GOOD LUCK TRYING TO BEAT THIS MARKET / BARRON´S MAGAZINE

Good Luck Trying to Beat This Market

By Mark Hulbert
.

Good Luck Trying to Beat This Market / Photo: Getty Images 


Stock-picking is difficult regardless of whether it’s a “stock market” or a “market of stocks.”

I’m referring to the age-old distinction between periods in which individual stocks are highly correlated with each other and times when stocks march to the beat of their own individual drummers. When they become more highly correlated—as they have been during the correction that began in late January and which continues to gather steam—analysts proclaim that it’s a monolithic “stock market” and conclude that stockpickers will have a particularly hard time beating the overall market.

In contrast, when stocks’ correlations are low, as they were for most of last year, analysts say that it’s a “market of stocks.” At such times, many assert, stock-picking has a decent chance of success.

The analysts are only half right. Index funds beat the vast majority of stockpickers in both kinds of market environments. Last year, for example, 63.1% of U.S. large-cap funds lagged the Standard & Poor’s 500 index, according to S&P Dow Jones Indices, even though 2017 supposedly was a year in which stock-picking was relatively easy.

Analysts who focus on stocks’ correlations often fail to realize that those correlations rise when the market falls, and vice versa. The only way a stockpicker could exploit that information would be if he knew in advance when the market is about to rally or decline. It’s no help to be told, after the fact, that correlations have risen or fallen.

One reason why correlations respond inversely to the market is statistical, according to Kristin Forbes, a professor of management and global economics at the Massachusetts Institute of Technology. In an email, Forbes told me that, because of the way correlation coefficients are calculated, increased volatility—which inevitably happens when the market declines—more or less automatically leads to heightened estimates of stocks’ correlations, even when nothing has really changed. The increased correlations we saw during the market’s recent correction were in part caused by this statistical idiosyncrasy.

Another reason why correlations increase when the market declines is behavioral: When fear grips investors, they tend to sell first and ask questions later. Such panic is far different than the prevailing mood during bull markets. As Jon Markman, editor of the Strategic Advantage advisory service, recently wrote, “During bull markets stocks tend to rise at a leisurely and thoughtful pace, like an 80-year-old couple out for a walk in the Florida sunshine.”

Consider stocks’ correlation with the S&P 500. As last year’s bull market took stocks to ever-higher levels, the Cboe S&P 500 Implied Correlation Index dropped to some of its lowest levels ever. But during the stock market’s correction in late January and early February, that index spiked upward.

This shouldn’t have been a surprise, once we understand the statistical and behavioral factors underlying changes in correlation. Analysts who in recent weeks have breathlessly proclaimed that we are now in a monolithic “stock market” are in fact telling us little more than that the market had declined.

What we would need to know in order to profit from this information is foreknowledge of what stocks’ correlation with the S&P 500 will be in subsequent weeks. But such foreknowledge is elusive. Periods of high or low correlation rarely persist for very long. In fact, they more often than not are quickly corrected. That means the trading environment tends to become a “market of stocks” at the very time that analysts announce it has become a “stock market”—and vice versa.

No wonder it’s so difficult to use stocks’ shifting correlations with the overall market to beat the market.

The proof of the pudding is in the eating, of course. For each year since 2007, which is when the Cboe first began calculating implied correlation indexes, I measured the proportion of the 500-plus portfolios monitored by the Hulbert Financial Digest that beat the overall market on a risk-adjusted basis. This proportion was no higher when correlations were below average than when they were above average. In all cases, furthermore, that proportion was well below 50%.

It’s not an accident that the strategies that have made the most money over the long term pay no attention to short-term or even intermediate-term changes in stocks’ correlations with the overall market. Warren Buffett, CEO of Berkshire Hathaway and widely considered to be the most successful investor alive today, advises us to “only buy something that you’d be perfectly happy to hold if the market shut down for 10 years.” If the market were to shut down, needless to say, there would be no way to even calculate stocks’ correlations with the overall market.

The bottom line? You should give up if you’re basing your stock-picking strategy on whether you think it’s a “market of stocks” or a “stock market.” The odds are overwhelming that your strategy will lag a simple index fund.