Experience can be helpful – at work, but also in everyday life. In this post, we showed that prior work experience outside the investment industry helps mutual fund managers. Other studies find that experience as a fund manager may also help. E.g., fund managers who have successfully weathered a recession profit from this experience in the future. If experience is generally beneficial, this makes it easy for investors to pick the best fund manager: just choose according to fund manager tenure.
This hypothesis is at the heart of the 2012 paper “The best mutual fund managers: Testing the impact of experience using a survivorship bias free dataset” by Porter and Trifts. The authors analyze a sample of single fund managers with a track record of at least 10 years. These managers have a lot of experience: Mean tenure is 14.5 years, the longest tenure is 52 years. As performance measures, Porter and Trifts compute marked-adjusted compound annual returns (MACAR), Jensen alpha, Carhart alpha, and nominal compound annual return (NCAR).
Star manager: Peter Lynch
Not surprisingly, Fidelity’s Peter Lynch is one of the best mutual fund managers. In spite of his short-ish tenure (13 years at Fidelity Magellan), his market-adjusted compound annual return is 12.75% and his career market-adjusted cumulative return is about 380.46%. Table 1 shows that Lynch is among the top 4 managers, irrespective of the measure. But he’s the exception rather than the rule: Other highly succesful managers only convince on some performance measures, and not on others.
||Jensen Alpha Rank
||Carhart Alpha Rank
|Lynch, Peter||Fidelity Magellan||1||4||3||1|
|Montgomery, John||Bridgeway Ultra-Small Company||3||3||12||2|
|Antoian, Edward||Delaware Growth Opportunities A||9||17||1||4|
|Royce, Charles M.||Royce Heritage Svc||11||12||17||12|
Table 1: Ranking of fund manager by performance measure. Source: Table 2 of Porter and Trifts (2012).
Is longer fund manager tenure better?
To analyze the impact of fund manager tenure on performance, the authors run a linear regression of performance on tenure and a number of control variables. The main result: longer fund manager tenuresignificantly decreases performance! This result also holds in a number of robustness checks: When analyzing returns for different time frames, or different performance deciles. Only for manager who already underperform (i.e., the ones in the bottom performance percentiles), the relationship between performance and tenure becomes insignificant.
|Regression of performance on tenure||Tenure loading
|Full sample||-0.0007 (-2.70)|
|Top 50%||-0.0013 (-3.80)|
|Top 25%||-0.0027 (-4.31)|
|Top 10%||-0.0046 (-3.21)|
|Bottom 50%||-0.0005 (-1.33)|
|Bottom 25%||-0.0007 (-1.24)|
|Bottom 10%||-0.0028 (-1.93)|
Table 2: Impact of tenure on performance. Source: Table 3 of Porter and Trifts (2012). T-statistics are in parentheses.
So: good managers become worse with time, and bad managers do not become any better.
Why do fund managers slack?
Porter and Trifts argue that the negative relation between performance and fund manager tenure may be due to mean reversion. Top managers work well early in their careers and outperform their peers. But what goes up must come down – at some point, efficient markets catch up with the fund managers. In effect, top-performing managers earn high returns early in their careers. This is followed by poorer returns afterwards.
The take-away: Some skills do not improve with age
In a nutshell, the results of Porter and Trifts imply that managers do not get better at generating performance as they age. Note that this is not evidence against the value of experience. Porter and Trifts are unable to rule out alternative explanations, such as less effort of older managers. However, the implication for investors is clear: Avoid fund managers with longer tenure.