Does Blame Predict Performance?
As an econometrician and a fund-of-funds portfolio manager, I spend much time researching quantifiable metrics to help me identify managers who can outperform consistently. There is, in fact, a rich body of literature exploring different manager selection criteria. Academic papers have considered portfolio manager attributes, such as tenure, the CFA designation, advanced degrees, and even SAT scores; they have also examined fund characteristics, such as portfolio turnover, expense ratios, and assets under management. Practitioners, especially investment consultants, have additionally focused on more nuanced and qualitative elements such as investment philosophy, compensation scheme, turnover of key professionals, ownership structure, and succession planning.
Ironically, perhaps, most people have given up on the hope that past positive alpha can predict future outperformance with any reliability.1 Some might even go as far as asserting that manager outperformance is mean-reverting due to cyclicality in styles and “luck.”
Some of the above-mentioned attributes may provide very incremental information on the true quality of the manager. However, most econometricians, asset owners, and investment consultants confess (although not all publicly) that effective methods for picking top quartile performers remain elusive. As one of my friends at a large Middle Eastern sovereign wealth fund famously proclaimed, “We are convinced that managers who can consistently deliver alpha exist. We are, however, also convinced that we do not know how to find them.” Perhaps, then, the science of manager selection really is about winning what Charley Ellis calls “the loser’s game.”
As my high school basketball coach was fond of reminding me, “If you can’t improve your shooting mechanics, you can still improve your field goal percentage by not forcing bad shots.” His advice is equally relevant to the investment industry: To improve your odds of outperforming, screen out the negative alpha managers. If an investor focuses on eliminating the lower quality managers from his selection universe, the odds for achieving outperformance, in the long run, would be much improved—even if hiring the best managers from the screened short list is still a crapshoot.
So, how does one win in a loser’s game? In this article, I argue that you can significantly improve your odds by employing simple rules for identifying and eliminating underperforming managers.
Predicting Long-Term Underperformance
First of all, we already know quite a bit about the predictors of poor long-term investment performance. High portfolio turnover, high expense ratios, and low active weights (Cremers and Petajisto, 2009; Sebastian, 2013) are quantifiable metrics that tend to predict underperformance in the long run. Qualitatively, anecdotes suggest that high turnover in the professional ranks, lack of organizational alignment due to poor compensation design, or deficient inter-generational transition planning also hurt long-term investment results. Both finance academics and investment consultants have been working hard on identifying quantitative and qualitative attributes which might predict underperformance.
However, given the reported negative median and average underperformance for active managers, investors will have to work much harder in screening out low quality funds and managers just to get the expected alpha for the screened universe up to zero. For example, the average mutual fund underperforms by 1.6% net of fees; screening out high fee funds merely brings the average active return above ?1%. Additionally, many low quality investment organizations are savvy enough to respond to RFPs and interviews carefully so as to tick most of the boxes on a consultant’s due diligence report. The cynical perspective is that asset managers are far more adept at solving the challenge of gathering assets from asset owners than solving the challenge of producing alpha for asset owners.
Bill Gross’ departure is a wakeup call for active management fans.
High interest rates in emerging markets are paid, not earned, for currency-hedged ETFs like HEEM and DBEM.
There's something very personal about choosing the right robot to manage your investments.
The PIMCO investigation has turned a spotlight on bond pricing.