Coming Soon: A 300-Percent Increase in Foreclosures


At Calculated Risk, Tom Lawler, a real estate economist and former risk policy veep at Fannie Mae, tries to figure out how many people have actually lost their homes to foreclosure, short sales or deed-in-lieu desertions. The answer: Not enough. Lawler (who is now living the life of Riley on a Virginia farm) says the number of foreclosures that have been completed so far is a drop in the bucket compared to the number of loans that have gone bad:

On the other hand, the above numbers could well OVERSTATE significantly the number of homeowners who lost their primary home either to foreclosure or to a short sale. A "significant" % of completed foreclosure sales has been completed foreclosures on non-owner-occupied homes, though estimates vary as to what that % has been. In addition, not all short sales have involved homeowners "involuntarily" leaving their home, but who instead wanted to (for economic or other reasons) move and who were able to negotiate a short sale with their lender.

So what is the right number for folks who lost their residence to foreclosure, a short sales, or a DIL? I don't rightly know.

It is pretty clear, however, that overall foreclosure moratoria, foreclosure delays, modifications, and other workout activity continued to keep the number of homeowners who "lost" their homes to foreclosure massively lower than one would have expected given the delinquency/in foreclosure numbers.

So what will this mean when the last moratorium is lifted, the last show-me-the-note lawsuit gets thrown out of court, and the last loan modification has failed? Well by that time you'll probably be able to buy property on a planet orbiting some nice warm star in Constellation Cygnus. But there could be roughly three times as many homes on the market as there are now. Lawler points to 1,445,000 completed foreclosures and short sales at the end of 2010, compared with 4,296,01 mortgages that are past due by 90 days or more.

Getting a handle on the shadow inventory is more than just a way to fill up the time between Sunday afternoon looky-loos. You should feel for these foreclosed people because they've lost their jobs – even though in most cases they haven't lost their jobs. You should be worried about how foreclosure drives up neighborhood crime – even though it doesn't. And be afraid, be very afraid of the failure of the Home Affordable Mortgage Program to keep hardworking American working families who work hard in America from losing their homes – even though the HAMP is actually designed to buy time for the banks. And if you really want to drop tears as fast as the Arabian trees their medicinal gum, read up on HAMP's underwhelming numbers, ineffectiveness, and costly efforts to limit redefaults.

Speaking of redefaults, the OCC/OTC Mortgage Metrics report [pdf] is out for the third quarter of 2010, and the results are barely less horrific than they were in the previous report. Just under half (47.6 percent) of modified loans are 30 days or more overdue (strong likelihood of permanent default) within a year; more than a third (36.7 percent) are 60 days overdue (near certainty of permanent default) and nearly a third (29.8 percent) are 90 days overdue (for God's sake, get out of the house). Just to be clear: These are mortgages that the bank – frequently with backstopping from the taxpayers – has already fixed up once.

Redefault rates are showing gradual improvement, but the numbers are still pathetic. A quarter of all modified loans go bad after six months. To get even B-minus performances you have to make serious reductions in the borrower's monthly payment. A third of bad borrowers who have had their payments reduced by up to 10 percent still default again; more than a quarter default after having their payment reduced between 10 percent and 20 percent. And 14.6 percent who have had their mortgage payments cut by more than 20 percent still manage to default within half a year. That goes beyond bad personal finance and becomes an achievement you have to respect.

Loans modified under HAMP actually have much better performance than other modifications. But the numbers are so poor all around that you shouldn't hold your breath waiting to hear arguments for the success of the program. At this point the consensus that HAMP has failed seems pretty much unassailable. Which is just as well: If HAMP actually worked as promised, it would cost us almost a trillion dollars.