«Barriers to Federal Home Mortgage Modification Efforts During the Financial Crisis Patricia A. McCoy August 2010 MF10-6 Paper originally presented at ...»
Programs to Refinance Distressed Loans Failed Despite a substantial investment of time and money, both federal programs to help delinquent borrowers refinance into FHA loans were an abysmal failure. There were two main reasons why both of them foundered. First, servicers’ participation in the programs was voluntary in nature. Second, and related to the first, the federal government insisted that participating servicers and investors take sizeable principal write-downs as the condition of making FHA refinance loans.
The government had ample justification for not paying more to pay off loans than the underlying collateral was worth. Doing so would have saddled the government with unwanted
98 Internal Revenue Service (2008). See also Internal Revenue Service (2007).
27 losses while rewarding lenders and investors for making loans in inflated amounts. Nonetheless, servicers and investors proved adamantly opposed to taking the required write-downs.
Given the mortgage industry’s implacable resistance to reducing principal, any government attempt to revive a refinancing program will have to grapple with hard choices.99 For a voluntary program to work, the government would probably have to reduce any principal write-down to a trivial level or eliminate it outright. Alternatively, the government could compel servicers to sell loans to the government at a mandatory haircut and then refinance some or all of those loans into new FHA loans. That type of compulsion seems highly unlikely in the current political climate.
Loan Modification Programs Show More Promise, But Face Significant Hurdles The Right Model for Loan Modifications From 2007 to the present, the government’s approach to loan modification programs evolved. Initially, the government limited its involvement to coordinating and facilitating loss mitigation efforts by private industry. After that effort failed to gain traction, the Obama Administration augmented that approach with subsidies designed to change servicer incentives.
To a lesser extent, the Administration pursued isolated strategies to increase servicers’ cost of automatic foreclosure.
Model 1: Government Coordination And Facilitation Ultimately, the HOPE NOW model, in which the government limited its intervention to moral suasion and coordination of private industry, failed to produce sustainable loan modifications in any significant amount. Coordination alone was not enough to alter the incentives faced by servicers. Nor was it enough to streamline workouts to handle such large volumes of distressed loans. Although HOPE NOW marginally increased workouts, most of those workouts had a high risk of redefault because they increased mortgage payments for borrowers who could not even manage their existing monthly payments.
Nevertheless, the tremendous coordination efforts undertaken by HOPE NOW had a positive effect on the borrower end, by improving the response rates of distressed borrowers.
Servicers for private-label securitized mortgages have long complained that reaching distressed 99 In March 2010, for example, the Obama Administration announced a new set of liberalized FHA refinance rules for borrowers with underwater loans. Department of the Treasury (2010b).
28 borrowers can be difficult or impossible. Although its magnitude is unclear, the problem is very real. Some borrowers understandably hide their heads in the sand in the hopes of delaying foreclosure. Others have given up or don’t trust their servicers.100 Outreach activities by HOPE NOW and other groups including NeighborWorks America helped tackle this problem by putting homeowners in touch with housing counselors. Still, more could be done in this regard. Servicers for private-label RMBS usually do not hire “door knocker” firms, probably due to doubts that investors would reimburse them for the fees incurred.101 In contrast, Fannie Mae and Freddie Mac have begun hiring these services, which drive to delinquent borrowers’ homes and knock on the door. Door knockers have had substantially more success in reaching borrowers to broach loan workouts, with rates as high as 90%.
Coordination could also help alleviate the frictions posed by junior mortgages. Many first-lien servicers cannot even locate the holders or servicers of junior-lien loans in order to discuss subordination. A centralized government registry of servicers for first- and junior-lien mortgages could help solve this coordination problem.102 Model 2: Federal Subsidies to Servicers HAMP tackled the fundamental task of altering servicers’ revenue and cost calculus through a carrot in the form of cash subsidies. It succeeded in making capitalizations less common and interest rate reductions more common. This is a worthy accomplishment, given how important lower monthly payments are to reduced redefault rates.
