The world has changed after the Great Recession, but some of the fundamental tenets of investing still apply and probably always will. Exhibit A is mean reversion, a theory that says that prices fluctuate around an average of historical prices or some other measure of value. True, you can’t count on mean reversion to apply like clockwork, particularly in the short run. In fact, there’s no way to know if mean reversion will apply at all in the years and decades ahead. Par for the course in finance: everything’s always open for debate. But a small library of empirical analysis suggests that we should take mean reversion seriously. In other words, buy low and sell high is a worthwhile investment strategy, even if the details are messy.
For example, buying stocks when trailing dividends are historically high, and selling when yields are relatively low, has delivered above-average returns vs. buying and holding a basket of stocks over long periods of time. Will this work in the future? No one really knows the answer, but there’s a lot of history that suggests we should be reluctant to dismiss the idea entirely.
Consider Professor John Cochrane’s November 2008 op-ed in The Wall Street Journal, in which he briefly reviews the history of current dividend yield and subsequent stock market performance since 1945. Buying when yield is high, and selling or paring equity allocations when yield is low, has been a powerful strategy for capturing the equity risk premium in something more than average doses. In fact, lots of researchers have come to similar conclusions over the years. As Cochrane notes,
When prices are low relative to dividends, subsequent seven-year returns are likely to be high. Stocks do not follow a “random walk.” More deeply, price declines above and beyond declines in dividends over the following year have entirely rebounded. This finding is confirmed by 30 years of research, ranging from “behaviorists” such as Robert Shiller and Richard Thaler to “efficient marketers” such as Eugene Fama and Ken French, to “economists” such as John Campbell and myself. The same pattern also appears in price/earnings, book/market and other ratios, and in many other markets.
And let’s not forget that buying stocks when Cochrane’s article appeared (the dividend yield was relatively elevated at the time) would have delivered unusually high returns. Since Cochrane’s op-ed was published, the S&P 500 has climbed nearly 50%. Coincidence? Maybe, although there’s a strong case for arguing that expected returns were unusually high in late-2008, when the financial crisis was raging and prices were crashing.
But no matter how times the mean reversion idea seems to work, there’s always fresh criticism that this time is different. In the wake of the Great Recession, a number of analysts have argued that it’s going to be tougher to earn above-average returns from mean reversion. They may be right—the future, after all, is still uncertain, and Mr. Market’s rule book is always in some degree of flux. But GMO’s James Montier isn’t ready to throw in the towel just yet. “In order for mean-reversion-based strategies to work, it is not required that the mean be realized for long periods of time, but that markets continue to behave as they always have, swinging pendulum-like between the depths of despair and irrational exuberance, or, from risk-on to risk-off,” he writes in a research essay published last month (“In Defense of the “Old Always,” Dec. 2010 via gmo.com) “As long as markets display such bipolar disorder and switch from periods of mania to periods of depression, then mean reversion should continue to merit worth as an investment strategy.”
Montier, a member of GMO’s asset allocation committee and author of Value Investing: Tools and Techniques for Intelligent Investment, goes on to advise,
History is littered with the remains of proclaimed, but unfulfilled, new eras. Exhibit 6 [see chart below] shows the long-run history for the Graham and Dodd P/E for the U.S. market. Over this time, we have witnessed some quite remarkable, and quite appalling, things – the deaths of empires, the births of nations, waves of globalization, periods of deregulation, periods of re-regulation, World Wars, revolutions, plagues, and huge technological and medical advances – and yet one thing has remained true throughout history: none of these events mattered from the perspective of value!
Montier concludes: “So, rather than throwing out the handbook of investment, investors may well be better advised to stay true to the principles that have guided sensible investment since time immemorial.”
Indeed, the empirical evidence continues to stack up in favor of these principles. Or, in the words of financial economists, expected return fluctuates. Why does it fluctuate? That’s a bit tougher to nail down, although the mounting evidence leans in favor of looking to macroeconomic factors for an answer. That’s not surprising, although the details are intriguing, as several recent economic studies reveal. In an upcoming post, I’ll take a closer look at some of the latest research that relates macroeconomic volatility to risk premia.
Meantime, let’s recognize that the new normal has limits. The world that existed before 2008 is gone, but some of the old rules still apply when it comes to evaluating investment opportunity.