Livinglies's Weblog

Given how well “shock and awe” worked in the Iraq war, I’d see the Administration’s use of that expression in the context of the mortgage mess as a Freudian slip.

I must confess to being surprised at the report by Shahien Nasiripour of Huffington Post, namely that the Administration is pushing for an even more aggressive-looking mortgage modification program than has been rumored. The reason I’m surprised is that this effort, even though it appears misguided on several fronts and falls far short of what is needed, represents an upping of the demands being made against banks. That is contrary to both the Obama Administration’s past behavior of making great sounding promises and walk them so far back as to wind up in a different country, and of inconveniencing the banks terribly much. But Shahien is an able reporter, so I’m sure he has the facts right.

The scorecard thus far appeared to be that the state attorneys general were the only group moving forward against the foreclosure fraud, but the bold promises of criminal prosecutions were quickly recanted. Instead, a 27 page outline of their settlement demands was leaked. As we discussed, it was a disappointment. Virtually all of it merely insisted that banks obey existing law. It has only two new requirements. One was ending dual track (if a bank is entering into a modification discussion or program with a borrower, it cannot keep moving forward in parallel with a foreclosure). The other was “single point of contact,” meaning having one person at the bank serve as case manager and be the interface with the borrower. We deemed that to be operationally unworkable even if the banks had their records and systems working well. And if they got those in order, borrowers would not need a designated person to make sure a modification request was handled properly.

There was also a rumor, which was connected to the AG negotiations, that the banks would be asked to make mortgage modifications at their own expense, and the number $20 billion was bandied about. The AGs and the Federal regulators seemed to be collaborating closely, which we also objected to; the state and Federal issues are very different. The idea that the banks would be pressured to make mods has gotten a huge amount of pushback in the media and from Republican legislators; there appears to be a full bore PR salvo underway.

Now notice all these ideas are being evaluated in a vacuum. We don’t know what liability the banks would be released from (the legal term is what form of release they would receive). Nor do we or the regulators have an even remotely adequate understanding of all the bad stuff the banks did. The media and anti-foreclosure attorneys have reported on various abuses, most importantly, servicer driven foreclosures, in which the borrower has either made all his payments, or perhaps been late on one or two, and impermissible application of payment, fee pyramiding and junk fees quickly drive a minor arrearage that most borrowers could correct into a foreclosure.

So despite my caviling, if the release covered only robo-signing and false affidavits, this deal (the 27 page term sheet plus a commitment to do mortgage mods) would be a very good deal for homeowners. But if it was a broad waiver, it would be a steal for the banks.

With that as an overlong but necessary background, the latest development looks like a ratcheting up of the effort against the banks, and perhaps a shift in who is in the driver’s seat among the Federal regulators. It had appeared that originally the Treasury was leading the cross-regulatory Foreclosure Task Force; it was the Treasury’s Michael Barr who spoke before the Financial Stability Oversight Council to launch it officially last November. Even then we deemed it to be an exercise in window-dressing that would make the bank stress tests look tough. It went from bad to worse when John Dugan of the Office of the Comptroller of the Currency, the most bank friendly regulator, spoke at recent Congressional hearings and indicated that the task force reviewed 2800 loan files of delinquent borrowers (from the bank side only; as we have stressed, independent verification was impossible given the compressed time frame for the whitewash exams) and found all bank foreclosures to be warranted. Needless to say, those who have been paying attention to this story saw the results as proof of the lack of interest in getting to the bottom of bank abuses. And the OCC playing a prominent role seemed to be further confirmation.

So here are the highlights from the Huffington Post story:

The Obama administration is seeking to force the nation’s five largest mortgage firms to reduce monthly payments for as many as three million distressed homeowners in as little as six months as part of an agreement to settle accusations of improper foreclosures and violations of consumer protection laws….

The modified mortgages could cost the five financial behemoths — Bank of America, JPMorgan Chase, Citigroup, Wells Fargo and Ally Financial — as much as $30 billion…

t also could lead to reduced mortgage payments or lowered loan balances for nearly two-thirds of the 4.7 million delinquent homeowners who have yet to fall into foreclosure, according to data provider Lender Processing Services.

The aim is to ensure the number of assisted borrowers is spread throughout the country, and that banks modify both expensive and inexpensive mortgages, people involved in the talks said. Banks also would likely forgive mortgage principal in situations where a pre-determined formula dictated that it was the best way to modify a home loan. Balances on second mortgages and home equity loans — of which nearly half of all outstanding loans are owned by BofA, JPMorgan, Citi and Wells — would also have to be written down.

That would then kick-start the healing process needed to clear the large overhang of repossessed and soon-to-be-foreclosed homes that’s depressing house prices and sapping consumer confidence.

