Wednesday, January 27, 2010

Mortgage Servicing Performance I: Underwater and That Social Trust Thing

Original posted on RortyBomb:

Steve Waldman of interfluidity has an interview up at mortgage calculator where he discusses, among many other things, his statement: “I think that the moral thing for most borrowers to do, under present circumstances, is to default on loans when it is in their financial interest to do so.” You should check it out.

There’s a lot of arguments for “strategic defaults” which we will leave to the side for right now. The counter-argument is that lending has an element of “social trust” norms build into it, a trust that isn’t easy to replicate and once broken it is very difficult to rebuild. The small, difficult to quantify but ever present, elements of these norms are what form the glue of many of our credit markets. Lenders trust borrowers, and borrowers can trust lenders, and this takes the rough edges off the credit market.

Recap: we have a problem where many homeowners, maybe one out of four, are underwater, which means they owe more than their home is worth. Many others, with 18% underemployment and loan resets, are having trouble making their monthly payments. Being underwater is very correlated with defaulting on a loan, and foreclosures destroy value for the house itself and has an externality destroying value for neighbors. Meanwhile there’s also an agency problem that has exploded: as we discussed last summer, mortgage backed securities were never designed to handle large number of write-downs; as the inventor Lewis Ranieri warned for years, nobody has a fiduciary responsibility to make sure principal gets written down in mortgage backed securities. For normal mortgages at a regular bank you writedown before you foreclosure; but none of the gee-whiz financial engineers stopped to think how you’d do that in a mortgage backed security.

Because of all these problems the Obama administration decided to start a program called the federal Home Affordable Modification Program (HAMP) last year. Long story short, we subsidize banks who make adjustments to the amount people pay per month. Obama is doing this do this because of the externality of waves of foreclosures, the threat of abandoned neighborhoods and to “nudge” the servicer of the loan, who have a twisted incentive to get paid through fees as things get worse for investors, to do the right thing.

Underwater and Social Trust

I think we can quantify this trust thing a little bit. Let’s get into the meat of a new report issued by the State Foreclosure Prevention Working Group: Analysis of Mortgage Servicing Performance (pdf) (h/t Shahien Nasiripour for the pointer).

Remember there’s two problems: too much of a loan principal that is underwater, and too high of a payment. Most agree that the first is the bigger problem for both parties, but the second is easier to modify, politically, from the point of view of the government. Now people (including myself) thought it would be great if banks would decrease the principal of the loans they are modifying in addition to the payment, which would reduce the risks of default and help get mortgage prices to adjust quicker, but they could instead just leave them the same. That’s their choice. But given that the borrower is going out of their way to try and make the payments, it would be a nice social trust kind of thing to reduce the principal slightly instead of leaving it the same, right?

So what has happened? From Analysis of Mortgage Servicing Performance, more than 70% of the modifications resulted in an increase in the loan balance. Not staying the same, and certainly not decreasing, but piling on more principal for the loan:

I’ve seen a lot of bad things during this financial crisis, but this is the most disgusting thing I’ve seen so far. At a time when one out of four homeowners are underwater, banks are using a mortgage modification program to pile on more debt on these loans. They do this even when it’s well known the correlation between the level of being underwater and default.

How? From the report: “Servicers routinely capitalize delinquent interest, corporate advances, escrow advances and attorney fees and other foreclosure-related fees and expenses into the loan balance when completing a loan modification.” So fees allow them to make it look like they are doing their clients a favor, while all they are really doing is running them in a big circle.

(If any risk quants or people who work in this area want to contact me, I can keep it completely off the record, but wtf? Is this a statistical juking? This has to increase the redefault risk, but is there a way that this makes the numbers in the medium term look better on expected values? Is this the servicers going completely rogue? Or is this simply profit maximization as sociopathic behavior? One can’t simply “nudge” a sociopath….)

And for the social trust argument, these are not random people. These are people who, instead of walking away from their responsibilities, are burning time, money and energy to credible signal to the bank: “I’m in over my head with this mortgage but I want to do right by it because it’s an obligation I made to you. Instead of simply walking away or trying to short sell, is there any way we can work this out so I can still pay you whatever I can? I gave you my word and that means something to me.” And the bank uses their signal that they want to do the right thing to fuck these borrowers the hardest, piling on as much debt as possible on these guys as they can get away with. The borrowers are trying to come up from the pool to get a breath of air, and the bank grabs them by the head and pushes them as far underwater as they can get them. And we are subsidizing this.

