Senate Committee on Banking, Housing, and Urban Affairs • April 10, 2008 • State & Local
Douglas W. Elmendorf • Former Director, The Hamilton Project; Former Senior Fellow, Brookings Institution
Chairman Dodd, Ranking Member Shelby, and Members of the Committee, I appreciate the opportunity to appear before you today.
This hearing addresses a set of serious and interrelated challenges now facing the American economy. The latest data on employment, spending, and production suggest that the economy is very likely in recession. Housing construction has fallen more than half since its peak and has yet to level out, much less to resume an upward trajectory. The turmoil in our credit markets has lasted much longer and become much more severe than most observers predicted even six months ago. And, in the absence of further policy action, several million families will probably default on their mortgages in the next few years and lose their homes to foreclosure.
Designing effective policy responses to the housing and mortgage problems is not easy. In particular, it is neither feasible nor appropriate for the government to ensure that all homeowners, regardless of their mortgages or overall financial situations, are able to stay in their homes. However, it is both feasible and appropriate in my view for the government to reduce the number of homeowners who will lose their homes in the next few years. Policy actions in this direction would also have favorable effects on the broader economic problems we face today.
The first section of this testimony summarizes briefly the causes of the current housing and financial crisis, and the second section reviews the significant problems we now confront. The next section of the testimony presents the case for greater government involvement in the mortgage market today, acknowledging the legitimate arguments of those who prefer to let market forces play out unhindered, but emphasizing the strong justifications for a vigorous government response under current circumstances. The following section turns to specific policy options, commending some provisions included in the compromise Senate housing bill but urging additional timely action — especially regarding an expansion of the FHA’s role in helping families with negative home equity to refinance their mortgages.
It is now apparent that many individuals and institutions took greater financial risks than they intended during the past several years. Excessive risk-taking occurred through both the risk of the assets purchased and the degree of leverage used to finance those asset purchases. Numerous factors contributed to this over-reaching, but the following ones seem to have played central roles.
The first key factor was macroeconomic conditions, especially the low level of short-term interest rates. Because the recovery from the 2001 recession was slow, and because inflation was low and declining for part of the recovery, the Federal Reserve kept the federal funds rate very low in the first part of this decade. The nominal federal funds rate was below 3 percent continuously from September 2001 through May 2005 after being below 3 percent for only five months of the preceding decade. The real funds rate was negative for several years, the longest sustained period of a negative funds rate since the mid-1970s. I have argued elsewhere that the course of monetary policy was largely justified by the outcomes for inflation and unemployment, but the policy nonetheless contributed to the excessive leverage. In particular, low short-term interest rates induced greater short-term borrowing. They also seemed to encourage investors to take greater risks in a “reach for yield.” A natural way to take greater risks in such an economic environment was to use additional leverage to buy more assets.
At the same time, house prices were rising rapidly, partly because of fundamentals that justified higher house prices in at least some parts of the country and partly because of the borrowing behavior just described. This house price appreciation fed on itself: Price appreciation made housing a more desirable asset for households, which encouraged further leveraging in order to buy more houses. Price appreciation also kept delinquency rates on mortgages very low, which encouraged lenders to provide more credit for housing. The induced housing demand generated further increases in house prices.
The second key factor was a wave of financial innovation during the past decade that created new products and institutions. These products and institutions addressed real needs but also carried substantial disadvantages that were not widely understood at the time, including a low degree of transparency, poor alignment of incentives, and greater difficulty in resolving credit problems. To take these disadvantages in turn:
· One disadvantage is that the new products and institutions had a low degree of transparency and a high degree of complexity. Some of the new products were nontraditional mortgages, including interest-only mortgages, negative amortization mortgages, and mortgages with teaser rates. The products were apparently not well understood by many who borrowed money this way and by many who lent money this way, and this problem was made worse by the expansion of mortgage credit to people with weak credit histories and other risk factors such as very low initial equity or undocumented income. Other new products were unconventional credit-market instruments, particularly derivatives on asset-backed securities that had complicated payoff patterns. The new securities went several steps beyond the basic mortgage-backed securities that had been widely used and traded for several decades, and they presented several transparency challenges: They were intrinsically more complicated than basic asset-backed securities; even supposedly sophisticated investors had little familiarity with them; and there was little track record of their performance. These problems were compounded because that limited track record was exclusively from a period of rapidly rising house prices, which disguised a multitude of sins. Yet more complexity and less transparency was introduced to the financial system through products that purported to protect or insulate investors from risk, such as credit default swaps, bond insurance, and shifting liabilities off balance sheets onto structured investment vehicles (SIVs) or other entities.
