RSS feedThe previous two posts on alpha and beta being used as a measure of performance may lead you to think that alpha and beta would be a terrific way to choose investments. What’s not to like? Two easy to understand metrics which will allow you to take a look at a manager or a stock, assess its “riskiness” and determine whether or not it produces the return you are hoping for. In fact, alpha and beta were created for just this purpose as part of Modern Portfolio Theory, a theory on optimizing portfolios that won Harry Markowitz a Nobel Prize.
Alpha and beta do, however, have some faults. The issue with using these two metrics to quantify the overall performance of one’s investment portfolio lie in the fact that beta, an undiscriminating measure of volatility, is equated to risk. But, what is risk? For most investors, risk only relates to downside volatility. Heck, we’d love to have a stock which doubles the upside of the stock market. Another deeper issue with alpha and beta is that alpha assumes that volatility and return are correlated. While I haven’t done any study on my own, two simple examples seem to shoot very quick holes in the theory. First, taking a look at speculative stocks like tech stocks over the last 10 years, it would seem that despite having a high beta, they do not necessarily outperform in the long run (Just look at the NASDAQ). Another example would be those of so-called “value” stocks which on average produce lower beta, but often outperform (Tweedy Browne).
Alpha and beta are helpful, but not be all and end all tools of evaluating your portfolio performance. They are backward looking and also period dependent calculations and, as a result, do not usually have much predictive value unless you have a lot of historical data as well as guarantees of a very stable investment strategy. For example, ETFs or asset classes in general. Thus, they are powerful statistics for asset allocation uses and long-term strategies but not necessarily the type of statistics which should drive your near term investment decisions. They are, however, helpful as “quick look” statistics and when used in conjunction with other metrics like Sortino ratio and Sharpe ratio, which I’ll discuss in future posts, they can tell you a lot about the risk-adjusted performance of a given portfolio strategy.
