How Household Debt Caused the Great Recession: A Q&A With Amir Sufi

A talk with the ‘House of Debt’ co-author and University of Chicago economist on the roots of the financial crisis, the problem of student loans, and how to talk about the issue of debt.

Photo: University of Chicago Press

Recently Thomas Picketty’s Capital in the 21st Century, a 700-page tome about capital, wealth, and inequality, its design consciously echoing Marx’s Capital, reached the bestseller list. Even with an audience normalized to the subject by the success of pop-econ books, the immense anticipation for and success of the book caught many in the dismal science by surprise.

Before the publication of Picketty’s book, his research on inequality laid the groundwork for its release, buoyed by the increased attention given to the subject in the wake of the Great Recession. But it wasn’t always so. When the housing bubble popped and the financial markets spiraled out of control, the pillars of the American economic system—and not its vast foundation—received the bulk of the attention from economists, policy experts, and politicians.

In their forthcoming book House of Debt, University of Chicago economist Amir Sufi and Princeton economist Atif Mian revisit the run-up to the Great Recession—they’re also blogging about related topics here—arguing that the central cause was a massive increase in household debt, concentrated among those least able to incur it.

I spoke with Sufi about their thesis, how Americans think about debt and its implications for public policy, why we should be wary about student-loan debt—and what to do about it.

In 2009, NPR reporter Adam Davidson told Elizabeth Warren that none of the economists he talked to emphasized household debt in regards to the Great Recession. Is that the sense you got from your profession, and why’s that the case?

I think that academics, people who are working in the research sphere, have a bit more patience, perhaps—they’re willing to think a bit out of the box. Within research circles, there was a cognizance that the financial crisis itself couldn’t be the explanation for everything, and how massive the recession was. In general, I think economists who were working in policy circles were focused primarily on the financial crisis and not the longer-run buildup of household debt, and in particular the housing collapse.

I don’t know exactly why. I think we have to remember that economists are subject to what’s going on in the financial press as well. Every day on TV you’re hearing “financial crisis,” “stock market collapsing,” “banks are failing,” the President is going on TV and saying “if we don’t save the banks, than the economy is going to collapse.” Even economists are subject to the notion that whatever’s acute and right in front of you, you’re going to blame for what we’re going through.

That’s a big reason why there was so much focus on the financial crisis, and very little on the household debt crisis.

It seems like, at the time, the financial crisis was the bottleneck, and household debt was much harder to wrap one’s mind around.

I don’t know if I agree with that. One of the things you see in the recession that should have been more on the minds of all economists and policy-makers was the tremendous collapse in household spending. And we make the argument in the book that that occurred by August 2008, before Lehmann Brothers. Economists were too quick to jump to the conclusion that the decline in spending was, in itself, due to the financial crisis, and not to this tremendous collapse in house prices.

It is surprising to me that the collapse in house prices was not more salient in the mind of people when they’re thinking about why people are cutting back spending so dramatically.

There’s also an intellectual history here that I think is quite important, and it’s related to research coming out of the Great Depression. Mervyn King, the former governor of the Bank of England, made this point when he was giving the European Economic Presidential Address in 1994. Irving Fischer was the primary economist who argued that household debt was crucial for understanding the severity of the Great Depression—he was discounted because he said some very silly things about how the stock market’s never going to collapse, in 1928-29. King says it very well, “Irving Fischer was a very bad forecaster but he was a very great economist.”

His legacy was discounted, and instead, the Milton Friedman/Ben Bernanke view of the Great Depression, that it was the collapse in the banking sector that precipitated the Great Depression, from an intellectual history point of view led economists to focus on banks more than focusing on household debt. That’s another reason why economists weren’t really as focused on the household debt issue as they should have been.

One of the interesting things reading this book was the timeline. I graduated in 2004, and people were saying “you have to buy a house, because that’s what you’re going to retire on.” I recall looking around at that time and seeing houses in middle-class neighborhoods that were going upwards of $300,000, and thinking “this just isn’t affordable.”

