The Dimensions of Active Management

The Myth of the Absolute-Return Investor

“…Realized alpha will always have a substantial random component, but for the skillful manager, the mean of the distribution will have risen. We discuss these issues in Waring and Siegel (2003); we don't mean to be glib in skipping over some technical details. Our comments apply in the context of the single-index model, the market model, or the capital asset pricing model, with the caveat that we allow for a positive expected value for alpha under the conditions just stated.…”

“…The reason lies in the proven lack of efficiency of portfolios that are subject to the longonly constraint: For a given level of skill, portfolios constrained to be long only deliver only a fraction of the alpha of an unconstrained or market-neutral portfolio. This principle is explained fully in Grinold andKahn (2000a, 2000b) and Clarke, de Silva, and Thorley (2002), and it is summarized in Waring and Siegel (2003).…”

“…The original version of Sharpe's capital asset pricing model presumed that the expected alpha would be zero, but for those of us who believe in active management, this alpha could conditionally have a nonzero positive expected value if the two conditions of inefficiency and skill discussed in Waring and Siegel (2003) were satisfied. Positive realized alphas might well be had simply through luck, but positive expected alphas require special skill, skill that is sufficient to beat the rest of the skillful crowd that plays the markets.…”

Performance Measurement via Random Portfolios

“…As in [Waring and Siegel, 2003] the investor is (or should be) doing a portfolio optimization. The investor is selecting weights for a variety of active and passive funds.…”

“…Thus investors who use active managers need to try to evaluate the quality of the active managers available to them. [Waring and Siegel, 2003] argue that the investor is faced with a portfolio optimization problem where the assets are the fund managers.…”