The Oser View

The Meaning of the Mortgage Market Meltdown

The following is taken from a talk delivered recently by Ellen Seidman Director, Financial Services and Education Project, New America Foundation and an Executive Vice President of ShoreBank Corporation.  Her remarks are the best summary of the present state of the subprime mortgage that I have seen.

Just released Census data for 2006 tell us that there are about 75 million household who own their own home, a 67.3% homeownership rate.  The rate is 73% for non-Latino whites, 46% for African Americans, and 49% for Latinos.  About 68% of these homeowners carried a mortgage, and about 26% of those also carried a second mortgage, a home equity line, or both.

About $10.4 trillion of 1-4 family mortgage debt is outstanding.  Estimates of how much of this is sub-prime vary, but a good estimate is that about 20% is sub-prime, i.e., it carries an interest rate at least 200 basis points above comparable Freddie Mac prime loans, and another 13% is “Alt-A.”.  Most of these are adjustable rate mortgages (ARMs), and many have such additional risk factors as:

• Initial teaser rates—of which about $263 billion reset between January and August of this year, with an additional $700 billion scheduled to reset between now and the end of 2008, which is estimated to be a total of at least 2 million loans.
• Options to pay interest only, or even less than that, for some period
• No escrows
• Prepayment penalties
• No or low documentation, even for wage earners (often called “liar loans”)
• Simultaneous second loans (“piggy-back” loans) that mean there is effectively no equity in the house

Whereas in 2003, about 9% of the single family mortgage debt issued was subprime ($342 billion), by 2006 that had skyrocketed to 23% ($644 billion).  More than half that was refinance.  In fact, the Center for Responsible Lending (CRL) estimates only 11% actually went to first-time homebuyers to buy their home.

Although subprime loans are only about 20% of outstandings, by the end of 2006, 60% of new foreclosure filings were of subprime loans.  CRL estimates that 20% of all subprime loans originated in 2005 and 2006 will end with foreclosure or other involuntary home loss (e.g. deed in lieu of sale).  Even Lehman Brothers thinks the rate for the 2006 cohort will approach 30%.

Both foreclosures and their precursor, delinquencies, continue upward.  By August 2007, foreclosures were up 115% from the prior year, and 36% from the prior month.  In the second quarter of 2007, almost 15% of subprime loans (17% of subprime ARMS) were more than 60 days overdue, and over 9% more were either 90 days or more overdue, or in the process of foreclosure.  Both are way up from 2006, and the increase is much higher than for prime loans.

State and local differences are large and important.  Nationally, in August 2007 one out of 510 households was in foreclosure, but this ranges from one in 27,940 in Vermont to 1 in 165 in Nevada.  The highest foreclosure rates are in Nevada, California, Florida, Georgia, Ohio and Michigan, in that order.  These states divide into essentially two groups: those where house prices ran up spectacularly at the beginning of the decade and are now declining almost as fast (Nevada, California, Florida and Georgia) and those where the broader economy is in deep trouble (Ohio and Michigan).

To put the first group into perspective, nationally, real house prices rose 86% between the fourth quarter of 1996 and the first quarter of 2006, while real rents rose only 4%.  Since the peak, real house prices have already declined 3.4%.  The national situation is not good.  However, the situation in the boom states is worse.  For example, from June 2000 to June 2006, prices in Los Angeles went up 135%; in the year since, they have declined 5.1%.   For Miami the equivalent numbers are up 140% and down 4.8%.

The result of this kind of house price run up is that, notwithstanding strong state economies, homeowners are paying a huge share of income for housing.  In California, with a median home price of $536,000 (compared to the national median of $185,000), more than half of the homeowners pay more than 30% of their income for housing, and 22% pay more than 50%.  If those payments become unsustainable—which can happen for any number of reasons, including a mortgage reset but also such normal human events as death, divorce, disability and unemployment—the price decline makes it unlikely they will be able to get out of the situation by selling or refinancing. 

In the other group, prices didn’t increase as much, but they have nevertheless also turned negative.  And moreover, these states have high and increasing unemployment rates, increasing the likelihood that someone hit with higher payments will not be able to make them, and will also not be able to refinance or sell.  In July 2007, with the national unemployment rate at 4.6%, Michigan’s was 7.2% and Ohio’s was 5.8%.  House prices in Detroit were down an astounding 11% from the second quarter of 2006 to the second quarter of 2007 and continue to fall; one out of 87 households in Detroit is in foreclosure.  In Cleveland, house prices declined 3.6% during the same period, but appear to be stabilizing.  Nevertheless, foreclosures in Cleveland have left about 10% of the city’s single-family homes vacant.