However, HAMP boosted interest rate reductions at the expense of capitalizations mostly in the context of temporary loan modifications. The larger goal – more permanent loan modifications with lower monthly payments – has remained frustratingly elusive, although the recent numbers have encouragingly improved.
The government has tinkered with various solutions to this problem, including streamlined borrower documentation, on-site visits to servicers, threats of sanctions, and public shaming.103 These efforts contributed to the rising level of permanent loan modifications.
100 See Mullainathan, Schoar and Tantia (2010).
101 Cordell et al. (2008), at 10.
102 Kiff and Klyuev (2009), at 27.
103 For instance, the latest HAMP report lists the rate of active loan modifications by name for 22 individual servicers. Department of the Treasury (2010c). Meanwhile, with respect to compliance oversight and sanctions, the Treasury Department did not plan to roll out remedies or penalties for servicers who violated HAMP guidelines until April 2010. Government Accountability Office (2010), at 11.
29 Nevertheless, the key question remains: is the current level of federal subsidies enough to overturn the strong disincentives that servicers face to making sustainable loan modifications on a permanent basis? And if not, what will it take?
Bringing permanent modifications to scale faces a difficult set of challenges. First, economic trends may cause servicers to become even more skittish about the cost-effectiveness of large-scale loan modifications. If serious delinquencies continue to rise, so may redefault rates. Similarly, if home values fall in individual localities or unemployment rises, that will tilt the NPV calculus further toward immediate foreclosure. While HAMP cannot tackle all of these problems, it might be able to improve redefault rates if – and that’s a big if – it can convince servicers to make permanent loan modifications that lower monthly payments.104 The best way to lower redefault rates is to reduce principal for borrowers with underwater mortgages. Research shows that negative equity combined with unaffordable payments is a leading driver of defaults, particularly once negative equity hits 20%.105 If underwater borrowers can regain some equity, they will have more incentive to stay current on their loans. But as we have seen, servicers fight principal reductions tooth-and-nail because these reductions cut into their monthly servicing fees. They further resist principal haircuts because accounting rules saddle them or their trusts with immediate losses for permanent loan modifications via mandatory write-downs, which can then eat into their capital.
From the larger standpoint of the economy and society, it would be better for financial institutions and trusts to move forward by recognizing these losses now. But so long as regulators give servicers, banks, and trusts discretion to avoid immediate recognition of losses, most will avoid taking unnecessary write-downs at all cost. In fact, one reason why the March 2010 revisions to HAMP made principal write-downs optional, not mandatory, may have been to avoid any appearance that the capital of the nation’s frail banking system was being further eroded.
104 The government also needs to consider taking borrowers’ total debt into account – including credit card, car payments, and other non-real estate debt – and not just mortgage debt when computing the 31% DTI ceiling. Currently, HAMP ignores non-mortgage debt when lowering mortgage payments below 31% of gross income. In this sense HAMP operates with blinders, meaning that heavily indebted borrowers may have difficulty even meeting HAMP-modified payments. This throws the sustainability of many permanent modifications into question, in turn, since over half of all permanent modifications under HAMP involve borrowers with total DTIs of over 50%. Government Accountability Office (2010), at 15-16; Office of the Special Inspector General for the Troubled Asset Relief Program (2010), at 15-16.
To date, the Treasury Department has instead approached the problem by requiring servicers to tell borrowers with a total DTI of over 55% to get HUD-approved housing counseling. However, the government does not track the number of borrowers who actually obtain that counseling because it considers it overly burdensome. Nor does the government measure the efficacy of counseling in reducing redefaults. Government Accountability Office (2010), at 16, 20.
105 Bernanke (2008), at 4.
30 Addressing this problem will require strong medicine. One way would be a fundamental overhaul of the accounting rules or a decision by federal regulators to require financial institutions to take those write-downs. Tweaking HAMP’s requirements will not achieve either step, however.
