day and desperate buyers (“greed”)  the next – making them a useful balancing mechanism in financial markets.

But some types of hedge funds clearly benefit from market volatility while others don’t.  In today’s installment of our “Alternative Viewpoints” column by a member of the CAIA Association, we feature an interesting study by Mikhail Munenzon of the relationship between the VIX and hedge fund returns.  What you are about to read may surprise you or it may confirm your intuition.  But regardless, it will likely help to clarify your understanding of the relationship between the volatility and hedge fund returns.

Special to by: Mikhail Munenzon, CFA, CAIA

Do hedge funds like volatility?  This question arises whenever volatility spikes (see related post).  So it has probably crossed the minds of at least a few hedge fund investors over the past month.

In a statistical analysis published in this recent paper, I found that different VIX states result in very different risk adjusted performance for several hedge fund strategies.  In addition, I found that correlations among strategies are unstable and non-linear, leading to highly concentrated diversification benefits during times of market stress.  I also demonstrate that at certain states, correlations are very high between traditional and alternative investment strategies and performance characteristics are very similar (a phenomenon experienced firsthand by many hedge fund investors in May 2010).  Finally, I found that superior, long term performance of such strategies relative to traditional asset classes is not due to higher returns in good times, but rather better preservation of capital in bad times.

The study used hedge fund data from the Center for International Securities and Derivatives Markets (CISDM) at the University of Massachusetts, along with data for the following traditional asset classes: equities – SP 500 Total Return Index (SPX); bonds – JPM Morgan Aggregate Bond Total Return Index (JPMAGG); commodities – SP GSCI Commodities Index (GSCI); real estate – FTSE EPRA/NAREIT US Total Return Index (NAREIT).

I found significant deviations from normality in various VIX states or ranges (e.g., VIX under 20, between 20-25, 25-30, 30-35, 35-40 and over 40) and for the full sample, which are not fully captured by traditional risk metrics such as volatility (see Table 1 – click on graphic below to open Excel file).

Convertible arbitrage and event driven strategies, for example, have large tails, as described by their large negative skewness and large kurtosis, making volatility an incomplete measure of potential losses. On the other hand, CTAs have positive skewness and relatively small kurtosis, suggesting that losses are likely to be limited and returns are likely to surprise above the mean.

But intriguingly, I also found that different VIX states result in very different risk adjusted performance for all strategies, which suggests that those investors looking to find consistent, absolute returns in hedge funds are likely to be disappointed.

The figure below shows the cumulative returns of various hedge fund strategy indexes from Jan. 1, 2002 to the end of January this year…

But the cumulative return of these strategies on days when the VIX was below 20 looked like this…

These returns aren’t nearly as high.  (Note this chart ends in 2007 since the VIX spent most of 2008 and 2009 above 20.)

But when I examined only those days when the VIX was above 40, this was the result (note: various other VIX ranges are included in my research paper.)

As you can see, only CTA, macro and equity market neutral strategies provide cumulatively positive returns across all states.  However, emerging market and equity long/short strategies, which experience significant cumulative declines at stress points, outperform strongly when VIX is under 20 while CTA, macro and equity market neutral lag meaningfully in the same state. Moreover, only CTA responds well to a rising VIX, improving its percentage of positive months in a particular VIX range (see Table 3 below – click to view Excel file).

Generally, returns for alternative investment strategies decline significantly as VIX and start turning negative once VIX rises 30. This result is similar to the behavior of traditional asset classes, though their performance starts to suffer once VIX crosses above .

I also found that correlations among strategies are unstable and non-linear, leading to highly concentrated diversification benefits at the times of market stress, which a broad set of exposures is likely to negate (see Table 2 below – click image to view Excel file).

With almost perfect correlation at extremes, most alternative strategies are much more highly correlated with each other and traditional asset classes than the full sample may suggest. For instance, at high VIX levels, CTA, macro and equity market neutral can play an important role in protecting capital as markets crash due to their negative correlation with traditional assets and other alternative strategies. This focus on protecting capital is crucial to the long term success of alternative strategies.  So superior, long term performance of such asset classes is not due to higher returns in good times, but rather much better preservation of capital in bad times.

The diversity of behavior of alternative investment strategies among themselves and with traditional asset classes suggests that such strategies may add significant value in portfolio construction and risk management but each strategy’s role and value may vary significantly depending on an investor’s goals, risk tolerance and investment environment scenarios (particularly the market volatility).

Unfortunately, we don’t know which volatility states will dominate in the future or how long they may last.   But this analysis suggests that a greater awareness of the current investment environment, its implications for risk adjusted performance and commensurately flexible investment policies would help investors produce more consistent results.

Be Sociable, Share!

One Comment

  1. CF
    June 22, 2010 at 10:12 pm

    Regarding the VIX returns represented in Table 1, I do not see how you came up with the VIX returns with VIX >40. I did the same exercise, and while my arithmetic mean for the entire period came out almost identical, I found that the VIX arithmetic mean return when VIX began the period at 40 or above was -10%. I used both calendar months – of which there were only 8 when VIX began the period >=40 – and daily data, where I used 22 trading days as a month (yes..oversampling the data, but was curious to see what it would show.) Both methods showed an arithmetic mean return of about -10%.

    Again, I used the VIX starting figure as the ‘state’ variable, as I don’t believe the VIX ending figure would be useful because of lookback issues. Can you please explain how you calculated the VIX returns, and in turn, if there are any knock on effects as to how the returns for the strategies were calculated. Thank you. CF

Leave A Reply

← Memo to Hedge Fund Marketers: Wait, there is hope... From Renaissance to Rates of Return: "Art Beta" →