Quantifying the Ambition and Impact of the New Mortgage Program

Click here for a printer-friendly version of this article

The Obama Administration has proposed a new mortgage forgiveness plan that aims to reduce foreclosures by promoting default. Essentially, the Obama plan would encourage banks to recognize that the outstanding principal balance of the mortgages they issued (or acquired) is too large relative to the underlying value of the property that collateralizes the debt. In exchange for allowing for a de facto default on the existing mortgage debt, the government would pay banks for the amount of debt they forgive according to a defined schedule and use the Federal Housing Administration (FHA), which happens to be insolvent, to issue guarantees against the new mortgage.

For too long, the housing debate had centered on cash flow rather than assets. Policymakers and analysts focused on the relationship between a household’s income and its monthly mortgage payment rather than the relationship between the outstanding mortgage balance and the market value of the house. As Federal Reserve researchers have found, subprime adjustable rate mortgage (ARM) resets – loans that started at a low “teaser” rate and then reset to a fixed rate that was several percentage points higher – received disproportionate attention and thus distracted researchers from focusing on the bigger issue: rising losses on mortgage assets as a result of the dramatic decline in the market value of housing relative to outstanding mortgage debt.

Most home loans are “non-recourse,” either in law or practice, which means that the borrower is not personally liable for the mortgage debt and can walk away from the house without any additional obligations. For this reason, the market value of a mortgage depends on the market value of the property that serves as the underlying collateral. Think of a lender that charges 6% for an 80% loan-to-value ratio (LTV) mortgage and 7% for a 95% LTV loan. When the price of a $200,000 house with an 80 LTV mortgage falls to $165,000, the economic value of the mortgage has obviously deteriorated because the borrower continues to pay at 6% for a risk the lender normally requires 7% to accept. Thus, even if the borrower continues to make payments, the economic value of the loan has been impaired by the decline in the value of the collateral.

From 1998-2006, the total face value of outstanding residential mortgages increased by more than $6 trillion. The financial system was easily able to accommodate over $670 billion in net new mortgage debt per year because the total market value of US residential property increased even more rapidly – by $13.9 trillion – over the same period, leaving the home equity-to-home value ratio roughly constant. However, starting in early 2007, the value of the US housing stock began to decline (including the value of newly constructed properties). In the two years (eight quarters) ending on April 1, 2009, the market value of US residential real estate fell by $7.3 trillion while outstanding mortgage debt grew by another $600 billion over that period. As a result, the ratio of home equity-to-home values fell by one-third and the US financial system faced asset impairment on a systemic basis.

The Obama plan would use TARP funds to pay banks to take the outstanding mortgage balances down in cases when the borrowers are underwater. But this does not address impaired mortgages in cases where the house value had not fallen below the mortgage balance. What kind of mortgage forgiveness would be required to take outstanding mortgage balances to a sustainable level?

*in millions of dollars

From 1998-2006 the average home equity-to-home value ratio was 59%. This ratio was remarkably stable during this period, with a quarterly high of 61% and a quarterly low of 57%. Assuming this 58% is a steady state equilibrium, achieving this ratio in 2010 would require a $3.5 trillion reduction in outstanding mortgage balances.

Obviously, such a swift resolution will be better implemented over a longer horizon, but the current mismatch is so great that if the total market value of residential real estate grows by 4% per year, home equity would not reach 59% of assets until 2020. Mortgage balances are likely to fall from current levels. The question is whether that will come from principal reductions on existing mortgages or a decline in access to mortgage credit going forward. Part of the reduced access to credit would be welcome, as large down payments would be required to buy homes. But part of the decline would obviously harm would-be borrowers’ ability to own homes – and the political system has historically been unwilling to allow this to occur.

The Obama Administration is right to view principal reductions as the key to reducing foreclosures and limiting bank losses, but the scale of the problem is far larger than the proposed solution. The latest housing bailout plan is unlikely to have any better results than past efforts. Nor is it clear that the small positive benefits from the program outweigh the problems associated with obvious moral hazard of debt forgiveness. The bottom line is that current economic and policy trends point to de facto government ownership of housing market credit obligations for the rest of the decade.