· Another disadvantage of this wave of financial innovation was that the new products and arrangements worsened the alignment of incentives between the people making and advising about financial decisions and the ultimate investors. Such principal-agent problems are endemic in financial markets and institutions, but they were exacerbated in recent years by financial innovation. One example is mortgage brokers who were compensated for the volume of transactions they initiated and had little incentive to monitor the quality of loans they made. Another example is credit ratings agencies that are paid by the sellers of securities rather than by the buyers; as securities became more complicated, investors’ reliance on the agencies’ judgment increased. A further example is investment bankers who benefited from selling securities and did not bear the consequences of poor investments.
· Yet another disadvantage was that the new products and institutions presented greater difficulties in working out problems. For example, as discussed further below, modifying mortgages sometimes makes more sense for lenders than foreclosing on properties. The decision to modify a mortgage is administratively easy when the mortgage lender is also the servicer; the decision is administratively difficult when the lender is different than the servicer; and it is even more difficult when the lenders are numerous and hold different tranches of a mortgage pool that includes the mortgage in question.
The U.S. economy now faces four serious and interrelated problems.
First, the economy is very likely in recession. Employment fell in each of the past three months, retail sales and industrial production both declined in the latest reports, and consumer sentiment has slipped to its lowest level in five years. Moreover, financial conditions are hardly conducive to spending: Since the middle of last year, stock prices have dropped roughly 10 percent, and house prices have dropped between 2 percent and 7 percent depending on the index one consults; the consequent reduction in household wealth is about $3 trillion. And despite the three percentage point cut in the federal funds rate since September, most interest rates paid by households and businesses are down only a little since the middle of last year or are actually up a little: For example, rates on 30-year fixed-rate conforming mortgages are down only 60 basis points, and Baa-rated corporate bond yields are up 20 basis points.
Second, housing is overbuilt and overpriced. Construction has fallen more than half from its peak and shows no signs of bottoming out, nor has the number of unsold new single-family houses substantially diminished. Moreover, both house-price futures and analysts’ estimates of sustainable house-price levels point to further sizable declines in house prices.
Third, the financial system is reeling, and lending to households and businesses is impeded. Uncertainty about the value of mortgage-backed securities, and especially about the value of complex derivatives of those securities, induced a general reassessment of financial risk, going well beyond the subprime mortgage market and beyond the residential mortgage market altogether. The resulting uncertainty about the solvency and liquidity of many financial intermediaries has led these institutions to try to reduce the risk and augment the liquidity of their balance sheets. Those steps in turn have pushed down the price of risky or illiquid assets and pushed up the rates charged for borrowing by households and businesses, as just noted.
Fourth, absent further policy action, several million families will likely default on their mortgages in the next few years and lose their homes to foreclosure. In some cases, this will occur because resetting mortgage rates push monthly payments out of people’s reach. However, declines in short-term interest rates since last year have reduced the magnitude of this problem. In more cases, foreclosures will occur because falling prices push house values below mortgage amounts, and people struggling to make their mortgage payments decide to stop struggling. Economist Mark Zandi of Moody’s has projected that 14 million families may end up with negative equity in the next two years and that 2 million of them will lose their homes. Foreclosures are clearly costly to homeowners in both personal and financial terms, and foreclosures are costly to borrowers, who may recover no more than half of the mortgage principal. Foreclosures are also costly to neighborhoods, communities, and cities, especially when the foreclosures are concentrated in geographic areas as they often are.
These four problems have generated a number of significant policy responses. Congress and the President agreed on a tax rebate that will be distributed in coming months to roughly 130 million families. The Federal Reserve has slashed the federal funds rate by 3 percentage points since September. In addition, the Fed has fulfilled its role as “lender of last resort” by providing a good deal of additional liquidity to financial institutions through a series of creative new lending arrangements and by organizing the sale of Bear Stearns to JPMorgan. All of the fiscal and monetary actions I have just described have been appropriate in my view, but they are not the focus of my remarks today.
Congress and the Administration have also recognized the importance of policies that tackle the housing and mortgage mess directly. In the fall the Administration expanded eligibility for refinancing into mortgages guaranteed by the FHA, the Federal Housing Administration. The Treasury Department coordinated an agreement among industry participants to freeze mortgage-rate resets in those cases where borrowers are unlikely to meet the higher payments. Congress raised the threshold amount for mortgages covered by Fannie Mae, Freddie Mac, and the FHA. And this week, of course, the Senate is acting on a collection of proposals directly aimed at the housing and mortgage markets.
