Boston Fed: No Way Out of Foreclosures

Tim Cavanaugh | July 10, 2009

Here's a deceptive headline from the Boston Globe: "Lenders avoid redoing loans, Fed concludes." Actually, this remarkably straightforward study [pdf] from the Federal Reserve Bank of Boston begins with the (old) news that lenders have "performed payment-reducing modifications on only about 3 percent of seriously delinquent loans."

The meat of the discussion paper "Why Don't Lenders Renegotiate More Home Mortgages? Redefaults, Self-Cures, and Securitization," by Manuel Adelino, Kristopher Gerardi, and Paul S. Willen, is an exploration of why loan modifications, or renegotiations, are so rare.

Short answer: Renegotiations are a much-worse-than-advertised deal for lenders. (If you believe people stuck in losing situations should be encouraged to get on with their lives, they're a bad deal for borrowers too.) 

The study finds about 45 percent of renegotiated loans end up delinquent again, a rate that sounds high but is actually considerably lower than the redefault rate found by the Office of the Comptroller of the Currency and others. But the Boston Fed finds something more encouraging in its "self-cure" analysis: If you don't help troubled borrrowers out at all, 30 percent of them end up getting out of the jam on their own.

There are plenty of other interesting factoids:

• The difference in loan-mod rates between securitized loans and straightforward loans is "statistically insignificant." So much for the advantage of doing business with your friendly neighborhood bank, as well as the canard (treated at length in this study) that fear of bond-investor lawsuits are holding up renegotiations.

• Something that should have been obvious but (to me at least) wasn't: In "a world with rapidly falling house prices," the truism that renegotiating is a better deal for the lender is not true, because of the one-in-two chance that there will be a redefault and the lender "will now recover even less in foreclosure."

• The "self-cure" risk — the likelihood that the lender is wasting money on a modification because the unlucky borrower would have solved the problem on his or her own — makes it even harder to argue that loan modification is advantageous (or less disadvantageous) than foreclosure: "One must take into account both the redefault and the self-cure risks, something that most proponents of modification fail to do."

If you're better than I am at equations featuring Greek letters, you'll find even more red meat in the 41-page pdf. It's hard to imagine a more clear argument against the Obama Administration's $75 billion "Making Home Affordable" program — though the authors do make some noise about how the best interests of "society" might make it worth persisting in the failed loan-mod experiment. At Seeking Alpha, Matt Stichnoth dismisses as "insane" co-author Willen's suggestion that the government would be better off just giving money directly to borrowers. He's right that it's insane to expect deadbeats to stop being deadbeats, but the dirty secret of all forms of welfare is that it's almost always cheaper and more effective for the government just to give the money away rather than setting up cockamamie schemes like this one.

One housekeeping note: Though I continue to encourage wider and more pointed use of the term deadbeat, I find that 30 percent self-cure rate heartening. Even among troubled borrowers, plenty of people retain the sense of shame that helped our Olde Tyme ancestors build a great country.

Comments

Fluffy|7.10.09 @ 5:05PM|#

The reason there aren't more modifications is pretty simple: to save the people who aren't part of that 30%, you would have to substantially cut principal balances. Having Obama write a check for $1000 is not a sufficient inducement to offer someone a $50000 principal reduction. This is not rocket science.

If the government wanted modifications, they should have allowed the lenders to fail. Post-liquidation, it would be possible for the new lender to move to more aggressively offer modifications than the existing lender can. This is true for the pretty simple reason that when you pick up a loan in liquidation for 10 cents on the dollar, if you give the borrower a 30 point haircut and the loan becomes a performing asset again, you win big time. A lender who acquired a loan for 10 cents on the dollar will behave much differently than one who acquired it at par. This is also not rocket science.

By propping up the zombie banks, the government created zombie loans.

|7.10.09 @ 5:11PM|#

"Even among troubled borrowers, plenty of people retain the sense of shame that helped our Olde Tyme ancestors build a great country.

