Here's a valuable new study that answers some questions I've brought up:
Foreclosures in States and Metropolitan Areas: Patterns, Forecasts, and Pricing Toxic Assets
William H. Lucy and Jeff Herlitz
Department of Urban and Environmental Planning, School of Architecture, University of Virginia
National housing price declines and foreclosures have not been as severe as some analyses have indicated, and they are not as important as financial manipulations in bringing on the global recession. Most foreclosures have been concentrated in California, Florida, Nevada, and Arizona, and a modest number of metropolitan counties in other states. In fact, 66 percent of potential housing losses in 2008 and subsequent years may be in California, with another 21 percent in Florida, Nevada, and Arizona, for a total of 87 percent of national declines in these four states.
The methodology used here is to assume prices fall back to their ratio to incomes in 2000. Of course, 2000 was a prosperous year, so that would be a pretty soft landing.
California had only 10 percent of the nation’s housing units, but it had 34 percent of the foreclosures in 2008. California was vulnerable to foreclosures, because the median value of owner-occupied housing in 2007 was 8.3 times median family income, while the 2007 national average was only 3.2, and in 2000 it was lower still at 2.4.
They're using RealtyTrac.com's foreclosure statistics as of November 2008. Also, they're using "family income" rather than the more usual "household income" income, which makes the housing price to income ratios less extreme (I believe the peak home price to household income ratio in California was 11X). They assume that if you are just a household, not a family, you probably shouldn't be buying a house, which seems sensible.
Another vulnerability to foreclosures was seen in the Los Angeles metropolitan area, where more than 20 percent of mortgage holders in each county were paying at least 50 percent of their income in housing related costs.
But even in California, enormous variations existed among jurisdictions, such as in the San Francisco metropolitan area, where Solano County had 3.69 percent of housing units in foreclosure in November 2008, while only 0.24 percent of housing units were in foreclosure in the City of San Francisco, a 15 to 1 difference.
The exurban frontier got hit hardest. A lot of people in San Francisco have been there a long time, long enough to pay off even 30 year mortgages sometimes. Heck, they may have inherited the family mansion from a robber baron great-grandfather. But 80 miles out of town, there was nothing but dirt until recent years, so everybody has a mortgage. And everybody is scraping to get by. You wouldn't live that far out of town if you had other options. As I've said, the second quartile got the most overstretched and then hit hardest: the people trying to keep their kids out of the underclass.
... Potential housing value losses from 2008 foreclosures in 50 states, if values decline to year 2000 levels, were less than one-third of the $350 billion that has been provided to banks and insurance companies to cope with losses in mortgage backed securities.
Right, although there are lot more shoes left to fall. Subprimes went into foreclosure first. The Alt-A loans that are between subprime and prime in supposed quality start resetting in 2009 and finish resetting in 2012, so we're looking at a lot more foreclosures even after subprime calms down.
And then there's all the damage to come from the economic downturn, which will no doubt take down a lot of prime borrowers, too. Normally, foreclosure waves follow recessions. This time, foreclosures set off the recession.
Damage to the balance sheets of large banks and AIG occurred not mainly from losses on foreclosed residential mortgages, but because of borrowing short-range to buy long-range derivatives and from selling credit default swaps insuring derivatives backed by mortgage payments. These financial manipulations had high speed forward gears, but when the housing bubble burst, the banks and AIG discovered they had neglected to create a reverse gear with which they could separate foreclosed properties from some forms of mortgage backed securities. Obstacles to disentangling toxic components of mortgage backed securities magnified many times the actual housing value declines.
In Australia they call leverage "gearing."
... Foreclosure rates among states were highly skewed. The 2008 national foreclosure rate was 0.79 percent of 2007 housing units. Only seven states exceeded that rate, with an eighth, Idaho, tying it. The seven states exceeding it had considerably higher rates, led by Nevada 4.10 percent, California 2.57 percent, Arizona 2.26 percent, and Florida 1.99 percent (Table 1).
The Sand States.
The top 10 foreclosure states were in the West, except for Florida, Illinois, and Connecticut.
