Every month, the New York Stock Exchange (NYSE) publishes total margin debt figures for the U.S. stock market. Margin debt refers to the amount of money borrowed by investors to purchase equities, represented in the NYSE as a percentage of total gross domestic product (GDP). This is an important figure, at least for markets, since total margin debt has a very strong correlation with the S&P 500. The relationship suggests that stock prices may be heavily dependent on borrowed money.

In April 2015, total margin debt reached an all-time high of 2.83% of GDP. This surpassed prior peaks from 2013, 2007 and 2000. Starting in May 2015, however, margin debt figures slowly retracted. In other words, investors started to pay off brokerage loans faster than they borrowed money.

Many market insiders worry that the retreat in margin debt is a harbinger of a market collapse. Looking at historical data, you can see why they are concerned. However, investors should be warned against taking bold action without more information.

Previous Cycles Point to a Large Correction

In March 2000, the S&P 500 reached an all-time high of 1,550. Total margin debt hit its own all-time high during the same month. Both figures turned bearish in subsequent months, however, and each declined more than 50% before 2003. Statistics rarely tell the full story, and it is difficult to see which trend was causal and which was dependent, but it is a striking coincidence.

A similar coincidence materialized during 2007. In early 2007, total margin debt surpassed $400 billion for the first time. This represented greater than 2.5% of GDP. The Great Recession was just around the corner, and it appears borrowers may have seen the signs faster than the general market. Margin debt figures started falling by mid-year, while the S&P 500 hit a new peak at 1,576 during intraday trading on Oct. 1, 2007. By early 2009, margin debt and the S&P 500 had fallen more than 50% again.

The S&P 500 hit a monthly peak of 2,134 on May 1, 2015, just one month after the total margin debt peak of 2.83% of GDP. Relying just on nominal figures and assuming a similar trend as in 2000 and 2007, it would seem the S&P 500 could fall to less than 1,100 before mid-2017. Total margin debt could fall to approximately 1.4%.

Some Problems With the Relationship

There are a lot of reasons to question the simple relationship between margin debt and stock prices, even when the correlation is extremely tight over a long period of time. The most immediate issue is that the NYSE measurement on margin debt is time-lagged by several weeks, whereas stock prices update almost instantaneously during day trading. At best, investors react to margin debt data once per month. Given the wealth of data published around the clock, as well as the relatively uncelebrated nature of margin debt numbers, it is difficult to see how public margin debt figures could actively cause changes in equity prices.

If there is any causality between margin debt and stocks, the trigger is probably less conscious. Settlements between brokers and margin borrowers almost certainly put downward pressure on equity prices, since borrowers sell stock to square their debts. So it is possible a margin sell-off could ignite a larger bear movement.

However, the S&P 500 was worth more than $18.8 trillion in March 2016. Total margin debt was just $435.8 billion, or 2.3% of the total market capitalization of the index. Even if the impact of margin debt is magnified due to investor psychology or momentum trading reasons — a difficult argument since total margin debt is not a well-known figure — it is still unlikely to be the driving force behind equity price movements. Margin debt could still be a useful indicator, even if its correlation with the S&P 500 is only coincidental. There is probably a much more satisfactory underlying causal factor (or set of factors) for each data set, and perhaps both are driven by the same factor.

Interested investors are well advised to seek out a plausible causal chain. It is rarely a good idea to trade according to trend-based predictions without a logical causal mechanism. Even if you can't identify the root cause, though, it seems reasonable to assume that markets are not going to experience a continuous rally in 2016 and 2017. After all, whatever is driving margin rates south might also take stock prices down with them.