- Tim Cavanaugh, Puritan

max hats|7.10.09 @ 5:19PM|#

The study finds about 45 percent of renegotiated loans end up delinquent again, a rate that sounds high but is actually considerably lower than the redefault rate found by the Office of the Comptroller of the Currency and others. But the Boston Fed finds something more encouraging in its "self-cure" analysis: If you don't help troubled borrrowers out at all, 30 percent of them end up getting out of the jam on their own.

Please correct me if I'm reading this paragraph wrong, but you appear (to me) to be saying that renegotiated loans default at 45 percent or higher, while 70 percent of non renegotiated loans presumably default. How is this an argument against renegotiation?

hmm|7.10.09 @ 5:22PM|#

There is shame in not doing what you have agreed to do. Regardless of the circumstances. You fulfill your obligations, even if it takes the next 80 years. That isn't puritanical. It's ethical and moral.


Good study and read so far.

|7.10.09 @ 5:25PM|#

people retain the sense of shame that helped our Olde Tyme ancestors build a great country

...that was weird: as I read that, I heard it in the voice of Eddie Albert, with a fife playing "Battle Hymn to the Republic" in the background.

Indeed, what would the people in American Gothic do?

I get a bad feeling, though, that I'm not going to hear this story reported on Planet Money.

Someone more hip than I to the legal and financial implications might comment on just putting liens on the property. Early on in the disaster (pre-TARP $7*10^11 freak out), I had heard of banks just accumulating unpaid loans as liens on the property, I guess with the idea of "you ain't payin', but we sure as hell don't want the house back NOW, so we'll just put our marker down." Seemed like a slick stop-gap measure (uh, where the "gap" might be 15 years -- or 28 in Midland TX).

Then it just disappeared off the radar.

Invisible Finger|7.10.09 @ 5:55PM|#

How is this an argument against renegotiation?

Do the math.

LarryA|7.10.09 @ 6:00PM|#

"Lenders avoid redoing loans, Fed concludes."

Anything in the study looking at whether borrowers are clamoring for renegotiation? I presume the lender can't force one.

max hats|7.10.09 @ 6:01PM|#

Do the math.

Okay. 45 percent default rate is lower than 70. Therefore renegotiation is good, provided banks do not want their lenders to default. That is, as best I can tell, not the point of the argument.

Where am I going wrong?

robc|7.10.09 @ 6:06PM|#

Do the math.

To do it, there are some numbers we dont know (maybe they are in the article, but I havent read it) so Im going to make them up:

Profit of self-cure: 50k
Profit if current foreclosure: -20k
Profit if renegotiate and pay off: 20k
Profit if regenotiate and fail: -50k

If you dont renegotiate:
50K*.3 + -20k*.7 = 1k

If you do renegotiate:
20k*.55 + -50k*.45 = -11.5

Those numbers are complete BS, but I think you can see the point, max hats, If the cost of renegotiating is high enough, increasing payouts from 30% to 55% isnt enough.

robc|7.10.09 @ 6:09PM|#

Where am I going wrong?

You failed to "do the math". :)

There are more than just two numbers in play.

max hats|7.10.09 @ 6:10PM|#

For sure, I can see why renegotiated terms can lead to lower profits. But the article, despite being really sure it is making that point, is not. What data it does provide seems to hint to a contrary point.

|7.10.09 @ 6:12PM|#

ChrisH,
The banks already have liens on the property. A mortgage is secured by a Deed of Trust, which equates to a lien on real estate. If the borrower defaults on the loan, the DoT gives the bank the right to take possession of the property.

If the borrower defaults, the asset(the loan) becomes a non-performing asset. Essentially, it's an asset that isn't providing any benefit to the bank. The bank can't use the money that's tied up in the mortgage unless they sell the place. Banks don't want real estate, they want cash. So having a non-performing asset just sucks up income that could otherwise be generated.

There's no way for the bank to hasten the pace of recovering the funds they loaned out. They have to sell the place, often at a loss to them, in order to recoup the money they lent out.

robc|7.10.09 @ 6:16PM|#

What data it does provide seems to hint to a contrary point.

