The authors’ working hypothesis was that the “turnover of subadvisors provides sharper tests of any underlying board and sponsor monitoring because these data are heavily weighted toward involuntary turnover. Departures by in-house managers are more likely to be voluntary because good performance gives in-house managers better opportunities, such as joining hedge funds.” About 15% of mutual funds employ subadvisors.

Their study, which was on actively managed mutual funds, covered the period from 1995 through 2009 and contains a large number of departures: 11,405 for internal managers and 695 for subadvisors. The following is a summary of Kostovetsky and Warner’s findings:

• There’s a dramatically stronger inverse relationship between subadvisor departures and lagged returns.

• The turnover-lagged performance relationship is that return performance lagged up to five years predicts departures. Firings don’t tend to occur because of short-term (one- or two-year) underperformance.

• There’s no evidence of improvement in return performance related to departures. Manager turnover has no effect on future return performance, regardless of the horizon examined.

• Fund flow improvements are associated with the departure of poor past performers. Turnover is associated with economically significant increases in future flow for poor past performers, and especially for subadvised funds. This suggests that investors pay attention not only to past returns, but to management changes as well. This evidence is also consistent with rational action on the part of fund boards, so long as manager identity is relevant to investors, even though turnover doesn’t lead to improved return performance. Boards appear to closely monitor performance in part because investors chase returns and expect an improvement if management of a poorly performing fund is changed.

The authors further found that management turnover at actively managed funds, on average, was almost 19% a year. Approximately 75% of the turnover was due to departures, while the remaining exits were due to fund closures. After three years of managing a fund, 47% of managers in the worst-performing quintile leave, while 36% of the fund managers in the best-performing quintile leave. After six years, the corresponding figures are 73% for the worst performers and 54% for the best performers. This highlights another risk of investing in actively managed funds. There’s a significant likelihood that top-performing managers will depart. However, the authors did find that the percentage of top-performing managers that departed subadvisors was much lower, although it was still at 20% after six years.

The authors also found that the average four-factor alphas (after expenses) over the previous year were -1.6%, which is just reinforcement that active management is a loser’s game. It might also be of interest that more than 70% of fund managers had advanced degrees (such as an MBA or a Ph.D.) and that the average SAT score was 1250. In other words, they don’t fail because they aren’t smart. Instead, it’s that the competition is too tough.

Kostovetsky and Warner provide evidence demonstrating not only that there is high turnover at actively managed mutual funds, but also that — while the turnover does improve fund flows for the fund sponsor, indicating that investors not only believe there will be an improvement in performance, but that alpha will be generated — there is no evidence that reality matches investor expectations. The lack of improved performance shouldn’t come as a surprise to anyone familiar with the literature, which demonstrates that there’s a lack of persistence in outperformance in mutual fund returns beyond the randomly expected.