Good vs. Lucky: Assessing Portfolio Manager Performance

There are three intangibles that all good portfolio managers have, says Jacques Lussier, CFA, but factor-based benchmarks are still the best way to distinguish the effective managers from the lucky ones.

Lussier, the CEO of Ipsol Capital, is a proponent of statistically reliable performance because investors often mistake good fortune for alpha.

“I believe that the performances of managers . . . are way more impacted by luck than we would like to admit,” Lussier told Jason Voss, CFA, during a recent Take 15 interview. “Some managers have good performance because they have good process, but some managers have good performance simply because they were lucky or because of a single glimpse of genius never to be repeated again.”

Since portfolio managers’ outcomes are imperfect proxies for their abilities, their processes might be more illuminating to investors. Lussier looks for three qualities in particular when determining whether a manager can generate repeatable alpha. The first is intellectual curiosity. While our understanding of performance drivers and sources of risk is constantly improving, few managers stay abreast of the current literature. Those who do, who are curious, have an advantage over their peers.

Second, managers must be able to distinguish the signal from the noise. According to Lussier, usually no more than 10 factors account for 90% of the variance a manager can explain. Everything else is noise.

Third, the best portfolios are simple and intuitive to investors. Managers embracing the same investment philosophies and levels of risk can build wildly divergent portfolios. But the managers with the least complex portfolios tend to understand the issues best.

Judging managers qualitatively is only necessary because available benchmarks fail to distinguish between luck and alpha. Ideally, investors could evaluate managers using matrices that captured portfolios’ abilities to meet clients’ long-term goals, but no such methods exist. The next best approach is to measure performance against benchmarks that capture market trends and risk tolerance, but these often fall short as well. Factor-model-based benchmarks, which relate systematic sources of returns to the performance of the benchmark, have the most potential to isolate statistically reliable alpha. The industry, however, is reluctant to use more sophisticated benchmarks.

“There’s a resistance to use more evolved factor-based benchmarks because if you do that you would eventually be able to explain way more of the source of performance of managers and a lot of managers would not like their investors to know what their sources of alpha are, because they might find out they’re not exactly alpha,” Lussier said.

Investors need to understand that even good processes do not always lead to good short-term performance, because luck and noise can also be strong determinants of short-term results. Clients do not, however, need to pay exorbitant fees for a manger with good process, according to Lussier.

If you liked this post, don’t forget to subscribe to the Enterprising Investor.

All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.