The only relevant data in the article is 3%. That doesnt just hint but pretty much flat out declares that redoing the loans is not in the banks' interest. While banks are stupid, I dont think they are that stupid. If it was obviously profitable, they would be doing it.

Actually, maybe they are that stupid.

Some dude|7.10.09 @ 6:37PM|#

Why Don't Lenders Renegotiate More Home Mortgages?

Lenders are counting on bailouts? I am guessing, but if lenders knew they were truly on their own, they would do whatever they could to reduce the losses on each mortgage.

max hats|7.10.09 @ 6:38PM|#

Yes, but that 3% is modified by saying it is "3 percent of seriously delinquent loans." Presumably a seriously delinquent loan is substantially different than a typical delinquent loan, but we have no idea how, or in what rate they occur. Maybe the fraction of total delinquent loans that are being renegotiated is 95. Maybe it's 1%, but includes a whole 3% of the bottom third.

Basically, the article is several impenetrable paragraphs that, ultimately, say nothing. Also the requisite shot at Obama (more like. . . NOBAMA!!!1!).

robc|7.10.09 @ 6:40PM|#

max hats,

It defined it in the article, IIRC. 60 days delinquent.

robc|7.10.09 @ 6:41PM|#

The Fed's study found that only 3 percent of seriously delinquent borrowers - those more than 60 days behind - had their loans modified to lower monthly payments

Huh, the 3% and the definition were in the same sentence.

max hats|7.10.09 @ 6:45PM|#

oof

Still, absent any other data, this paragraph:

The study finds about 45 percent of renegotiated loans end up delinquent again, a rate that sounds high but is actually considerably lower than the redefault rate found by the Office of the Comptroller of the Currency and others. But the Boston Fed finds something more encouraging in its "self-cure" analysis: If you don't help troubled borrrowers out at all, 30 percent of them end up getting out of the jam on their own.

Is not doing the author any favors.

wuestefuchs|7.10.09 @ 6:57PM|#

I actually ended up pursuing a short sale on a home, and my biggest regret is not walking away sooner. I ended up with serious health problems around the same time that the economy went sour. Expecting the economy (and my health) to turn around any day, my wife and I ended up eating our savings and saddling ourselves with consumer debt in an attempt to hold onto the house.

If we had just stopped then, we would be in much less debt, and the bank would have sold the property for more. All of our pushing off the inevitable ended up helping only the eventual buyer (which is just fine).

That is just the thing about this though. Someone who exercised better judgement than we did will be able to enjoy that house at a reasonable price. If the government had stepped in to "help", that person would still be waiting for housing prices to become reasonable and I would be saddled with an overpriced house.

I am not ashamed of walking away. I would be ashamed, however, of TAKING money from other hardworking people who had no part in the contract.

If it were truly in everyone's best interest that contracts never be broken, then we would have debtors prison in this country.

With the way that housing prices have plummeted, it may even be in the best interest of someone who is fully employed to walk from a home. I see it as part of the ongoing correction. As noted above, the banks might even be willing to renegotiate if the government would changing the rules all of the time.

The worst thing for wealth and production that could be done, however, is for the government to subsidize stagnation. The money saved on lower mortgage payments will let people spend more money on consumables, which is what will help the economy right now anyways. It is just part of the natural shift.

robc|7.10.09 @ 7:08PM|#

max hats,

If the redefault rate is even higher than 45%, then that makes the authors point (that the banks are losing money on renogiations) even stronger. Not weaker.

Tricky Prickears|7.10.09 @ 7:15PM|#

A lender who acquired a loan for 10 cents on the dollar will behave much differently than one who acquired it at par.

What, one whose stock price went from $40 to $1.25 and is into the gov. for 30 or 40 billion?

I know, I know.

max hats|7.10.09 @ 7:24PM|#

hate to belabor such a minor point, but, this:

But the Boston Fed finds something more encouraging in its "self-cure" analysis: If you don't help troubled borrrowers out at all, 30 percent of them end up getting out of the jam on their own.

