Foreclosure-gate updates

Joseph Mason on the Myth of Good Servicers

Posted: 20 Dec 2010 12:37 AM PST

By Thomas Adams, an attorney and former monoline executive, and Yves Smith

Joseph Mason, the Hermann Moyse, Jr./Louisiana Bankers Association Professor of Finance, Louisiana State University, has a post up at Housing Wire that not only struck both of us as more than a tad off beam, but even elicited critical e-mails from real estate industry participants. In addition, at a couple of junctures is it so unclearly written as to be difficult to parse.

The post is misguided from at least three perspectives. First, Mason claims that his take on servicing as of October 2007 was so correct that there is virtually nothing to be added. That is tantamount to saying a recommendation for urban planning for New Orleans made pre-Katrina is the pretty much the only thing worth considering now. Like New Orleans, the servicing industry has been hit by devastation, in this case a level of foreclosures that has overwhelmed the industry, that with the benefit of hindsight should have been anticipated. Mason’s 2007 paper did foresee a large increase in delinquencies, but his estimate of the cost of the crisis was $150 billion, consistent with the prevailing “subprime is contained” forecasts.

Moreover, his view of borrowers was based on the subprime ARM resets of 2007 and 2008. Many of those borrowers were simply not viable once a reset hit. Many of them have already lost their homes. One of his premises in that paper, that a mod might not leave the investors any better off, is quite different now, when loss severities (losses as a percentage of mortgage amount) are now averaging over 70%.

While older policy prescriptions to address servicing problems that were recognized pre the foreclosure crisis are no doubt worth considering, the failure of his piece to factor in the impact of a storm surge of foreclosures and the other effects it has had, such as overwhelming court systems in judicial foreclosure states, is a fatal oversight.

The second flaw in the post its failure to deal candidly with servicer economics. Peculiarly, the post tries to create an urban legend of “good servciers” who can be held up as a model for business practices for the rest of the industry. We’d like to know exactly who these “good servicers” are, since we believe that the economics of the industry make it impossible for a servicer not to hemorrhage cash in the current environment unless it cuts corners or worse. For instance, Wells Fargo, which has sanctimoniously claimed that it is better than other servicers, has also falsely stated in Congressional briefings that it did not engage in robo-signing. In addition, industry practices in foreclosures are standardized to a great degree by the use of outsourcers to manage foreclosures, the biggest common platform being Lender Processing Services, which handles more than half the foreclosures in the US.

While Mason’s paper correctly pointed out that servicers had nowhere near the capacity to deal with the increasing requests for mods that would result from subprime mortgage resets, he fails to tie it back to badly designed fees structures and resulting flawed incentives which lead them to treat servicing as a factory, routinizing activities where ever possible to allow for the use of low-skilled staff. That is antithetical to what is necessary for mods to have a high success rate, which is individualized appraisal of borrower incomes and expenses.

Third is that his attacks on the various iterations of government mortgage mod programs are wide of the mark. There is no question they have not worked as advertised; we’ve been critics of each one as soon as they were announced, and they’ve all delivered predictably poor results. But Mason appears to have an allergic reaction the premise that the failure of the servicers to do mods means official prodding is necessary:

The government’s Home Affordable Modification Program, or HAMP, and private modifications have failed consumers and investors, as they have not only substantially duplicated costs in the foreclosure pipeline but also have been found wanting as potentially good modifications have been foreclosed upon anyway.

This is an unsupported statement for something that could easily be proven by data if true. Yes, HAMP has underperformed massively (we thought the “kick the can down the road” so called “permanent mods” were the most troubling design flaw). But the biggest problems appear not to result from program design (which as Mason neglects to mention, had substantial industry input) but gaming and poor implementation. Even Treasury has acknowledged that servicers abused HAMP; the lawsuits launched against Bank of America by the attorneys general of Arizona and Nevada allege multiple instances of failure to live up to HAMP guidelines. The biggest snafus included: widespread “dog ate my homework” loss of consumer documents necessary to move forward; failure to suspend foreclosure processes while mods were under consideration; false assurances to consumers that they could ignore foreclosure notices.

And how have these and private modifications failed, exactly? This charge is a drive-by shooting. What costs have been duplicated? The Congressional Oversight Panel complained of the poor metrics and reporting on HAMP, so burdensome reporting is unlikely as a major culprit. What good mods were foreclosed on anyhow? And again, to the extent this occurred, how much was it due to the servicer rather than HAMP per se? HAMP, after all, was strictly voluntary. HAMP is far more likely to have exposes the level of incompetence and abuse in private mod programs rather than create a unique new set of bad behaviors (again, the Arizona and Nevada suits cover both pre and post HAMP malfeasance).

As much as the analysis and forecast in Mason’s 2007 had merit, his policy views seem distorted. He seems to think mods are bad, and refis are somehow better. But mods, or as one might more accurately put it, debt restructuring, has a long and proud history. It is rational, self-interested creditor behavior when dealing with a stressed borrower to determine whether he worth more to you dead or alive. Banks routinely gave mods to homeowners who were in trouble if they thought they were viable in the long haul.