Alternatively, HAMP could subsidize principal write-downs to vitiate their effect on servicing fees and on capital.106 The March 2010 revisions to HAMP started down this road.107 But because principal haircuts remain at servicers’ discretion under those revisions, it is necessary to ask whether the March 2010 subsidies are large enough to induce those writedowns. The subsidies may need to be larger – substantially larger – to achieve that goal.
Alternatively, the government could offer to subsidize the cost to servicers and investors of any subsequent redefault. Finally, it is time to consider introducing the threat of a mandatory haircut to principal if servicers do not agree to voluntary haircuts themselves.
Model 3: Increasing the Cost of Rushing to Foreclosure The current amount of subsidies may not be sufficient to yield permanent loan modifications in optimal amounts. For that reason, the government should consider steps to increase the cost to servicers from needless foreclosures.
The first is more data and transparency. One of the untold stories of foreclosure prevention efforts so far has been the remarkable effect of exposés and shaming. Alan White threw the government’s rosy statistics on loan workouts by national banks, thrift institutions, and HOPE NOW into doubt by publishing proof that the majority of those workouts increased borrowers’ monthly payments rather than lowering them. His seminal research and the press coverage that it sparked led to renewed public pressure for more sustainable modifications.
More recently, government data about the scarcity of permanent loan modifications under HAMP focused attention on ways to increase permanent modifications. Public outrage over TARP put pressure on major banks to step up loan modifications as well.
106 Kiff and Klyuev (2009), at 23, for instance, propose having the government pay half of any principal write-down.
107 In a similar vein, there is a question whether HAMP’s subsidies are big enough to compensate servicers for the re-underwriting they must do to process permanent modifications. Until recently, borrowers could qualify for HAMP trial modifications without submitting paperwork, which eased the underwriting burden of servicers at the trial modification stage. At the permanent modification stage, however, servicers had to gather documentation from borrowers, verify their eligibility for HAMP, and underwrite their ability to handle a permanent modification. This latter task cost considerably more, raising questions whether HAMP subsidies are sufficient to cover the expense.
In January 2010, to facilitate conversions to permanent modifications, the Treasury Department started requiring servicers to verify borrowers’ eligibility for HAMP before starting trial modifications effective April 15, 2010. Government Accountability Office (2010), at 6-7, 11-12; Office of the Special Inspector General for the Troubled Asset Relief Program (2010), at 13. Because this did not alter the financial incentives faced by servicers, the result will probably be to reduce the number of new trial modifications.
31 None of this would have happened without publicly available data on outcomes and reporting. Data reporting needs to be standardized and expanded to include all key fields bearing on sustainable outcomes. The Government Accountability Office has reported in that regard that the Treasury Department still does not systematically track critical metrics, such as servicer compliance, denials of trial modifications, or the default rate on trial and permanent loan modifications. Nor has the government established benchmarks for performance.108 Second, the federal government should make foreclosure contingent on good faith loan modification attempts. The March 2010 changes to HAMP tackle this objective in two ways: by forbidding foreclosures from going forward while HAMP negotiations are pending and by requiring servicers to certify that borrowers are not eligible for HAMP as a prerequisite for foreclosure.109 An even stronger measure would be for Congress to enact legislation requiring servicers to go to court-supervised mediation and bargain in good faith before they can proceed to foreclosure.110 Both Philadelphia and the State of Connecticut, for example, mandate courtsupervised mediation before a servicer can foreclose on a loan.111 Finally, it is time to enact federal bankruptcy cram-down legislation for owner-occupied homes. Such legislation would insure equal treatment for all secured loans.112 More importantly, it would alter servicers’ cost calculus by forcing them to factor in the risk that a bankruptcy judge would write down the value of the home.113 The prospect of a cram-down would help counteract the resistance to principal haircuts in a way that nothing else has to date.