Maximum Cumulative Underperformance: A New Metric for Active Performance Management

Abstract

We propose a new metric to monitor an active manager’s performance: ‘maximum cumulative underperformance’ (MaxCU). MaxCU measures the maximum cumulative underperformance an actively managed fund can experience over a given timeframe. The link between MaxCU and its key drivers—the fund’s tracking error, information ratio, investment horizon—varies dramatically by the prevailing return environment. Holding the manager skill and tracking error constant, MaxCU can be multiple times greater in a bull market than in a bear market. We uncover the complex mapping between the drivers and MaxCU with the help of Boosted Regression Trees. MaxCU insights from this paper are useful for investors allocating to active managers. Ex ante, MaxCU can help inform the optimal size of allocation to an active strategy. Ex post, MaxCU can help inform the relevant performance threshold for manager retention/replacement.

Keywords: Underperformance, Tracking Error, Drawdown, Manager Selection, Active Allocation