His push for mods to be classified as new loans is to allow banks to recoup new fees and so incentivize them to restructure debt. But the logic is faulty. There is nothing to restrict refis now; in fact, this was the way, up through the crisis, that a lot of subprime borrowers were kept alive. Treating mods as refis would require them to come out of the current MBS, if in one. Under the current structures, this would require the trust to pay off the full principal balance of the existing loan. This appears to be the basis of the argument used by Bill Fry’s Greenwich Capital in their litigation with Countrywide. The case was dismissed on procedural grounds, since Greenwich didn’t have the required 25% of certificate holders to initiate litigation. Currently, the trusts don’t have the capital to pay off the modified loans to the trust, thus we are in a quandary.

The part that Mason failed to anticipate in 2007 and seems to miss now is that servicing is a completely underwater business. Tom Adams had a limited experience with one predatory servicer, Fairbanks. Fairbanks was a servicer for distressed portfolios – that means they had three issues most other servicers did not: very high concentration of borrowers in foreclosure, high servicing costs and no lending arm, to provide another source of revenue. People hired Fairbanks to work out bad portfolios – they claimed not to even want good performing portfolios.

Because Fairbanks was a stand alone servicer, and because servicing portfolios amortized quite rapidly back then, they were forced to come up with creative ways to produce additional fee income – hence aggressive late fees, junk fees, quick foreclosures, etc. Like today, Fairbanks depended on foreclosures so they could recoup their fees, which would not have been available via a principal modification.

After Fairbanks’ settlement with the FTC, other servicers of “high risk” portfolios made clear that they were different from Fairbanks, because they only serviced pools in which they (or their parent) had an equity investment in the related MBS. Also, their parents had several other sources of revenue. One such servicer indicated that it did not believe that servicing was, on its own, a profitable business. (it also typically required a variable fee for its distressed portfolios which allowed it to charge more than market rate. When asked them back in 2005 or 06 what it thought the break even servicing cost for distressed pools, it said over 100bps. Today, the level would clearly be even higher).

Today, servicers are forced into the same economic analysis that Fairbanks was – how to make money in servicing highly distressed portfolios, but without the ability to refinance “re-performing” borrowers into new loans. The contractual servicing fees only make sense if most of the pool is performing and very little work. So they need to find other sources of income to make the work profitable, or at least reduce the losses.

Servicing fees are written into the contracts – they can’t be increased. But today, servicing requires hundreds of more bodies for collection, loss mitigation, HAMP analysis etc. The servicing fee doesn’t cover the cost of these new bodies. So they need to use junk fees – which are reimburseable from the proceeds of the foreclosure – to raise additional income.

All servicers are facing this problem today. As we know, today there are few if any “good” servicers – the ones contending with high levels of delinquencies all have issues.

None of them have any flexibility in raising their servicing fees, so they all face the Fairbanks dilemma.

The biggest fix for servicing would be to raise servicing fees. There is no way to do that under the current contracts. New regulations, tons of delinquent borrowers – needing attention, hand holding, and loads of extra work – have dramatically raised the cost of servicing – there is no other way around it.

Servicing fees are uneconomic because the industry systematically bid their fees down to get more business on the theory that larger portfolios increased servicing “efficiencies” and so they could get easy money from lots of low maintenance borrowers. They left themselves no margin for error or the possibility that dramatically more borrowers would become delinquent. And now, they are stuck. If the lenders had done a better job lending or, servicers had done a better job pricing the cost of servicing, then they wouldn’t be in the mess they are in now.

The underlying problem is that servicers are in conflict, basically at war, with homeowners, investors and the economy. It is a Gordian knot that will either be solved by regulatory action, by the courts and foreclosures or by massive rebellion by investors and homeowners. But the servicing contracts, perversely, have the effect of putting the needs of the very few ahead of the needs of the many.

The best solution is meaningful principal mods for viable borrowers. With mortgages showing average losses of 70% or greater, there is plenty of leeway for deep reductions that still leave both investors and borrowers ahead. Anna Gelpern and Adam Levitan pointed out in a 2009 paper that the contracts themselves, the pooling and servicing agreements, make private resolution well nigh impossible, which means some sort of intervention is necessary:

Rewriting PSAs will not resolve today’s financial crisis. Yet voluntary foreclosure prevention initiatives are unlikely to succeed as long as contract rigidities persist. The continuing foreclosure epidemic also holds an important lesson for the future: even where contract rigidity makes perfect sense for the parties, pervasive rigidities can have catastrophic consequences for financial stability and for society.

There will also need to be a mechanism for recapitalizing banks that will suffer large writedowns on their second mortgage portfolios. The industry will of course try to characterize this process as the unnecessary “creation” of losses to try to blunt the well-deserved political backlash. In fact, the losses already exist; the nature of this exercise is to start recognizing them for the benefit of the economy, and figure out how to distribute the costs.

Dealing with the hit to servicers is rounding error compared to the balance sheet losses the biggest US banks need to recognize. And the truly important task is not servicer economics (although that needs to be addressed too) but how to restructure the major banks when they come back to the trough after paying executives and staff multimillion dollar compensation.