Friday, February 29, 2008

Bonds better than hedge funds? Not so fast…

Are bonds better than hedge funds? Only if you are fighting the environment of the last few years. The comments below have been making headlines in some of the financial trade publication. These comments from the top Merrill Lynch strategist are from Institutional Investor.

“Whereas hedge funds were an effective diversifying tool in the late 1990s, there is very limited diversifying effect today,” adding that “the efficient frontier” for HF and stocks is now “virtually a straight line.” While hedge funds offer limited diversification benefits these days, says Bernstein, “by adding just a small proportion of bonds to an all stock portfolio, investors now appear able to reduce their risk.” He notes there have been noticeable changes since the tech bubble burst eight years, where investors found comfort in the non-correlation value of hedge funds, art and non-U.S. stocks. During that time, he points out, correlation in these three sectors to the Standard & Poor’s 500 increased dramatically. “Investors should view high quality bonds similarly to their view of hedge funds and non-stocks eight to 10 years ago,” according to Bernstein.

Bonds have always been a good diversifier for stocks. Stocks are a real asset while bonds are a nominal asset. They will perform differently over the business cycle, so the value of stocks or bonds will change with economic condition and with expected inflation. You can have diversification but return can be hurt by holding an asset which is uncorrelated to stocks. Some would say that if you cannot predict the future, then the returns profile does not matter, but any type of analysis based on efficient frontiers implicitly is a bet on a view. The view is that the futures will be the like the past. The expected returns, volatility and correlations used for the efficient frontier will be similar to the past. That assumption may be a good one but a simple look at correlation between stocks and bonds may make this view troubling. A simple correlation between the S&P 500 and 10-year yield changes provides some interesting information on the stability of this correlation. The correlation for the 5 year rolling periods using monthly data from 1990 to the present show a strong negative correlation between stocks and yield changes from 1990 until the Fall of 2001 at which time the correlation moved to positive. The correlation peaked in early 2006 and then moved close to zero. This is not a stable relationship.

This is one of the classic problems of optimization. You use the correlation and returns of the past, so you are saying that the investment scenario of the past will be the best alternative for the future. Could that be the case? Perhaps, but not likely. Let's look at the current market environment for bonds and stocks. We have an environment of stagflation. Slow growth and increasing inflation. Stock and bonds will not be correlated in this environment because stocks are a real asset while bonds are a nominal asset. The correlation will be low or negative. You will get diversification but you may not be better for it. The easing monetary policy will be good for stocks if growth can be achieved but the focus on economic growth will do nothing for your bond portfolio. Is this the type of diversification you want to have?
Let’s focus on the specific charge by Merrill Lynch that hedge funds are not a good diversifier relative to bonds.
Note that this was the period when the cash rate was close to one percent and the yield curve was extremely steep. Many hedge funds try to minimize risk and provide a return slightly over LIBOR. If the focus of hedge funds is on market neutral the returns will be alpha plus the risk free rate of return because the market exposure or beta has been taken out of the equation. The alpha has to be higher than bond yields plus the price changes. Or more precisely, alpha has to be higher than the price appreciation of the bonds and the current yield above the risk free rate of return.

Bonds would do better than the low risk hedge funds which have to move above the hurdle between short and long-term rates. If you do not jump this hurdle then you will underperforms bonds. What may be a more precise explanations is that hedge funds are less attractive during a steep yield curve environments when you can clip bond coupons at low cost. But even this story has some holes. If bonds outperformed hedge funds, it may have been because the time period examined included a recession when bonds will outperform many other asset classes. Bond volatility was also exceedingly low during this period which made them attractive.

What seems to be the case in hedge fund land is that there is a growing diversification of hedge fund types. No one index can describe hedge fund returns; consequently, it is hard to say that hedge funds in general are poor relative to bonds. Market neutral may have been a poor diversifier, but something like global macro or credit-based funds may actually have done well. Of course, the diversity of hedge funds means that the work necessary to understand hedge funds is much greater than that required for bonds. If you don’t want to do the home works or want a simple hedge then buy bonds.
I am not trying to be an apologist for hedge funds. They should defend their returns and risk, but I do want to point out the fallacies of making judgments based on the impact of selective environment.