This is pretty bizarre. It reads like HAMP 2.0. Notice that the banks are NOT being required to make principal mods. The story simply states, “reduce monthly payments”. So the $30 billion is presumably for a combination of servicer costs, payment reductions, and some second mortgage writedowns (since the Administration has stressed that these modifications are to come out of the hide of banks, I am curious as to how a bank would compensate a securitization trust for a first mortgage mod).

But $30 billion for 3 million homeowners, even assuming every penny went to principal mods, is a mere $10,000 per borrower. If you assume the bottom end of the target participation range (1 million), the maximum dollar amount ($30 billion) and modest budget for servicer costs (10% of mod amount), the highest average you could expect is $27,000 per borrower. That’s helpful but unlikely to be outcome changing for borrowers in distress. So this exercise appears to be about maximizing participation rather than really rescuing anyone. So this exercise appears also to be a stress test 2.0: that the Administration can uses this initiative as a way to talk up real estate and put a floor under the housing market.

Why is this a terrible idea?

First, there is good reason to believe that mere payment reduction plans don’t work when the borrower is upside down. Homeowners are not dumb. Why should they struggle to keep a home if in the end they will still face negative equity if they need to exit in the next few years? What is keeping a lot of these homeowners in place is probably inertia: they like the house, their kids are in local schools, moving is disruptive, and exiting the house involves a lot of hassle and probable adverse impact on their credit record. A payment mod does not change the basic equation. By contrast, the one party known to have tried deep principal mods, distressed investor Wilbur Ross, has reported far lower redefault rates than for other types of mod programs.

And a lot of borrowers are upside down. A recent CoreLogic report found that as of fourth quarter 2010, 11.1 million homeowners had mortgage debt in excess of the value of their house. Moreover, the negative equity for those upside down by more than 50% was $450 billion.

So what is a puny $30 billion max (which will include servicer expenses) accomplish? By itself, nothing except some modification theater. In combination with principal mods, which would come from reduction in principal balances by investors, you could see a positive outcome. As we have stressed, when banks foreclose, the losses to investors are 70% and rising as home values continue to fall and foreclosure defense attorneys are making headway in local courts making arguments based on chain of title issues. All but a tiny sliver of subordinated bond holders would welcome deep principal mods. When you are looking at 70%+ losses, 30% to 50% would look like a screaming bargain.

Second, servicers have every incentive to make sure mods fail. They don’t get paid to mod. They do get paid to foreclose. Their income is based on fees based on principal balances (which is one of many reason they’ve rejected principal mods) plus fees they earn for various activities performed in foreclosures. Tom Adams estimates that servicing is costing 125 basis points today, versus income of 50 basis points coming from regular servicing fees based on principal balances plus 30 to 50 basis points based on late, junk, and foreclosure related fees. So having borrowers fail is economically attractive to servicers.

Third, servicers have never been any good at mortgage mods. Tom Adams again:
Giving a modification to a borrower, principal or otherwise, is basically underwriting a new loan. Obviously, many of the lenders have proven that they were not very good at that. However, at least they had staff and “guidelines” for making the loans.

Servicers have neither guidelines nor staff for loan underwriting. Principal modifications were just not contemplated by the securitization model.

I’ve visited dozens of lenders and servicers over a 20 year period and the only company I saw that had a real policy for modifications was Household Finance (now a part of HSBC). Their stated plan was a perpetual debt model (”generational”). They aggressively offered modifications, sometimes even for moderately delinquent borrowers. They claimed about a 25% re-default rate (I looked at data that more or less confirmed this). Of course, left unsaid was that they didn’t always mind re-defaults as they were an opportunity for additional servicer fees on a loan that was going south either way (investors wouldn’t have wanted to hear that).

The next closest thing to a modification plan was Litton, which was an advocate of short sales based on their confidence in their own valuation of the loans. Litton only serviced loans on which they were the residual holder, so they had an economic incentive similar to third party investors, as long as their was value in the residual (which is pretty unlikely now, for most deals).

As far as servicer factory floors – rather than sweatshops, they bore a resemblance to college dorms – young staff with a high turnover rate (20-40% in good times), lots of calling campaign contests, decorations, balloons, morale boosters. Typical call center stuff, though the mortgage servicers were more aggressive with the morale stuff than credit card, student loan, etc.

Very different from commercial loan servicing, where the concept of -re-underwriting, modification, workouts etc. are much more a part of normal business.

Note that Goldman is now trying to sell Litton….not that there is anyone who could possibly want to buy it.

Law professor and securitization expert Adam Levitin has argued that servicers should not do mods, that the task needs to be assigned to a third party. There have been approaches to compensate for the lack of servicer skills in this area, including having mortgage counselors play a prominent role as well as the NACA approach, where an independent group verifies and uploads key borrower documents and works with borrowers to prepare a household income and expenses spreadsheet which is a key input to a loan re-underwriting. But absent a new approach, why should a repeated failed experiment of unmotivated servicers doing mods lead to different outcomes? I much prefer his not quite a joke solution of having the banks spend the then rumored settlement amount of $20 billion on Legal Aid. The threat of borrowers chipping away at banks enough to develop class action theories or prove out the New York trust theory discussed on this blog (which would pave the way to asteroid-hitting-the-finanical system suits against trustees) might change their incentives.

