Misleading Study on Servicing Settlement

The bank-funded study on the supposed economic costs of the state AGs proposed settlement with mortgage servicers is a patchwork of unsupported assumptions all designed to lead to one conclusion – that the big servicers who funded the study ought to get a free ride.

Financial journalist Felix Salmon has already done a devastating take on the problems with this study. Iowa Attorney General Tom Miller pointed out that contrary to the study’s assumptions, a goal of the settlement is to speed up legitimate foreclosures, which have ground to a halt in many states due to the cloud of legal problems surrounding servicer behavior. Georgetown Law professor Adam Levitin reminded us that the servicers should expect to pay a cost for breaking the law and depriving both homeowners and investors of legal protections they were entitled to. And both Levitin and economics blogger Mike Konczal provided the evidence that servicers haven’t been willing or able to perform sustainable modifications on their own.

This ‘study’ takes three easy steps to make the servicer settlement look like a bad deal:

• First, assume away the problem: The authors start by assuming there’s no problem we need a legal settlement to fix. This completely ignores the endemic problems with servicer performance and incentives. Incredibly, the report assumes that servicers are already making all the mortgage modifications that would be economically efficient – that save money as compared to a foreclosure. But we’ve seen beyond doubt that the servicers lack both the proper incentives and the competence to do economically rational mortgage modifications.

• Next, make up some costs of the settlement: Then, they mischaracterize costs to the servicers – in other words, transfers from the big banks to the rest of us – as costs to the economy as a whole. Other assumed costs result from a completely unsubstantiated claim of a 25 percent increase in so-called “strategic defaults” due to a settlement. They also simply assume that a settlement will delay future foreclosures by close to a year – despite the fact that a settlement would speed appropriate foreclosures by ending servicer delays in processing modification requests
• Finally, ignore the benefits of a settlement:  Of course, the major reason to ask servicers to do a better job preventing foreclosures is that foreclosures have big social and economic costs. The authors’ own extremely conservative assumptions in this paper imply that the 2 million loan modifications targeted by a settlement would prevent at least 700 thousand foreclosures. The latest research finds social costs of $80,000 per foreclosure. So the authors own assumptions imply that the settlement agreement would create economic benefits of $56 billion. But you won’t find those benefits mentioned anywhere in this study.

Add it all up, and there’s nothing there. More detail below…

Assume away the problem. The report assumes that servicers are already making all the mortgage modifications that would be economically beneficial. (“That servicers have not modified more loans indicates that, under their NPV analyses, additional modifications would not result in higher payouts for investors, despite the benefits of avoiding a protracted and expensive foreclosure process” – page 7). But the entire issue is the failure of the servicers to do economically rational mortgage modifications. Servicers lack both the proper incentives and the competence to perform modifications passing a Net Present Value (NPV) test (this test shows that the modification would save money as compared to a foreclosure). The government mortgage modification (HAMP) program has provided data on the reasons why servicers fail to perform modifications when borrowers apply. Just 6 percent of modification applications are turned down because they fail the NPV test and don’t make economic sense – but over one fifth are turned down for paperwork reasons.

Make up the costs. The report estimates various costs of a settlement. They start by describing the requirement that servicers transfer $25 billion in principal reductions to selected borrowers.   The 25 billion penalty is a transfer from banks to homeowners that reflects estimated excess servicer profits from breaking the law. Then the authors add an admitted guess as to the cost to servicers of complying with the law and their contracts by providing adequate servicing of loans. Finally, they assume that every dollar of these costs will be passed on to future home buyers in the form of higher interest rates – none will come from bank profits.

They authors also add on additional costs resulting from a guesstimate of a 25 percent increase in “strategic defaults” induced by a settlement.  This figure is completely made up and has no substantiation (see footnote 47 of the paper – “We use a 25 percent increase in strategic defaults for illustrative purposes only. It is unclear how much strategic defaults would increase under the settlement”).  The assumption that the settlement would lead to any increase in strategic defaults ignores the heavy costs to homeowners of strategic default – humiliation, the risk of home loss, and a ruined credit rating – as well as the fact that the modifications contemplated under the settlement are carefully targeted to homeowners who demonstrate a real need for the assistance. Indeed, the authors’ main substantiation for the idea that strategic defaults will increase is a study of tiny cash loans to poor, high-risk borrowers in South Africa, not home mortgage lending to families in the United States (see pp. 9-10 of the study). The authors then assume that each and every strategic default will lead to a costly foreclosure (see footnote 48 of the paper), despite the fact that their only rationale for claiming increased strategic defaults is that making loan modifications more accessible will tempt homeowners to default in order to get a modification and keep their home.

In a final leap of logic, the authors then assume unspecified costs due to foreclosure delays supposedly created by the settlement. The only substantiation for these supposed delays is a long list of the various timeframes set out in the settlement term sheet, which generally require servicers to evaluate borrowers for a loan modification without delay. But the authors assert that the imposition of those limits on servicer delay would cause delay. Delays occasioned by the servicers’ malfeasance, such as the submission of false documents in the foreclosure process, are also attributed to the settlement.   There are only two timeframes listed by the authors that would delay a foreclosure:  the requirement that borrowers denied a loan modification be given 30 days to challenge the denial, and the three month trial modification period.  But this trial modification period will only last three months if the borrower is making payments:  otherwise servicers are free to proceed with a foreclosure.    A settlement should actually increase the pace of legitimate foreclosures, by forcing servicers to modify or foreclose in a timely way, as opposed to the endless stalling, losing documents, and servicer-initiated delay we see now–which benefits servicers, but not homeowners or investors, and by clearing aside legal barriers to appropriate foreclosure and by.

Ignore the benefits. The authors generate their costs based on an estimate of 2 million mortgage modifications for homeowners created by the settlement (see footnote 43). Elsewhere in the paper, they cite a figure that 30 percent of mortgages in default self-cure without modification (see page 9 of the paper) and that 50 percent of modified mortgages re-default (see page 3 of the paper). These figures are likely too high (government data clearly shows that the re-default rate has been well under 50 percent for properly designed mortgage modifications). Nevertheless, taking the authors’ own assumptions implies that the 2 million additional mortgage modifications created by the settlement will lead to 700 thousand foreclosures averted.

So the settlement will prevent at least 700 thousand foreclosures. A recent study by the Urban Institute shows that each foreclosure leads to economic costs to society of about $80,000 each. So the 700 thousand foreclosures averted by the settlement would create economic savings of $56 billion. Unlike the ‘costs’ estimated in the report, these benefits are actual increases in total economic output – not savings to servicers. These economic benefits – generated from the authors own assumptions and figures – are nowhere mentioned in the report.