Though I shouldn’t be too surprised. I mentioned before when it came to the walking-away problem simple game theory tells us, “If you are convinced that the other party will cooperate with you no matter what (i.e. never walk away), that’s when you fuck them over the hardest (or more politely, that’s when you ‘don’t co-operate’). It’s only through the threat of non-cooperation that you can actually secure cooperation.” So the people who really want to meet their obligations are the ones you screw with as much principal as possible, and the ones who threaten to walk away are the ones you have to take seriously. If we had mortgage cramdowns, everyone would suddenly be someone the banks would have to take seriously in these situation, instead of a mark who still believes in things other than profit maximization…

Part 2:

So we talked about a recent report about mortgage modifications, the Analysis of Mortgage Servicing Performance, here. I asked back in August whether or not the Mortgage Modification plan has failed, and nothing in this report, which is the latest in data collection by 12 state attorneys general and three state banking supervisors, makes me think otherwise.

If you are also reading the on the ground journalism, say this excellent article by Mary Kane, the closer you get to it the more of a disaster it looks:

The voluntary plans by servicers, however, were called “extend and pretend” plans by critics, who said servicers simply were setting up repayment plans with late fees and other charges rolled into them, without ever actually reducing a borrower’s debt. Re-default rates on those loan modifications have been high as a result. Making Home Affordable has been more aggressive about lowering a borrower’s monthly payment, and the government is pressing servicers to switch to using its program — one reason why Making Home Affordable permanent loan modifications are lagging behind. In addition, some borrowers simply can’t qualify for the government’s program because they are too far underwater on their mortgages.

So I’ll ask the obvious question. Is the current system predatory? I’m not a lawyer, and I know there isn’t a good working definition of what constitutes predatory lending. But let’s take a look at two separate charts in the report. One we’ve already discussed:

70% of people who go through mortgage modification have an increase in their balance. Here’s another chart that should worry you:

Within 1 year almost 45% of modified loans become “delinquent.” 7-9% reach 60-day delinquency within just the first three months after modification, which for all intents means the borrower didn’t make the first post modification payment. What will the 2 and 5 year markers look like?

So we have a financial technique offered to consumers that (a) piles on principal unexpectedly through hidden fees and surcharges and (b) has a reasonable expectation that the consumer won’t be able to make the payments. That doesn’t sound good, does it? Lawyers in the audience, would you consider this predatory? How do those arguments proceed?

It isn’t simply a matter of nobody being able to pay the loans. They find that with significant principal reduction you can get the re-default rate from 45% to 34%, though that is still very high. I am not sure what constitutes significant principal reduction, though I doubt it is on the order of the % decline in the neighborhood value, as the Zingales and Posner Slate article argued as a benchmark for principal reduction.

This is a long, stressful process, and it appears to leave nobody better off, except perhaps the servicers who get some nice fees and a subsidy from the government. This one or two year process is a period where the borrower could be taking the difference on their loan and a cheaper rental rate and start rebuilding their savings and financial lives (and ultimately, the economy).

That said, there is a norm asymmetry here, that is being exacerbated by unreasonable expectations from government officials as to the effectiveness of mortgage modification, a process that is subsidized by our taxpayer dollars. I get the sense that many are worried this is all they’ll get on this front from the administration, so better to try and get it working better than pushing for new efforts.

Where to Go?

Think Progress has a very important article on where the debate should go. Mortgage cramdowns would make banks have to take a writing down process more seriously. As Megan McArdle has written, the law around second-leins and securitization is a mess for these kinds of principal writing down situations, and could be modified. Many banks are probably worried about solvency issues ahead of the issue of properly writing down balances before foreclosure. If a mortgage is 95 cents on the books, but only worth 50 cents, given how weak banks are doing it might be worthwhile to let some foreclosure to 20 cents if they can keep the majority at 95 cents. There are FDIC workarounds for this that can be proposed or implemented. A version of Right to Rent could be pushed more aggressively. And in what I’d like to see some serious brain effort put into, taxpayer dollars could be used to subsidize short sales instead of ineffectual modifications.

Interestingly, as the document points out, we’ve hit the ceiling for the rate at which we can process foreclosures.

See how the “foreclosure closed” marker can’t scale at the pace of delinquency? There’s room for action there as well.

Currently we are trying to pay the banks to take care of this themselves. Given that so much innovation of late in consumer finances has gone into the process of making quasi-predatory items for consumers, it doesn’t surprise me that the fruits of this partnership are ineffectual at best. Now would be exactly the right time to push a stronger agenda, with bolder experiments, to fix this broken market.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.