Not So Fast, or So Simple

Simple mortgage modification programs may encourage a high rate of strategic default for those whose homes aren’t at risk of foreclosure.
August 23, 2011 | Research Feature
Print this page

Millions of homes in the United States have been lost to foreclosure, and several years in, there are few signs that the crisis will ease up any time soon. Millions more borrowers may yet default on their mortgages and face foreclosure. Meanwhile, policy makers have yet to reach a consensus on the most effective solution for assisting borrowers in order to avert these pending foreclosures.

Offering mortgage modification to at-risk borrowers is not just an act of goodwill. Foreclosures cost borrowers, lenders, and society as a whole. A neighborhood where many homes go unmaintained after being foreclosed can see property values drop. As its tax revenue sinks, local services like schools and fire departments strain to meet demand and the quality of life for all residents suffers. For lenders, the foreclosure and resale process may, in some cases, be more costly than modification.

Yet large-scale modification efforts have not succeeded in preventing millions of foreclosures in the United States, as many proponents of such programs had claimed, in large part because of challenges in implementing such programs in a simple and transparent way. It’s easy to see why. It can be difficult to identify which borrowers need help. While an estimated 10 to 15 million households look, on paper, like they are at risk of missing a payment and going into default, most of these households continue to make payments. When these at-risk borrowers do default, behavior varies. Some default as soon as they go underwater — when the value of their home drops below what they paid for it. Others wait, possibly deterred not only by the threat of losing their homes, but also by the prospect of a blemished credit score that will endanger prospects for securing a future mortgage, credit card, or student loan.

To help at-risk borrowers avoid foreclosure, mortgage modifications are focused on decreasing monthly payments for the borrower. Extending modifications to those who are already delinquent on their mortgages is one option. If a bank cannot identify which of its at-risk borrowers will default, missing payments is surely a signal that borrowers want help and would accept it. The complication with this seemingly straightforward solution though, is that it risks strategic behavior on the borrowers’ side — that is, borrowers missing mortgage payments with the intention of attaining a modification even if they could afford to make their payments — potentially triggering a wave of delinquencies by borrowers who would otherwise continue to pay their mortgages.

This question is at the crux of the policy debate, says Professor Chris Mayer, who has advised the debate and testified before Congress throughout the foreclosure crisis. “Opportunists might respond to fixed rules strategically, and the people who do that are likely to be the more sophisticated people who don’t need the help,” Mayer says.

Policies that use more elaborate verification processes seem more likely to limit strategic behavior by borrowers — checking income, employment, savings, and the like, it’s thought, will discourage opportunists seeking unneeded mortgage modifications. But verification is costly, and, because it takes more time than simpler programs, modifications may come too late for the borrowers who most need it. If verification rules are too stringent or complicated, many people who probably should get help might not qualify. Critics have pointed to the Home Affordable Mortgage Program of 2009, citing its abstruse rules as a reason for its limited effectiveness.

“The nightmare scenario for every policy maker is that you try to help five or six million at-risk borrowers and you put 40 million borrowers into delinquency because everyone starts missing payments in order to default and reap its benefits,” says Professor Tomasz Piskorski, whose expertise includes real estate finance and asset securitization and who frequently collaborates with Mayer. “So do we want to have simple programs that extend help quickly but risk strategic behavior, or do we want to spend more money and time on costly verification procedures to limit strategic behavior?”

Mayer and Piskorski, working with Professor Edward Morrison of Columbia Law School and Business School doctoral candidate Arpit Gupta, with support from the School’s Milstein Center for Real Estate, sought a way to determine the extent to which people are, in fact, willing to game a simple modification program.

When Countrywide Financial was sued by a large number of states’ attorneys general in 2008 for deceptive lending practices, the firm agreed, as part of the subsequent legal settlement, to offer a simple, broad-based mortgage modification program to borrowers. Countrywide financed approximately one-fifth of all subprime home loans in the United States in the early 2000s, and a number of features of the settlement presented an ideal test case that allowed the researchers to approximate a randomized experiment. The modification program was likely unanticipated by most borrowers and was available to all Countrywide subprime borrowers with certain types of loans. Since other lenders also offered the same type of loan, the researchers could rule out that there was anything unique about the terms Countrywide gave its subprime borrowers versus what other banks gave borrowers who took out the same types of loans. Countrywide’s plan had very simple eligibility requirements with little screening or additional documentation required — any borrower who missed mortgage payments was considered eligible for help (although not all received it).

The researchers were able to gain unprecedented access to detailed individual (but anonymous) credit data from Equifax and mortgage default data from BlackBox Logic for millions of borrowers. Linking the sets of data allowed the researchers to make two different comparisons: they could observe the behavior of Countrywide borrowers and their credit ratings before and after the modification program was announced, and they could compare Countrywide borrowers who were offered the modification program with other lenders’ borrowers who were not offered a modification option to see how both groups behaved over the same period of time.

The researchers attributed strategic default to instances where Countrywide borrowers who were eligible for the modification program defaulted (defined as missing at least two mortgage payments) but otherwise comparable borrowers from a different lender did not similarly miss payments. Mayer and Piskorski’s results indicated that 20 percent of these Countrywide borrowers — who they estimate would have otherwise stayed current on their payments if not for the modification offer — went into delinquency, compared with similar borrowers who did not have the option of entering a simple mortgage modification program. In addition, the borrowers most likely to respond strategically were those already in the best financial shape of all borrowers, those with the most available credit and lowest loan-to-value ratios. “The results suggest that there is a sizable group of households for whom moral considerations and the threat of bad credit are not sufficient deterrents to strategic default,” Piskorski says.

That 20 percent may not fully reflect the potential for strategic default. “Our research looked at just a few months after the program was announced, so it likely captured people who follow the news closely and responded fairly quickly,” Piskorski continues. “In a national, broadly advertised program the response is likely to be larger, partly because over time people could advise family and neighbors to default, and the effect would trickle down.”

Mayer adds, “20 percent is a number that says strategic behavior is not the only thing we should think about when pursuing modification programs. If you want to target your resources most effectively, it may mean spending time and effort to find out who will benefit most from mortgage modification.”

Christopher J. Mayer is the Paul Milstein Professor of Real Estate in the Finance and Economics Division at Columbia Business School and co-director of the Richard Paul Richman Center for Business, Law, and Public Policy at Columbia University.

Tomasz Piskorski is associate professor of finance and economics at Columbia Business School.

Christopher Mayer

Professor Mayer is Paul Milstein Professor of Real Estate and Finance and Economics at Columbia Business School. His research explores a variety of topics in real estate and financial markets, including housing cycles, mortgage markets, debt securitization, and commercial real estate valuation. Dr. Mayer is also a principal at Longbridge Financial, a new and innovative company focused on developing and delivering reverse mortgage products...

View full profilePersonal Website

Tomasz Piskorski

Tomasz Piskorski is the Edward S. Gordon Associate Professor of Real Estate at Columbia Business School and the Paul Milstein Center for Real Estate. He is also a Research Associate at the National Bureau of Economic Research and served as a board member of the Finance Theory Group. He received a M.S. in Mathematics from New York University Courant Institute of Mathematical Sciences...

View full profile

Read the Research

Christopher Mayer, Edward Morrison, Tomasz Piskorski, Arpit Gupta

"Mortgage Modification and Strategic Behavior: Evidence from a Legal Settlement with Countrywide"


Download PDF View abstract/citation
View Research