The reason we wanted to investigate the beta exhibited during 2008 – and especially during the fall crash – is that investors and traders use beta as a measurement of how risky an asset is relative to the market, with the goal of diversifying or hedging away that beta-adjusted risk. Our concern is whether beta – which is often derived by comparing a set of returns over a long time frame - was an effective predictor of risk during the turmoil of late 2008.
This is relevant to investors with large stock portfolios, and it is also relevant to options traders: a portfolio of stock or option positions that can be examined on a beta-adjusted basis is much easier to hedge, since hedges can be placed on some one or handful of underlying assets or indexes, rather than on each underlying asset represented in the portfolio. For instance, a portfolio that is long 100 deltas/shares in each of AAPL, GE, X, WFC, WMT, and CPB will be ill-served by simply shorting 600 deltas/shares in SPY, the ETF that tracks the S&P 500. Intuitively, while some of those stocks may closely track the S&P 500 on a given day, most of them will not, which means a simple 1:1 delta hedge will not provide adequate protection. Instead, we would be better served by finding the beta of each underlying and then hedging our exposure on that basis. (Clearly, hedging six positions in each underlying is not burdensome; but what if your portfolio has exposure to sixty different names, or six hundred? You’ll need either automated algorithmic software, several clerks, or the ability to hedge in a smaller subset of underlying assets.)
So we posed two questions:
- How did stock betas change during 2008 versus the prior five year period? Did the old maxim hold true during the October crash (i.e., did all betas “go to one”)?
- Would a beta-adjusted portfolio hedge have worked as expected during the crash of 2008?
We addressed the first topic by calculating the beta on the six stocks listed above as well as on some other popular tickers (RUT, NDX, XLF, SSO) for three periods: monthly returns during 1997-2007, daily returns during 2008, and daily returns during the latter part of 2008. We will post our results in the next part of this series.
Photo courtesy Flickr user mrjoro.