Content area
Full Text
B eta represents an asset's sensitivity to the movements of the broad market and is a measure of the nondiversifiable, systematic risk embedded in the investment. It is defined as the covariance with the market divided by the variance of the market. A beta of less than one indicates that either the asset has a lower volatility than the stock market or the asset is just as volatile but is uncorrelated with the broad market.
Higher-risk investments deserve higher expected returns to compensate for the extra risk, or so theory tells us. Historically, this has not always been the case for U.S. and other developed-market stocks. This "beta anomaly," which is now well established by academics, has started to gain traction with investors, especially in the wake of the Global Financial Crisis. In the first part of this article, we explore the beta anomaly in the academic literature and provide an empirical analysis for stocks in the United States, developed markets, and emerging markets. Our primary finding is that beta is not a strong predictor of expected returns.
We do find beta to be useful for the systematic diversification of a portfolio. The principle of diversification asserts that assets should be spread across many different investments to reduce the risk of loss. The simplest diversification approach is to equal weight the portfolio across all the potential investments.1 This approach leads to the highest diversity and least concentration risk, but it ignores the fact that some of the assets may be highly correlated--not as "different" as one might have hoped. Often investors will group securities into countries and sectors and diversify across these classifications with the idea that securities within the grouping are more similar to one another than those outside the grouping.
In modern portfolio theory, the covariances of the assets are estimated and a mean-variance or minimum-variance optimal solution is calculated. A well-known weakness of optimization is that estimation errors are magnified by the process. Instead, we propose a systematic set of rules, using country and sector classifications and estimates of beta, which result in a diversified portfolio with lower risk. There are still unavoidable errors in the estimation of beta, but in this approach, they are not magnified by the process of...