|
Residential Investment and Finance
Economic activity in the past two years has been restrained by the ongoing contraction in the housing sector, and that restraint intensifi ed in the second half of 2007. Home sales and prices softened signifi cantly further, and homebuilders curtailed new construction in response to weak demand and elevated inventories. In all, the decline in residential investment reduced the annual growth rate of real GDP in the second half of 2007 by more than 1 percentage point, and the further drop in housing starts around the turn of the year suggests that the drag on the growth of real GDP remains substantial in early 2008.
The downturn in housing activity followed a multiyear period of soaring home sales and construction and rapidly escalating home prices. The earlier strength in housing refl ected a number of factors. One was a low level of global real interest rates. Another was that many homebuyers apparently expected that home prices would continue to rise briskly into the indefi nite future, thereby adding a speculative element to the market. In addition, toward the end of the boom, housing demand was supported by an upsurge in nonprime mortgage lending—in many cases fed by lax lending standards.1 By the middle of the decade, house prices had reached very high levels in many parts of the United States, and housing was becoming progressively less affordable. Declining affordability and waning optimism about future house price appreciation apparently started to weigh on the demand for housing, thereby causing sales to fall and the supply of unsold homes to ratchet up relative to the pace of sales. Against this backdrop, prices began to decelerate, further damping expectations of future price increases and exacerbating the downward pressure on demand.
House prices decelerated dramatically in 2006 and softened further in 2007. In many areas of the nation, existing home prices fell noticeably last year. For the nation as a whole, the OFHEO price index declined in the second half of the year after rising modestly in the fi rst half; that measure had risen 4 percent in 2006 and about 9? percent in each of the two years before that.2 In the market for new homes, the constant-quality index of new home prices fell 2? percent over the four quarters of 2007. Moreover, many large homebuilders reportedly have been using not only price discounts but also nonprice incentives (for example, paying closing costs and including optional upgrades at no cost) in an effort to bolster sales of new homes and reduce inventories.
In all, the pace of sales of existing homes fell 30 percent between mid-2005 and the fourth quarter of 2007, and sales of new homes dropped by half. Builders cut production in response to the downshift in demand; by the fourth quarter of 2007, starts of single-family homes had fallen to an annual rate of just 826,000 units—less than half the quarterly high reached in early 2006. Nonetheless, the ongoing declines in sales prevented builders from making much progress in paring their bloated inventories of homes. In fact, although the number of unsold new homes has decreased, on net, since the middle of 2006, inventories have climbed sharply relative to sales. Measured relative to the average pace of sales over the three months ending in December, the months’ supply of unsold new homes at the end of December stood at nine months, more than twice the upper end of the narrow range that had prevailed from 1997 to mid-2005.
The contraction in housing demand and construction was exacerbated in the second half of 2007 by the near elimination of nonprime mortgage originations and a tightening of lending standards on all types of mortgages. Indeed, large fractions of banks that responded to the Federal Reserve’s Senior Loan Offi cer Opinion Survey on Bank Lending Practices reported that they had tightened lending standards over this period. Nonetheless, interest rates on prime conforming mortgages have declined on net: Rates on conforming thirty-year fi xed-rate loans dropped from about 6? percent last summer to just above 6 percent at year-end. This year they dipped as low as 5? percent but have recently moved back up to about 6 percent, within the range that prevailed for much of the 2003–05 period.3 Rates on conforming adjustable-rate loans have also fallen signifi cantly over the past several months and now stand at their lowest level since the end of 2005. Offered rates on fi xed-rate jumbo loans, which ran up in the second half of 2007, have recently declined somewhat, on net.4 Even so, spreads between rates offered on these loans and conforming loans remain unusually wide.
The softness in home prices has played an important role in the ongoing deterioration in the credit quality of subprime mortgages. The deterioration was rooted in poor underwriting standards—and, in some cases, fraudulent and abusive lending practices—which were based in part on the assumption that house prices would continue to rise rapidly for some time to come. Many borrowers with weak credit histories took out adjustable- rate mortgages (subprime ARMs) with low initial rates; of those loans originated in 2005 and 2006, a historically large fraction had high loan-to-value ratios, which were often boosted by the addition of an associated junior lien or “piggyback” mortgage. When house prices decelerated, borrowers with high loan-to-value ratios on their loans were unable to build equity in their homes, making refi nancing more diffi cult, and also faced the prospect of signifi cantly higher mortgage payments after the initial rates on the loans reset.
Subprime ARMs account for about 7 percent of all fi rst-lien mortgages outstanding. Delinquency rates on subprime ARMs began to increase in 2006, and by December 2007, more than one-fi fth of these loans were seriously delinquent (that is, ninety days or more delinquent or in foreclosure). Moreover, an increasing fraction of subprime ARMs in the past few years have become seriously delinquent soon after they were originated and often well before the initial rate was due to reset.5 For subprime ARMs originated in 2006, about 10 percent had defaulted in the fi rst twelve months, more than double the fraction for mortgages originated in earlier years. Furthermore, the path of the default rate for subprime ARMs originated in 2007 has run even higher. For subprime mortgages with fi xed interest rates, delinquency rates have moved up signifi cantly in recent months, to the upper end of their historical range.