Designing effective policy responses to the housing and mortgage problems is not easy: Many analysts and policymakers have struggled during the past six months to develop effective forms of government intervention and have been disappointed by a lack of appealing options. Still, the government can and should do more. In the next section I present the general case for further government involvement, and in the following section I turn to specific policy options.
History shows that the best way, by far, to organize economic activity in order to maximize people’s material well-being is through markets and private property. One hallmark of market-based economies like ours is that people generally make their own economic decisions — what to buy and sell, what to save and borrow. This system is sustainable only if people bear the consequences of those decisions. Therefore, some analysts and policymakers have asked the very legitimate question of why the government should become more involved in the mortgage market rather than letting market forces play out by themselves.
This skepticism about a greater role for the government can be elaborated along several dimensions:
To start, skeptics can note, foreclosures are an unfortunate fact of life in this country. Even in good times, many families end up with mortgages they cannot sustain. For example, Federal Reserve Chairman Bernanke noted in a recent speech that foreclosure starts in 2005 and 2006 were under 1 million per year and that more than half of foreclosure starts typically result in sale of the property — suggesting that about half a million families actually lost their homes to foreclosure in each of those years. With the sharp deterioration in underwriting standards during the past few years, still more families presumably ended up in mortgages that are unsustainably large even with government help. Trying to keep these families in their current homes would, so the argument goes, simply prolong their struggle with high mortgage payments and prevent other families with stronger economic positions from buying and living in those homes.
In addition, skeptics can argue that many families who will lose their homes are not especially deserving of government help. People with negative equity in their houses will be disproportionately those who bought houses without putting much money down or who refinanced and withdrew equity to support other consumption. These people are not actually losing much housing equity and have enjoyed a comparatively nice lifestyle. It is unfair, so the argument goes, to help homeowners who are defaulting and facing foreclosure while not helping people who kept renting rather than taking out mortgages beyond their reach or people who are also stretched to meet their mortgage payments but are making the sacrifices to do so. Moreover, helping borrowers and lenders who entered into contracts that are now unworkable will create so-called “moral hazard” by encouraging unduly risky borrowing and lending in the future.
These arguments contain some truth. However, and I want to emphasize this point, they are not the whole truth. Despite these legitimate concerns, the government has a crucial part to play in resolving the current mortgage mess. Let me explain why:
First, the government has long had an active role in housing and housing finance. This role stems partly from the view that homeownership encourages responsible citizenship and strengthens people’s ties to their neighbors and communities. It also stems partly from the view that financial markets do not always conform to economists’ idealized conception of markets: Asymmetric information between borrowers and lenders, leveraged financial institutions that are vulnerable to “runs” when savers’ confidence falters, and the possibility of contagion in the financial sector all justify government involvement. For these reasons and others, the federal government has granted tax deductibility for mortgage interest and excluded most house-price appreciation from capital gains taxes; it has fostered mortgage lending through its regulation of savings institutions; it has established the Federal Home Loan Bank System and the Federal Housing Administration; it has created Fannie Mae and Freddie Mac and provided an implicit guarantee to their securities; and so on. We now face challenges in housing finance that are unprecedented since the Depression of the 1930s, and it natural to think that the government’s role should be increased under these very unusual conditions.
Second, government policy never does — nor should — follow free-market principles absolutely. In all areas of economic policy, we balance the need for people to bear responsibility for their decisions with the goal of protecting vulnerable members of society. The families facing foreclosure appear to be a tremendously varied group: Although some struggling mortgage borrowers do not deserve our sympathy, many others were victims of predatory lending practices, entered into mortgage contracts they could not fully understand, or took risks on their mortgages to escape unpleasant or dangerous rental housing. These families do deserve our sympathy and our help. To be sure, some of these families would not own their current homes if risks had been recognized fully during the past several years, but government policy can ease their transition to a world with appropriate recognition of risks.
