Atif Mian and Amir Sufi’s recent book, House of Debt, has received much praise from such prominent New Keynesians as Paul Krugman, Carmen Reinhart, and Christina Romer. Mian, of Princeton University and a former visiting scholar at the Federal Reserve Banks of New York and San Francisco, and Sufi, of the University of Chicago and the National Bureau of Economic Research, offer plenty of theoretical and empirical evidence for their analysis of the recent financial crisis and policy proposals for avoiding future crises. However, they often stretch (sometimes well beyond the breaking point) that evidence, leaving me unconvinced of many of their ideas.

Household debt and spending / The book opens with a compelling story of plummeting sales and layoffs during mid-2008 at Monaco Coach, a manufacturer of recreational vehicles with a heavy presence in northern Indiana. Mian and Sufi use that vignette to illustrate the broader calamity of the 2007–2009 “Great Recession” and promise “an evidence-based explanation for why the Great Recession occurred and what we can do to avoid more of them in the future.” That, they say, is something that “laid off workers at Monaco, like millions of other Americans who lost their jobs, deserve.”

Their opening chapter presents a lengthy time series of the ratio of U.S. household debt to income. It shows a steady climb from the early 1950s to about 2007, as both consumers and lenders grew increasingly comfortable with financing consumption. But then the housing bust came and there was a dramatic decline in the ratio as lenders and borrowers retrenched.

The authors are highly critical of the steep increase in household debt that occurred between 2000 and 2007. They present a wide variety of both U.S. and international evidence that deep downturns like the Great Depression and the Great Recession were preceded by such increases in household debt followed by large drops in household spending. In the case of the Great Recession, the spending decline began well before the massive government interventions in financial institutions that started in September 2008. The authors criticize the current financial system for “concentrat[ing] risk squarely on the debtor” and, in a populist vein, argue that the system “works against us, not for us” and “fosters too much household debt.”

The authors do some good analysis early on in presenting the distribution of the roughly $5.5 trillion in losses in home values during the Great Recession. They break down homeowners into five groups based on net worth as of 2007, revealing that, going into the Great Recession, the poorest homeowners were highly leveraged and had their net worth heavily concentrated in home equity (about $4 out of every $5). As a result, they were particularly devastated by the ensuing housing crash. The richest homeowners were the mirror-opposite, with comparatively low leverage and a low concentration of net worth in home equity (about $1 out of every $5).

The authors should have seized on that point to discuss how ill-advised the Clinton administration’s interventionist housing policies were during the 1990s, as well as the George W. Bush administration’s continuing those policies in the 2000s. The policy objective was to entice low net worth individuals into homeownership through expanded leverage. Much later in the book (chapter 6), Mian and Sufi do mention the “historic” rise in homeownership starting in the 1990s, but somehow they fail to make the connection to government policies and the role they played in that dramatic increase. They only decry the current allocation of risk in debt.

Stretching the evidence / Mian and Sufi structure the book in a manner that lays out a number of important questions and responds to them with dozens of case studies that seem to support the authors’ policy conclusions. Some of the case studies are supported by hypothetical examples, others by academic research the authors find favorable to their viewpoint, and some present their viewpoint on current policy issues. This viewpoint consistently calls for a heavy dose of government intervention and redistribution in the system of consumer and mortgage debt: limiting reliance on nonjudicial foreclosures, forcing write-downs of household debt, mandating changes to consumer debt and mortgage contractual provisions. A major problem for me is that their case studies are quite often flawed, which undermines the credibility of their analysis and the arguments for the policy changes Mian and Sufi advocate. Although it is not possible to deconstruct all of their case studies in this short review, a few examples will illustrate my point.

One of the first case studies in the book concerns the “harshness of debt” and the supposed one-sided nature of it. Supposedly, debt disfavors debtors because they are in a “first losses” position, but favors the lenders and savers who fund it. Mian and Sufi offer the hypothetical example of a home purchased for $100,000, with the borrower taking out a loan of $80,000 and making a down payment of $20,000. An aggregate 20 percent drop in home prices is then assumed, which wipes out the borrower’s equity. The authors claim that the lenders and savers’ position “improves” after the price drop because they now “own 100 percent of the home.” But surely that is not an improvement: Before the 20 percent crash, lenders and savers had an $80,000 first claim on the home. After the crash, they still have just an $80,000 first claim on the home, but that claim is now at greater risk because of the increased likelihood that the homeowner, who no longer has an equity stake, will walk away from the mortgage.

The authors’ main point in this example is that the system is rigged against borrowers, and so lending contracts should be rewritten. But this narrative is disingenuous in its explanation of the allocation of risk in borrowing arrangements. Like any “equity” participant in a deal, borrowers take on risk, but they also have the upside potential from an “up” market and most housing markets have been up over the past few years since the worst of the crash. Lenders and savers assume comparatively less risk, but they also have a cap on their upside: the principal they will get back is limited to what they loaned, with no equity kicker. Mian and Sufi give nary a mention to that concept.

