The Double Dip in Housing
Click here for a printer-friendly version of this article
The first imperative of economic policymaking when faced with unwelcome developments is to “do something.” Sitting idly by while some unfortunate statistic trends upward is simply not an option because it provides political opponents with a platform of attack. Unfortunately, in many circumstances it is not obvious that “doing something,” which usually means a large outlay of government resources, will help solve a problem. That is precisely the situation in which the federal government finds itself with respect to the housing market.
This summer, the policy response to the mortgage crisis in the United States is entering its third year, yet all recent data point to a double dip rather than sustainable growth. The lack of success cannot be blamed on timidity. Since 2009, the government has made 3.2 million mortgages eligible for modification and committed more than $1.5 trillion to the mortgage markets when one includes the direct Treasury investments in the government-sponsored enterprises (GSEs) to keep mortgage credit available to borrowers and the Federal Reserve’s purchases of GSE securities to lower effective mortgage rates. Yet, the net effect of these policies has been to delay the day of reckoning, as they have failed to address the fundamental supply-demand imbalance that can only be cured over time, or through debt forgiveness on an unthinkably large scale.
Initially, policymakers like Fed Chairman Bernanke misdiagnosed the problem, believing that the spike in delinquencies would be limited to poorly underwritten subprime mortgages and therefore “contained.” The early narrative was that the increase in delinquencies was the result of adjustable rate mortgages resetting from low “teaser” rates to unaffordable levels for borrowers who shouldn’t have been given a loan in the first place. The conventional, conforming mortgage loan market was seemingly unaffected. Indeed, some policymakers began pressuring the regulator for Fannie Mae and Freddie Mac to remove existing regulations on the GSEs’ risk-taking. Their regulator obliged in 2008, just as it became increasingly clear that Fannie and Freddie’s losses would dwarf those of all other institutions.
In 2008, policymakers discovered that subprime loan delinquencies were just the first wave of the crisis. The poor quality of the underwriting – low down payments, low FICO scores, high debt-to-income ratios, etc. – drove the entire residential housing market to an extent that was underappreciated in 2007. When individuals who would otherwise be denied credit are given a subprime loan, they are able to bid on a property that would have otherwise not existed. Basic microstructure suggests that this new bid will push up the price of the existing property. The longer-term result will be more new home construction to accommodate the increase in the number of prospective buyers.
Existing homeowners believed to be entirely untouched by subprime mortgages benefited from this house price appreciation, which was partly triggered by the credit-fueled increase in demand. Subprime loans also impacted households that bought homes with traditional mortgages during 2004-2007 by forcing these buyers to pay much higher prices than would have prevailed in the absence of subprime loans. The increase in the availability of credit increased the value of residential real estate collateral, which amplified the increase in the availability of credit, which further increased the value of residential real estate, as depicted in the classic credit cycle models.
As the volume of subprime loans fell substantially in 2006, home prices stopped rising. By early 2007, subprime mortgage defaults began spiraling upwards. This led to a further contraction in credit availability, which led home prices to fall further. This constricted demand in two ways. First, marginal borrowers could no longer access mortgage credit, which meant their bids were removed from the market, placing downward pressure on prices. Second, the “house flippers” or speculators could no longer sell their properties at a profit, which prevented these investors from recycling profits into more investment. Based on survey data, the investor share of house purchases peaked in 2005 and declined each year to 2009. An underappreciated aspect of the housing boom was the extent to which investors in residential properties committed “occupancy fraud” by claiming the investment property as a primary residence to get lower interest rates. As it became clear that sales prices were divorced from equivalent rents, sales prices plunged and landlords found themselves implicitly subsidizing their tenants (paying mortgages well in excess of rents).
As demand waned, the problem of excess supply became more obvious. Newly constructed single-family homes reached a record of 1.7 million in 2005. More problematic was that the next three highest years on record for new home construction were 2003, 2004, and 2006. There was no demographic reason for the spike in construction, as forecasts for household formation implied that only about 700,000 new homes per year were required to meet the needs of the growing population. Worse, the new construction was not evenly distributed across the country but rather concentrated in areas like Florida, Arizona, Nevada, and California. As 62% of foreclosures were concentrated in the same states that experienced the most construction, the result was a supply glut that caused prices in these areas to plummet.
Falling prices caused delinquencies to increase across the board. When prices fall by 40% from their peak as unemployment rates increase, even the most conservative loans go bad. By the end of 2009, more than 15% of all mortgage loans were delinquent or in foreclosure. And this figure does not include previously foreclosed properties from earlier in 2009 or in 2008. Even the most conservative portfolios reached unthinkably high default rates. Freddie Mac, for example, has “non-credit enhanced” mortgages with 80% loan-to-value ratios and high (greater than 700) FICO scores. By February 2010, 3.2% of these mortgages were delinquent. This seems low until one considers that default rate on these loans was less than 0.25% in 2005 and 2006, meaning the February default rate was nearly 14-times greater than forecast. The New York Times found that high-end mortgages are now defaulting at an accelerating rate, further cementing the fact that declining house prices have left no portion of the mortgage market untouched.
In April, LoanPerformance released a white paper that credits the home-buyer tax credit, the Federal Reserve MBS purchases, and federal foreclosure prevention programs – Home Affordable Modification Program (HAMP), Home Affordable Refinance Program (HARP), and Home Affordable Foreclosure Alternatives (HAFA) – with helping to stabilize the market. But the paper also projects a return to house price deflation (and more foreclosures) without continued government support. And the effective cost of the continued government support would be greater today than in the recent past because “any further extension, while still positive, would exhibit diminishing returns.”
To date, the tax credits and foreclosure mitigation programs have stabilized the market by “restricting supply and shifting home sales that would have occurred after the market’s eventual stabilization from the future into the present.” (See: LoanPerformance Pg. 4) The tax credit provided an artificial boost to demand: prices stabilized because sales that would have occurred in summer were moved to spring. The HAMP, HARP, and HAFA programs artificially restricted supply, as homes that would have come on the market via foreclosure were kept off of the market through a mortgage modification. In neither case was this support sustainable. Demand is falling back below trend even with the extension of the tax credit because the credit did not create new buyers – it simply changed the timing of sales that would have occurred anyway. Similarly, supply (and distressed sales) is rising again, as only about 25% of trial mortgage modifications started have made it to the permanent modification stage (when the modification can still fail).
The week of July 21 is a big one for residential real estate market statistics. On Tuesday, the Census reports on new residential construction, on Wednesday the LoanPerformance housing price index is released, and on Thursday both the Federal Housing Finance Administration (FHFA) monthly price index, and the National Association of Realtors (NAR) existing home sales report are released. Hopefully this battery of numbers will show stabilization in the market. But if it does not, Congress must resist the urge to “do something” and delay the eventual market recovery that will only come about through a policy of benign neglect.