Third, the effects of turmoil in the housing and mortgage markets are felt well beyond the families that borrowed too much and the financial institutions that lent too much. Concentrated foreclosures lower property values throughout the communities in which they occur, hurting every family trying to sell its home in such areas. Wild gyrations in financial markets pose risks to everyone’s savings, including many people who were not trying to increase their leverage to squeeze out a higher return. The weakening of the overall economy hurts many workers who lose their jobs and cannot find new ones. Indeed, the downside risks to economic activity are especially pronounced now, and continued distress in the housing and financial sectors could launch a reinforcing downward spiral in which financial turmoil begets economic weakness, which causes further turmoil, and so on.
Fourth, mortgage markets are not functioning in a normal manner now. For example, many families that could easily obtain mortgage credit just a year or two ago now have great difficulty obtaining mortgages, in part because some of the largest mortgage lenders have suffered massive losses and are struggling to maintain their viability and because many types of mortgage-backed securities are viewed especially negatively in financial markets. Regulatory policy should have been employed more vigorously to reduce the swing in the financial pendulum toward laxness in lending, and government guarantees can be used selectively to reduce the swing toward stringency in lending.
Fifth, our standard approach to mortgage securitization severely limits the likelihood that servicers will modify large number of mortgages in ways that will prevent defaults. Mortgage servicers and lenders have several legitimate reasons to avoid writing down principal, which may be the most effective way to induce borrowers who have negative equity to stay current on their loans: Other borrowers will want the same deal, which greatly raises the cost, and some borrowers will default later even with a principal write-down, which raises the cost as well. But other factors imply that yet fewer loans will be modified than is optimal from the perspective of lenders. One obstacle is the dispersion of ownership through securities and derivatives. Although the pooling and servicing agreements (PSAs) generally give servicers the authority to make modifications that are in the interest of the lenders, the degree of latitude varies across contracts, and ownership of different tranches creates different incentives for different investors; all of this makes modifications a judgment call, which opens the door to legal challenges. In addition, some servicers who are willing to accept lower payoffs on some mortgages will want borrowers to obtain new mortgages from other lenders, and the current problems in mortgage markets make that very difficult for some borrowers.
With these points in mind, I now turn to specific policy options.
The compromise housing bill being debated in the Senate this week includes several provisions that will help to improve conditions in mortgage markets in the near-term.
One such provision is the appropriation of additional funds for mortgage counseling. Many families who lose their houses to foreclosure never contact a credit counselor or their mortgage servicer in advance. Yet, counseling by local organizations, and the interaction with mortgage servicers that results, has had a high success rate in the past. Therefore, it makes sense to appropriate additional funding for this purpose just as quickly as counseling organizations can build their capacity and use the funds effectively.
Other valuable provisions of the bill include the augmenting of funds for state and local governments through the increase in private-activity bond authority and the increase in Community Development Block Grants. These funds will help these governments to facilitate mortgage refinancing as a means of avoiding foreclosure, and to maximize the value of homes that have been foreclosed upon while minimizing the negative spillovers. Through both channels, some of the negative consequences of the prospective foreclosure wave will be staunched, especially in localities where foreclosures might be highly concentrated.
Despite these important provisions, however, the compromise housing bill falls short of what is needed in my view.
One major deficiency is that the bill does not include reform of the bankruptcy law to allow judges to write down principal amounts owed on mortgages. A careful examination of the merits and demerits of such reform is beyond the scope of this testimony. In brief, though, the reform clearly would not come without cost. As opponents argue, it would induce an increase in bankruptcy filings and would likely have some detrimental effect on the future supply of mortgage credit. However, if eligibility for “stripdowns” in bankruptcy were carefully limited, as is the case for the proposals that have received the greatest attention in Congress, then the effect on future credit supply would probably be quite limited as well. Moreover, this reform has the key advantage of targeting mortgage relief to those families that are in the most perilous economic circumstances and for whom relief is most appropriate — a targeting that is difficult or impossible to achieve through most other policies for addressing current mortgage problems. Therefore, I have reluctantly come to support this reform, and I urge Congress to pass it into law.
The other main deficiency of the compromise bill is that it does not include a significant expansion of the role of the Federal Housing Administration. Two templates for such an expansion have been circulated widely in Congress — one by Chairman Dodd of this Committee, and the other by Chairman Frank of the House Committee on Financial Services. Under their similar proposals, eligibility for FHA-guaranteed loans would be broadened to help more families refinance their mortgages when they have negative equity in their homes. Such an expansion would be an appropriate and important step forward for several reasons. (As this testimony was being finalized, news reports stated that the Administration would shortly announce an expansion in FHA eligibility to help families with negative equity in their homes. However, the scope and details of the Administration’s proposal are unclear at this writing.)
