Size Matters, Part III: The Academic Research
Over the past two weeks, we’ve showed how market structure favors small funds in Japan and Europe. We hypothesized that assets under management and performance are negatively correlated. Below are the charts that show this relationship in Japan and Europe for backtests of simple value strategies.
Figure 1: Fund Returns vs. Strategy Capacity in Japan (LHS) and Europe (RHS)
Source: Verdad research
Our backtests suggested that we could explain ~90% of the variability in performance with just the single variable of assets under management.
But do these simulations match the real-world performance of actual investment funds?
We looked to academic research to see if the finding held true for mutual funds and hedge funds. This performance finding was replicated in a study of hedge funds, where “hedge fund investors should be wary of hedge fund asset size before investing.” The paper includes this chart showing performance versus size for long/short equity fund managers.
Figure 2: Lagged Returns vs. Fund Size for Long/Short Equity Fund Managers
Note: x-axis, assets in natural log. Source: Mila Getmansky, “The Life Cycle of Hedge Funds: Fund Flows, Size, Competition, and Performance.”
The paper shows that AUM explained 90% of the real-world performance of actual funds. In other words, there was very little difference between what our backtests suggested we should have seen and what the author found in real fund performance numbers.
This is also replicated in the research on mutual fund performance. A group of prominent finance academics studied the mutual fund data from 1962 to 1999 and found a strong relationship between fund size and performance. “We first document that fund returns, both before and after fees and expenses, decline with lagged fund size, even after accounting for various performance benchmarks,” they write. “Whereas a small fund can easily put all of its money in its best ideas, a lack of liquidity forces a large fund to have to invest in its not-so-good ideas and take larger positions per stock than is optimal, thereby eroding performance.”
We argue that the reason small funds produce higher returns is simple. There are massively more small stocks than large stocks, and this dramatically expands the selection opportunity (and valuation spreads) small funds have access to. In the charts below, we can see this dramatic difference in market structure.
Figure 3: Stocks by Size
Source: Capital IQ
Small funds that can invest in small stocks thus have many more investment options.
A fund’s size is one of the most important determinants of returns potential. This makes logical sense and is borne out by the data we’ve shown you, which suggest that returns should deteriorate very quickly with increasing scale, with very little alpha generation above about $200M in AUM in a strategy (note that this number comes from public equity funds, not fixed income, global macro, or private equity where size and volume constraints are different).
Our research—and the academic research cited above—could help explain why so few active managers outperform the market. Capital chases good performance, and so funds that do very well tend to have the opportunity to scale well past the $200M mark. It’s human nature to attribute success to one’s own skill and genius, so few consider the dramatic impact this change in scale has on the opportunity set, the competitive set, and ultimately the ability to generate alpha.
Despite the reams of supporting data, these arguments are not something widely shared in marketing presentations or bank research notes. Given that the most powerful voices in the finance world are the biggest fund managers, this is perhaps not surprising.
How do we understand this silence from the asset management industry about one of the most important predictors of future returns? Professor Chengdong Yin of Purdue University wrote a paper that developed a novel hypothesis for exactly why this is the case. “I show empirically that managers’ compensation increases when fund assets grow, even when diseconomies of scale in fund performance exist,” he writes. “Thus, managers’ compensation is maximized at a much larger fund size than is optimal for fund performance.” Rumor has it that Professor Chengdong Yin is now working on research showing that cars with higher horsepower engines go faster.
Verdad is different. We follow the evidence where it takes us in both our investment process and our business model. And the evidence is very compelling that if alpha generation is our goal, we have to constrain our fund sizes. That’s why we closed our Japan fund to new investors when we reached $50M in assets under management. That’s why we plan to close our Europe fund at a similar level and our global fund at around $125M. We’d rather build a few Ferraris that went very, very fast than spend our lives mass-producing Nissans.