Jack M. Guttentag The Mortgage Professor

While plans for saving the financial system are being formed, modified, and dissolved by the day, the common thread through every plan seems to be to assist financial firms holding depreciated assets, most of which are mortgage-related assets. The cost is now unofficially estimated at $2 trillion or more. In contrast, aid to distressed borrowers is a second priority, with costs estimated at $50 billion to $100 billion.

There are two things about this approach that scare me. The first is that the cost of assisting financial firms in trouble is completely open-ended. The ultimate cost will depend on future declines in the value of their assets. If home prices continue to decline -- and at this point there is no end in sight -- asset values will continue to decline and $2 trillion may not be enough.

The second scary thing is that the government's priorities are wrong: The first priority should be assisting distressed borrowers rather than assisting distressed firms. This judgment is based on relative cost and effectiveness, not on a preference for "folks over firms." It costs less to convert a home loan in default to good standing than to offset the decline in its value to the firm that owns it. Furthermore, curing defaults relieves downward pressure on house prices, which will result in both fewer defaults and lower costs per default in the future.

Assisting Borrowers Rather Than Lenders Will Cost Less in the Short Term: Under current economic conditions, firms lose about 50 percent of the unpaid loan balance on home loans that are foreclosed. On a $100,000 loan, for example, the firm nets about $50,000 from sale of the property less the foreclosure expenses. It would cost the government $50,000 to make the firm whole.

An alternative is to direct government assistance to the borrowers in default by writing down their loan balances. In some cases, the default is not curable even with a balance write-down of 50 percent. In most cases, however, a balance write-down of less than 50 percent will cure the default by making the loan affordable and by increasing the borrower's equity. At a guess, the required write-down will average 25 percent, which would cut the government's cost in half.

Assisting Borrowers Rather Than Lenders Will Cost Less in the Long Term: Aid to firms has a negligible impact on the supply-demand balance in the housing market. In contrast, each cured mortgage default results in one less house being offered on the market, which brings us one step closer to a bottom in house prices.

The major cause of the decline in asset values at financial firms is house price declines, which increase both mortgage defaults and the losses per default. Mortgage defaults, in turn, put downward pressure on home prices by adding homes acquired through foreclosure to the inventory of unsold homes. The major focus of public policy ought to be to break this vicious circle by curing all mortgage defaults that are curable.

To allow this vicious circle to run its course -- which could take years -- while keeping financial firms on life support throughout, is a disaster in the making.

Clearly it is much more difficult to cure several million defaults than to place a few thousand firms on life support. But the government should do what needs to be done, not what is easy to implement.

My over-simplified illustrations might leave the impression that I am proposing that the government finance wholesale write-downs of loan balances. That would be an exaggeration. Igor Roitburg and I spent several months developing a plan for curing defaults that would mobilize the considerable human and institutional resources needed to do the job (see Breaking the Back of the Financial Crisis, on my Web site).

Balance write-downs are only one part of this plan, and the government would pay only a part of the write-downs required. The plan is probably deficient in many respects, but it is directed at what needs to be done, which is to target mortgage defaults -- not the losses that result from such defaults.