Implies a default rate way higher than 45%, and that is what the author is trying to insinuate is a better option. It is true that successfully renegotiated loans can be less profitable. But in that case, there's no point in even mentioning it at all.

Tim Cavanaugh|7.10.09 @ 7:47PM|#

To start with, Max Hats, the 30 percent and the 45 percent are apples and oranges: One treats the universe of people who have not had loan mods, the other counts the universe of people who have. We can debate the fairness or unfairness of loan mods, but we can't pretend these are comparable figures (i.e., "We'd rather get 55 percent successful mods than only 30 percent successful mods"). If anything in the paragraph you don't like led you to believe that that was the point, I regret the ease with which you were confused.

Also, the redefault rate is, as noted, probably higher than 45 percent.

As for what this means to a lender: You're deciding whether to give a delinquent borrower a haircut. There's a one in two chance that in addition to losing that income upfront, you will still lose the loan, and the property you end up owning after foreclosure will be worth less than it will be if you foreclose right now. On the other hand, you can do nothing: Two chances out of three you'll end up owning a house (which no bank wants to do, I'm a banker not a realtor dammit, blah blah blah) in the very near future, and you can then begin the difficult process of cutting your losses and selling it. And one chance out of three you will win (or break even, really) because the borrower will find a way to keep paying. As noted in the Boston Fed study (which I suggest you read, given your frequently repeated calls for more information), this introduces another variable into the renegotiation decision: the likelihood that even if you do the loan mod, and even if the borrower doesn't redefault, you still would have been better off doing nothing.

Finally, when I said "encouraging" I meant that it's encouraging to learn that nearly one out of every three bad borrowers still has the decency to keep paying on a deal he or she made.

hmm|7.10.09 @ 7:58PM|#

The Boston Fed gave us the 1992 discrimination in loans and bank redlining paper which has turned out to be a fucking nightmare government ran with. While grouping all economists together from one Fed is a mistake, the source is not the most pure of sources.

Tim Cavanaugh|7.10.09 @ 7:59PM|#

Sorry to be snippy, Max Hats. Further questions welcome.

hmm|7.10.09 @ 8:23PM|#

That was a pretty good layman explanation.

nicely done.

The argument boils down to drawing out a problem and incurring more problems and loss or getting the headache over with and moving on. I imagine the argument becomes even stronger when you start discounting cash and considering future inflation. Hell right now the bank may be better off with land it may start losing value at a slower rate than dollars in the near future.

hmm|7.10.09 @ 8:24PM|#

That and land doesn't go to zero, dollars can.

Tim Cavanaugh|7.10.09 @ 9:22PM|#

Hell right now the bank may be better off with land it may start losing value at a slower rate than dollars in the near future.

You know, right now I'm pretty much at streetcorner-preacher-with-a-bullhorn level in my real-estate-must-and-will-lose-another-50-percent evangelizing, but that's a damned interesting point.

robc|7.10.09 @ 10:43PM|#

You know, right now I'm pretty much at streetcorner-preacher-with-a-bullhorn level in my real-estate-must-and-will-lose-another-50-percent evangelizing, but that's a damned interesting point.

Maybe this is true for morons who live on the coasts, but for those of us in middle america that only had a boomlet to begin with, eh, we've already given it all back.

hmm|7.10.09 @ 11:07PM|#

It was a big bubble. I've seen retail space numbers that are down right scary with respect to the amount of needed space compared to available space. But like I said before, land doesn't go to zero, dollars can. Even a chunk of dirt in Detroit is worth something.

The problem for banks is liquidity, other wise they wouldn't care about holding land. That and the rules they have to use to report land. Oddly enough those rules were different than reporting derivatives based on land assets. Go figure, regulation failure.

Dan|7.11.09 @ 12:02AM|#

The difference in loan-mod rates between securitized loans and straightforward loans is "statistically insignificant."

The study is useful in that it demonstrates that for any type of lender modifications are economically undesirable, which refutes the benefits of Obama's foreclosure plan.

However, it doesn't necessarily refute the benefits of dealing with a local lender. This study (which is limited to Ohio) concludes that delinquency rates are correlated not to whether a lender was regulated or unregulated, but to whether it was local or out-of-state. I'd be interested to see if these results hold throughout the country.