Another example of flawed analysis is the authors’ hypothetical of a married couple in their late 50s, approaching retirement. The authors assume that the couple gets caught up in the above example: they have 20 percent equity in their home and the market undergoes a 20 percent price collapse. Because of the collapse, they “no longer have sufficient wealth to cover their planned spending in retirement. As a result, they cut spending in order to build up savings.” Here the authors compose a sad narrative in order to justify their policy proposals. But their hypothetical does not withstand close scrutiny; Mian and Sufi’s numbers capture few if any “real world” examples.

The average first-time homebuyer is in his mid-30s. One reason this age is not higher is that lenders are hesitant to offer a 30-year mortgage to someone who will likely enter retirement before the loan is paid off. So someone who was a first-time homebuyer in his 30s will have been a homeowner for 20–25 years when he reaches the age of Mian and Sufi’s hypothetical couple. At that point, unless the couple has been “using their home as an ATM” (i.e., taken out a second mortgage or home equity loan), they should have equity in the range of 50 percent or more, not the mere 20 percent used by the authors that was so easily wiped out in a down market.

Write-downs / The authors then transition from hypotheticals to specific policy issues. They apply equally flawed logic to a critique of Sen. Bob Corker’s (R–Tenn.) opposition to legislatively requiring principal write-downs of underwater mortgages—that is, banks (and their depositors and investors) would be made to forgive a portion of the outstanding loan. The write-down would certainly benefit the underwater homeowner, but what about the lenders?

Corker realizes that his state’s residents, who were not heavily swept up in the housing bubble, would be cross-subsidizing residents in states whose typical mortgage is deeply underwater (Nevada, Florida, Arizona). Mian and Sufi acknowledge that Tennessee

mostly avoided the housing boom and bust…. [H]ouse prices in Tennessee rose by only 25 percent, far below the 60 percent growth in California and Florida…. Households in Tennessee came into the recession with debt levels well below the national average.

But then they go through a convoluted and largely unrelated discussion regarding the indirect effect of economic weakness in a state with underwater mortgages on a state like Tennessee (which has a high concentration of auto manufacturing jobs). Their redistributionist argument that Tennessee residents should support write-downs for underwater mortgages is unconvincing, especially given that over time those mortgages have been getting less and less underwater (or even getting above water) as home values have recovered.

Another tendency of the authors is to make overly broad statements regarding the mortgage market. For instance:

Foreclosures are bad for everyone. They kick families out of their homes, depress house prices, and elicit major losses for lenders, who typically want to avoid foreclosure except in extreme circumstances.

Okay, sure, foreclosures are bad. But that does not mean that government intervention intended to prevent foreclosure must be better. Yet, Mian and Sufi seem to believe that is the case, arguing that there must be some sort of “market failure.” Their reasoning ignores the social benefits of foreclosure: without it and the added financial security it provides, most lenders would not even consider extending funds for a home purchase.

On that issue, Mian and Sufi personally (and rather viciously) attack Edward DeMarco, former head of the Federal Housing Finance Agency that oversees Fannie Mae and Freddie Mac, arguing that he impeded legislation to require write-downs. The authors seem not to understand that DeMarco’s agency is obligated to conduct its operations “in a manner [that] maximizes the net present value return from the sale or disposition of such assets” and “[that] minimizes the amount of loss realized.” They only seem to care about their redistributionist viewpoint and the fact that “principal forgiveness would have resulted in a more equal sharing of the losses associated with the housing crash.”

Conclusion / The final chapter of House of Debt has the curious title of “Sharing.” It demands a series of presumably compulsory changes in the precise terms and conditions governing contracts in the consumer and mortgage market to address the inflexibility of debt contracts. Those changes include “risk sharing” in student loans (“recent graduates should be protected if they face a dismal job market upon completing their degrees. In turn, they should compensate the lender more if they do well”); and “shared responsibility mortgages” (“[1] the lender offers downside protection to the borrower, and [2] the borrower gives up 5 percent capital gain to the lender on the upside”). In tandem, the authors decry the fact that “the government thus pushes the financial system toward debt financing, even though debt financing has horrible consequences for the economy,” but imply that if their preferred contract provisions are imposed, this pushing can continue.

House of Debt provides some useful questions for consideration in the wake of the recent financial crisis. The authors are right that government policy puts a collective “thumb on the scale” in favor of debt financing. However, Mian and Sufi’s suggested policy responses of even more government micromanagement of the terms of consumer and mortgage lending is completely off base. That is probably why only their fellow Keynesians believe those proposals have merit. If these are truly desirable provisions, then there are plenty of market incentives for them to be adopted voluntarily. The fact that they have not been adopted, and that Mian and Sufi want government to force them on borrowers and lenders, is insightful.