How to design a better loan modification program

Posted by Barbara Kiviat Friday, February 13, 2009 at 12:17 pm

10 Comments

Step 1. Know what has—and hasn't—worked in the past.

Friends of the Curious Capitalist will be familiar with how much I like to go on and on about how there's no good data showing which sorts of mortgage rewrites keep struggling borrowers out of foreclosure in the long-term. And how it would be nice, before we spend $50 billion trying to solve the nation's foreclosure crisis, to have some evidence about what's likely to work and what's likely to only be a very expensive way of putting off dealing with the problem for a few more months.

So here's some good—though much-belated—news. The folks who regulate nationally chartered banks and thrifts are going to start releasing data detailing if modifcations increase, keep the same, or decrease a borrower's monthly payments, and then how well those modifications fare over time. (I won't go into the details here, but a lot of modifications, counterintuitively, don't lower what a borrower pays each month.) In March, we're supposed to get data on what's happened to loans in each category since receiving modifications in the first and second quarter of 2008.

This makes me happy. Make no mistake: it's unforgivable we weren't doing this this time last year, but better late than never. I especially like the newfound focus on the 60-day re-default rate. Early on, the folks at the OCC and OTS, the agencies bringing us this data, liked to talk about the 30-day re-default rate, which really isn't the right one to look at, because even people who are never going to wind up in foreclosure often mail in their mortgage checks a few days late (especially those with subprime loans).

But I still have issues. From what the OCC and OTS have said, it's not clear to me that they're going to tell us if, for example, a person who gets a permanent reduction in principal is more or less likley to keep his house in the long-run than, say, a person whose monthly payment is similarly reduced by an interest rate cut. There are reasons to believe that different pathways will have different long-term outcomes. Nearly 20% of mortgage holders owe more than their house is worth. Even if your monthly payment goes from unaffordable to affordable, you might still default because of the psychological weight of paying more for your house than you could get were you to sell it. Maybe in this situation keeping the interest rate high but forgiving some principal balance would be a better way to go.

The agencies have been collecting data on "modification type" since at least October 2007, but since one of the fields available to banks and servicers was "combination," the numbers don't come out clean. The OCC and OTS announced more in-depth reporting standards (PDF!) last month, but I'm not sure if detailed data on the specific types of modifications being made will be included in the March report. I really hope so. I'm getting a little tired of writing about this.

UPDATE: The folks at the OCC tell me that they will, in fact, have data on the types of modifications being made in next month's report. Very glad to hear it. Especially since CNBC is telling me that the Administration will roll out its big foreclosure-prevention plan next Wednesday. Let's learn from the data, people.

In other news... If you want some broader thoughts on how to fix the housing market, check out the story I have in this week's Time magazine (there's a woman praying—and being healed?—on the cover). The piece begins:

Here's a thought: let's have the government do something to fix the housing market. Now what would that be?

Ideas are flying around Washington. They include tax credits and cheaper mortgages for home buyers as well as leaning on lenders to rewrite mortgage terms for struggling borrowers. But maybe we should ask what, exactly, fixing the housing market means--and prepare ourselves for the limits of what these policies can actually accomplish. [Click here to read the rest.]

Barbara!