A frustrated trader on the floor of the New York Stock Exchange in 2014. (Photo by Spencer Platt/Getty Images)

There is little that Democrats and Republicans agree on when it comes to American banking. Yet most of them believe that the cause of the financial crisis was a shift in banking practices that resulted in more poor and middle-class households obtaining mortgages, even if they could not afford them — especially the costly, risky mortgages known as subprime loans.

Some economists have been skeptical of this broadly accepted theory, however.

“It's very hard to argue, as some people did, that the bubble was caused by subprime” lending, said Paul Willen, one of the authors of a new working paper from the National Bureau of Economic Research. “There’s no evidence that that’s true, and it’s a very implausible story.”

Willen, an economist at the Federal Reserve Bank of Boston, argues that rates of debt and default increased among the rich, as well.

Researchers who take the standard view see this skepticism as little better than an apology for elite hubris and an excuse for inaction on reforms. That anyone would make the claims in the new paper disturbs Atif Mian, an economist at the University of Princeton. In his words, Willen’s thinking is, “from a practical perspective, quite dangerous.”

It has become an unusually passionate dispute in the realm of banking policy, a field that was — until the financial crisis — a bland affair of acronyms and inscrutably detailed charts. Now, researchers are on edge, knowing that a wrong move by Congress could lead once again to disaster.

Living beyond their means

Mian and a colleague, Amir Sufi at the University of Chicago, have argued that banks, confident they had a way to manage the risks, disproportionately increased lending to poor borrowers. To make their monthly payments, these households began skimping on other expenses, slowing the pace of economic activity overall.

Republicans argue that the government was an important reason for the change in banks’ behavior, pointing to federal programs designed to help poorer Americans buy homes. 

“There was a really significant long-term lowering of credit standards that was driven at least in part by government programs,” said Stephen Oliner, an economist at the conservative American Enterprise Institute.

He and other economists point to additional factors as well, such as more lenient lending standards, the development of complex financial products based on payments from homeowners and outright fraud.

All these theories focus on excessive subprime lending as the cause of the crisis. The authors of the new paper — including Willen and his collaborators at the Federal Reserve in Boston, Christopher Foote and Lara Loewenstein — say that view is too narrow. They argue that risky bank behaviors like securitization, public programs and lending standards can explain only why poorer Americans borrowed more and do not account for substantial increases in household debt throughout the economy.

These researchers show that, during the boom in the housing market, the share of people taking out mortgages for the first time was stable or declining. In other words, the authors argue, the buyers bidding up the price of houses were not people with limited incomes.

Instead, they were more affluent buyers who already had access to conventional loans and were buying more expensive houses. Willen contends that laxer lending on the part of banks simply allowed first-time buyers to keep up with skyrocketing prices.


Indeed, by the time subprime lending began to accelerate in 2004 and 2005, housing prices were already on a tear. The rapid increase in subprime loans came too late to have been the real cause of the boom, Willen, Foote and Loewenstein argue.

Zip codes and credit scores

Part of the dispute is about how to measure the increase in debt in different groups of Americans.

The essential question is whether access to loans increased disproportionately for borrowers who fundamentally could not afford to buy houses. Mian and Sufi  found that borrowers who had low credit scores before the boom increased their debt more quickly than did borrowers with high scores.

Willen and his colleagues at the Boston Fed counter that more debt for borrowers with bad credit scores at the beginning of the cycle does not necessarily show an increase in debt among poorer borrowers. They note that people with bad credit scores tend to get their finances in order after a few years anyway. They pay off what they owe, reduce their expenses or take on more work. Once their credit scores improve, they are able to borrow more.

While individual borrowers' circumstances might change over time, making it difficult to keep track of who is taking out more debt, the character of neighborhoods generally remains stable, the three researchers argue — and they did not find that debt increased disproportionately in poor Zip codes. During the boom, defaults followed suit, with substantial increases in all kinds of neighborhoods.


In a typical Zip code with an average household income of $25,000, the annual default rate roughly doubled from 6.2 percent in 2001 to 14.4 percent in 2009, according to Foote, Loewenstein and Willen. In a more affluent Zip code with an average income of $53,000, however, the rate roughly tripled, from 2.3 percent in 2001 to 7.2 percent in 2009.

While defaults in poorer neighborhoods accounted for the bulk of foreclosures, borrowers in wealthier neighborhoods tended to take out more expensive mortgages, so more money was at stake when one of them defaulted.

'A tautology'

This statistical puzzle is difficult to resolve, but there seems to be an even more profound disagreement between the two groups of economists.

Pressed for their own theory about what, if not subprime lending, caused the boom and the bust, the Boston Fed group refer to that ancient “irrational exuberance,” those same “animal spirits” that have disrupted markets across the centuries. This kind of reasoning verges on the unscientific, says Mian, who argues that explaining where that inexplicable exuberance comes from should be the goal of economic research.

The theory advanced by Willen, Foote and Loewenstein is “quite difficult to prove or disprove,” Mian said. He argued that economists should do better than to say that markets produce booms and busts, and therefore the housing market had a boom and a bust.

“That’s almost a tautology,” Mian said. “If you just are hanging your hat on that kind of hypothesis, it’s not really saying anything.”

Willen, though, is worried about saying too much — that is, asserting more than he and other economists know about the true causes of the catastrophe.

“Our understanding of asset prices is still quite primitive,” he said. “We don’t have a really good explanation for why house prices went up as much as they did.”

Financial, seismic engineering

This philosophical disagreement has practical implications for policymakers in Washington, who, for the second time in a decade, are considering major changes to the rules that govern the country's financial system. President-elect Donald Trump has said he will undo the comprehensive Dodd-Frank financial reforms that Democrats enacted in 2010.

Mian argues that regulators should scrupulously monitor banks’ activities to prevent the kinds of risky practices that he believes contributed to the financial crisis. On the reasoning that the causes of the crisis are still unknown, Willen contends policymakers should instead try to prepare the system to withstand the next collapse, which he suggests might be as unpredictable as it is inevitable.

For instance, he said analyses of how financial institutions would fare in a crisis — such as the so-called “stress tests” mandated by the Dodd-Frank law — can help bankers and policymakers prepare for the worst.

Willen compared financial regulation to seismic engineering. “No one tries to prevent earthquakes,” he said. Rather, the goal is to ensure buildings can remain standing despite them.

Policymakers might be able to do both at the same time. Alan Blinder, the former vice chairman of the Federal Reserve under President Bill Clinton, said that the decisions made on Wall Street merit scrutiny, even if subprime lending was only one aspect of the broader problems in the market.

“There was a big housing bubble, a craze, and house prices just went through the roof,” Blinder said. “That by no means exonerates Wall Street.”

Contrary to Rep. Jeb Hensarling (R-Tex.), chairman of the House Financial Services Committee, and his Republican allies on Capitol Hill, Blinder warns that another major financial overhaul could give bankers too much leeway.

“It would be a horrible mistake to dismantle Dodd-Frank,” he said.