Behavioral Risk & Rebalancing

Central bankers don’t usually deliver insight on matters of portfolio management, but San Francisco Fed President John Williams bucked the trend in a speech on Monday. Discussing “Bubbles Tomorrow and Bubbles Yesterday, but Never Bubbles Today,” his focus, of course, is on monetary policy and what a central bank can and can’t do when Mr. Market suffers from an extended bout of irrational exuberance. It’s an old topic but a perennially relevant one since what we call “bubbles” are likely to stalk the market landscape for something close to eternity. Identifying these beasts in real time in the cause of distinguishing the general article from an imposter is tricky. But recognizing how the crowd has a habit of aiding and abetting misguided behavior on this front reminds us why the critical task of rebalancing is so hard for so many investors.

Williams zeroes in on this challenge by looking at dramatic increases in asset prices through the lens of the Gordon model, which was originally presented more than 50 years ago in a paper by Myron Gordon. “Two explanations are possible based on changes in economic fundamentals,” Williams explains. “One is an upward shift in the expected growth rate of future dividends. The second is a reduction in investors’ expected returns on the asset.”
The first explanation has been dismissed based on the historical record. “With respect to U.S. stocks, over history, the price-to-dividend ratio is uncorrelated with future real dividend growth,” Williams observes (for example, see John Cochrane’s study “The Dog That Did Not Bark: A Defense of Return Predictability”). That leaves changes in expected returns as the explanation, but here too the empirical record is harsh, at least from the perspective of a “rational” interpretation.

The evidence from surveys of investors’ expected returns is directly at odds with the implications of standard asset price theory. For one, stock market investors tend to expect high future returns when the price-to-dividend ratio is high, contrary to the theoretical prediction of a negative relationship between rational expected returns and the level of asset prices relative to dividends. A picture tells the story. Figure 2 shows Gallup survey results on the relationship between the S&P 500 price-to-dividend ratio and investor optimism regarding stock market returns over the next year. Gallup asks whether people are optimistic, pessimistic, or neutral about future market returns. The figure reports the difference between the number saying they are optimistic or very optimistic and those saying they are pessimistic or very pessimistic. As the figure shows, periods of high stock valuations, such as the late 1990s and mid-2000s, are when investors were more optimistic regarding future stock gains. And during periods of relatively low valuations, such as the early 2000s and the period of the global financial crisis, investors had relatively low expectations of stock market returns. The positive relationship between current stock prices and expected future returns is consistent across a variety of surveys and alternative model specifications.

The chart above helps explain why the average investor finds it so difficult to rebalance in a timely and productive way. Even professionals managing multi-asset-class funds have a tough time beating a passive asset allocation benchmark through time, as my research shows.
The lesson here is that much of what bedevils investors in earning a respectable risk premium over medium-to-long-time horizons is behavioral risk. We are, collectively, our own worst enemy when it comes to intelligently managing asset allocation. This has less to do with a lack of technical skills or intelligence about how markets work, and more about our own internal issues with recognizing, in a timely manner, the distinction between short-term momentum and longer-term reversion-to-the-mean factors. When one or the other dominates, our brains don’t easily embrace the idea that the other could revive. Or in the jargon of economics, investors are too often procyclical at critical turning points when they should be acting in a countercyclical (contrarian) manner.
Rebalancing, in short, is hard. But it’s also where all the action is. That’s a reminder that most of our attention should be, must be, focused on doing a better job at rebalancing our portfolios. In other words, quite a lot of what passes for intelligent analytics is missing the boat. Recognizing that so few of us excel on this front, in fact, explains why the opportunities are so attractive. The bad news is that only a relative handful of investors will benefit from superior rebalancing decisions.