A Dangerous Combination: Bubbles and Debt
I still remember that day in the spring of 2000. I was sitting in my dorm room at Stanford University when I heard people shouting in the hall. It turned out that all the commotion was about the stock market crashing -- in particular, technology stocks. In fact, it had been crashing for days, but only now had people decided that this was The Big One.
But the real question is: Why didn't the economy crash as well? Sure, about $8 trillion in notional wealth vanished and we had a recession 2001. Yet the decline in gross domestic product was minuscule, the contraction lasted only two quarters, and the unemployment rate peaked at just 6.3 percent.
Now compare that with the bursting of the housing bubble in 2007-8, and the Great Recession that followed. Unemployment rocketed to 10 percent, and GDP took the biggest and longest plunge since the Great Depression. But if you look at the actual size of the bubbles -- at the amount of notional wealth that evaporated in the crashes -- they were in the same ballpark. Household net worth fell more as a percentage of GDP in 2008 than in 2000, but the difference was not that huge.
Why is it that some bubbles hurt the economy a lot, while others hurt it only a little?
One possible reason is that interest rates hit zero after 2008, which prevented the Federal Reserve from being able to do more to fight the recession. But that just leads us back to the question of why the 2008 shock was so bad that it made rates hit zero in the first place!
Another reason, which has attracted increasing amounts of attention in recent years, is the difference between equity bubbles and debt bubbles.
Many people have homed in on this explanation. Robert Hall, a leading macroeconomist, wrote that since debt bubbles damage the financial system, they endanger the economy more than equity bubbles, which transmit their losses directly to households. Financial institutions lend people money, and if people can't pay it back -- because the value of their house has gone down -- it could cause bank failures. In contrast, banks weren't that exposed to tech stocks. This idea -- that credit bubbles disrupt the financial system -- is at the heart of a lot of the macroeconomic models of credit-driven business cycles that have become popular since the crisis.
That's the theory. Now there's evidence. Economists Oscar Jorda, Moritz Schularick, and Alan Taylor recently did a historical study of asset price crashes, and they found that, in fact, debt seems to matter a lot.
Jorda, Schularick and Taylor look at 17 countries since 1870 and track equity and house prices. As their measure of the amount of credit in the economy, they use bank lending -- this stays very close to the idea that bank disruption is the reason credit busts are so dangerous. To make a long story short, they look at what happened to the economy of each country after each large drop in asset prices. Then they separate these bubble bursts by the amount of bank lending in the economy and add some controls for other factors.
Here, in two graphs, is what they found:
The green line represents the average path of GDP after the bursting of a bubble with low credit (like the 2000 tech bubble). The red line represents growth when a bubble bursts with high credit. The blue represents the average recession.
As you can see, bubbles make recessions longer, and credit worsens the effect.
But the effect is bigger when the bubble is a housing bubble, rather than an equity bubble. Why? Hall's bank disruption explanation isn't entirely satisfying. For some reason, bubbles are bad even when there is low credit in the economy. And for some reason, housing bubbles are worse than equity bubbles across the board.
One possible answer, not suggested by Jorda, Schularick and Taylor, is that housing bubbles hit a very different slice of people than equity bubbles. Stocks tend to loom larger in the portfolio of a rich person, while a middle-income or lower-income person's wealth is often almost entirely tied up in housing. Less wealthy people could be more likely to suffer from wealth effects -- the tendency to cut consumption when personal net worth declines.
But in any case, the message is clear: Bubbles and debt are a dangerous combination.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Noah Smith at firstname.lastname@example.org
To contact the editor on this story:
James Greiff at email@example.com