Prisoner’s Dilemma of No Principal Reduction

One of the reasons housing policy during the crisis was such a failure was because homeowners could see the policy coming down the pike and adjust their behavior accordingly– mostly in the direction of defaulting on their loans. For example, borrowers who found out that Countrywide had agreed to a mortgage modification for anyone who was 60 days delinquent made the rational choice of trying to become 60 days delinquent. Homeowners in judicial foreclosure states (New Jersey and Florida, for example) where kicking people out of their homes was pretty much impossible just stayed put, not paying their mortgage but not getting kicked out of their house, with the result that the downturn was initially less severe but the recovery was most likely slower.

The counterargument is that even if policies that kept borrowers in their homes slowed recovery, they also prevented the spillover effects from foreclosure and contagion from getting truly out of control (“each foreclosure filing leads to an additional 0.3 to 0.6 completed foreclosures within a 0.10 mile radius.”) Depending on banks’ losses on foreclosures, these spillover effects could put them in a prisoner’s dilemma where banks would be better off if all loans in an area were marked down, to reduce the incentive for presently underwater borrowers to default, or if foreclosures were temporarily halted in general, but each individual bank is better off foreclosing. At least that’s what I thought in 2011 and 2012; here’s an (unpublished) op-ed I wrote then arguing the case.

Depending on whom you ask, the $25 billion foreclosure fraud settlement negotiated this week between the attorneys general and the biggest banks is a victory for troubled borrowers, who will finally get some help—or a win for the banks, who will escape their just deserts. Truth is, it’s a win for both borrowers and banks.

The benefits to borrowers are easy to see. Families that have undergone foreclosure suffer worsened health and employment prospects, increased risk of homelessness, and lasting damage to their credit. Their neighbors also suffer—increased crime from vacant buildings and blight, lowered property values and greater difficulty relocating for a job, and strong incentives to default on their own loans and invite foreclosure themselves, if lowered property values have left them “underwater”—owing more than their homes are worth.

But foreclosures also cause big trouble for mortgage servicers and banks. Foreclosing and reselling a delinquent property is enormously expensive, and estimates show that lenders routinely lose half of the value of the loan. A glut of foreclosed properties is trouble for lenders in other ways, lowering the value of the homes they are trying to sell and encouraging more and more borrowers to stop paying their mortgage, as they see more of their friends and neighbors doing the same.

Many readers are, no doubt, taking out the world’s smallest violins to play mournfully for these poor troubled banks. But the question remains—if foreclosure is so expensive, why don’t more lenders renegotiate home mortgages? The Obama administration has tried a slew of programs and dangled big payouts to entice lenders to reduce principal and change the terms of loans, and overall, the programs haven’t worked, for the simple reason that the lenders only renegotiate the smallest sliver of delinquent loans, continuing with foreclosure on the vast majority. If foreclosures are such a bad deal for banks, why are they being so perverse?

The reason is that the banks are trapped in what game theorists call the Prisoner’s Dilemma. You know the story. Two bank robbers are caught by the police, brought to separate interrogation rooms, and given a choice. They can rat on each other, or they can stay shut. If neither one spills the beans, there is still enough evidence to lock them up on a lesser charge, and they both go to jail for two years. If they both rat on each other, they both are convicted of the bank robbery, and they both go to jail for ten years. If one squeals and the other stays shut, the one who stays shut gets twenty years in jail and the squealer walks away free.

You can picture one of the bank robbers’ fevered brain: if the other guy rats, I’m better off ratting, so that I get ten years instead of twenty. If the other guy stays shut, I’m better off ratting, so that I go free rather than get two years. No matter what my buddy does, I’m better off selling him out. The result? According to the logic of game theory, the two robbers will both rat each other out, even though it means they both get ten years instead of the two years they would have gotten if they both had stayed shut. In the absence of a repeated game (or an enforcement mechanism like an organized crime syndicate that will rub squealers out), rational bank robbers will squeal on each other (or ‘defect’, as the game theorists say) rather than cooperate.

What does this have to do with mortgage holding banks, who (as many readers might say), have robbed from the nation instead of being robbed from themselves? The key is to recognize who gets helped most when a bank forgives principal on a loan and avoids a foreclosure. It isn’t the bank that does the principal reduction– it is the other banks that hold delinquent loans in the same area and are intending on foreclosing. Suddenly, thanks to the bank doing principal reduction, foreclosure becomes much more profitable, since there is no longer such a glut of foreclosed homes on the market and a repossessed property can be disposed of more quickly and lucratively.

The banks would have most likely been better off if (way back in 2008) they all collectively reduced principal on their most overvalued loans, taken their losses up front, and stopped the fire-sale before it began. But for any individual bank at any individual time, it was better to foreclose than to reduce principal or renegotiate terms. Like the bank robbers ratting on each other even if they would have been better off both staying shut, the banks defected on each other by liquidating their entire portfolio as quickly as they could (particularly in states like Nevada or California that permitted rapid foreclosure). The result was that property values plummeted and encouraged far more borrowers to default, with even greater losses for the banks.

In the case of the bank robbers, an enforcement mechanism (like the Mafia) can help them to avoid defecting on one another and to escape the longer jail times that would result. In the case of the banks, the agreement with the state attorneys general could play the same role. By forcing lots of lenders to renegotiate loans simultaneously, the banks could be doing better for themselves, as well as for the families who thankfully can now stay in their homes.

5 thoughts on “Prisoner’s Dilemma of No Principal Reduction

  1. While I can’t fault your analysis of the outcome, but I am not sure about your analysis of the way the banks got to their actual course of action. I think there is a danger in analyzing large institutions as if they were individuals making decisions with imperfect knowledge. The decisions of the big banks, even though there may be one CEO, are a product of collective decision-making and inertia. I suspect that their bureaucratic rules, which were never made for an outlier event like the housing crisis, had them take the simplest route with each individual borrower, and they couldn’t react in time to the situation in the aggregate. Goal displacement I suppose.

    BTW, where do you publish Op-eds? Nevermind, in today’s climate anonymity is best.

    Liked by 1 person

    1. Yep- institutional sclerosis is a good guess as to how organizations behave in the real world. I had a little bit of a window into public finance agencies trying to deal with the housing crisis, and as they described it, they were just trying to avoid getting on the cover of the newspaper for doing something stupid- they didn’t have a really strong sense for how to proceed.


  2. It is not clear whare any government agency gets the power to force the alteration of contract terms on anyone any time, unless specifically authorized by duly enacted legislation. A messy free market (which includes States that have different enacted contract law) seems preferable to me, and most who give it serious thought.


    1. Yeah, I’m open to the idea I was wrong, for that reason. I’ve also wondered if the prisoner’s dilemma was one of the reasons we’ve seen so much bank consolidation, however, and if it created conditions for a lot of behind the scenes coordination among banks. I was also curious if markets with a single dominant lender were more likely to reduce principal, which I think was in fact the case.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s