With respect to individuals, it’s important to remember that for most low-wealth people, and most minorities, virtually all their net worth is represented by home equity.  For example, according to 2004 Survey of Consumer Finances the median net worth for the bottom quartile by wealth is $1,700; within that group, the median value of houses owned is $65,000, and median debt on it is $60,700.  For non-whites or Hispanics, the median net worth is $24,800; within that group, the median value of houses owned is $130,000 and debt on it is $63,900.  For these people, and indeed for most of the middle class, losing their house means losing everything they’ve built up, their credit standing and their hope for the future.   

The situation is exacerbated by the fact that high-cost loans have been made disproportionately in low-income and minority communities.  Just-released 2006 Home Mortgage Disclosure Act (HMDA) data show that 53.7% of home purchase loans (52.8% of refinance loans) to African Americans and 46.6% of home purchase loans (37.7% of refis) to Latinos were high cost, compared to 17.7% of purchase loans (and 25.7% of refis) for whites.  In census tracts with incomes under 50% of the area median, 46.5% of loans were higher-priced (compared with 18.3% for tracts over 120%); in tracts with over 80% minority population, incidence was 46.6%, compared with 21.7% for tracts with a minority population of less than 10%.  Other sources indicate that over 1/3 of both interest only and payment option borrowers had incomes under $70,000 and over 12% had incomes under $48,000.

It is also important to understand the effect of all this on communities.  During the 1980s and 1990s, major strides were made by cities, banks and credit unions, community development corporations, community development financial institutions, and other community based organizations around the country in stabilizing and bringing back communities that had seen substantial disinvestment in the 1960s and 1970s.  This was especially the case in the Northeast and Midwest, but other neighborhoods also benefited.  Home-ownership was a major strategy in this effort, not only for individual wealth building, but for civic engagement, crime prevention, and physical rehabilitation.  Some of those same communities are now being hard-hit by foreclosures.  These include Slavic Village in Cleveland, Back of the Yards in Chicago, and North Minneapolis.

A foreclosed house is more than a financial statistic; it generally turns into a vacant house, which in turn reverses gains in community stability, results in increased crime, costs city governments money in both increased expenses and decreased taxes, and reduces nearby property values.   The best estimate of total costs per foreclosed and vacant house to local government is about $20,000.  Dan Immergluck at Georgia Tech, one of the best researchers on this topic, has shown that a single foreclosure lowers the value of homes within one city block by .9% and by 1.4% in low and moderate income communities.  In another study, he has also demonstrated the close, positive relationship between foreclosures and increases in violent crime.

One of the implications of this is that, while I think it is totally appropriate to focus attention, and especially any outside funds, on owner-occupants, and in particular lower income owner-occupants and those who appear to have been misled, confused or only mildly excessively optimistic, the community effects are there no matter who owns the house.  In many ways, the fact that in 21% of California serious delinquencies and new foreclosures and 32% of those in Nevada are in investor-owned homes (compared to 13% in the rest of the country), makes the problem for communities in those states harder, not easier, because these are the people who will walk away quickly, from multiple properties, and create vacancies and no tax revenues. 

Housing Slump Continues

The Commerce Department reported that sales of new homes dropped in August to an annualized rate of 795,000 units.  This is the lowest level in nearly 10 years and a decline of more than 40% from the July 2005 peak.  The median price of new homes sold dropped 7.5% to $225,700, the largest one-month price decline since 1970.  The average price fell 8% to 292,000 and the inventory of homes for sale climbed to 8.2 months.  Today’s data mirrors Tuesday’s report on existing home sales, which fell to an annualized rate of 5.5 million units, its lowest point in six years.  Existing home sales peaked in mid-2005 at 7.2 million units. Putting it together, total home sales dropped from an annualized rate of 8.55 million at the summer 2005 peak to just 6.30 million this summer.  According to a private survey, the S&P/Case-Shiller index, home prices in 20 major metropolitan areas were down 3.9% for the year ended in July.  Prices in some formally hot markets in the Southwest, Florida and the DC area are down more than 6%. 

