The Oser View

Home Sales, Home Prices

New home sales increased 2.6% to an annualized rate of 858,000 units in March, up from a pace of 836,000 in February. It was the first increase in sales since December. Sales of existing homes, however, fell to an annualized rate of 6.12 million units in March, a huge drop-off from February’s 6.66 million unit pace. It was the biggest one-month percentage decline since 1989 and the lowest level of monthly sales since mid-2003.

The data for new and existing home sales are reported differently. The Commerce Department counts a new home sale when a sales contract is signed; the National Association of Realtors counts an existing home sale at closing. Because there can be many a slip—cancellations, failure to secure a mortgage, cold feet—between signing a contract and closing, the new home sales data is viewed more skeptically than the existing sales data. Also, the difference in timing means that erratic weather patterns can further complicate analysis. Unusually cold February weather is credited with preventing people from house-hunting and depressing March closings on existing homes. Unusually warm March weather is credited with luring people into house-hunting and increasing contracts for new homes.

Meanwhile, home prices continue to decline. The highly-respected S&P/Case Shiller Home Price Index reported an average decline of 1% in 20 major metropolitan areas between February 2006 and February 2007. Over the last six-months, the decline was 2.42%. In some areas home prices are falling precipitously. In Boston and Detroit prices are down more than 5% in the last half year, and they are down more than 4% in Washington DC, San Diego, and Cleveland. This is probably good news for Washington, where home prices have rocketed up 40% in three years. It’s not good for Cleveland (up 2% in three years) and especially Detroit (down 2.4% in three years).

What does it all mean? It means that the real estate markets are slowly correcting themselves from previous excesses. It means that all sorts of extraneous things can affect any one month’s statistics. It means real estate remains highly localized, with national trends smoothing out significant local fluctuations. It means there are still an awful lot of houses being bought and sold in this country.

The Bad Year

Every business has its jargon, and the mortgage business is no exception. My favorite mortgage jargon is the word “vintage” to describe a year’s production. Mortgage professionals speak of 2003 and 2004 as good vintages and 2000 as a poor vintage. Mortgages made in 2006, young as they are, are well on the way to being the worst vintage on record.

Differences in real vintages are the unpredictable result of weather, disease, pests, and other vicissitudes attendant on agriculture generally. Poor performance in a local mortgage market may also have causes outside of mortgagors’ control. But a bad national vintage during a period of high employment and a healthy economy is no act of nature. The well-publicized problems in the mortgage industry today are directly attributable to actions deliberately taken by the major players in late-2005 and 2006. As interest rates rose, mortgage refinancings declined. Mortgage companies, desperate to maintain production and market share, loosened and sometimes entirely abandoned prudent credit standards, creating what has become known as the sub-prime meltdown.

In this month’s commentary we will examine the barren harvest of the 2006 vintage. I will be arguing that the problems occurring today in housing are far more those of Wall Street than Main Street. In doing so, it may seem that I am being insensitive to families who are being foreclosed out of their slice of the American dream through some combination of bad luck and unexpected financial problems. There are many such, but—and this is the main point—there are neither more nor less such families in the 2006 vintage than in other years. What makes last year so exceptional is that the wrong mortgages were made to the wrong people for the wrong reasons.

The wrong mortgage has an adjustable interest rate and an unrealistic re-payment schedule. The wrong people are those with poor or non-existent credit histories. The wrong reason is to purchase a first home with no money down and limited knowledge of the financial demands of home-ownership. A standard measure of mortgage performance is the percentage that has ever been 60 or more days delinquent. Just nine months after issuance, close to 20% of 2006 adjustable-rate, sub-prime, purchase money mortgages are already in this category. The surprise is not the 20% level—all vintages except 2004 have hit or surpassed it—but the speed with which it was attained. No other type of mortgage and no other group of mortgagees comes close.

The mortgage business in the United States is huge, with total outstandings somewhere north of $10 trillion. About 10% of this total is labeled “sub-prime,” but it’s an elastic distinction among the many gradations of credit history, loan-to-value, debt-to-income, and willingness and ability to pay that constitute mortgage underwriting. By 2006, the definition had been stretched to include people just emerging from bankruptcy, with no savings, no experience of homeownership, and, who in many cases were not even required to verify their incomes. It’s reasonable to assume that many 2006 sub-prime borrowers had very low levels of financial literacy, since this is true of Americans in general. A 2005 survey by the National Council on Economic Education found that working-age Americans scored an average grade of “C” on a 24-item financial questionnaire containing such posers as, “Where do most people derive the largest portion of their personal income?” In another recent study of 1,700 Baby Boomers, less than 20% correctly answered a very simple question on compound interest(1). But financial illiteracy should not be confused with stupidity. Credit-scoring is now well-established and well-understood, and fast, easy methods of improving people’s scores have developed. In 2006 they were extensively utilized to help qualify potential home buyers, a practice the mortgage industry calls “soft fraud.”

