Countless disclaimers tell us that “past performance is no guide to future performance”, but in mutual funds, many investors think it is. And not only do they use past performance as a guide, their actions are backed by several seminal academic studies from the 1990s – notably M. Carhart’s 1997 paper “On Persistence in Mutual Fund Performance”. However, Yale University academics J. Choi and K. Zhao recently updated Carhart’s work in their new paper “Did Mutual Fund Return Persistence Persist?” Their answer is no.
This article is only available to Macro Hive subscribers. Sign-up to receive world-class macro analysis with a daily curated newsletter, podcast, original content from award-winning researchers, cross market strategy, equity insights, trade ideas, crypto flow frameworks, academic paper summaries, explanation and analysis of market-moving events, community investor chat room, and more.
Countless disclaimers tell us that “past performance is no guide to future performance”, but in mutual funds, many investors think it is. And not only do they use past performance as a guide, their actions are backed by several seminal academic studies from the 1990s – notably M. Carhart’s 1997 paper “On Persistence in Mutual Fund Performance”. However, Yale University academics J. Choi and K. Zhao recently updated Carhart’s work in their new paper “Did Mutual Fund Return Persistence Persist?” Their answer is no.
The Study
The authors of the new paper first replicate the results of the 1997 paper, which covered the period 1963-1994, and then update it to include the 1994-2018 period. The study is structured to track the performance of mutual funds based on their past one-year returns. More precisely, the total universe of mutual funds is split into ten buckets, or deciles. The top decile contains the top performing funds and the bottom contains the worst. The rankings are performed every January and are rebalanced to equal weights every month. This provides us with ten time-series of the monthly returns of the ten decile portfolios.
They then compare (regress) the performance of each decile to the overall market’s excess return (one-factor alpha) and also to the market’s excess return, size, value, and momentum styles (four-factor model). This allows them to determine whether the mutual funds statistically outperform the market and whether they outperform common style factors.
The Results
For the original sample period of 1963-93, Choi and Zhao find identical results as Carhart: there is evidence of past performance leading to future outperformance. This is especially the case when looking at the returns of the funds and also their performance compared to the market (one factor). The spread between the top decile and bottom decile performance was found to be around 9% per annum. And, there was a similar outperformance versus the market. However, there was less evidence of any outperformance relative to all four factors.
Updating the returns for the 1994-2018 period, the authors find no statistically significant outperformance in the spread between the top and bottom deciles. Compared to the four-factor model, the alpha is actually negative. Moreover, the top decile showed negative alpha compared to either the one-factor or four-factor models. That is, top performing funds would later underperform the factors.
The authors delve into the data further and find that the period 1963-1979 was the big driver of the statistical significance of past performance. Looking at just 1980 onwards, the alphas all become insignificant. Another way of looking at this to track the rolling ten-year window on the one-factor alpha. Using the long top decile – short bottom decile funds, they find the one-factor alpha trends down and becomes negative in 2012 (Chart 1).
They also find that past winners saw their returns become less correlated with favourable styles (such as momentum or value), which were also less profitable by themselves, and their style-adjusted performance worsened.
Figure 1: Ten-year Rolling One-factor Alphas
Source: “Did Mutual Fund Return Persistence Persist?”- J. Choi and K. Zhao
Bottom Line
There appears to be little evidence that mutual funds with strong past returns will deliver superior future returns. This contradicts past academic work. One reason appears to be that mutual funds are not able to switch to the best style factors and also style factors themselves have performed less well of late. The golden era of past performance guiding future performance appears to be the 1960s and 1970s.
(The commentary contained in the above article does not constitute an offer or a solicitation, or a recommendation to implement or liquidate an investment or to carry out any other transaction. It should not be used as a basis for any investment decision or other decision. Any investment decision should be based on appropriate professional advice specific to your needs.)