Marcin Kacperczyk, a Professor of Finance at the University of British Columbia in Vancouver, has spent the better part of his career searching for ways to identify skillful active managers among the universe of mutual funds. He believes he has now found a couple of key metrics that predict future performance, using publicly available data. We spoke with Kacperczyk about his research, and how advisors can identify funds that will have superior performance results.
By way of background, Kacperczyk grew up in Poland, and received a Masters degree from the Warsaw School of Economics, followed by a PhD in Finance from the University of Michigan. With a research focus on active investment management, he has studied the performance of institutional investors, but the mutual fund analysis has been the key to his findings. He asks two basic questions, "Does it make sense to invest in active managers?" and "Are investors naïve in their belief that it is too difficult to identify skillful managers, or is there money to be made in mutual fund investing?" Kacperczyk acknowledges that the body of research demonstrates that active managers do not, on the average, beat passive, indexed portfolios. "But the average does not mean all," Kacperczyk notes. He believes that some managers are persistently good and some persistently bad, and that there are ways to identify which managers fall into each groups.
Kacperczyk's first study, in 2004, co-authored with Clemens Sialm (U. Texas) and Lu Zheng (UC Irvine), looked at the level of diversification in mutual funds. Many investors, including Warren Buffett, claim that too much diversification is costly to the active manager – it is too hard to gain an edge vis-á-vis an index and it is too costly to track a large number of positions. Therefore, more concentrated funds should demonstrate better results, in terms of risk-adjusted performance. Kacperczyk defined a measure that reflects the level of concentration in an active fund, capturing the difference in sector allocations between the fund and its benchmark, market portfolio. Larger differences imply more concentration. The data ranked the performance of funds relative to their level of concentration, and showed that the 10% of funds that were most concentrated outperformed the 10% that were most diversified by 2% per year. Kacperczyk offers this study as evidence that a key characteristic of skillful active managers is that they hold more concentrated portfolios, reflecting the fact that they have superior information about a smaller number of positions in which they are willing to place larger bets.
Next, Kacperczyk, and his co-author Amit Seru (U. Chicago), looked more closely at the question of whether skillful managers posses superior information. "Ideally, we want to know whether managers posses valuable private information," Kacperczyk notes, "but this is impossible to observe." Instead, Kacperczyk believes that investors can infer the amount of private information a manager possesses by their reliance on public information. If a manager relies heavily on public information, then he must lack private information, and vice versa. Kacperczyk chose sell-side analyst recommendations as the source of public information, and measured the degree to which mutual fund holdings reacted to upgrades and downgrades issued by analysts. If a mutual fund increased or de-creased its position in a security following an analyst's upgrade or downgrade, this indicates reliance on public information. Kacperczyk defined a metric, the Reliance on Public Information (RPI), to measure this effect, and showed that a greater reliance on public information correlated with poor performance. The better performing funds rely less on public information, implying that they have superior private information.
The study in which Kacperczyk places the greatest investment value is his most recent, entitled "Unobserved Actions of Mutual Funds," also co-authored with Sialm and Zheng. The methodology behind this study is stunningly simple. Kacperczyk began with the hypothesis that skillful active managers must add value through transactions in their portfolio, and devised a way to measure the effect of these transactions. Today, virtually all mutual funds publicly disclose their holdings on a quarterly basis. Each quarter, Kacperczyk measured the actual performance of the fund, and compared this to the hypothetical performance if the fund had simply held it positions at the end of the prior quarter. The difference between the former and the latter is defined as the Return Gap; a positive Return Gap indicates an active manager is adding value, and a negative Return Gap implies the manager is destroying value.
Positive Return Gaps are highly predictive of future fund performance, Kacperczyk claims. Kacperczyk insured that these results are not influenced by other aspects of the funds, by adjusting for common risk factors, as well as for market capitalization, growth or value orientation, and momentum. Furthermore, only one quarter’s worth of Return Gap calculations are necessary; advisors do not need to look at a history of quarterly Return Gap calculations. Return Gaps are also highly persistent, in that a positive Return Gap will be predictive of positive future performance over a number of periods into the future. "Of the three studies I have done on this topic, this one has the strongest predictive results," notes Kacperczyk. "If managers do something bad, such as poorly managing expenses or transaction costs, it will show up in future performance."
In general, it does not make sense to measure the Return Gap for an index fund. But Kacperczyk notes that some index funds, including some Vanguard funds, engage in securities lending, and in these cases the Return Gap can measure the effectiveness of such strategies.
The investment industry has yet to adopt the Return Gap, or any of the other metrics Kacperczyk has studied, as a tool for fund measurement or selection. But Kacperczyk believes it is a matter of time before his data finds commercial applications.
What does Kacperczyk's research say about the debate between active and passive management? "There is money to be made in active management, but it is not easy," says Kacperczyk. "I am a strong proponent of active management, even though the data says that active managers on the average do not beat the index. The key is finding ways to consistently identify the good active managers."
Source: Advisor Perspectives