My testimony today will address two specific questions. The first is: why have the Obama and Bush Administration efforts, along with those of the mortgage industry, to reduce foreclosures had so little impact on the overall foreclosure numbers?
The second question is: given what we know about why previous efforts have had such little impact, what are our policy options?
In answering both these questions, I rely on an extensive body of academic literature, the vast majority of which has been subjected to peer review, which has examined the determinates of mortgage delinquency and default. Foremost among this literature is a series of recent papers written by economists at the Federal Reserve Banks of Boston and Atlanta, in particular the work of Paul Willen, Christopher Foote and Kristopher Gerardi. My testimony owes a considerable intellectual debt to this research.
Why haven’t previous efforts stemmed the foreclosure tide?
The short answer to why previous federal efforts to stem the current tide of foreclosures have largely failed is that such efforts have grossly misdiagnosed the causes of mortgage defaults. An implicit assumption behind former Treasury Secretary Paulson’s HOPE NOW, FDIC Chair Sheila Bair’s IndyMac model, and the Obama Administration’s current foreclosure efforts is that the current wave of foreclosures is almost exclusively the result of predatory lending practices and “exploding” adjustable rate mortgages, where large payment shocks upon the rate re‐set cause mortgage payment to become “unaffordable.”
The simple truth is that the vast majority of mortgage defaults are being driven by the same factors that have always driven mortgage defaults: generally a negative equity position on the part of the homeowner coupled with a life event that results in a substantial shock to their income, most often a job loss or reduction in earnings. Until both of these components, negative equity and a negative income shock are addressed, foreclosures will remain at highly elevated levels.
Given that I am challenging the dominant narrative of the mortgage crisis, it is reasonable to ask for more than mere assertions. First, if payment shock alone were the dominate driver of defaults then we would observe most defaults occurring around the time of re‐set, specifically just after the re‐set. Yet this is not what has been observed. Analysis by several researchers has found that on loans with re‐set features that have defaulted, the vast majority of defaults occurred long before the re‐set. Of course some will argue that this is due to such loans being “unaffordable” from the time of origination. Yet according to statistical analysis done at the Boston Federal Reserve, the borrower’s initial debt‐to‐income (DTI) had almost no predictive power in terms of forecasting subsequent default.
Additionally if payment shock was the driver of default, the fixed rate mortgages without any payment shocks would display default patterns significantly below that of adjustable rate mortgages. When one controls for owner equity and credit score, the differences in performance between these different mortgage products largely disappears. To further illustrate this point, consider that those mortgages generally considered among the “safest” — mortgages insured by the Federal Housing Administration (FHA), which are almost exclusively fixed rate with no‐prepayment penalties and substantial borrower protections, perform, on an apples to apples basis, as badly as the subprime market in terms of delinquencies.
The important shared characteristic of FHA and most of the subprime market is the widespread presence of zero or very little equity in the mortgage at origination. The characteristics of zero or negative equity also explain the poor performance of most subprime adjustable rate mortgages. Many of these loans also had little or no equity upon origination, providing the borrower with little equity cushion when prices fell. Recognizing the critical role of negative equity of course raises the difficult question as to what exactly it is that homeowners are losing in the event of a foreclosure.
Central to the arguments calling for greater government invention in the mortgage market is that many, if not most, of the foreclosures being witnessed are “unnecessary” or avoidable. Generally it is argued that investors and loan servicers do not face the same incentives and that in many cases in would be better for the investor if the loan were modified, rather than taken to foreclosure, but still the servicer takes the loan to foreclosure.
The principal flaw in this argument is it ignores the costs to the lender of modifying loans that would have continued paying otherwise. Ex Ante, a lender has no way of separating the truly troubled borrowers, who would default, from those that would take advantage of the system, if they knew they could get a modification just by calling. As long as potentially defaulting borrowers remain a low percentage of all borrowers, as they are today, it is in the best interest of the investor to reject many modifications that might make sense ex post. In addition, lenders may institute various mechanisms to help distinguish troubled borrowers from those looking to game the system.