Macroeconomics as a field has generally downplayed the importance of wealth distribution. As a result, the only way debt can matter in a macroeconomic model is if the distribution somehow matters. What I mean by that is, a macroeconomist who doesn’t appreciate the distribution of wealth will say, “well, look, for every debtor there’s a creditor, and if house prices rise, sure, the debtor loses, but ultimately the fundamental asset is the same asset: it’s a house. Now the creditor owns it instead of the debtor, why should that matter for anything?”

That’s even expressed in Ben Bernanke’s work on the Great Depression. He actually mentions Irving Fischer, and says, “no one really takes Irving Fischer’s argument seriously, because we don’t really think the wealth distribution can matter so much.” That’s something that now, with the work of [Thomas] Piketty and others, you’re starting to see macroeconomists caring more about wealth distribution. So there will be a heightened appreciation for things like household debt.

From the research and figures in your book, that household debt—especially at lower-income levels—begins to really take off around 2002. Do you have a sense of why that was?

We have a pretty pessimistic view about the decision-making for low-credit-score borrowers. In my view, they were more or less passive, in the sense that what you see, and you see this in evidence from credit-card lending even in the late ’90s, there seems to be a pretty large segment of the population that, if credit is made available to them, they will take it.

I don’t think anything really dramatically changed about the demand for credit by these lower-income, low-credit-score borrowers. All of a sudden they were being offered it. We do hint that maybe they were being tricked, some of them at least, into not understanding what they were taking on.

What we think is really moving is not so much the borrowers, who all of a sudden wanted more debt, or wanted to consume more, what really moved was the willingness of lenders to provide that credit to the borrowers.

It seems like lenders should have the responsibility that bartenders have—you have to cut people off when they’ve had too much.

That’s just not the way our financial system works, unfortunately. There’s a lot of evidence on that. For example, there’s an old saying in the financial industry that you don’t make money on the guys who pay back all the time. Credit card lending is really where you see that—where you make money off people who are late payers, who have to pay fees. There’s a financial incentive, oftentimes, for the financial industry to… I think the way to say it is, they are looking for what I would call the marginal borrower who is willing to pay a very high interest rate and may need to pay fees, because that ultimately can be quite profitable for the bank, even though it may not be in the best interest of the customer.

And I think that’s a lot of what happens in the payday lending market, and a lot of what happens in these subprime market, where the lender is willing to take that risk, even though from the perspective of the consumer, it’s not obvious the consumer should be taking on that loan.

In the book you touch on the cultural aspect of debt, that there is a zero-sum-game approach to debt—if someone incurs it and doesn’t pay it off, they should be responsible for it in full. The “inflexibility of debt” is the term you use. But one point you make is that it’s not necessarily a zero-sum game.

The central argument we’re making is that, when there’s an aggregate collapse in the economy, or house prices, there is a need to renegotiate debt contracts because they unfairly force too much of the risk and the losses on one party. Because they impose such large losses on one party, they ultimately bring down the whole economy, and that’s a central argument we make in our book.

The argument is that, actually, if you can forgive some debt, everyone ends up being better off. Including, actually, ironically enough, the lenders who are actually lending the money.

For example, John Maynard Keynes—a lot of intellectual history is based on his General Theory of Employment, Interest, and Money. But actually, his Economic Consequences of the Peace, which is the book that really made him famous, makes this point in the conclusion very strongly, when he’s talking about enforcing debts on Germany. And he’s basically saying, look, at the end of the day, if you enforce these debts on Germany, and make Germany pay this much, it’s going to end up coming back and hurting you, as the United Kingdom and France, more than if you just forgive some of it. And I think he obviously proved to be very prescient in that, given what happened in the years after.

