Assessing the Market Timing Performance of Managed Portfolios

A number of techniques have been proposed to measure portfolio performance and to distinguish between performance due to forecasting security-specific returns and performance due to forecasting market-wide events. We show theoretically and empirically that it is possible to construct portfolios that show artificial timing ability when no true timing ability exists. In particular, investing in options or levered securities will show spurious market timing. These types of securities will also induce the negative correlation between measured selectivity and timing ability found by others. We suggest specification tests to help distinguish between spurious and true timing ability. In addition, the tests can be used to distinguish between different models of the manager's reaction function.