Sunday, February 10, 2008

Let's Talk about Walking Away

by Tanta on 2/10/2008 01:58:00 PM

Much has been reported, on this blog and elsewhere, about claims by industry participants that affluent (or at least fully solvent) borrowers who could afford to pay their mortgages are walking away from them, particularly in California where lenders cannot pursue deficiency judgments on purchase-money loans. I have seen two media articles (here and here) in the last few days actually crediting us—a blog!—for being on this and related stories, so I'm going to take this opportunity to talk about where all the reporting on this subject might go from here.

I certainly appreciate the willingness of professional media outlets to attribute this blog. It suggests that editors are getting over their preference for misleading information from named sources with institutional affiliations and a book to talk up (Hi, NAR!) over solid information from anonymous bloggers who are accumulating credibility the hard way (by, you know, being credible). So I don't want the following to sound like carping from a Blogger Who Is Never Happy. I mean this to be quite constructive.

I actually believe that reporters should never abandon their skepticism anywhere, including here. This is not simply because you must evaluate us as a source: it is because the ideal result of your hanging out here is that you take what we come up with and improve it, by bringing to the table reportorial skills and access that we may not necessarily have. If we can function most of all to give you the background knowledge in how these things work, and sufficient exposure to the issues to help you know what questions to ask of your sources, we're happy to get "scooped" by you. We've done our part.

The "walking away" story is a great place to think about this idea. What we have, so far, is a series of industry insiders making a general claim that "ruthless default" is on the rise. What we do not have, so far, is any rigorous quantification of the extent of this problem, or even any really detailed definition of what "a borrower who could afford to pay" is. We have no one offering baseline measures (what, for instance, a lender's analytical models might have predicted is the "normal" level of walking away), and hence no clear sense of the magnitude of the "change" in borrower behavior and attitudes (not to mention much rigor in distinguishing between the two). Hence, we don't yet really know if it's a change in borrower behavior as much as a change in definitions, servicer data collection and interpretation, or media exposure. Or a handful of anecdotes that are being pluralized into "data."

Let me give you a for-instance: here's the claim from a recent WSJ article:

And some borrowers, even those who can theoretically afford to keep their homes, realize they owe much more than what comparable houses in the neighborhood are selling for -- and think that prices won't rebound anytime soon. So they're walking away, according to anecdotal reports as well as recent statements by top executives of both Wachovia and Bank of America. [My emphasis]
But what made them "realize" this? How do we know what they're thinking about future price recovery prospects?

We do have some data indicating that borrowers in general are not, on the whole, likely to be highly-informed about the current value of their homes unless they are actively trying to sell or refinance.
NEW YORK (CNNMoney.com) -- Despite numerous reports showing home values in historic decline, more than three out of four homeowners believe their own home has not lost value in the past year, according to an online survey.

The survey was conducted by Harris Interactive for Zillow.com, a Web site that gives estimated home values.

The survey of 1,619 homeowners found 36% believe their home has increased in value, and another 41% believe their value has stayed the same. Only 23% believe their home has lost value. . . .

Moore said it's important to remember that only a small fraction of homeowners try to sell their home in any given year, and unless they are trying to get new financing or a home equity line of credit, there's no reason most will be confronted with the decline.
Stories--for what they're worth--from the Option ARM world suggest we have at least a few borrowers who are not even very clear on how much they currently owe until they try to refinance, let alone what their home value is.

So is the claim here that people recognize that they are increasingly "underwater," attempt to sell or refinance, fail at doing so, and then decide to "walk away"? If so, how did they really know how far underwater they were before they tried listing the property or getting a refinance appraisal? If we are talking about a subsection of the borrower pool who religiously monitors such things as the current comparable sale data in their neighborhoods, and who are unsentimental enough to realize that their own homes aren't "special cases," then how big is this group? I'm quite sure that our commenting community is overrepresented there, but how representative are we of the wider world?

Or are we talking about borrowers in neighborhoods with high vacancy levels (such as unsold new developments) or already-high numbers of for-sale signs planted along the street, who cannot help but notice that either nobody wants to or is able to move in or everybody else seems to want out?

Mish at Global Economic Analysis reports the following anecdote from a reader:
The first house in our subdivision was foreclosed about 9 months ago. That wasn’t a walk away; that was a get out notice from the sheriff. A few months later, there was another house that had a for sale in the front lawn, and the owner moved out a few weeks later. Another house went on the market one day and the owner loaded up a U-haul the next and drove away at 4am. And since September, we have seen six more homes that are “for sale” but the owner is long gone. . . .