It is also claimed that the process of securization has driven a wedge between the interests of investors and servicers, with the implication that servicers would be happy to modify, and investors would prefer modifications, but that the pooling and servicing agreements preclude modifications or that servicers fear being sued by investors. The first fact that should question this assumption is the finding by Boston Fed researchers that there is little difference in modification rates between loans held in portfolio versus those held in securitized pools. There is also little evidence that pooling and servicing agreements preclude positive value modifications. According to recent Credit Suisse report, less than 10 percent of agreements disallowed any modifications. While the Congressional Oversight Panel for the TARP has been critical of industry efforts, even that Panel has found that among the sample of pools it examined with a 5‐percent cap on the number of modifications, none of the pools examined had actually reached that cap. If few pools have reached the cap, it would seem obvious that the 5 percent cap is not a binding constraint on modifications. In many instances the pooling agreements also require the servicer to act as if the servicer held the whole loan in its portfolio, raising substantial doubts as the validity of the “tranche warfare” theory of modifications.
A careful review of the evidence provides little support for the notion that high transaction costs or a misalignment of incentives is driving lenders to make foreclosures that are not in their economic interest. Since lenders have no way to separate troubled borrowers from those gaming the system, some positive level of negative value foreclosures will be profit‐maximizing in the aggregate.
Is cramdown the answer?
The high level of foreclosures has left many policymakers and much of the public understandably frustrated and searching for answers. One “solution” that has been regularly presented is to allow bankruptcy judges to reduce the principle balance of a mortgage loan to reflect the reduced value of the home, the so‐called “cramdown.” For a variety of reasons, I believe allowing cramdowns would have adverse market consequences while also providing little real relief to borrowers.
Given the unemployment‐driven nature of most foreclosures, and the inability of unemployed individuals to put forth a repayment plan under Chapter 13 of the bankruptcy code, it appears that cramdowns would do nothing for those most in need, the unemployed.
As proponents of cramdowns point out, vacation and investment properties can currently be subjected to cramdown. This raises the question: why aren’t the significant number of foreclosures involving investment properties being resolved via bankruptcy rather than the foreclosure process? The most likely reason is that property speculators realize that even a reduced mortgage value is likely to exceed the home value in the near future. With home prices still declining, a crammed down mortgage would be underwater in few months. The incentive facing most speculators is often to simply walk away and let the home fall into foreclosure. This would not be a significant problem if investment properties did not constitute approximately 40 percent of current foreclosures.
At this point, it is worth reflecting on these two points: cramdowns do little or nothing to help the unemployed and speculators can already pursue that route, but largely choose not to, as it isn’t in their economic interest. With speculators making up about 40 percent of foreclosures, and the unemployed likely making up to around 50 percent, it becomes apparent that at minimum cramdowns will do little to help at least 90 percent of borrowers currently in foreclosure.
The main function of a cramdown would be to serve as reduction in outstanding principle, thereby lowering the monthly payment. Even significant payment reductions may not offer long‐term solutions. According to the most recent OTS/OCC mortgage metrics report, of those delinquent borrowers seeing a payment reduction of 20 percent or more 37.6 percent were again delinquent twelve months later. Continuingly re‐modifying the same loan is not a solution for the borrower, investor, or lender.
We often use the term “speculator” to refer to purchasers that do not intend to live in the home and often quickly “flip” the home to make a quick profit. That definition is useful, but far too narrow. Many borrowers purchasing a home for occupancy did not do so solely for the consumption benefits of homeownership, but also for the investment returns. They were both consumers and speculators. As these speculators were generally not offering to share potential gains with their lenders, it is not clear why they should be allowed to share their losses.
Of the remaining borrowers, who were neither pure speculators nor unemployed, many of these borrowers invested little of their own cash in the home purchase. Once again, the empirical evidence demonstrates that minimal or zero downpayments on the part of borrowers are the leading mortgage characteristic in terms of predicting default. If borrowers, who have placed no money of their own at risk, are allowed to reduce their losses via cramdown, while also reaping any future appreciation, we are only encouraging future speculation in our housing markets. We should not act surprised if the next housing cycle of bubble and bust is even worst than the most recent.
Proponents of cramdown have also misrepresented the treatment of vacation homes and investor properties during a Chapter 13 bankruptcy. While the current Bankruptcy Code does allows secured debts other than those secured by a principal residence to be crammed down; if they are crammed down, the debtor is required to pay off the entire amount of the secured claim within the three‐to‐five year duration of the Chapter 13 plan. The debtor does not have 30 years to pay off a modified mortgage as the original loan term may provide. The borrower in these instances is required to pay the entire amount of the secured mortgage by the end of their payment plan. This is one of the reasons many owners of investment choose to walk way rather than seek bankruptcy protection.
