The problem with housing price bubbles is that they are rare. That can be regarded as good news. They only have to be dealt with every decade or so. The bad news is that while they all have individual characteristics, their rarity makes consequences hard to judge.

The most recent San Francisco Bay Area experience accompanied the more widespread real estate and economic downturn of the early 1990's. The magnitude of that problem was a hit of about 25% on value at the depth of the problem in 1993. At that point about half of all transactions in Redwood Shores were foreclosures.Values had recovered to their 1989-90 levels by about 1996. About six years of difficulty and at the worst, wiped out the equity of those who bought fully leveraged at the top of the previous cycle. Today, of course, prices are on the order of triple their 1989 levels. That bubble also corresponded to the denouement of the Savings and Loan Crisis and may have been exacerbated by the Tax Act of 1986 and the Gulf War.

The previous down cycle was related to Paul Volcker's taming of inflation through double digit interest rates and lasted about four years, 1980 to 1984. In that case, there were few sales absent assumption of a low interest rate mortgage and sellers had to fund part of the sale themselves, creative financing. Nominal prices didn't fall much, but their "cash equivalency" was dubious. Even foreclosing lenders might have made "loans to facilitate" distress sales, inflating the nominal prices.

Nor was that the only credit related downturn. In the mid '70's, corresponding to the Arab Oil Embargo, there was a recession that affected real estate as well. Again, the magnitude of the problem was partly disguised by creative financing.

Before that, the Vietnam war and the US leaving the Bretton Woods agreement, demonetizing both silver and gold had repercussions. In the late 1960's residential lending suffered from fraud that intensified the fallout. A friend owned a tape measure which was 48 feet long delineated in 50 increments. It was issued by a savings and loan which failed in that crash.

So, how is it likely to be this time? First, there has to be some triggering event. Everyone has been anticipating that rising interest rates would be the trigger, but so far they haven't risen and the yield curve says that markets don't anticipate their rise. Markets can be wrong, of course, about the future. So far there are several elements which have kept long term rates low. One is that our trading partners, notably China, have been recycling their trade surpluses into our government debt. Why might that change? Well a Chinese general suggested the other day that China might find it necessary to nuke Los Angeles. Maybe it wouldn't have to be so serious, but it would be unprecedented if we were able to run a trade deficit of the present magnitude indefinitely without serious side effects.

And how bad will it be? The magnitude of loss was concealed in the 1970's and '80's by an abundance of long term assumable fixed rate mortgages. Banks shared the pain with sellers who could not readily cash out of their properties as their low yielding loans stayed on the books while their cost of funds climbed as a result of inflation. But fraudulent appraisal and sloppy underwriting created problems and necessitated the closing and merger of numerous, mostly savings and loan, lenders. This time around we have mortgage companies and the ultimate lenders are generally investors in mortgage based derivatives. The World Trade Center was doomed on impact, but it took some time to figure that out.

Some derivatives' value will be adversely affected if there are excess defaults. Default and foreclosure happens when the loan amount exceeds the property value. So, for instance, watch the loan to the head of Fannie's appraisal quality control. His house was reported for appraisal purposes to be about 3000 square feet and five bedrooms when it actually is closer to 2000 square feet with three bedrooms. Others will suffer if long term fixed rate loans at 6% or so stay on the books longer than presently predicted. The first is a matter of securing the loan. Losses will be taken, first, on the marginal top prices, the most recent loans. Not only will they not be "seasoned" from a credit point of view, they will also not have any inflationary cushion. Maybe they will even be found to be worth less than their loans. The second is a matter for financial markets. Previously the pain was felt by lenders. Large numbers of savings and loans were merged in each recession of the last half century. Now there are few portfolio lenders in banking. Instead, insurance companies and pension funds seem to be the major investors in mortgage backed securities.

Will the combined problem be enough to bring down the house that Fannie built?