Foreclosure rates were low in most states in 2008. In three-fourths (38) of the 50 states, foreclosure rates were below 0.50 percent (1 in 200). In one-half of the states (25), foreclosure rates were below 0.25 percent (1 in 400). And in 11 states, foreclosure rates were below 0.10 percent (1 in 1,000) (Table 2).
...From 2000 through 2007, the relationship between housing values and annual incomes widened. In 2000, the average 50-state ratio of median value of owner-occupied housing to median family income was 2.4 to 1. By 2007, this average 50-state ratio had increased to 3.2 to 1 (Table 3).
So, that was a one-third increase in price to income ratio from 2.4 in 2000 to 3.2 in 2007, which doesn't sound outlandish. Of course, some of the income growth from 2000 to 2007 was derived from the Housing Bubble.
In 12 states, the ratio of [median] house value to [median] income exceeded 4.0 to 1, led by California at an extraordinary 8.3 to 1. The other 11 states exceeding 4.0 to 1 ratios were Hawaii, Nevada, Massachusetts, New York, New Jersey, Rhode Island, Maryland, Arizona, Florida, Oregon, and Washington.
... High foreclosure rates were influenced, but not controlled, by population growth. Of the 10 states with the highest foreclosure rates in 2008, six were in the top 10 population growth states from 1990 to 2000 and from 2000 to 2007 (Nevada, Arizona, Colorado, Utah, Idaho, and Florida) (Table 1). California, which was second in its foreclosure rate, was 18th in population growth rate, but first in the number of new residents.
High population growth would lead to high housing prices if supply lagged behind demand. Housing values to income ratios were higher than the national average in each of the top 10 states in foreclosure rates (Table 1). But six states in the top 10 in house value to income ratios (Hawaii, Massachusetts, New York, New Jersey, Rhode Island, and Maryland) were not in the top 10 in foreclosure rates or in population growth rates. These six states also were high household income and family income states, making high housing costs more manageable.
People in Massachusetts have a little more cushion, plus they tend to have relatives with cushions who might help them through a bad spell. The modern computerized system for evaluating creditworthiness don't seem to have direct measures of how much relatives could help out in an emergency, whereas the old relationship banking system for getting mortgages was less efficient, but bankers knew a lot about who was related to whom in their markets. So, today, an immigrant from a peasant family in Guatemala who is making $50,000 per year in America is assumed to be just as creditworthy as somebody making $50,000 per year who has lots of American relatives. But, when the crunch comes, the Guatemalan probably won't be able to collect much passing the hat back in Guatemala.
Foreclosure processes in four states—California, Florida, Nevada, and Arizona—constituted 62 percent of the U.S. total in 2008.
Wow, 62% in four states. And prices were higher in those four states, so total dollars defaulted must be enormous.
Within three of those four states, foreclosures were concentrated in a few metropolitan areas. In Nevada, Clark County, which constitutes the entire Las Vegas Metropolitan Area, contained 88 percent of Nevada’s foreclosures but only 72 percent of Nevada’s population. The two counties, Maricopa and Pinal, which comprise the entirety of the Phoenix Metropolitan Area, included 91 percent of Arizona’s foreclosures and 63 percent of its population. In Florida, the metropolitan areas of Miami, Orlando, and Tampa-St. Petersburg contained 62 percent of Florida’s foreclosures and 53 percent of its population. In California, foreclosures were more widely dispersed, as the metropolitan areas of Los Angeles, Sacramento, San Diego, and San Francisco contained 81 percent of California’s population and only 63 percent of its foreclosures.
The Central Valley of California got hit hard by foreclosures because dreamers had decided that it was really the exurbs of San Francisco and Los Angeles if they just closed their eyes and wished hard enough.
... If all the listed foreclosures and preforeclosures became repossessions, then these value reductions would cost $95 billion in California, $10 billion in Florida, $5 billion in Nevada, and $4 billion in Arizona, a total of $114 billion (Table 7).
This estimate overstates the crisis dimension of foreclosures.
But then there are all the foreclosures to come that Wall Street is finally worrying about.