For mortgages made to higher-quality borrowers (prime and near-prime mortgages), performance weakened somewhat in 2007, but it generally remains fairly solid. Although the rate of serious delinquency on ARMs has moved up, that on fi xed-rate loans has stayed low. Serious delinquencies on jumbo mortgages—which often carry adjustable rates—have crept up slightly from very low levels.
The credit quality of loans that were securitized in pools marketed as “alt-A” has declined considerably. Such loans are typically made to higher-quality borrowers but have nontraditional amortization structures or other nonstandard features. Some of the loans are categorized as prime or near prime and others as subprime. The rate of serious delinquency on loans with adjustable rates in alt-A pools currently stands at almost 6 percent, far above the rates of less than 1 percent seen as recently as early 2006. The rate of serious delinquency on fi xed-rate alt-A loans has also increased in recent months.
The continued erosion in the quality of mortgage credit has led to a rising number of initial foreclosure fi lings; indeed, such fi lings were made at a record pace in the third quarter of 2007. Foreclosures averaged about 360,000 per quarter over the fi rst three quarters of 2007, compared with a rate of about 235,000 in the corresponding quarters of 2006. As was the case in 2006, more than half of the foreclosure fi lings in 2007 were subprime mortgages despite the relatively smaller share of such loans in total mortgages outstanding. In some cases, falling prices may have tempted morespeculative buyers with little or no equity to walk away from their properties. Foreclosures have risen most in areas where home prices have been falling after a period of rapid increase; foreclosures also have mounted in some regions where economic growth has been below the national average.
Avoiding foreclosure—even if it involves granting concessions to the borrower—can be an important lossmitigation strategy for fi nancial institutions. To limit the number of delinquencies and foreclosures, fi nancial institutions can use a variety of approaches, including renegotiating the timing and size of rate resets. A complication in implementing such approaches is that the loans have often been packaged and sold in securitized pools that are owned by a dispersed group of investors, which makes the task of coordinating renegotiation among all affected parties diffi cult. In part to address the challenges in modifying securitized loans, counselors, servicers, investors, and other mortgage market participants joined in a collaborative effort, called the Hope Now Alliance, to facilitate cross-industry solutions to the problem.6 Separately, the Federal Reserve has directly responded in a number of ways to the problems with mortgage credit quality (described in the box entitled “The Federal Reserve’s Responses to the Subprime Mortgage Crisis”).
Most commercial banks responding to the Federal Reserve’s January 2008 Senior Loan Offi cer Opinion Survey indicated that loan-by-loan modifi cations based on individual borrowers’ circumstances were an important part of their loss-mitigation strategies. Almost twothirds of respondents indicated that they would consider refi nancing the loans of their troubled borrowers into other mortgage products at their banks. About onethird of respondents said that streamlined modifi cations of the sort proposed by the Hope Now Alliance were important to their strategies for limiting losses.
All of the factors discussed above—the drop in home sales, softer house prices, and tighter lending standards (especially for subprime and alternative mortgage products)—combined to reduce the growth of household mortgage debt to an annual rate of about 7? percent over the fi rst three quarters of 2007, down from 11? percent in 2006. Growth likely slowed further in the fourth quarter.
Consumer Spending and Household Finance
Consumer spending held up reasonably well in the second half of 2007, though it moderated some in the fourth quarter. Spending continued to be buoyed by solid gains in aggregate wages and salaries as well as by the lagged effects of the increases in household wealth in 2005 and 2006. However, other infl uences on spending have become less favorable. Job gains have slowed lately, household wealth has been damped by the softening in home prices as well as by recent declines in equity values, and consumers’ purchasing power has been sapped by sharply higher energy prices. Moreover, consumer sentiment has fallen appreciably, and although consumer credit has remained available to most borrowers, credit standards for many types of loans have been tightened.
Real personal consumption expenditures (PCE) increased at an annual rate of 2? percent in the third quarter, a little above the average pace during the fi rst half of the year; in the fourth quarter, PCE growth slowed to 2 percent. With the notable exception of outlays for new light motor vehicles (cars, sport-utility vehicles, and pickup trucks)—which were well maintained through year-end—the deceleration in spending in the fourth quarter was widespread. PCE appears to have entered 2008 on a weak trajectory, as sales of light vehicles sagged in January and spending on other goods was soft.
Growth in real disposable personal income—that is, after-tax income adjusted for infl ation—was sluggish in the second half of 2007. Although aggregate wages and salaries rose fairly briskly in nominal terms over that period, the purchasing power of the nominal gain was eroded by the energy-driven upturn in consumer price infl ation in the fall. Indeed, for many workers, increases in real wages over 2007 as a whole were modest, once again falling short of the rise in aggregate labor productivity. For example, average hourly earnings, a measure of wages for production or nonsupervisory workers, increased only ? percent over the four quarters of 2007 after accounting for the rise in the overall PCE price index. Moreover, for some workers, real wages actually declined: Real average hourly earnings in manufacturing edged down about ? percent last year, while for retail trade—an industry that typically pays relatively low wages—this measure of real wages fell about 2 percent.