The cultural norm that we have about debt is something along the lines of, “you should have to pay back your debt because you incurred it, so you should have to work harder, or you should do whatever you have to do to pay it back.” That argument kind of goes out the window when there’s an aggregate shock that’s beyond the control of anyone. Like we say in the book: should we blame people for being born in 1988, and therefore being 22 right when the Great Recession is starting, and having student debt? That’s not their fault. There’s no notion of a cultural norm, where you should have to pay even though the unemployment rate for college graduates is 20 percent.

That’s the kind of argument we’re making—when there are these aggregate shocks it doesn’t make sense to impose those losses on the debtors.

Even if there are good economic and historical arguments to be made, it does seem to run against the rhetoric of debt.

The argument I’m making about aggregate shocks maybe isn’t appreciated when people make these moral arguments. I think it’s interesting—even the Old Testament, and the Jewish tradition of having these debt jubilees when there’s a massive downturn—there’s a moral tradition of debt forgiveness in times of aggregate despair.

Of course, we have religious traditions about what we think of debt; sometimes they’re not very specific what exactly is it about debt that’s considered wrong or immoral, but we have usury prohibitions in Christianity and Islam. There is a moral tradition that’s outside the purview of stuff we’re talking about in the book, but I think we can tap into that a little bit when we’re thinking about this issue.

One of the things I find most interesting in looking at the history, and realizing that it seems like debt forgiveness was considered more socially acceptable after the Great Depression, after the Panic of 1819. We have a quote in the book from the Panic of 1819 which is just startling: a senator saying, “yeah, these farmers got a little out of control, they paid too much, but we all got a little out of control, and as a result, we should forgive this debt.” Can you imagine a senator saying that today?

I don’t think it’s something that’s immutable, or constant, or static—this Rick Santelli notion that how dare we ever forgive debt. I think there’s something specific about our time where that has become less acceptable. That’s not really my area of expertise for why that’s occurred, but I don’t think we should think of it as static. It could change.

In looking out for future bubbles, what indicators should we be looking to?

Real estate is always dangerous. That’s one of the lessons that we draw upon. Real estate bubbles that are fueled with debt, you should always be worried about. Canada’s now having a slowdown in the housing market; we’ll see how that works out. There just seems to be this very strong predictability from real-estate driven leverage cycles.

The other thing we look at very carefully—and that’s something I’ve been blogging a little bit about, the subprime auto loan market, for example—do you see an increased willingness of lenders to lend to certain borrowers even though we don’t really think those borrowers have better income prospects. What we call a supply-side shift. It’s not that the fundamental economic circumstance of the borrower has improved, but somehow the lender is much more willing to lend than they were before.

That to us is not conclusive evidence that something’s wrong, but at least it’s important for people to think about why the lenders are behaving this way. Why are they now willing to lend to someone they were not willing to lend to before, even though nothing about that individual has changed.

Today, you could say, and we wrote a couple blog posts about subprime auto issuances that got a lot of pushback, where people say, “look, lenders overreacted in 2009-2010, and they weren’t willing to lend to anybody. So of course, we see them more willing to lend to the same guy even though the guy doesn’t have higher income, but that’s just us returning to normalcy, not a bubble or a disaster.” And I think we have some sympathy for that. But in general, that’s what we look for in our data when we look for troublesome circumstances.

Do you have any concerns about how heavily large investors are getting into the housing market?

The very first blog post we did was on this point, and I gave a talk last year called “Will Housing Save the U.S. Economy?” The housing market rebound has been driven primarily by investors buying up foreclosed properties, and what does that mean for the macro economy?

My personal view is that I don’t think this represents necessarily a bad thing. If I’m given the option of a homeowner having to lever up a lot of debt, and therefore take on a lot of risk, versus that homeowner renting, I’d say that I prefer that individual renting. Take whatever money you were going to put down, and instead invest it in some broad securities, some broad financial portfolio. That’s what finance tells people they should do. I don’t think from that perspective it’s as worrisome.