Cramdown is often presented as simply a way to put pressure on lenders to negotiate, or to “bring them to the table.” It is no more appropriate, in a free society, to use the coercive stick of the state to bring lenders to the table, than it would be to use that stick to bring borrowers to the table. A government focused on the common good, the general welfare, does not choose sides in private disputes.
Less tangible, but perhaps more important in the cramdown debate is the message it sends to market participants, particularly investors. It has long been established in law, and in common sense for that matter, that the body of law relevant to and existing at the time of a contract enters into and comprises part of that contract. To change by legislative fiat the terms of contracts that have already been agreed to is to change the contract itself. I fear if the cramdown were to become law, we send a signal that any private agreement is subject to being re‐written depending on which way the political winds are blowing. This is a sure recipe to reduce investment and the overall reliance of market participants on contract. In order to rebuild public trust in both our markets and our government, I believe Congress should affirm its own trust in the voluntary decisions of private parties. To do otherwise is to weaken the very bonds that make a free and civilized society possible.
In speaking of investors, it is also important to remember that cramdown is not simply an issue of taking from lenders and giving to borrowers. As bad as that would be, it is made all the worse as the ultimate investors in mortgage related assets that will suffer losses rather than the largest banks. As the largest banks are mostly just servicers and not the ultimate investor, they will pass along any losses from cramdown to investors. As we have seen in the recent auto restructuring, often these investors are not large corporations or wealthy individuals; they are pension funds representing the retirement savings of millions, usually retired state and local government employees. I have yet to hear a compelling reason why retired teachers and firefighters should be forced to bear the burden of irresponsible borrowing and lending.
I am concerned that inherent in the title of this afternoon’s hearing is the assumption that if voluntary modifications are not working, we must look to coercive solutions. The force of the State must be applied to those unwilling to see the light. This assumption should trouble anyone who values a free society. I urge Congress to look for only those solutions that are voluntary.
Some voluntary alternatives to consider: encouraging bank regulators to give lenders more flexibility to lease out foreclosed homes to the current residents. Typically banks come under considerable pressure from their regulators not to engage in long term property leasing or management, as that activity is not considered a core function of banks. I believe we can avoid the larger debate of banks being property managers by giving banks greater flexibility in retaining properties with non‐performing mortgages as rentals, preferably to current residents.
In order to separate out deserving borrowers, who are trying to get back on their feet, from those simply walking away from a bad investment, Federal lending entities, such as FHA and the GSEs, should engage in aggressive recourse against delinquent borrowers who have the ability to pay, but simply choose not too. We should make every effort to turn away from becoming a society where legally incurred debts are no longer obligations to be honored but simply options to be exercised.
In concluding my testimony, I again wish to strongly state: the current foreclosure relief efforts have largely been unsuccessful because they have misidentified the underlying causes of mortgage default. It is not exploding ARMs or predatory lending that drives the current wave of foreclosures, but negative equity driven by house prices declines coupled with adverse income shocks that are the main driver of defaults on primary residences. Defaults on speculative properties continue to represent a large share of foreclosures. Accordingly for any plan to be successful it must address both negative equity and reductions in earnings. Cramdown fails on both accounts. I thank you for your attention and welcome your questions.
Foote, Gerardi, Goette and Willen (2009) “Reducing Foreclosures: No Easy Answers,” National Bureau of Economic Research working paper 15063.
Cordell, L., K. Dynan, A. Lehnert, N. Liang, and E. Mauskopf (2008). The Incentives of Mortgage Servicers: Myths and Realities. Finance and Economics Discussion Series, Federal Reserve Board 46.
Foote, C., K. Gerardi, L. Goette, and P. S. Willen (2008). Just the facts: An initial analysis of subprime’s role in the housing crisis. Journal of Housing Economics 17 (4).
Foote, C., K. Gerardi, and P. Willen (2008). Negative equity and foreclosure: Theory and evidence. Journal of Urban Economics 6 (2), 234–245.
Gerardi, K., A. Shapiro, and P. Willen (2007). Subprime outcomes: Risky mortgages, homeownership experiences, and foreclosures. Federal Reserve Bank of Boston Working Paper 07–15.
Hunt, J. (2009). What do subprime securitization contracts actually say about loan modification. Berkeley Center for Law, Business and the Economy.
Kau, J. B., D. C. Keenan, and T. Kim (1994). Default probabilities for mortgages. Journal of Urban Economics 35 (3), 278–296.
Sherlund, S. (2008). The past, present, and future of subprime mortgages. Finance and Economics Discussion Series 2008–63, Federal Reserve Board.