First, the FHA’s traditional mandate is to assist individuals underserved by the traditional mortgage market, and it has many years of experience in doing so. Given the pullback in private mortgage lending and securitization, it is natural to increase the FHA’s role as a counterweight. Although under normal circumstances, the FHA helps only borrowers who are current on their loans, last fall the administration expanded the program to include adjustable-rate borrowers who had been making timely payments but became delinquent following interest-rate resets. With negative equity now becoming a key contributor to rising foreclosures, an expansion of FHA programs that addresses borrowers with negative equity is the logical next step.
Second, these proposals are appropriately selective in the families they help. Although every foreclosure can be painful for the families involved and for the neighborhoods and communities in which they live, not every family can afford to stay in their current homes with a reasonable amount of government help. The FHA expansions that have been put forward recognize this hard truth, and they are explicitly limited to owner-occupiers that satisfy solid underwriting standards and represent good credit risks at their new mortgage amounts.
Third, the proposals have been constructed carefully to limit eligibility to circumstances where loans can be refinanced at low or zero cost to taxpayers. The proposals do not simply throw open taxpayers’ wallets to help anyone who would prefer to make smaller mortgage payments. Instead, they require servicers of existing mortgages to take substantial write-downs of the principal amounts owed, and they ensure that the FHA shares in any renewed house-price appreciation. The low expected cost to the government means that these proposals are not bailouts in the sense of providing large amounts of taxpayer money to get borrowers or lenders off the hook.
Fourth, these proposals provide an important incentive for servicers to reduce principal amounts owed. Chairman Bernanke and others have urged mortgage servicers to consider writing down principal amounts in the many cases where that approach will generate more value for investors than foreclosure. Mortgage servicers have not traditionally pursued this type of workout and probably lack standard procedures for doing so. In addition, they may be especially reluctant to mark down principal in cases where they would continue to hold the mortgage and thus be exposed to its various risks. The proposed legislation addresses these problems by offering a safe harbor against legal liability for servicers who participate in the program and by providing an FHA-guarantee for the new mortgage that facilitates its purchase by someone else. Importantly, participation in the program would be voluntary for servicers and lenders, so this approach would not restrict future credit supply.
One remaining obstacle is the prevalence of second liens. Schemes to refinance first mortgages into more appropriate ones typically cannot go forward without re-subordination of the original second liens. This has reportedly proven difficult when the holders of the second liens are different from the holders of the first liens. Looking for ways in which the government could help to coordinate this process should be an important goal.
Lastly, I should note that I am skeptical about further proposals to use some type of auction process for bringing more loans to the FHA. The appeal of such auctions is clear: They appear to provide a mechanism for the government to take timely action on a large number of mortgages. However, it is unclear how such auctions could distinguish effectively among families and mortgages in the ways that are needed. For example, if the government buys pools of mortgages offered at the largest discounts, and if servicers know more about their customers than the government and are sophisticated in using that information, then the government will end up buying the riskiest mortgages. This selection problem would expose the government to additional risk and expense. That said, the auction idea deserves further study, and I would be pleased to see it included in legislation on that basis.
After the current housing and financial crisis has passed, reducing the probability of its recurrence must be a high priority for policymakers. Financial markets will always experience swings between confidence and fear, but appropriate changes to our system of financial regulation and oversight can reduce the frequency, magnitude, and broader consequences of such swings.
At this time, however, we do not have the luxury of choosing between the messy policy options available to us and an idealized world in which such ad hoc policies are not needed. Instead, we must choose between messy policy options and inaction — and the cost of inaction is very high. I urge this Committee and other Members of Congress to move beyond the policy responses already in law or under consideration as part of the compromise Senate bill. In particular, a measured expansion in the role of the Federal Housing Administration as proposed by Chairman Dodd and Chairman Frank would contribute importantly to reducing the size of the coming foreclosure wave.
To be sure, this proposal and others focused on the mortgage mess are not silver bullets for our economic problems: They will not completely prevent a rise in foreclosures, halt the decline in house prices, restore stability in financial markets, nor avert a recession. However, they can reduce the scale of these broader problems, helping to avert an overshooting of housing prices and helping to stabilize the prices of risky financial assets. Moreover, they can do so with limited repercussions for future mortgage lending and risk-taking, and at fairly low cost to taxpayers.
Thank you very much.
A periodic newsletter of events, policy briefs, and working papers from The Hamilton Project.