From 1997 through 2006 the average foreclosure rate was 0.42 percent of mortgage loans, about one-third of the 2008 rate (HUD 2008, 73). It had become the normal cost of being in the mortgage business. Consequently, the foreclosure crisis should be considered, at most, the number and rate of foreclosures above the previous decade’s norm.
Right. That's an important point: that what's relevant is not the absolute foreclosure rate but the unexpected foreclosure rate.
An extreme perspective on pricing mortgage-backed toxic assets can be acquired by projecting 2008 foreclosure losses if housing prices decline to year 2000 ratios of housing values to family income. Calculating declines in the 34 states above the year 2000 national ratio of house values to family incomes (2.4) in 2007, the loss from lower house values would be about $143 billion. In all 50 states, the decline to year 2000 house values would be about $145 billion, with 87 percent in four states—California $95 billion (66 percent), Florida $18 billion (13 percent), Nevada $6.5 billion (5 percent), and Arizona $5.5 billion (4 percent) (Table 8). Declines of $1 billion or more also would occur in Illinois, New Jersey, New York, Massachusetts, Colorado, and Washington (Table 8).
Eight of the 12 states with house value to family income ratios above 4.0 had low foreclosure rates—Hawaii, Massachusetts, New York, New Jersey, Rhode Island, Maryland, Oregon, and Washington (Appendix 1). Consequently, the example above based on returning house value to family income ratios in 2000 exaggerates potential toxic asset losses in most states.
...The financial crisis was triggered by sub-prime mortgages, no down payment mortgages, resetting adjustable rate mortgages, and by some low income home buyers being manipulated by unscrupulous mortgage initiators. Herman Schwartz (2009,Chapter 8) identified a 16 percent default rate after nine months on 2007 subprime mortgages as launching the insolvency of several important lenders (including Countrywide and IndyMac) in 2007.
In addition, the financial crisis was caused by house value to income imbalances in a few states and a modest number of counties and metropolitan areas, by easy credit to support these imbalances, and by MBSs and subsequently by credit default swaps which ostensibly spread risk and reduced risk, but which actually greatly increased risk. They created an inflexible structure which neither lenders, packagers, central banks, nor national governments were able to access easily to repair the underlying delinquent mortgage payments. ...
... This inaccessible financial system was encouraged by home ownership policy goals. Frederick Eggers (2001) described the Clinton Administration goals as follows: “…the Nation’s home ownership rate actually declined in the early 1980s. Between 1985 and 1994, the home ownership rate remained virtually unchanged (at 64 and 65 percent)….In late 1994, President Clinton set as a national goal to raise the home ownership rate to 67.5 percent by the end of 2000. Beginning in 1995, the home ownership rate has risen almost steadily until, by the third quarter of 2000, it was 67.7 percent—surpassing the President’s ambitious goal….HUD used its oversight of Fannie Mae and Freddie Mac to encourage those entities to reach out to low-income borrowers and areas underserved by the private market.”
As an important part of his concept of the United States as an Ownership Society, President George W. Bush set a goal in 2002 of increasing home ownership by 5.5 million minority families. “We want everybody in America to own their own home,” President Bush said in October 2002 (Ferguson 2008, 267). Niall Ferguson (2008, 267) summarized Bush’s strategy: “Bush signed the American Dream Downpayment Act in 2003, a measure designed to subsidize first-time house purchases among lower income groups. Lenders were encouraged by the administration not to press sub-prime borrowers for full documentation. Fannie Mae and Freddie Mac also came under pressure from HUD to support the sub-prime market. As Bush put it in December 2003: ‘It is in our national interest that more people own their home.’”
Financial manipulations became overly clever and difficult to reverse. But they served public policy goals, which, in general, were supported by successive Democratic and Republican Administrations, members of Congress, federal agencies, and government sponsored entities (Fannie Mae and Freddie Mac). As the home ownership rate descends from its peak of 69.2 percent in 2004, the appropriate home ownership rate or range should be revisited. Based on more than 110 million owner-occupied dwellings in 2008, a four percent reduction to 65 percent home ownership would reduce owner-occupants by 3.5 million. The 64 to 65 percent home ownership rate was sustained for two decades without engendering a financial crisis. That experience is one place to look for guidance.