Performance Persistence in Institutional Investment Management

Abstract

Using new, survivorship-bias free data, we examine performance persistence in 6,260 institutional portfolios managed by 1,475 investment management firms between 1991 and 2004. Unlike retail mutual funds, persistence in winner domestic equity portfolios is significant and economically large for up to one year. Loser portfolios, conversely, do not persist. International portfolios exhibit similar patterns, and fixed income portfolios persist up to three years. We find that better-performing portfolios offer performance-based fees and most-favored-nation clauses more often, but also charge higher fees. The magnitude of fees is insufficient to eliminate excess returns. Top performers draw an influx of assets from plan sponsors, and in the year following such inflows, alphas sharply decline.