On the whole, household balance sheets remained in good shape in 2007, although they weakened late in the year. The aggregate net worth of households rose modestly through the third quarter, as increases in equity values more than offset the effect of softening home prices. However, preliminary data suggest that the value of household wealth fell in the fourth quarter, and as a result the ratio of household wealth to disposable income—a key infl uence on consumer spending— ended the year well below its level at the end of 2006. Nonetheless, because changes in net worth tend to infl uence consumption with a lag, the increases in wealth during 2005 and 2006 likely helped sustain spending in 2007. In the fourth quarter, the personal saving rate was just a shade above zero, about in line with its average value since 2005.
Overall household debt increased at an annual rate of about 7? percent through the third quarter of 2007, a notable deceleration from the 10? percent pace in 2006; household debt likely slowed further in the fourth quarter. Because the growth of household debt about matched the growth in nominal disposable personal income through the third quarter, and net changes in interest rates on mortgage debt to that point were small, the ratio of fi nancial obligations to disposable personal income was about flat.
Consumer (nonmortgage) borrowing picked up a bit in 2007 to 5? percent, perhaps refl ecting some substitution of consumer credit for mortgage debt. The pickup in consumer debt was mostly attributable to faster growth in revolving credit, a pattern consistent with the results of the Federal Reserve’s Senior Loan Offi cer Opinion Survey. Banks, on net, reported easing lending standards on credit cards over the fi rst half of 2007 and reported little change in those standards on net over the second half of the year. In contrast, signifi cant fractions of respondents in the second half of 2007 reported that they had tightened standards and terms on other consumer loans, a change that may have contributed to a slowing in the growth of nonrevolving loans over the fi nal months of 2007. Average interest rates on credit cards generally moved down in the second half of the year, but by less than the short-term market interest rates on which they are often based. Interest rates on new auto loans at banks and at auto fi nance companies have also declined some in recent months.
Indicators of the credit quality of consumer loans suggest that it has weakened but generally remains sound. Over the second half of the year, delinquency rates on consumer loans at commercial banks increased, but from relatively moderate recent levels. Meanwhile, delinquency rates at captive auto fi nance companies increased somewhat but are well below previous highs. Although household bankruptcy fi lings remained low relative to the levels seen before the changes in bankruptcy law implemented in late 2005, the bankruptcy rate rose modestly over the fi rst nine months of 2007.
The issuance of asset-backed securities (ABS) tied to credit card loans and auto loans (consumer loan ABS) has remained robust. Spreads of yields on consumer loan ABS over comparable-maturity swap rates have moved up considerably since July; the rise pushed spreads on two-year BBB-rated consumer loan ABS to almost double their previous peaks in late 2002. Spreads on two-year AAA-rated consumer loan ABS jumped to between 60 basis points and 100 basis points after having been near zero for most of the decade, perhaps in part as a result of investors’ general reassessment of the risk in structured credit products.
1. Nonprime mortgages comprise subprime and near-prime loans and accounted for about one-fourth of all home-purchase mortgages in 2006. Near-prime mortgages are generally less risky than subprime mortgages but riskier than prime mortgages; they may require limited or no borrower documentation, have nontraditional amortization structures or high loan-to-value ratios, or be made on investment properties.
2. The index is the seasonally adjusted purchase-only version of the repeat-transactions price index for existing single-family homes published by the Offi ce of Federal Housing Enterprise Oversight.
3. Conforming mortgages are those eligible for purchase by Fannie Mae and Freddie Mac; they must be equivalent in risk to a prime mortgage with an 80 percent loan-to-value ratio, and they cannot exceed the conforming loan limit. The Economic Stimulus Act of 2008, signed into law on February 13, retroactively raised the conforming loan limit for a fi rst mortgage on a single-family home in the contiguous United States from $417,000 to 125 percent of the median house price in an area, with an overall cap of $729,750. The new conforming limit will be in effect through the end of 2008.
4. Jumbo mortgages are those that exceed the maximum size of a conforming loan; they are typically extended to borrowers with relatively strong credit histories.
5. The initial low-rate period for most subprime ARMs originated in the period from 2005 to 2007 was twenty-four months. Roughly 1? million subprime ARMs are scheduled to undergo their fi rst rate reset in 2008. Even with the recent declines in market interest rates, a notable fraction of those subprime ARMs are scheduled to reset to a higher interest rate.
6. The Hope Now Alliance (www.hopenow.com) aims to increase outreach efforts to contact at-risk borrowers and to play an important role in streamlining the process for refi nancing and modifying subprime ARMs. The alliance will work to expand the capacity of an existing national network to counsel borrowers and refer them to participating servicers, who have agreed to work toward cross-industry solutions to better serve the homeowner.
|
