"A Home Price Firewall"

Martin Feldstein presents his plan to reduce mortgage loan defaults:

A Home Price Firewall, by Martin Feldstein, Commentary, Washington Post: Home prices are down 20 percent from their peak in 2006 and are falling rapidly... Experts predict an additional 15 percent decline during the coming year...

The danger is that home prices could spiral further down, hurting millions of homeowners and pushing the economy into a deep recession. ...

I believe the federal government should create a firewall to prevent too great a fall in housing prices. ... This can best be done through a program of mortgage replacement loans.

Such a program might be structured this way: The federal government would offer all homeowners with mortgages the opportunity to replace one-fifth of their existing mortgage (up to some dollar limit) with a government loan. This loan would carry a substantially lower interest rate than the individual's mortgage (reflecting the government's cost of funds). It would be a full-recourse loan that would have to be repaid regardless of what happens to the borrower's mortgage or home. By law, it would take priority over all non-mortgage debt.

Such a mortgage replacement loan would eliminate the potential incentive to default for almost all homeowners who now have positive equity. In doing so, it would limit the number of foreclosures that could contribute to a downward spiral.

Consider how the program would work for someone who has a $360,000 mortgage on a home worth $400,000, a 90 percent loan-to-value ratio. A 15 percent drop in prices would push that homeowner into a negative equity position, because the house's value would be only $340,000. But if one-fifth of that $360,000 mortgage ($72,000) were converted to a loan from the government, the mortgage loan be $288,000. As a result, the 15 percent decline in housing prices would still leave the homeowner with $52,000 in positive equity -- the difference between the reduced house price of $340,000 and the new mortgage of $288,000. There would be a strong reason not to default.

A program of mortgage replacement loans would act as a circuit breaker to reduce the number of defaults... In doing so, it would stop prices from overshooting on the way down in the same way they did on the way up. ...

[T]his program ... would not involve any increase in government spending or in the deficit..., it would not reward people who made high-risk purchases and now have high negative equity..., it would help participants meet their monthly payments. And ... substitution of government bonds for a portion of outstanding loans would provide substantial liquidity in the credit market, which could be used to support new lending.

Yet all these advantages would be incidental to the primary purpose of the mortgage replacement loans: reducing the amount of non-recourse mortgage debt and therefore reducing the risk of a damaging downward spiral in home prices and the economy. Congress needs to act quickly: Home prices fall every week, increasing the number of homeowners who have negative equity -- and an incentive to default.

Here are my comments from the last time this proposal came up:

...I can't imagine it actually happening and if it did I have no idea how many people would participate so it's hard to say if it would make a substantial difference. There is one aspect he doesn't mention. I'm not quite sure how the government avoids default risk without substantial loan collection costs, and even with aggressive and costly collection not all default can be avoided. I suppose the government could take future tax returns, Social Security payments, etc. in the case of default, but each additional restriction the government puts in place to protect itself will make the loans less popular and limit the program's effectiveness. The government could also avoid any losses from defaults by setting the interest rate high enough (spreading the losses among borrowers), but the higher interest rate would also limit participation. In the end, it's hard to imagine the government not taking some losses from this program, especially if participation is widespread...

But any government program to limit foreclosures will cost something, so the fact that the program will cost the government money is not, in and of itself, a reason to object. If you agree that there is a need to limit foreclosures by putting a floor under prices (and you may not, though note that Feldstein's concern is overshooting and the distortions that follow), then the question is how to reach the goal of limiting foreclosures efficiently so that net benefits are as large as possible. Given the concerns expressed above about how many people would be eligible and want to participate in the program, I still need to be convinced this program is the best we can do.