Fourth, the six month timetable is nuts. Servicers are factories. As the late Tanta pointed out, it takes servicers six months to implement the software changes associated with meaningful new initiatives. Even if they did a full court press, the most they could compress it to is probably four months.

Although a lot of the chaos of HAMP mods appeared to be servicer “dog ate my homework” loss of borrower-submitted documents, there is every reason to believe that a lot of the screw-ups reflected deep-seated operational problems. Servicers are working with platforms, both software and procedural, that are already deficient and cracking under the volume of delinquent loans. Asking them to do something different, on an aggressive timetable, and in high volumes is just about certain to create a complete train wreck.

Even though we are deeply skeptical, the dynamics are curious indeed. The HuffPo account states that the Department of Justice is leading the negotiations with the banks, and HUD, the Treasury, and the FDIC are on board. The OCC, which recently seemed to be in the driver’s seat, has apparently been marginalized. And the upping of the rumored amount to be extracted from the banks, $30 billion (admittedly a maximum, we’ll believe that when we see it) is markedly higher than the earlier $20 billion that elicited all sorts of noise.

Even though the Foreclosure Task Force’s exam was cursory, and managed to find that all foreclosures were warranted, save in a very limited number of cases when an “intervening event or condition” took place. Nevertheless, that review found legal violations (and the language suggests they go beyond the poster child of robosigning). Of course, a literature search or database query of court filings would have shown the same thing. But Walsh’s testimony in February made no mention of Federal violations (click to enlarge):

So what is the Administration’s source of leverage against the banks? In theory, it has a ton, starting (as we have pointed out in meetings with the Treasury) violations of REMIC, the IRS rules that govern securitizations (the investors would be charged but the violations result from bank failures to adhere to their representations in the pooling and servicing agreements; they have a basis for litigation, and this is a nuclear weapon level of threat). We raised it twice in an August meeting with Treasury when officials, including Geithner, piously maintained that there was little they could do about servicers. The questions about using IRS violations to bring servicers to heel were pointedly ignored. And we knew then that the issue had already been raised directly with a senior enforcement officer at the IRS who knew the REMIC rules and was initially very interested. The result? The report back was that the matter had gone over to the White House, which said it did not want to use tax as a tool of policy. Ahem, didn’t Obama swear to uphold the laws of the land?

But the bottom line, and it certainly has been consistent with the Administration’s posture, is that it sees its authority over the banks as being narrow. But the Huffington Post article mentioned consumer law violations, and the 2003 FTC/HUD action against miscreant servicer Fairbanks was based on a broad range of violations. Perhaps the powers that be revisited some of the thinking behind that action. One can only assume they have a real smoking gun; this sudden show of spine (even if the effort falls vastly short of a sound course of action) is very much out of character (although Treasury has been bloody-minded in its Volcker rule negotiations with banks, so this is not completely without precedent).

The Administration’s argument may also be that if the banks do widespread mods, they can also get consumers to waive their rights to litigate. That may be the real rationale for a broad-but-shallow strategy. No Federal or state governmental body can waive a private party’s right to seek recourse. But do the banks buy that they have real liability from chain of title issues? They appear to be in deep denial on this front, given the lack of investor lawsuits. But we are told that the reason that those who have studied the question haven’t acted isn’t that they think they have a weak case, but if they prevailed, it would blow up the banking system, which isn’t exactly in their interest. But if they came up with a more limited basis for action, they might well proceed if only to pressure servicers to do meaningful principal mods.

But even with this new desire by the officialdom to press forward, it isn’t clear the other moving parts will line up. The Administration is also pushing the state attorneys general to wrap up their settlement. But that group appears to be fracturing, with defections expected on the Republican side and probable among some Democrat AGs as well (the article mentions New York’s Eric Schneiderman as a possible holdout; we are also told the Nevada AG Catherine Masto is not keen about the deal). The banks also want a pound of flesh to come from Fannie and Freddie, which makes sense given that we have gotten reports from readers of HAMP mods being approved by servicers and nixed by the GSEs.

This is all very curious indeed. My gut (and it could prove to be dead wrong) is that there is no negotiating space between the banks and the Administration, that the bid and offer are too far apart. The haste on the part of the Administration to wrap things up is not likely to help them in the absence of a real threat; undue eagerness to strike a deal is usually a sign of weakness. But the $30 billion may also be on the table to give room to negotiate down for the banks to save face. Since the Obama Administration has never been very good at negotiating, the results even on a level of bargaining are likely to be underwhelming.