After reading your work, I began to examine the attitudes of my neighborhood. The first foreclosure was one of those borderline families you often write about. They were in over their heads and couldn’t afford the house they were living in. Most likely mortgage reset. In any event, the scuttlebutt around the neighborhood was one of scorn, shame, and embarrassment. No thought was given to the negative impact to the value of all our homes in this subdivision. With each subsequent “pre-foreclosure,” people’s attitudes softened about their ex-neighbors. Gone was the Scarlet F; it was replaced with empathy, understanding, and even compassion. Maybe the attitudes have changed because people now realize that the value of their homes have fallen off a cliff. They don’t have time to shame their ex-neighbors when they are worrying about their wealth is being vaporized or a $400 natural gas bill or car payment or the kids or whatever.
This anecdote links borrower distress, listed homes and vacancies in the immediate (visible, visceral) area, and shifting attitudes (not, interestingly, a direct report of the attitudes of the "walkers," but of the neighborhood witnesses thereto) in a way that seems difficult to untangle. I have, for instance, personally moved my household by starting out at 4:00 a.m. in a U-Haul, but I'm a coward who doesn't want to drive a long truck she's not used to handling around the city during morning rush hour, and who prefers driving in the dark in the familiar area and still having daylight at the new area. Are we even sure that what we are witnessing is "furtive" moving?

My guess is that servicers do not have the kind of information that would let us answer these questions directly. The term "jingle mail" is a fine joke (the mail "jingles" because the borrowers are just sending the keys to the servicer instead of a mortgage payment), but we need to bear in mind that it's a joke. True "walk-aways" do not call or write to the servicer to inform them of their intent to stop payment permanently and wait to see how long it takes to foreclose. (There are always some borrowers who request a deed-in-lieu when they are in distress, but that's not really what we mean by "jingle mail" or "walking away," which implies that the borrowers are letting the banks foreclose, not voluntarily surrendering the deed.) It isn't always easy, then, to identify intentional walk-aways in your foreclosure caseload.

In fact, it seems possible that the borrowers being labelled "walk-aways" are those who 1) do not respond to servicer attempts to contact them at the first or subsequent delinquencies, and/or do not respond to offers of loss-mitigation efforts (forbearance, modification, short sale, anything short of foreclosure) and 2) do not show financial distress as indicated by the servicer's review of a current credit report. If they aren't responding or cooperating, they aren't sharing details of their current income or expense situation with the servicer; it seems probable to me that the servicers are deciding that these folks could carry the mortgage payment, if they wanted to, because they have pulled a credit report and find no evidence of increased debt levels from origination or defaults on non-mortgage debt.

The no-contact borrower is a difficult one to make assumptions about, since any servicer will tell you that borrowers in true economic distress caused by circumstances well outside of their own control are quite often non-responsive: depression, shame, fear, and having had the phone cut off, among other things, often keep people who could be helped from getting help. A Freddie Mac credit policy expert notes:
Unfortunately, as detailed in a groundbreaking study conducted by Freddie Mac and Roper Public Affairs in 2005, 61 percent of delinquent borrowers did not know that there are workout options and significant percentages of those borrowers did not return lender phone calls out of embarrassment or a lack of faith that anything can be done to help them.
There is some difficulty, then, in deciding whether a "no-contact" problem involves shame or shamelessness. It is therefore unwise, it seems to me, to assume that all borrowers who seem to just "disappear" are "walk-aways."

Similarly, the assumption that a borrower's current credit report proves that they can carry the mortgage payment is fraught with difficulty. As we have seen in a number of recent reports, the customary assumption in mortgage servicing was that borrowers in distress would prioritize payments such that they would skip the credit cards, personal loans, and auto loans (in that order) before failing to make the mortgage payment. When that pattern held true, one could more confidently assume that a borrower current on all other obligations was probably able to make the mortgage payment.

However, we seem to have some observers suggesting that the divergence between mortgage and credit card delinquency rates indicates that borrowers are skipping the mortgage payment first and keeping the cards current when they are in financial distress. They may well, for instance, believe that in a temporary reduction of income, like a layoff, it's more important for them to keep revolving credit lines open for emergencies, and to keep their cars for transportation to work, than to worry about foreclosure on the mortgage, which they probably know will take some time and can probably be reinstated when they are employed again. Surveys still show, however, that the overwhelming majority of distressed borrowers indicate that they would prioritize the mortgage payment over other debt. An Experian study from last year accounts for the contradiction here by indicating that while prime borrowers still pay the mortgage first, subprime borrowers are more likely to keep the credit cards current and let the mortgage payment lapse. There does seem reason to question the automatic assumption that a clean credit report (in terms of non-mortgage debt) automatically means that the borrower in question could afford the mortgage payment but is choosing not to make it.

There is evidence that substantial numbers of foreclosure "cures" (loans with an initial foreclosure filing that do not result in ultimate foreclosure) are in fact due to borrowers making up missed payments, not to lender workouts. From the Conference of State Bank Supervisors' State Foreclosure Prevention Working Group:
The October data from Reporting Servicers shows that most mortgage payment delinquencies are resolved by action taken by the homeowners themselves. Of the loss mitigation efforts closed in October, 73% of all resolutions were due to the borrower bringing the account current.
This is not to suggest that workouts are unnecessary because most borrowers can find the money to bring their loans current; it is merely to recognize that there are borrowers out there bringing their loans current out of some source of funds other than sale or refinance. It would certainly be useful to know what that source is; I have yet to see any reported data on that, and servicers may not be collecting it consistently. Are borrowers suffering from temporary loss of income? Temporary increase in obligations? Are they paring household budgets down to ramen and bus fare in order to make up mortgage payments? Are they raiding retirement accounts to stay in their homes? If any of these things is true, that complicates the picture of ruthless walk-aways, since whether it is wise or not to make severe financial sacrifices to keep a home right now, if folks are making those sacrifices, they aren't thinking like the "rational agents" of options theory gone awry.

