How well do experts perform in timing individual stock trades? In their May 2007 paper entitled “Individual Security Timing Ability and Fund Manager Performance”, David Gallagher, Andrew Ross and Peter Swan define individual security timing ability as the proportion of potential returns obtained by a trader over the holding period and measure this ability for a set of active Australian equity fund managers. Using a unique database of daily trades and monthly portfolio holdings for 30 fund managers from January 1996 through December 2001, they conclude that:
- Across approximately 10,000 timing measurements, fund managers on average do not exhibit ability to time individual stock trades (see chart below), but some managers do consistently display superior trade timing.
- At the fund level, the 150 aggregate trade timing scores (30 funds across five years) are approximately normally distributed with a mean close to zero, again indicating that managers do not on average possess timing ability for individual stock trades.
- Funds with aggregate trade timing scores above (below) the median outperform (underperform) the S&P/ASX 300 Index by an average of 8.62% (-0.48%) annually. Fund-level timing scores are largely uncorrelated with both raw fund return and fund stock selection ability.
- Fund manager timing scores are higher for small-capitalization stocks, volatile stocks and stocks for which fund positions are large relative to market weights. These results suggest that stock entry and exit timing ability may relate to degree of informational advantage.
- Timing scores do not depend on fund size or trading frequency. However, growth fund managers exhibit better trade timing than managers for other fund styles.
The following figure, taken from the paper, depicts the distribution of of the roughly 10,000 timing scores derived from the entire sample. These scores are approximately normally distributed, with a mean near zero. The dispersion of the distribution reflects the large differences in actual trade outcomes when measured against both the best and worst potential returns that could have been achieved over the holding periods. A mean near zero indicates that fund managers on average do not successfully time their stock trades.
In summary, Australian stock fund managers on average exhibit no timing ability in executing trades of individual stocks, but some managers do consistently outperform others.
The authors attribute the distribution of trade timing scores and fund-level timing grades to differences in the abilities of the fund managers involved. It seems inconclusive whether the differences derive from ability or luck.
