Markowitz Versus the Talmudic Portfolio Diversification Strategies

Abstract

Although expected utility theory and the classical mean variance diversification theory of Markowitz assert that optimal diversification depends on the joint distribution of returns, investors tend to ignore these well-accepted theoretical approaches in favor of the naïve investment strategy promulgated in the Babylonian Talmud called the 1/3 rule (or the 1/n rule for n assets),which assigns an equal weight to each security in the portfolio. In testing the efficiency of the 1/n rule, the authors find that it outperforms the mean variance rule for individual small portfolios out of sample, but for large portfolios (i.e., institutional investors) the Markowitz strategy is superior. The advantage of the 1/n rule in the out-of-sample analysis is the absence of exposures to estimation errors.