I suspect--but have no data to prove--that servicers are also extrapolating from bankruptcy filing rates here, the assumption being that if a borrower were truly unable to make the mortgage payment and didn't want to keep the house, he or she would file for bankruptcy. But it is also not clear to me that a borrower's failure to file for bankruptcy necessarily means that the borrower's income is sufficient to carry the mortgage, given how stringent means-tests in BK have become. I for one am not willing to assume that borrowers are not in valid economic distress just because they don't meet BK filing requirements or are unwilling to try to survive on Chapter 13 budget plans.

Aside from the obvious reasons to care about the extent to which "walking away" is a significant part of the problem or just rare but annoying food for moralists, headline writers, and bankers who want to blame their borrowers' morals rather than their own for life's troubles generally and earnings reports particularly, there is the danger that more legislation will be passed--the 2005 bankruptcy reform springs to mind here--to combat "bad consumer behavior" while not coincidentally attempting to bail lenders out of stupid credit-granting practices.

Here's a proposal, made by a writer with whom I probably agree on a lot of things (but not this):
In the future, Congress should require California to allow lenders to garnish wages of affluent borrowers who walk away from their homes. It's dishonest to have it both ways: (1) federal tax money backstops investor and bank losses when homeowners walk away from homes, and (2) California law allows homeowners to walk away without liability - even if they have money to pay. It's not that the California statute is bad alone; it's that it's wrong for federal taxes to guarantee huge loans without homeowners guaranteeing those loans too.
The purpose of non-recourse laws is not, actually, to give borrowers the motive or the means to stiff lenders without penalty. That is not a social goal state legislatures get behind, as a rule. The purpose is supposed to be to prevent bad lending practices in the first place: if lenders know they are secured only by the real estate in a purchase transaction--not by any additional recourse to other assets--then, in theory, they will institute sane LTV limits and pay attention to decent appraisals. Quite obviously that didn't work during the boom. While nobody I know is thrilled with the idea of the taxpayers bailing these lenders out, I am also not thrilled with the idea of repealing recourse laws to allow lenders to make themselves whole out of anything left in the borrowers' pockets.

It may sound appropriately populist to be in favor of making the rich pay up, but then again we could use some hard evidence that people can, in fact, afford to pay before we go down this road. My own sense is that such a proposal would be more likely to garnish wages of already-struggling families than it would to force the fat cats to liquidate the jewelry collection to pay off Countrywide. And certainly, if we impoverish borrowers in order to stave off lender failure, then we taxpayers will have impoverished former homeowners to cope with.

I am trying to lay all these questions and concerns out in hopes that we move forward from popularizing the idea of "walk aways" to some slogging through the issues and hard numbers to try to get a more nuanced view. Personally, I'm willing to believe that we have a real increase in "ruthless defaults" in the prime mortgage book; it's too rational for too many people for it not to go on. But I'm not willing to take it as a matter of faith, and I'm impatient with pronouncements from banks and rating agencies that aren't backed up with better data. Unfortunately, I am also very willing to believe that servicers and servicing platforms are broken: that we aren't collecting the right kind of data, that contact efforts (rather than just responses) are inadequate, and that it's just plain easier for understaffed outfits to credit "borrower attitudes" for rising defaults than to send those defaulting files through a rigorous analysis process, one that looks carefully at the original loan underwiting and reverifies more than just a current FICO.

We did just go through this in 2005 with the credit card lenders, who preferred to talk about consumers being irresponsible with the credit they were granted, not lenders being irresponsible to the borrowers they spent a small fortune soliciting. While the legislative efforts we're seeing at the moment seem largely pro-consumer (eliminating taxes on debt forgiveness and increasing borrowing opportunities with the GSEs and FHA passed, cram-downs at least on the table), there's always the possibility of backlashes forming as discussion of "moral hazard" shifts from lenders who took too much risk to borrowers who loaded up the U-Haul at 4:00 a.m.

I therefore suggest to the media that we've done about all we can usefully do just by reporting on unsupported claims being made about walk-aways. Granting credence to claims about motivations and social attitudes simply because the person making the claim is in the industry seems a bit rich at the moment; we haven't even distinguished ourselves lately as mortgage lenders, which is what we're supposed to be, let alone as sociologists. It's time to demand the evidence and to analyze it in the context of other information we have about borrower distress and repayment patterns. Otherwise the danger arises that an "echo chamber" starts to create conventional wisdom about default behavior, which may be hard to challenge if it turns out to be a bit of an exaggeration.