Commenting on the news, Fannie Mae CEO Daniel Mudd said that home prices would fall by two to four percent this year and more in 2008.  He does not expect the housing slump to bottom until late next year, and even then it will take time to come back up.  Calling a bottom in any market is chancy, but I suppose if anyone should know, it’s the head of Fannie Mae.

Turmoil

Despite the reduction in the fed fund’s target rate, the financial markets remain in considerable turmoil, which we do not expect to be resolved quickly. The current crisis was triggered by higher than normal defaults in subprime mortgages, and that alone suggests that its resolution will be lengthy. The housing and mortgage markets move slowly. Talk of a “housing bubble” was in the air for more than a year before any substantial effect on mortgage originations or home values, let alone the financial markets, was felt. The turmoil has now gone well beyond housing. The trigger, after all, is not the crisis, anymore than a stray match is a forest fire. The credit markets are in the midst of a wrenching process of re-assessing risk appetite. Broadly speaking, this crisis is about the market value of various types of financial assets, not the ultimate non-payment of mortgages. Many of the assets in question are largely unknown to the general public. They trade in specialized markets and are often opaque, even to sophisticated institutional investors. The complexity of the assets themselves and the burgeoning of derivatives based on them form the perfect breeding ground for uncertainty and fear.

The re-valuation of assets has become a crisis of liquidity. In a sense it’s a mirror image of old-fashioned liquidity crises when anxious depositors stormed bank lobbies to withdraw their funds. This summer, liquidity has been drained from the banking system not from the loss of liabilities but from an excess of assets that must be funded. These assets include loans that would normally be sold into securitized pools but must now be kept on banks’ balance sheets—or not made at all. They also include assets that leveraged funds must sell to satisfy creditors. Take for example a hedge fund that raises money from institutions and wealthy individuals. The fund then borrows substantial sums to buy even more assets. This is fine as long as the assets appreciate in value or at least retain their liquidity. The double whammy hits when a hedge fund’s assets depreciate and become illiquid. This is what happened to the Bear Stearns funds that went bankrupt in July. Their assets were all triple-or double-A rated classes of securitized mortgage pools, but the pools consisted of subprime mortgages. The Bear funds could not meet the demands of investors who wanted their money back because the market for their assets had evaporated. The ultimate repayment of the assets was not the issue. Investors who hold them to maturity will receive 100 cents on the dollar in principal and all the interest that’s due them. But timing is everything.

The next step in the re-valuation came when problems in one asset class started bleeding into others. As assets with direct subprime connections became illiquid, funds (a term I am using to cover a wide variety of investment vehicles that hold financial assets) had to sell other, higher quality assets. This was the moment when contagion struck. Too many “good” assets owned by too many desperate sellers hit the market all at once. Naturally, the price of these assets fell. Other holders of similar assets found that they were worth less relative to benchmark securities than ever before. In effect, a new risk factor in these asset classes was discovered. They are still just as creditworthy, still just as likely to be repaid as formerly, but they have become outcasts. It will take a long time to win back their former status.

That’s not all. Because “good” assets of certain types can’t be sold, new ones of the same type can’t be produced. Who wants to add to positions that are already illiquid? Since these assets are often derivatives, the chain of illiquidity runs all the way back to the base assets. Consider for example, triple-A rated classes of pooled jumbo mortgages. The underlying loans are made mostly to high income borrowers with high credit scores and large down payments. The triple-A classes are supported by subordinated classes that absorb delinquencies and defaults. These securities are bullet-proof, but nobody wants them. The mortgage brokers that used to make the loans that went into the pools now can’t get funding and some have gone out of business. Borrowers have turned to their banks, which can still make the loans but now must keep them on their books instead of securitizing them. The capacity of banks to hold loans is limited, leading to credit rationing, which further reduces the options for even high net worth borrowers. Declining mortgage availability leads to fewer homebuyers, and that leads to more erosion in home values. In this way, because the credit markets ceased to function with their usual efficiency, what started as a problem for low-income homeowners has become a problem for high-income homeowners and perhaps for the whole economy.