Of course, there’s an entire industry promoting homeownership as the best thing ever, but I tend to think that those benefits are overstated. 

In terms of what it means for the macro economy, it is quite interesting. Because the housing rebound is primarily driven by investors, we shouldn’t expect it to have knock-on effects on spending like it did before. Because, to be frank, if you’re renting, your house price going up is a bad thing because you have to pay higher rents.

In the book, when you’re talking about debt forgiveness, the people you’re quoting—economists, government officials—when they’re talking about what’s a good idea, they’re talking in the past tense. “If we’d done more debt forgiveness, that would have worked.” Is it still too late for those solutions.

I basically agree; at this point, it’s too late. That’s not going to help a lot today. The last chapter of the book is rethinking the way we have the financial system operating. This model that we have now, which is what I’d call the credit-cycle of the world: lenders decide they want to lend more, they lend to a group of individuals that have very low income, very low credit scores, those individuals consume very aggressively, they buy cars, they have new credit cards, they spend. Then all of a sudden the lenders decide they don’t want to lend anymore. That is just unsustainable in the long run.

What we’re trying to propose in the last section in the book is a more sustainable financial system, that instead of just imposing risk on marginal borrowers, actually helps share risk. For example, the student-debt proposal we’re making, to make student debt contingent on aggregate employment outcomes, I think is very relevant today. Student debt has been exploding. If there isn’t an improvement in job market opportunities for young Americans, there’s just no doubt that defaults are going to skyrocket. So you want to think a little bit about whether you should write down student debt when the aggregate job market just looks so horrible.

If anything, the government has moved towards things like income-based repayment, which is very similar to our idea—if you don’t have income, you don’t have to pay the interest. A more systematic approach is needed in terms of thinking about how we should make student debt contracts contingent on aggregate labor market outcomes like unemployment.

The problem with student debt seems to be that it’s much less of a choice than buying a house. You can buy a house if you think it will be a good investment, but you can always rent. But a college education is practically a necessity if you want to move into the middle class.

That’s probably one of the strongest points in the book—does it make any economic sense to make an 18-to-22-year-old to bear this aggregate economic risk. It makes absolutely no sense. If there’s something the financial system should do, it is exactly to allow people to finance an education without taking on so much aggregate economic risk.

Of course, from an incentives standpoint, you want to make sure people understand that if people take on debt that they’re going to have to find a job, individually, that they can afford to pay back that debt. But if the aggregate economy collapses, and of course that’s not your fault at all, why should you be bearing that risk, instead of people with a lot of money who are financing that loan? It’s crazy, really.

That’s something that, with student debt, you see how little sense the financial system makes from the perspective of actually helping people finance an education without taking on so much risk themselves.

One thing that comes up in the book is “Helicopter Ben": that helicopter-drops of money would be the best way of getting the economy back on track, but it’s limited by the rules of the Fed. Is there any future for that?

No, essentially not. What we’re essentially saying is that the helicopter drops that make the most sense are targeted. They need to be targeted at the areas of the country where there’s too much debt. In that sense, once you have something like that, it really is a fiscal action, not a monetary action. I think that’s not going to happen.

We do talk about helicopter drops could help with inflation, but the experience that Japan’s going through now… a joke that we tell at the faculty lunch table is that inflation works so beautifully in macroeconomic models, but we actually have very little idea of how it works in the real world. What’s happening in Japan now is instructive. They’re doing what are essentially helicopter drops of money, they’ve been putting a lot of money into the economy. But what’s happening is that you’re getting inflation in goods prices, but you’re not getting wage inflation.

And that’s the worst-case scenario; that means real wages are going down, which is the opposite of what you want. I’ve always thought that these more targeted interventions, either on the fiscal side, or debt forgiveness, are much more advisable than relying on the central bank to try to use a sledgehammer to thread a needle. It’s a much harder game to play on the monetary side.

 

Share

Advertisement

Submit your comment