Tuesday, April 24, 2007

Veryan Allen: Hedge funds and interest rates

Hedge Fund Blog

23.4.07

The hedge fund industry has benefited in many ways from the low interest rates of the past decade. "Risk free" investments had insufficient returns to present much competition. Funding costs have been advantageous allowing cheaper exploitation of market anomalies. With investment grade bond yields so low and unhedged equity fund returns poor until very recently, there was natural demand for better risk/reward products to fill the vacuum. Unfortunately some popular alternative investment strategies will not do well should we enter a period of interest rate volatility.

While a lot of time is spent on the alpha versus beta debate, little consideration is given to another equally important but often ignored greek letter - rho; the sensitivity to changes in interest rates. Some hedge fund strategies are exposed to negative rho, that is depend on interest rates remaining low and benign. Ironically pension funds have positive rho as rising rates permit them to use a higher number to discount future liabilities. That is a pyrrhic victory because on the asset side of the balance sheet their portfolios of stocks and bonds often drop in such times. It therefore makes sense for investors to identify and diversify with strategies that tend to perform well in a RISING interest rate environment.

Leverage has been cheap for a long time. The infamous yen carry trade makes leverage almost free for those prepared to take on or remain ignorant of the fx risk. In steeper yield curve currencies, many borrow short term to invest in the long and often illiquid end of the curve. It is ages since we had genuine interest rate volatility and even longer since double digit risk free rates. The last time we had real US rate volatility was 1994, a year in which several "hedge funds" that had been considered skilled were revealed as not being skilled. The search for yield has reduced credit spreads enormously yet such spreads will not do well should turbulence return. Private equity and real estate is also very dependent on low borrowing and mortgage costs.

Hedge fund performance is not indexed to inflation. Should interest rates and bond yields rise this will present significant issues. In 1980s many risk free rates round the world were in double figures. Hedge funds in those days HAD to target much higher returns to be worth investing in. A few strategies do benefit from rising rates; for example managed futures and short biased funds have large cash balances. Put option prices drop as rates rise so some hedging costs reduce. But some hedge fund strategies are negatively exposed to rising rates, costlier credit, steeper yield curve twists and shifts and reduced and less reliable interest rate differentials between currencies.

Some fund of hedge funds target returns like LIBOR+500bp and not an absolute return. This sounds institutional but ignores what risk free means. In Japan that implies a pathetic 5.5% a year, in New Zealand that means 12.5%. Surely an absolute return strategy necessitates an absolute return target. It is not as though most hedge fund returns will magically ratchet up as rates rise. It would be better to aim for 10% or 15% depending on the risk profile than some relative benchmark. Some "market neutral" hedge funds tout their alleged dollar, market cap, beta and delta neutrality but are not rho neutral. And is a single digit target return even worth allocating to when you can get 5% without risk in the US and more elsewhere in some cases?

Hedge fund performance measures also have negative rho. Sharpe, Sortino and Information ratios change significantly depending on what risk free rate is selected. All go lower as the risk free rate in the numerator rises. Yield curves are pretty flat but oftentimes choosing the lowest point on the curve makes a significant difference. The same hedge fund performance will have very different reward to risk ratios depending on the rate and base currency. In the early 80s there were very good hedge funds around with negative Sharpe ratios because risk free rates were in the high teens.

Given the strong earnings and growth round the world, rising demand and commodity prices, it seems difficult to believe that an overdue bout of inflation or at least precautionary and preemptive interest rate hikes will not occur, sometime. While of course interest rates have fluctuated somewhat, it is a long time since true uncertainty and turbulence hit the markets. Most monthly economic numbers still show inflation to be relatively benign so perhaps globalization and "smarter" overheated growth fighting has permanently locked the inflation genie away. But sooner or later there will likely come a time when hikes are not a "well-discounted 0.25%" but become "surprise 0.50% and even 1% rises".

There is a good number of central bankers, CDO and CPDO structurers, credit derivatives traders, private equity spreadsheet junkies and negative rho positioned hedge fund managers that have little experience of such conditions. Parsing Bernanke and Fukui comments or ECB and Bank of England minutes will be of renewed importance should rapid and untelegraphed anti-inflationary moves become necessary. The bond and credit markets may not even wait for them.

Negative rho can be as dangerous as negative gamma but many investors don't even realise how short rho they are in their total portfolios. Rho has been safely ignored for a long time by many market participants. We entered 2007 with much commentary on the "Fed was done" with easing "later in the year" nonsense while now the directional probability seems much less clear and many countries are already in the process of renewed tightening. Stress tests for interest rate sensitivity are always necessary. The BEST hedge funds do this but NOT all hedge funds, sadly.

No comments:

Lunch is for wimps

Lunch is for wimps
It's not a question of enough, pal. It's a zero sum game, somebody wins, somebody loses. Money itself isn't lost or made, it's simply transferred from one perception to another.