|
Home > Industry Analysis > Research & Analysis > Dallas Regional Outlook, First Quarter 2002 |
|||
|
Dallas Regional Outlook, First Quarter 2002 |
||
|
Regional Perspectives
Region's Economic and Banking ConditionsPotential Exists for Housing Price Risk in Key Dallas Region Metropolitan MarketsSeveral of the nation's fastest-growing metropolitan economies in the past ten years have been in the Dallas Region. In addition to robust employment levels, housing activity in many of these markets also has been strong, spurred more recently by declining mortgage interest rates. However, at the same time, other economic sectors have slowed. Economic weakening in certain MSAs, as a result of either the September 11 attacks or the slowing national economy, or both, could undermine the strength of these housing markets. Moreover, concern exists about the downside risks to housing prices in metropolitan areas that previously had experienced strong employment growth and robust gains in home prices. This article considers the metropolitan areas of Denver, Austin-San Marcos,1 Dallas, Fort Worth-Arlington,2 and Houston and examines the susceptibility of these MSAs to housing price risk in the near term. These MSAs represent key economic centers within the Dallas Region and fit the profile of strong employment growth and accelerated gains in home prices (see Table 1). These MSAs were among the leaders in both categories from 1998 to 2000. However, employment growth and housing price appreciation slowed during 2001, new trends that are likely to continue at least through the first half of 2002. 1 The Austin-San Marcos MSA will be referred to as Austin in this article. 2 The Fort Worth-Arlington MSA will be referred to as Fort Worth in this article. Table 1[D]
Denver MSAOf the five MSAs, Denver's home prices appear most at risk to a weakening economy and prolonged downturn. From second-quarter 1999 to third-quarter 2001, housing prices in this MSA grew at a double-digit rate (see Chart 1). Robust employment growth and high levels of in-migration have supported this level of price appreciation. Since mid- to late 2001, however, the Denver economy has been affected adversely by the slowing U.S. economy and the effects of September 11. Year-to-year job growth rates have decelerated every month since April 2001, and employment declined 0.3 percent in fourth-quarter 2001 from a year before (see Table 2). Chart 1[D]
Table 2[D]
As a result of the U.S. recession, employment losses have spread beyond the metro area's ailing telecommunications, manufacturing, and travel industries to the transportation; communications; public utilities; finance, insurance, and real estate; and service sectors. Notably, job losses are occurring in industries that represent 55 percent of Denver's total nonfarm employment. Weakening in the broad service sector is particularly ominous for the metro area's housing market, as this sector employs the most workers (31 percent) and as employment did not decline on a quarterly basis during the U.S. recession from 1990 to 1991 or the state's recession from 1986 to 1987. In the aftermath of September 11, the air transportation industry in the Denver area lost numerous jobs; this MSA is a major United Airlines hub. The tourism sector, particularly the lodging and retail trade industries, also has been affected adversely. The results of a recent study by the Milken Institute estimate that 17,000 jobs will be lost in the Denver area this year as a result of the September 11 attacks.3 3 Ross Devol, Armen Bedroussian, Frank Fogelbach, Nathaniel H. Goetz, Ramon R. Gonzalez, and Perry Wong. January 2002. "The Impact of September 11 on U.S. Metropolitan Economies." Milken Institute. Denver's negative job growth rate is the lowest since the mid-1980s. Economy.com forecasts virtually flat employment growth for Denver in 2002, as a result of continuing problems in the high-tech, travel and tourism, financial services, and construction and real estate sectors. Rapidly rising home prices during a period of deteriorating employment growth are unsustainable, and could portend weak home price growth, or perhaps outright depreciation, to follow in the Denver housing market. Housing affordability in Denver declined considerably during the past three years, as the median home price soared (see Chart 1). The median price for an existing single-family home in the Denver area for the year 1998 was $152,200, 18.5 percent above the U.S. median. However, by third-quarter 2001, the median price for an existing single-family home in Denver had climbed to $224,000, 48 percent above the U.S. median. Despite lower mortgage interest rates during this period, the number of households in Denver that could afford a median-priced home fell 17 percent compared with a 6 percent decline for the nation. The decline in housing affordability could be particularly problematic for the first-time buyer and could dampen trade-up markets. Economists with the governor's Office of Planning and State Budgeting concluded