A fund’s investment performance is measured by its risk-adjusted return (alpha) which is the return of the fund minus the return of the fund’s benchmark. A fund’s perfect benchmark is a portfolio that (1) is fully diversified and (2) has precisely the same exposures to sources of risk (risk factors) in stock market returns as the fund. Forty years of capital markets’ research indicates that there are four risk factors in returns; a general market factor, size (capitalization) and valuation factors and a momentum factor. Hence a fund’s perfect benchmark is a diversified portfolio with exactly same exposures to the four risk factors as the fund.
The large majority of institutional investors use either the single-factor model or style benchmarking to benchmark and select actively managed funds.
A factor model measures a fund’s exposures to the factors in the model and produces a benchmark which is fully diversified and has exactly the same risk exposures to the model’s factors as the fund. The single-factor model contains the general market factor. Therefore, it produces a diversified benchmark which has the same exposure to the general market factor as the fund. However, because it ignores the size, valuation and momentum factors, the benchmark has different size, valuation and momentum exposures than those of the fund. Thus, the single-factor model’s estimate of a fund’s alpha is not accurate enough to satisfactorily measure an actively managed fund’s investment performance; for example, it tends to overestimate the alpha of funds that hold small capitalization and value stocks and underestimate the alphas of funds that hold large capitalization and growth stocks.
In style benchmarking, a fund’s benchmark is selected from a set of existing indices. Specifically, a practitioner selects a benchmark that, in the judgment of the practitioner, has similar exposures to the size