If it is true that local banks make more prudent loans, then by bailing out large financial institutions, the government is propping up lumbering, inefficient lenders at the cost of leaner, more decentralized banks.

Tim Cavanaugh|7.11.09 @ 1:05AM|#

I've seen retail space numbers that are down right scary with respect to the amount of needed space compared to available space.

And when that retail market does cycle up again, it's going to cycle to a much lower point than it attained in the past, because of those interwebs.

I'm only repeating conventional wisdom by saying the commercial real estate apocalypse has just begun.

Tim Cavanaugh|7.11.09 @ 1:09AM|#

If it is true that local banks make more prudent loans, then by bailing out large financial institutions, the government is propping up lumbering, inefficient lenders at the cost of leaner, more decentralized banks.

Well, Bank of Wyoming doesn't seem to have had any better survival skills than my own lumbering lender. ("Lumbering Lenders" should be the nickname of the offensive line of a college football team.)

|7.11.09 @ 3:34AM|#

The difference in loan-mod rates between securitized loans and straightforward loans is "statistically insignificant." So much for the advantage of doing business with your friendly neighborhood bank...

The paper only argues that renegotiation is almost always a losing proposition, regardless of the size of the institution. But it doesn't say anything about whether local banks perform better than large servicers in identifying those exceptional cases where renegotiation is warranted because the likelihood of delinquency is much higher than the average. Both large servicers and local banks have an incentive to identify these borrowers, but local banks are supposed to suffer from less of an information asymmetry problem than large servicers. I have no idea whether this is actually true -- it seems plausible, but who knows -- and the paper does not address this issue at all. (The authors suggest the LPS is detailed, but I assume it is detailed about the borrowers' status at the origination time of the loan, which is not relevant to the asymmetry issue.)

I'd note that these desperate borrowers are crucial, according to the Boston Fed's Double-Trigger theory (of which the paper you link is a part). They argue that foreclosures are driven by a double whammy of homeowners having negative equity and suffering a personal financial setback due to the recession. These people are not (necessarily) deadbeats, and most would like to keep their homes (Fed's argument, not mine) but they need some temporary assistance to get through a rough patch. That's why the Boston Fed's proposal (slide 10) is for the government to identify these people and offer them loans or grants. The government would try to identify the high-risk foreclosure pool in a by focusing on two highly-relevant criteria -- "Is this property underwater?" and "Is this borrower suffering a sufficiently significant financial shock?" Note that the government would be getting at the information asymmetry issue noted above in a very targeted manner, thereby speeding economic recovery. (At least that's the argument.)

Incidentally, from this talk (slides 3 & 4) I take it that this is a very personal issue for Paul Willen. He doesn't seem like a deadbeat to me.

hmm|7.11.09 @ 5:13PM|#

I'm only repeating conventional wisdom by saying the commercial real estate apocalypse has just begun.

I agree it has just begun. I don't think it is going to be apocalyptic or Detroit like. With the cost of property so low, and a lower rent will follow that, I can see things like a resurgence of the local hardware store, local market, local clothing store, smaller non chain restaurants and so on. While not a saving grace I can see a lot of startups taking advantage of lower overheads, starting with rent. That doesn't mean it isn't going to be ugly or that the worse isn't yet to come.

Banking in the US is a lost cause. The Canadians have it right, but the reason they have it right is not the one all the screaming tards keep pointing at.(legislation) They have it right because they are conservative by nature. They are traditional bankers who are more averse to risk than an eighty year old hemophiliac hypochondriac. The few conservative run banks in the US are usually private, and the semi-conservative run public banks have to contend with a public and system that demand short term profit damn the risk instead of long term profits avoiding the risk.

|7.12.09 @ 7:18PM|#

Yet another comment thread illustrating why progressives shouldn't be allowed to touch any aspect of the economy with a ten foot pole.

There should be some sort of court-ordered restraining order preventing the economically illiterate from setting economic policy.