Mean Reversion in Financial Markets: The Case for a Contrarian Approach to Investing
In a prior installment of Articles on Wealth Management Topics, we discussed value vs. growth stocks and I mentioned in passing that "... as is often the case in investing, returns revert to the mean." More simply put, returns in financial markets are often cyclical. That is, multi-year periods during which asset classes or investing styles or mutual fund sectors succeed in generating above-average returns are usually followed by multi-year periods of disappointing returns, and vice versa. Since this market behavior is a key tenet on which our contrarian investment philosophy rests, let's explore the academic research supporting it and the ramifications for successful long-term investing.
The Cyclicality of Stock Market Returns
In a recent research paper, NYU finance professor Aswath Damodaran examined several methods to estimate the magnitude of future stock market returns. Among the methods he studied, the worst was to extrapolate future returns from past returns. In fact, he found that future 10-year stock market return premiums are negatively correlated to historical stock market premiums. In other words, the better the relative performance of the stock market in recent years, the worse we should expect it to perform in future years, and vice versa. Stock market returns are cyclical.
Mean Reversion of Asset Class Returns
In a recent interview, Rob Arnott, CEO of quantitative investing firm Research Affiliates, discussed how asset class returns also tend to revert to the mean. Looking at each major asset class during the last 40-plus years, he and his colleagues found that whenever an asset class suffers an extraordinarily severe annual decline, it recoups those losses within 5 years 86% of the time. In his words:
"We believe markets seek fair value, which means mean reversion can be a powerful force in the capital markets. Contra-trading against the market's most extreme bets is, we believe, eventually rewarded, though it can take time."
The Case for Contrarian Mutual Fund Picks
Three researchers, Bradford Cornell, Jason Hsu, and David Nanigian, recently published a paper entitled, "The Harm in Selecting Funds That Have Recently Outperformed". In it, they compare the historical performances of two hypothetical investors, one who invests solely in mutual funds whose returns have been in the top 10% during the prior 3 years and one who invests solely in mutual funds whose returns have been in the bottom 10% during the prior 3 years. During the 22-year study period, buying the "loser" mutual funds, i.e. the recent poor performers, was a more profitable strategy than buying the "winner" mutual funds, i.e. the recent stellar performers, by more than 2% per year. Mutual fund performance reverts to the mean.
And yet, to their detriment, mutual fund investors continue to chase performance, to pour money into "hot" funds whose returns have shone recently and to redeem their investments in recent underperformers. Knowing this, Morningstar has for years published an annual "Buy the Unloved" study. The contrarian strategy they suggest within is to buy those categories of stock mutual funds which have recently experienced the greatest investment outflows. The implicit assumption in this strategy is that mutual fund investors will sell most aggressively those funds that have recently performed the worst. Morningstar has found that buying these "Unloved" funds instead has been a market-beating strategy through the years.
To some extent, Morningstar itself is responsible for the counterproductive, performance-chasing, investing habits of mutual fund investors. Investment flows into and out of a given mutual fund are highly correlated with that fund's Morningstar "star-rating". This despite the following admission from Morningstar's president of fund research Don Phillips in an article he wrote a few years ago:
"The Morningstar rating for funds is a grade on past performance. Period. No one at Morningstar ever claimed that the stars have predictive power or ever ran an ad telling investors to follow the stars to riches."
I wonder how many users of Morningstar fund ratings are aware: (1) that those ratings are based solely on the past performance of a mutual fund, and (2) that the recent past performance of a mutual fund is a terrible indicator of its future performance.
And one would think that those financial advisors who earn their livings by selling mutual funds must also bear a lot of the blame for the performance-chasing behavior of mutual fund investors. Can you imagine a broker making the following sales pitch?
Mr. and Mrs. Unsuspecting Investor, do I have an investment idea for you. This mutual fund has had terrible returns for the past three years and shareholders have been running from it like scalded dogs. How much can I put you down for?
Of course you can't imagine it. But research on mean reversion shows that this fund may well be a sounder investment recommendation than a fund representing the hottest investment fad of the moment, which is much more likely to be the one being peddled by brokers.
Even institutional plan sponsors - those whose very job it is to vet investment management firms in a professional and fiduciary fashion for pension plans, endowments and foundations - are susceptible to chasing performance, to being unaware of the cyclicality of investing styles. Amit Goyal and Sunil Wahal studied the hiring and firing decisions of 3700 plan sponsors over the course of ten years. They found that:
- plan sponsors hire investment managers after large positive excess returns in the years just prior to hiring, only to find that this return-chasing behavior does not deliver positive excess returns thereafter,
- plan sponsors terminate investment managers after underperformance, but that the excess returns of these managers after being fired are frequently positive, and
- if plan sponsors had stayed with the fired investment managers, their returns would have been larger than those actually delivered by the newly hired managers.
The Ramifications for Successful Portfolio Management
A variety of evidence suggests that the returns derived from stock markets, from asset classes, from investing styles, and from mutual fund sectors, revert to the mean over multi-year periods. This cyclicality in the financial markets has several implications for successful investing.
At a basic level, it should be clear that chasing performance is counterproductive. Picking funds or managers or investments based on their returns in recent years is more apt to be disappointing than rewarding.
At a deeper level, the evidence argues for a disciplined, consistent, patient and somewhat counterintuitive approach to portfolio management. Rebalancing to a static target asset allocation, for example, should add value over the long haul through the enforced process of continually harvesting "winning" investments to reinvest in recent "losers". Perhaps more importantly, the cyclicality of market returns also warns us of the dangers of becoming complacent now after years of positive returns.
About the Author
Paul Winter, MBA, CFA, CFP® is a Fee-Only financial advisor and fiduciary in Salt Lake City, UT. His independent wealth management firm, Five Seasons Financial Planning, provides professional portfolio management and objective financial planning services to individuals and families, and to their related entities including trusts, estates, charitable organizations, and small businesses.