Housing Bust Legacy? Too Much Fear of Debt
After the great American housing bubble blew up in 2008, taking the financial system and the economy with it, there was an often-bitter debate about the underlying causes. Why had housing prices been driven to unsustainable heights in the first place? Were Americans borrowing to overconsume, or were they just trying to make a quick buck? The answer matters because it will guide future policy.
One hypothesis -- let's call it the debt hypothesis -- was that homeowners chose to take on too much housing debt, and that this debt allowed prices to rise above reasonable levels. A duo of well-known economists, Princeton University's Atif Mian and the University of Chicago's Amir Sufi, have advanced this idea in a magisterial recent book, "House of Debt." The debt hypothesis is appealing both to many conservatives (who think government policy encouraged overborrowing), and to many liberals (who think that the unscrupulous finance industry tricked borrowers into piling on too much debt).
Why would Americans take out so much debt? Perhaps they were trying to replace their declining incomes. Median household income in the U.S. peaked in 1999. People probably have consumption benchmarks -- no one wants to be less well-off than their parents. Also, income inequality in the U.S. climbed substantially during the late 20th century, and people may have a desire to spend more when they see their richer neighbors spending more. This explanation has been advanced by some heavy hitters, such as former University of Chicago economist Raghuram Rajan, now governor of India's central bank.
The alternative hypothesis is that the housing bubble was simply a classic speculative asset bubble, in which people knowingly overpaid for houses because they expected other people to come along and pay even higher prices. This explanation -- let's call it the bubble hypothesis – says that the housing bubble wasn't fundamentally different from the late-1990s tech bubble, or any other bubble throughout history. This suggests that the inherent instability of financial markets, not the borrowing decisions of homeowners, was to blame.
This debate seems arcane, but it matters a lot for public policy. If household debt is a destabilizing force in the economy, maybe the government should act to limit the amount that people can borrow. But if the housing bubble was just a normal asset bubble, then such distortionary measures are likely to hurt more than they help.
So which is right? Economists have come up with evidence both for and against the debt hypothesis. In 2011, a group of economists from the Federal Reserve Bank of New York found that much of the increase in mortgage origination in the bubble years went to people who owned multiple properties, many of them house flippers who were speculating on price increases, rather than people buying homes to live in or hold as rental properties. That solidly supports the idea that speculation, not debt, was the culprit.
But Mian and Sufi offer evidence in the other direction. They show that in areas where income grew less during the mid-2000s, mortgage credit went up more. In other words, it looks like people were using debt to make up for falling incomes, giving support to the debt hypothesis. Lower income meant more debt, and a bigger housing bubble.
In January, however, opponents of the debt hypothesis released a rebuttal to Mian and Sufi. Economists Manuel Adelino, Antoinette Schoar and Felipe Severino show that when you look at individuals rather than zip codes, the pattern reverses -- people whose incomes grew more also borrowed more during the boom. Even more crucially, it was middle- and high-income borrowers who were more likely to default after the bubble burst. So if we restrict our attention to only the loans that went bad -- the loans that, in a perfect world, shouldn't have been made in the first place -- we see that debt wasn't a substitute for income. Instead, this research suggests that the housing bubble was a game played by the well-off, who used their increased incomes to speculate on rising price. In other words, the bubble hypothesis.
The fight isn't over -- Mian and Sufi shot back with a response questioning their rivals' methodology. Expect a return of fire. But to me, the evidence seems conclusive that there was a lot more at work in the housing bubble than a surplus of lending to lower-income Americans.
So what is the upshot for policy? We should think twice about having the government restrict the flow of lending. Although there are definitely cases in which lending is done in a fraudulent or predatory manner, which we should try to prevent, there is great danger in the idea that debt itself is a toxin that wreaks havoc on the economy. The long asset booms of the '80s, '90s, and 2000s made us complacent about debt, but the bursting of the housing bubble has made us too suspicious of it. We need to find a happy medium.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the author on this story:
Noah Smith at email@example.com
To contact the editor on this story:
James Greiff at firstname.lastname@example.org