What We Believe: The 10 Ps of Manager Selection – Performance
We analyze past performance of our prospective and existing managers in a detailed manner, but past performance in isolation is not predictive of future performance. We instead focus on analyzing past performance to be able to ask more informed questions about a manager’s decision-making and process improvements over time. To a large extent, we see performance as an outcome of the other variables.
We have found that great managers tend to focus more on process than outcomes, which in this case is performance. If the process and approach make sense, the results will come. We are careful about over-extrapolating a period of difficult performance to imply manager deficiency. Based on an article by Warren Buffett, V. Eugene Shahan wrote a famous article, “Are Short-Term Performance and Value Investing Mutually Exclusive?”
Shahan analyzed the performance of seven legendary investors, including Buffett and Charlie Munger. Over long time periods, each investor outperformed his benchmark by 8% to 17% annualized, which is impressive. However, these same investors underperformed the market between 8% and 42% during some of the years, and the worst three-year period of each manager (on a relative cumulative basis) ranged from -4% to -49%. Therefore, if one had selected some of the best investment managers of that time period, one of the managers may have underperformed its benchmark by 49% over three years. That same manager, over a 19-year period, generated excess returns of 16% vs. its benchmark. This example provides support for using past performance wisely.
Our team is focused on finding great managers, and even the best managers in the world may go through difficult periods. Managers who have recently outperformed often hold more expensive securities, while those who have recently underperformed tend to hold cheaper assets, creating risks to investors who fire an underperforming manager and replace it with a top-performing manager. In their 2015 paper, “Timing Poorly: A Guide to Generating Poor Returns While Investing in Successful Strategies,” Jason Hsu, Brett W. Myers, and Ryan Whitby reported that the average mutual fund investor earns 2% less than the mutual funds in which she invests, likely a result of market timing. In the past, we have hired managers that had recent returns that looked average on the surface, but we understood the reasons for the underperformance and had conviction in their process and approach. Thus far, these decisions have been vindicated.
Within this category of performance, we are most focused on assessing the likelihood of long-term investment success, which we define as having conviction that a strategy’s returns will meet our expectations. To do so, we spend a significant amount of time understanding the key drivers of go-forward returns and the risks that may affect those drivers. These drivers often link back to the underlying investment philosophy and strategy.
For example, managers with a tighter valuation discipline recently underperformed more growth-oriented managers for a number of years. Maintaining a deep understanding of a manager’s investment strategy and portfolio allows us to maintain conviction when it is out of favor. We also look at historical performance attribution over time, seeing if strong performance has been broad-based across investments or if only a select number of investments have driven fund performance.
In a Journal of Performance Measurement article, “Just Because We Can Doesn’t Mean We Should,” Dan DiBartolomeo considered performance attribution to be the process of disentangling component portions of the observed returns to draw conclusions about the strengths and weaknesses of the investment process. If a manager has a higher hit rate, or more broad-based performance attribution, it is more likely that the process is indeed repeatable.
Click here to read more about our other 10 Ps for evaluating high-quality investment managers: