A successful plan to stabilize the U.S. economy and prevent a deep global recession must do more than buy back impaired debt from financial institutions. It must address the fundamental cause of the crisis: the downward spiral of house prices that devastates household wealth and destroys the capital of financial institutions that hold mortgages and mortgage-backed securities.
The recently enacted financial rescue plan does nothing to stop this spiral. Credit will not flow and liquidity will not return to the banking system until financial institutions have confidence in the solvency and liquidity of other banks.
Because of the 20% fall in the price of homes since the bursting of the house-price bubble, there are now some 10 million homes with mortgages that exceed the value of the house. Residential mortgages are generally "no recourse" loans, meaning that if the homeowner stops making payments, the creditor can take the property but cannot take other assets or attach income. Individuals with loan-to-value ratios greater than 100% therefore have an incentive to default even if they can afford their monthly payments, and to rent an apartment or other house until house prices stop declining. When individuals default and creditors foreclose, the property is added to the stock of unsold homes. That depresses prices further, increasing the number and magnitude of negative equity houses.
The prospect of a downward spiral of house prices depresses the value of mortgage-backed securities and therefore the capital and liquidity of financial institutions. Experts say that an additional 10% to 15% decline in house prices is needed to get back to the prebubble level. That decline would double the number of homes with negative equity, raising the total to 40% of all homes with mortgages. The mortgages of five million homeowners would then exceed the value of their homes by 30% or more, which could prompt millions of defaults.
The process of default and foreclosure leading to price declines and further defaults could take house prices far below the long-term sustainable level. But even when prices seem low, prospective buyers will delay buying as long as they expect prices will continue to fall.
The financial rescue plan would bring back the confidence needed to revive the financial system only if the Treasury's asset purchases could eliminate the current impaired securities now held by the financial institutions, and if the remaining securities could be counted on to remain healthy. The legislation will do neither.
More than $700 billion is needed to buy all of the impaired securities. The impaired mortgage-backed securities reflect not only the negative-equity mortgages but also positive-equity mortgages with very high interest rates, adjustable rates, or negative amortization. Even if the government could purchase every troubled mortgage, the prospect of future price declines would contaminate the mortgage portfolios. As house prices fall, the value of mortgage-backed securities would fall further.
The impaired assets are not just mortgages but the complex derivatives based on those mortgages: the collateralized debt obligations, the various risk slices of those CDOs that, even if rated AAA, often have market prices close to 50% of their notional value. In addition, hundreds of billions of dollars of credit default swaps "guarantee" the value of mortgage-backed securities.
There are also important technical problems in using the $700 billion fund to buy impaired securities. The Treasury's preliminary idea was to use a "reverse auction," a method that works well when used to buy a single homogeneous security (like a firm buying back its own shares). But that is not feasible for buying the impaired securities, because of the enormous variety of underlying mortgages and of the almost limitless number of different derivatives based on those mortgages. The buyback will therefore involve a large number of arbitrary valuation decisions by the Treasury staff and their investment-banker advisers.
Because of the arbitrary pricing, many banks will choose not to sell some of their assets. Mark-to-market rules may force banks to write down the value of their remaining securities, further reducing their capital and subjecting them to margin calls that reduce their liquidity. Institutions that avoid marking their assets to market will simply cause a cloud of uncertainty about the value of their portfolio.
The features that Congress added to the initial Treasury plan do nothing to achieve sustained confidence in the financial institutions. They provide Congressional oversight, delay the use of funds, create partial government equity ownership in some firms and do other things to protect taxpayers. But they do not address falling home prices.
We need a firewall to break the downward spiral of house prices. Here's how it might work. The federal government would offer any homeowner with a mortgage an opportunity to replace 20% of the mortgage with a low-interest loan from the government, subject to a maximum of $80,000. This would be available to new buyers as well as those with mortgages. The interest on that loan would reflect the government's cost of funds and could be as low as 2%. The loan would not be secured by the house but would be a loan with full recourse, allowing the government to take other property or income in the unlikely event that the individual does not pay. It would by law be senior to other unsecured debt and not eligible for relief in bankruptcy.
The individual could repay the loan at any time or could refinance the remaining loan on more favorable terms as long as the principal did not increase. A 30-year amortization of the government loan would make the payments low, and a life-insurance policy would protect taxpayers if the borrower dies before the loan is repaid. If the homeowner chooses to accept the loan, creditors would have to accept the 20% mortgage repayment, reducing the monthly payments of principal and interest by 20%.
Consider a homeowner who has a mortgage equal to 90% of the value of his home. The 15% decline in the value of his house that may be needed to bring it back to its prebubble level would shift that homeowner into negative equity. Further price declines would make default attractive. But the 20% mortgage replacement loan would take the loan-to-value ratio to 72% from 90%, making it unlikely that prices would fall far enough to push him into negative equity. An interest saving that could be as large as $3,000 a year would provide a strong incentive to accept the mortgage-replacement loan, even if the individual thinks that he might temporarily have a moderate level of negative equity.
Although the total size of the mortgage-loan program might be as much as $1 trillion, this would not be comparable to other government spending or to a swap of government bonds for impaired assets. The government would instead have a fully offsetting claim on individuals who could be counted on to repay their low-interest government loans. The cash that the banks and other creditors would receive from the government to replace the existing mortgages would be available to finance new loans.
Mortgage-replacement loans cannot solve all the problems of the housing sector. Unlike the recent Frank-Dodd legislation, it does not provide help to those who now face foreclosure. Homeowners without mortgages would benefit only indirectly from the program because it would stop the decline in the value of their homes. But everyone would benefit from the overall economic effect of reviving the financial sector and the credit flows.
The recent financial recovery plan that Congress enacted will not rebuild lending and credit flows. That requires a program to stop a downward overshooting of house prices and the resulting mortgage defaults. The mortgage-replacement loan program may be the best way to achieve that.
Feldstein, Martin. “The Problem Is Still Falling House Prices.” The Wall Street Journal, October 4, 2008