Of course, temporarily improving one’s credit score is not the same thing as being able to manage credit consistently, particularly when the credit that needs management is a first-time mortgage with changing and unpredictable terms. And, when none of your own money is at stake, what really is the difference between renting from a landlord and “renting” from a mortgage company? And why should you worry about hurting your credit when bad credit didn’t prevent you from getting a mortgage? So, if the burden is too much, why not just stop paying the mortgage, live rent-free for eight or nine months, and go back to renting when the foreclosure is complete?

The biggest “why” of all is, “Why were so many bad mortgages made in 2006?” The simple answer is that the mortgage companies thought they could lay off the risk. They immediately sold everything they originated and pocketed the fees. They understood that they might have to buy mortgages back if too many suffered early payment defaults, but they gambled that the future would resemble the past. When it didn’t, some mortgage companies had insufficient capital to meet their commitments. About 40, so far, have been forced to sell out or declare bankruptcy.

There is a more complex answer to the question. Risk, like matter, can neither be created nor destroyed. All of the risk inherent in a financial transaction remains in the transaction throughout its life. The mortgage companies would not have been able to sell their adjustable-rate sub-prime mortgages without buyers willing to assume the risk. Here we must plunge into the arcana of Wall Street, where groups of mortgages are merely cash flow streams whose credit profiles are channeled to provide investors with a variety of risk/reward alternatives. The process of grouping or pooling mortgages is called securitization, and it’s an iterative process. Securitized mortgage pools can themselves be securitized and channeled to provide still other levels of risk/reward trade-offs. These re-securitizations are called collateralized debt obligations (CDOs), but even they are not the end of the line. The most sophisticated investors can hedge or double down on their risk/reward bets by buying or selling contracts on synthetic sub-prime mortgage indices called ABX indices. The original risk of the default and foreclosure of an individual mortgage still exists, but it’s been attenuated and gone underground into the secretive realm of hedge funds.

Some hedge funds made the wrong debt in their search for higher returns and their investors will suffer. There has been a shakeout of sub-prime mortgage companies, and even shares of major lenders have suffered. Countrywide’s stock is off 25% from its 12-month high and Washington Mutual is off 15%. This is where the pain is. Its cause was overly relaxed underwriting standards. The mortgages that resulted are defaulting in droves, but all other mortgages are performing much as usual. Lenders have tightened their standards in 2007, and consolidation will strengthen the survivors.

How much spillover into the housing market and the broader economy will occur? I believe the impact will be limited, although it will be intense for awhile. Sales of newly-built homes, which represent about 15% of the market, are off sharply, and home builders are suffering from the over-building begun in the boom years of 2003 through 2005. Foreclosures have the dual effect of increasing supply and lowering prices, which will exacerbate the problem. But with unemployment low and immigration not choked off, demand will catch up fairly quickly. Growth in median home prices has stalled and will decline in some areas, but this can hardly be viewed with much surprise given the unsustainable increases of the past few years. Nor should it be viewed with regret by anyone interested in affordable housing.

The minutes of Federal Open Market Committee’s March meeting stated that problems relating to sub-prime adjustable mortgages “likely would constrain home purchases by some borrowers, perhaps retarding the recovery in the housing sector…However, there was no sign of spillovers from the sub-prime market to the overall mortgage market.” It takes many months or even years for a mortgage to become delinquent, to default, for the home to go through foreclosure and to be re-sold. It’s still far too early to claim with certainty that the crisis has peaked. Perhaps the “signs of spillover” will emerge later this year or even in 2008. But I doubt it.

Note(1): Here's the question: "Let's say you have $200 in a savings account.  The account earns 10% interest per year.  How much would you have in the account at the end of two years?"  Did you get it right?

(Information on financial literacy is from "Financial Literacy and Retirement Preparedness," by Annamaria Lusardi and Olivia S. Mitchell, published in "Business Economics" Vol 42, No.1, pp 35-44. Data on mortgage delinquency courtesy of Countrywide Securities Corp.)

Get a Job? No Problem!

Non-farm payrolls increased by a higher-than-expected 180,000 jobs in March, and, as usual, previous months' estimates were revised higher: 113,000 jobs added in February and 162,000 in January.  The big surprise was a gain of 56,000 in construction jobs.  This reversed February's loss of 61,000, which appears to have been correlated with much colder than normal weather.  Job growth was also strong in retail trade (+36,000) and the education/health group (+54,000).  Manufacturing continues its job-bleed, losing another 16,000 positions in March.  This sector, once the core of American employment, now provides barely 10% of US jobs.  It's on the way to becoming nearly as irrelevant as agriculture as a source of employment.  In a separate report, the Labor Department announced that the unemployment rate dropped to 4.4%, matching a five-year low.  Not surprisingly, given these tight labor conditions, hourly and weekly earnings have been growing faster than the rate of inflation.  The year-over year change in hourly earnings is 4.0% and in weekly earnings, it's 4.4%.

In an abbreviated trading session, Treasuries sold off sharply.  The ten-year Treasury note yield hit 4.75%, its highest level since Janaury.  Today's data delays even further any expectations of a decrease in the fed funds rate.  The data also adds weight to the thesis that problems in the housing and mortgage industries will remain limited in their impact on the overall economy.