in the December 2001 Colorado Economic Perspective that housing in the state is overbuilt, and a substantial slowing in residential construction through 2002 is expected. As suggested by the following quote from this report, the Denver housing market, unlike that of the nation, may be slowing significantly: "The number of unsold homes in the Denver area was up 35.4 percent in November 2001, compared with November 2000, and new home sales in the Denver area fell 34.0 percent through the third quarter of 2001, compared with through the third quarter 2000."4 4 "Colorado Economic Perspective: State Revenue and Economic Projections Through FY 2006-07," Office of State Planning and Budgeting, December 20, 2001, page 46. Furthermore, the employment/permit (E/P) ratio5 in Denver, a measurement of a housing market's demand/supply balance, indicates that the current level of housing production far exceeds the amount justified by employment growth (see Chart 2). A balanced housing market in the Denver housing market is characterized by an E/P ratio of 1.57 new jobs to 1 additional housing permit.6 6 According to the Meyers Group, a residential real estate consulting firm, the point of equilibrium for a balanced housing market in the United States is 1.26, calculated by taking total employment over total households. Applying the same methodology to the Denver MSA, using 2000 annual data, yielded an equilibrium value of 1.57. Equilibrium values for the other four MSAs were 1.54 for Austin, 1.65 for Dallas, 1.31 for Fort Worth, and 1.53 for Houston. As the ratio decreases, fewer new jobs exist to support the level of housing construction. Denver's negative employment growth rate contributed to an E/P ratio of -0.12 as of November 2001. Chart 2[D]
5 The E/P ratio is calculated by dividing 12-month employment growth by 12-month total permits. Poor economic fundamentals (deteriorating employment conditions), declining housing affordability, and residential overbuilding are expected to contribute to considerable softening in home price growth in 2002. In addition, should the recession in Colorado become more severe than that of the nation, home prices could fall. Austin MSAAustin also experienced a sharp run-up in home prices beginning in 1999 (see Chart 3) as a result of several years of strong employment growth and in migration. However, employment growth rapidly decelerated during 2001 in this MSA, and year-over-year home price growth slowed from 15.4 percent in first-quarter 2001 to 8.2 percent in third-quarter 2001. Chart 3[D]
Austin's job growth rate was 5.2 percent during fourth-quarter 2000, ranking eighth among 289 MSAs,7 and the MSA reported an unemployment rate of 1.7 percent for that quarter. Both measures are indicative of a healthy economy. By fourth-quarter 2001, however, job growth was a modest 0.7 percent, one of the sharpest declines in growth among the nation's MSAs, and the unemployment rate had jumped to 4.6 percent. 7 The 289 MSAs are tracked by Haver Analytics with Bureau of Labor Statistics employment data. The Austin economy has been affected adversely by weakness in the personal computer, semiconductor, and telecommunications sectors, as well as by the serious decline in the dot-com industry. Deterioration in these high-tech industries began well before the current recession and exerted a more adverse effect on production, capital expenditures, employment, and earnings than weakening in non-high-tech sectors. Layoffs in the computer, communications equipment, and semiconductor manufacturing industries are spilling over into support and secondary industries, such as software and computer-related services, business and professional services, financial services, and retailing. More than one-third of the output of the Austin economy is related to the high-tech sector.8 8Standard & Poor's DRI for the U.S. Conference of Mayors and the National Association of Counties. Unlike in Denver, the median home price in the Austin MSA does not exceed that of the United States significantly. In 1998, the median home price in Austin was 3.3 percent below the U.S. average; by third-quarter 2001, it was only 2.5 percent above the national average. Median home prices in the Austin MSA lagged those of the United States through much of the 1990s and have surpassed those of the nation only recently. Moreover, housing affordability in the Austin MSA, despite a slight deterioration three years ago, remains comparable to the national average (see Chart 1). However, the Austin E/P ratio declined from 1.91 in November 2000 to 0.34 in November 2001 (see Chart 2), indicating a growing supply/demand imbalance in residential construction. The drop in the E/P ratio is attributable solely to weak employment growth. The Austin MSA added an average of 35,000 jobs annually between 1998 and 2000; however, in the year ending November 2001, the MSA added only 4,300 new jobs. During the same period, the number of residential building permits filed fell from more than 16,000 units to about 12,000 