The Conventional Wisdom on Margin Debt Is Wrong
(Bloomberg Opinion) -- As the pandemic hit last March, total margin debt – money borrowed to invest in the stock market – stood at $479 billion. Since that time, the stock market has boomed, and margin borrowing has only increased. This past January, it hit an all-time high of $798.6 billion.
In response, mainstream commentators and iconoclasts alike are sounding alarms about systemic risk and warning of a repeat of the crash of 1929, arguably the worst in the nation’s history. But they are citing some history while missing critical pieces of it, relying on an old narrative that has proven deeply flawed.
In 1934, the Senate Committee on Banking and Currency’s “Report on Stock Exchange Practices” fingered margin trading as a primary force behind the crash’s self-sustaining spiral. It cast margin traders as fools driven mad by the promise of “quick profits” and brokers as far too willing to waive margin requirements in a market that they falsely believed would never stop rising.
As a consequence, the margin traders occupied an “overshadowing position on the financial scene.” When the market peaked, the inevitable margin calls led to waves of uncontrolled liquidation, creating a feedback loop that drove the stock market into the ground.
The Securities and Exchange Act of 1934 passed on the force of this argument, handing power to the Federal Reserve for mandating margin requirements under Regulation T. But it was John Kenneth Galbraith who popularized the idea that margin debt was inherently problematic – as well as the notion that it could be used as a sign that the market was overheating. In 1955, he published “The Great Crash, 1929,” the most influential account of this turning point in history.
Ever since, researchers have been finding that the actual historical evidence doesn’t bear out his conclusions. In 1988, multiple researchers showed that brokers actually began raising margin requirements to extraordinarily high levels well before the actual crash. Instead of the 5% or 10% requirement that had prevailed at the start of the practice in the 1870s, far more stringent requirements prevailed by 1928. By the fall of 1929, many brokers had raised collateral requirements as high as 80% or even 100%.
Other researchers raised additional questions. A study in 1988 pointed out that most of the forced liquidations of margin positions transpired quickly, long before the stock market registered most of the declines that would represent one of the worst downturns ever.
In a 2012 paper, two researchers disagreed with the conclusions of the Senate banking panel and Galbraith, blaming the resiliency of the stock market in this period, not its instability. If margin debt was as destructive as predicted, they argued, the declines should have been far larger. In 2005, another set of researchers looking at margin debt and stock market volatility argued that the relationship was tenuous at best.
Then there’s the question of whether margin debt is a reliable indicator of the stock market’s direction. Here, too, the research runs counter to conventional wisdom. While it’s clear that rising margin debt mirrors the rising stock market, it’s hard to know when levels signal danger.
One way to measure this is to weigh margin debt relative to GDP or to the total capitalization of the stock market. Some market watchers have done that in recent weeks, claiming that not only is margin debt at record highs in nominal terms, it’s at all-time highs relative to GDP: approximately 3.6 percent.
Once again, the historical record disagrees. Back in 1929, margin debt was as high as 6% or even 8% of GDP. In other words, we’re not close to the kind of speculative fervor that preceded the crash that did so much to define our beliefs about margin debt.
None of this means that we should ignore rising levels of margin debt, or blithely assume that the bull market is going to keep going. But it does call into the question whether we should put so much weight on margin debt as a decisive factor.
That said, history does suggest two closely related ways of using margin trading as forward guidance. The first is inspired by what happened in 1928, not 1929: The fact that brokers began hiking margin requirements well in advance of the actual crash. It was their behavior, not the people speculating on margin, that correctly anticipated the coming disaster well in advance. When these rates start rising above historical averages, trouble likely beckons.
In a fascinating paper published last fall in the “Review of Finance,” several researchers pointed out that when it comes to using margin trading as a predictive tool, it’s not the amount of money that investors are sinking into the market that matters. Instead, look at the amount they’re holding back.
Using FINRA data, they constructed a measure of what they dubbed “margin credit” – the “excess debt capacity” in investor accounts. This balance rises when investors choose not to borrow against their gains; it falls when investors plow their gains back into the market. The researchers speculated that fluctuations in these balances had predictive value.
It apparently does. The researchers compiled data on margin credit across several decades and found that it was a “powerful predictor of future excess market returns.” But it also predicted downturns: As they put, “we can view each margin investor’s decision to accumulate margin credit as a reaction to an individual signal about overvaluation or upcoming market crash.”
It was just the latest reason stretching back to the 1930s to believe that rising levels of margin debt are a sign of something. But if you really want to know where the market is headed today, don’t stop at Galbraith.
In the succeeding years, the Fed adjusted this level up and down, but in 1974 left it at 50 percent, where it remains today.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
Stephen Mihm, an associate professor of history at the University of Georgia, is a contributor to Bloomberg Opinion.
©2021 Bloomberg L.P.