CHAPEL HILL, N.C. (MarketWatch) — Exuberant investors reacted to the U.S. stock market’s strong rebound in June by purchasing more stocks on margin — a lot more.

That means the bull market is alive and well, according to many analysts who believe margin debt is a telling leading indicator. (Margin debt refers to the use of borrowed money to invest.) But I’m not convinced. My analysis of the historical data shows that margin debt is a coincident indicator — it rises and falls along with the market itself.

The Dow Jones Industrial Average gained a stunning 7.2% for the month and the S&P 500 jumped 6.9%. That’s as margin debt rose by $27.5 billion to $596.3 billion, a 4.8% increase and the biggest monthly dollar advance and percentage increase since September 2016. So it’s not a surprise that margin debt would have jumped in June.

Still, how did market timers ever come to believe that margin debt is a good leading indicator? Therein lies a cautionary tale.

Margin debt’s Achilles’ heel as a leading indicator is that it appears to do a good job only after the fact. Proponents are guilty of what statisticians refer to as “hindsight bias.” Since this bias is widespread on Wall Street, it’s worth spending some time reviewing how it sabotaged the margin-debt indicator. Hopefully you will learn how to avoid making the same mistake yourself.

Take the bull-market top in October 2007, right before the Great Recession and one of the worst bear markets in decades. Because margin debt hit its peak in July of that year, three months in advance of that top, it certainly looks like a “win” for the margin-debt indicator.

In fact, however, there were a number of occasions in the 2002-2007 bull market when margin debt hit a peak and began to decline. In retrospect, of course, those drops appear to be short-lived and, therefore, nothing more than temporary hiccups. But that would have been impossible to know at the time.

To smooth out the monthly gyrations and focus on the underlying trend, many compare margin debt to a 12-month moving average. But when we do that, the margin-debt indicator no longer appears to have anticipated the top of the bull market in 2007. Total margin debt didn’t drop below its 12-month moving average until January 2008, when the December 2007 data were reported.

That’s worse than the 200-day moving average, which is a widely followed technical tool for determining the market’s major trend. The S&P 500, for example, broke below its 200-day moving average in November 2008.

No Warning in 2000

The margin-debt indicator also left much to be desired at the top of the internet bubble in March 2000. Though margin debt also hit its peak that month, we couldn’t have known until much later that the month’s total was a peak. Margin debt didn’t drop below its moving average until November of that year, when the October total was reported.

Once again, the simple 200-day moving average did better: The S&P 500 dropped below its 200-day moving average in September 2000.

To be sure, those are just two examples. But a more comprehensive analysis of the historical data back to 1970 reaches the same conclusion. I fed into my PC’s statistical package the monthly data on margin debt dating to 1970, and searched for any of a number of possible correlations between changes in the margin total and the stock market’s subsequent performance. I came up empty.

When I focused on correlations with the stock market’s trailing returns, however, a strong trend did materialize. That’s why margin debt is a good coincident indicator: When the stock market rises strongly, traders buy more stock on margin and total margin debt rises. By the same token, whenever the market plummets, those erstwhile enthusiastic investors receive margin calls or become dejected, and they reduce their margin balances.

The bottom line? The bull market may be alive and well. But we can’t know that it is because margin debt happened to jump so much in June.

The broader lesson: Be ruthless in your analysis of an indicator’s record. Exercise great care to make sure you don’t assume that you knew something in real time that wasn’t, in fact, knowable until well after the fact. One thing you can be confident about: Doing that will cause you to lag the market over time.

{"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish.