"Underwater, But Will They Leave the Pool?"

Richard Thaler on walking away:

Underwater, but Will They Leave the Pool?, by Richard H. Thaler, Commentary, NY Times: ...Why is the mortgage default rate so low? After all, millions of American homeowners are “underwater”... Yet most of them are dutifully continuing to pay their mortgages, despite substantial financial incentives for walking away...

Some homeowners may keep paying because they think it’s immoral to default. ... But does this really come down to ... morality? A provocative paper by Brent White, a law professor at the University of Arizona, makes the case that borrowers are actually suffering from a “norm asymmetry.” In other words, they think they are obligated to repay their loans even if it is not in their financial interest to do so, while their lenders are free to do whatever maximizes profits. ...

That norm might have been appropriate when the lender was the local banker. More commonly these days, however, the loan was initiated by an aggressive mortgage broker who maximized his fees at the expense of the borrower’s costs, while the debt was packaged and sold to investors who bought mortgage-backed securities in the hope of earning high returns, using models that predicted possible default rates.

The morality argument is especially weak in a state like California or Arizona, where mortgages are so-called nonrecourse loans. ... Nonrecourse mortgages may be viewed as financial transactions in which the borrower has the explicit option of ... walking away. Under these circumstances, deciding whether to default might be no more controversial than deciding whether to claim insurance after your house burns down.

In fact, borrowers in nonrecourse states pay extra for the right to default without recourse. In a report..., Susan Woodward, an economist, estimated that home buyers in such states paid an extra $800 in closing costs for each $100,000 they borrowed. These fees are not made explicit to the borrower, but if they were, more people might be willing to default, figuring that they had paid for the right to do so.

Morality aside, there are other factors deterring “strategic defaults,” whether in recourse or nonrecourse states. These include the economic and emotional costs of giving up one’s home and moving, the perceived social stigma of defaulting, and a serious hit to a borrower’s credit rating. Still, if they added up these costs, many households might find them to be far less than the cost of paying off an underwater mortgage.

An important implication is that we could be facing another wave of foreclosures, spurred less by spells of unemployment and more by strategic thinking. ... So far, lenders have been reluctant to renegotiate mortgages, and government programs to stimulate renegotiation have not gained much traction.

Eric Posner ... and Luigi Zingales ... have made an interesting suggestion: Any homeowner whose mortgage is underwater and who lives in a ZIP code where home prices have fallen at least 20 percent should be eligible for a loan modification. The bank would be required to reduce the mortgage by the average price reduction of homes in the neighborhood. In return, it would get 50 percent of the average gain in neighborhood prices — if there is one — when the house is eventually sold.

Because their homes would no longer be underwater, many people would no longer have a reason to default. ... Banks are unlikely to endorse this if they think people will keep paying off their mortgages. But if a new wave of foreclosures begins, the banks, too, would be better off under this plan. ...

This plan, which would require Congressional action, would not cost the government anything. It may not be perfect, but something like it may be necessary to head off a tsunami of strategic defaults.

I think that people in non-recourse states understood the option a bit differently. If medical costs wipe you out, if the demand for the widgets you produce falls permanently causing you to lose your job and also have trouble finding a new one, or if other things out of your control cause you to be unable to pay your mortgage, then you won't lose your car, furniture, heirlooms, etc. in a forced liquidation to pay of as much as possible of the remaining balance on the housing loan. Non-recourse protects you fro losing everything. But a change in the price itself wasn't part of the deal. You get to keep the upside, but have to eat the downside - that's how it worked and you knew that going in. At least, that's how I always understood the implicit deal (enforced in part by a fear of losing access to credit in the future, social norms, etc.).

If you were underwater and lost your job and had to move to get a new job, that was one thing, there was little choice but to default and use the protection embedded in non-recourse. But simply walking away when you were still employed and could still afford the mortgage was another. That wasn't the option embedded in the implicit contract. Following this implicit rule lowers costs for everyone, that's the sense in which, contrary to claims above, there's a financial incentive to follow this norm -- should I pay more for my loan so that you can speculate and then walk away from a bad bet (there are unrecoverable costs each time a default is socialized)? That's different than using non-recourse as a form of social insurance against contingencies beyond a household's control, and paying extra closing costs for that insurance.

The change in norm will occur when people begin to believe that they weren't just unlucky, but instead were duped or treated unfairly in some other way. If it wasn't just a bad bet, the kind you reluctantly pay when you lose, but was instead caused by some unfair factor that only becomes evident ex-post, then the norm begins to change as people begin to realize what really happened to them. They don't want to believe or admit to themselves that they were fooled into a loss rather than the victim of a fair bet, but that changes as the losses and resentment mount, and as the evidence that things weren't what they seemed comes into focus. And that does seem to be happening.