��ɫApp

In early 2009, in the depths of the mortgage meltdown, President Barack Obama launched a multi-billion-dollar effort to stem the flood of home foreclosures.

It was called the Home Affordable Modification Program (HAMP), and it aimed to help families keep their homes by offering incentives to banks and loan-servicing companies that modified mortgages of troubled borrowers.

The idea was to correct what economists call a “market failure,” because foreclosures can be a losing proposition for everybody involved. Not only do borrowers end up losing their homes, but a bank’s loss from a foreclosed mortgage can actually be higher than the cost of negotiating more favorable terms with the homeowner. Foreclosures also drag down the value of surrounding properties, creating wider losses by depressing the overall housing market.

HAMP offered modest financial incentives for banks and the companies that facilitate mortgages (servicers) to negotiate loan reductions with at-risk homeowners. The government would pay the mortgage lenders $1,000 for every loan they modified, as well as annual $1,000 “success payments” for every borrower who stayed current over the subsequent three years.

Did it work?

Yes and no, according to a sweeping  coauthored by Amit Seru, a senior fellow at the ��ɫApp Institute for Economic Policy Research and the Steven and Roberta Denning Professor of Finance at ��ɫApp Graduate School of Business. The study finds that HAMP did indeed produce 1 million additional permanent loan modifications and prevented about 600,000 foreclosures that otherwise would have occurred.

Organizational Lapses

Those are good results, but they fell well short of hopes. The Obama administration had hoped to spur as many as 4 million loan modifications and prevent millions of foreclosures.

The main reason for the shortfall, the study concludes, was surprising: Many big banks and mortgage-servicing companies apparently weren’t up to the job, even if it was in their financial interest. A significant share of those companies lacked the organizational capacity to renegotiate large numbers of loans and opted not to make internal changes that would have enabled them to take advantage of the program.

The new study is based on analysis of data from 60% of all mortgages that were outstanding in the United States at the time of the mortgage collapse. In addition to Seru, the authors were Tomasz Piskorski at Columbia University; Souphala Chomsisengphet, an economist at the Treasury Department’s Office of the Comptroller of the Currency; Itzhak Ben-David at Ohio State University; Gene Amromin at the Federal Reserve Bank of Chicago; and Sumit Agarwal at the University of Singapore.

To assess the impact of HAMP, the researchers compared borrowers who had essentially the same credit characteristics and the same kinds of loans but differed on meeting the program’s eligibility requirements. For example, the program wasn’t available to borrowers with mortgages above $729,750 or, initially, to borrowers who weren’t actually living in the mortgaged homes. The difference in loan modification rates between those otherwise similar borrowers was a measure of the program’s effect.

Discrepancies Between Lenders

In many respects, the researchers concluded, HAMP worked as its supporters had hoped. The 1 million loan modifications it spurred reduced borrowers’ monthly payments by an average of about 25%, or $400 a month. People who received the modifications were 12% less likely to go into foreclosure.

HAMP also appears to have helped stabilize housing prices. In regions where mortgage companies were actively using the federal incentives to negotiate loan modifications, home prices recovered slightly faster than average. Indeed, those regions also showed other signs of improving financial health: less delinquency on other kinds of consumer credit, such as credit cards and home equity loans, and stronger growth in the average household’s spending on big-ticket durable goods. Last but not least, the researchers found very little evidence that the government program “crowded out” private efforts at loan modification, implying that there was an expansion of renegotiation activity in the economy due to the program.

So why was the number of loan modifications so disappointingly low?

It turns out that there was a huge difference among banks in their willingness to take advantage of the federal incentives. Some mortgage servicers pursued four times as many loan modifications as others, for reasons that had no perceivable correlation with differences in their loan portfolios. The laggard companies held 75% of all the loans examined.

If the laggard mortgage servicers had been as active as the others, the researchers estimated, the number of permanent loan modifications would have reached 1.7 million, or 70% higher than it actually was.

Staffing Issues

The researchers found no evidence that the laggard lenders were held up by unusually troubled borrowers. They were simply less interested than other institutions in modifying loans, and they had been that way both before and after the Treasury started offering incentives.

On closer examination, Seru and his colleagues concluded that the laggard banks didn’t have as much organizational infrastructure for dealing with customers. The laggards generally had less staff for loan servicing, as measured by the number of loans per employee, and they spent fewer hours training employees. Their telephone call centers also had longer holding times and higher numbers of dropped calls.

The researchers say there were ways to solve the problem. For example, they say, the program could have made it easier for unprepared banks to transfer their loans to those with better infrastructure. In the commercial real estate market, it is common for lenders to automatically transfer non-performing loans to a special servicer that’s equipped to deal with work-outs.

“The moral of the story is that policymakers have to think about the nature and organization of the intermediaries,” Seru says. “Policymakers who want to encourage more renegotiation have to take into account that some banks just don’t have the organizational design conducive for such activity.”

This  first appeared in Insights by the ��ɫApp Graduate School of Business.