A further piece of the puzzle must be touched on, though it can’t be given justice in this short space. This is the ABCP problem. ABCP is the abbreviation of asset-backed commercial paper, heretofore a benign and little-known asset class. ABCP was invented by banks to provide clients with cheap funding while receiving favorable capital treatment for that funding. Like all commercial paper, ABCP is short-term, but it’s collateralized by longer-term assets and often guaranteed, to varying degrees, by large bank “sponsors.” Most ABCP is rated AAA and was considered so safe that it was a staple of money market funds. But ABCP, unlike ordinary commercial paper issued by large corporations for working capital, is not transparent. This summer, investors became concerned over ABCP whose collateral contained subprime mortgage pools or collateralized debt obligations (CDOs) that held subprime mortgages. The ABCP market suddenly shriveled and new issuance became extremely pricey. As a result banks have had to fund assets that would normally have become ABCP, creating yet further strains on liquidity. The amounts involved are not trivial. Before the crisis began, more than $1.18 trillion of ABCP was outstanding; as of yesterday the amount had fallen to $927 billion.

Conditions like these make a recession more likely. A recession is defined as two or more quarters of negative growth in gross domestic product (GDP). But there can also be “growth recessions.” These are slowdowns in the rate of growth below trend during an expansionary period. Trend GDP growth in the United States is about 3% a year. Growth is expected to be below trend for the rest of 2007 and for most, if not all, of 2008. Housing will be slow to recover, constraining liquidity for consumers. We project that interest rates will fall from current levels and remain low next year. At the short end of the yield curve—that is, from overnight to two years—the driver will be the Fed’s monetary policy. The Fed’s normal response to a slowing economy and reduced credit availability is to cut the Fed fund’s target, but the Bernanke Fed has been and will continue to be more cautious than the Greenspan Fed. Former Fed Chairman Greenspan has been widely and correctly criticized for lowering the fed funds target to 1% in June 2003 and keeping it there for a full year. At that time, the Fed was concerned that the economy was on the verge of deflation, something it had limited capacity and no experience in combating. Later revisions to the data showed that deflation was not as much of a threat as originally feared, but by then the damage had been done. Very low short-term interest rates were the catalyst for teaser ARMs and other so-called “affordability product” in subprime mortgages that sparked today’s financial crisis. The Fed will not make the same mistake twice in succession.

The Fed will be unenthusiastic, but will ultimately have to bring the fed funds target below 4% to prevent the growth recession from becoming a traditional recession. Our forecast calls for additional reductions of 25 basis points at each of the last two Federal Open Market Committee meetings of 2007. We expect further cuts totaling 50 basis points in the first part of 2008, bringing the fed funds target down to 3.75%.

Rates could go higher or lower than we project. A driver toward higher interest rates could be that rising food and commodity prices finally bleed through to core inflation. This month, crude oil broke through $80/barrel and wheat went over $9/bushel for the first time. The prices of industrial metals, particularly copper, zinc, and nickel, have tripled over the last few years. Largely this has come from increased demand from developing nations. To date, low labor costs abroad and increased productivity at home have prevented producers of finished goods from passing on rising raw material costs to consumers. But this could change quickly, and if it does, the Fed might well have to step in and raise rates in pursuance of its statutory mandate to maintain price stability.

A risk to the downside is that credit-dependent consumers will finally be forced to cut back on spending. As declining or even just stable home values become embedded, homeowners will find it more difficult to withdraw equity. Even if some modest appreciation occurs, tighter lending standards for home equity credit lines could have the same effect. Credit card debt will take up some of the slack, but ultimately consumer spending could slow or perhaps contract for a quarter or two. An outright contraction hasn’t been recorded for 16 years and would almost certainly signal a genuine recession and lower interest rates.

The Sound of One Shoe Dropping

Spillover effects of the financial crisis sparked by subprime mortgage delinquencies and defaults can no longer be denied: The other shoe—employment—finally dropped.  For the first time in four years, non-farm payrolls declined in August.  The economy lost 4,000 jobs last month, and job growth for both June and July was revised sharply downward, to a total of 137,000 from 218,000.  Manufacturing lost a crushing 46,000 jobs and construction lost 22,000.  Only the education and health care sectors, where 63,000 positions were added, prevented a complete debacle.

Treasury prices are rallying on the news, sending yields on T-notes maturing between 18 months and 39 months below 4%.  Nevertheless short-term borrowing rates, particularly for commercial paper secured by mortgage-backed bonds, remain close to 6%, indicating that the liquidity crunch is far from resolved.  As I write, the US stock markets have not yet opened, but futures are down and European markets are falling.

The Federal Reserve has been behind the curve in addressing the present situation.  Maybe this morning’s data will be a wake-up call.