units, underscoring the weakness in local job growth. Homebuilding activity slowed, and the number of unsold homes doubled between November 2000 and November 2001. Housing price growth began slowing in first-quarter 2001 and is likely to continue slowing through 2002, a trend particularly evident for high-priced homes. This sector of the real estate market has been hit hard by weakening in the high-tech sector and the failure of many Internet start-up companies. Economic forecasts developed by Economy.com and AngelouEconomics call for a slow regional recovery with historically weaker gains in employment, population, and retail sales growth. Although housing prices are not expected to depreciate this year, prolonged difficulties in the high-tech sector could change the outlook for home prices in this MSA. Dallas, Fort Worth, and Houston MSAsThe Dallas, Fort Worth, and Houston economies are discussed together because these MSAs are characterized by relatively moderate employment growth rates and do not show any serious signs of overbuilding. As a result, home prices in these housing markets are not at risk. Housing price growth for these three MSAs has been comparable to that of the United States in recent years (see Chart 3). Moreover, despite job growth rates that have slowed during the past year, these MSAs continue to add jobs faster than the U.S. average, and unemployment rates are at, or below, that of the nation (see Table 2). These MSAs have been affected by the U.S. recession and fallout from September 11. However, these areas have diversified during the past ten years, helping them withstand problems in the high-tech, energy, transportation, and travel-related industries. Moreover, the weakness in these economies appears to be cyclical rather than structural, suggesting that once the U.S. and global economies recover, stronger employment growth would be expected to resume. Housing affordability does not appear to be problematic for homebuyers in these MSAs (see Chart 1). Median home prices in Dallas, Fort Worth, and Houston were at least 10 percent below the U.S. median in 2001. Moreover, 1999 per capita incomes9 for all three MSAs exceeded that for the United States. Consequently, a greater proportion of households in these metro areas, compared with those in the nation as a whole, can afford a median-priced home. 91999 per capita income figures are the most recently available and were supplied by the Bureau of Economic Analysis. However, November 2001 E/P ratios for these MSAs were down substantially from year-earlier levels (see Chart 2). The Fort Worth and Houston E/P ratios were 1.30 and 1.11 in November 2001, down from around 2.00 a year before. The Dallas E/P ratio fell more precipitously from 2.77 to 1.13. Although these housing markets are not considered overbuilt, a softening in supply/demand conditions has occurred. Slowing employment growth and an increase in residential building permits are contributing to the lower E/P numbers. Employment growth and affordable housing are likely to support some housing price appreciation in these markets in 2002, albeit at a slower pace. Weakening demand relative to new supply could dampen housing price growth. Although prices are not expected to depreciate in the Dallas, Fort Worth, or Houston metropolitan markets, housing price growth in submarkets with employment concentrations in troubled industries, such as the telecom corridor in North Dallas, could be affected adversely. Mortgage Portfolios in Residential Lending Experience GrowthTrending with the Region's economic growth during the 1990s, residential mortgages10 booked by insured institutions headquartered in these five MSAs increased in volume and as a share of total assets. Mortgage loans11 grew 186 percent (merger-adjusted) in these MSAs during the ten years ending September 30, 2001, compared with 142 percent for the Region and 64 percent for the nation. The Houston and Denver banking markets experienced the most significant growth in mortgage portfolios, increasing 231 percent and 635 percent, respectively. Mortgage volume in Houston grew dramatically during the 1990s from a small initial base resulting from the depressed local economy in the late 1980s. Mortgage originations in Denver increased rapidly, largely because of major in-migration during the 1990s, motivating institutions to grow balance sheets and shift greater asset allocations into mortgages.12 to 29 percent as of September 30, 2001. 10 One- to four-family residential mortgages only, not including mortgage-backed securities (MBS). MBS can be diversified geographically and do not represent the same level of exposure as direct mortgages. 11 All banking data refer to insured institutions headquartered in that metro area, unless otherwise indicated. 12 Insured institutions in the Denver MSA increased mortgages as a percentage of total assets from 11 percent as of September 30, 1991, Thus, despite a slowing economy, mortgage portfolios among insured institutions in these MSAs have grown significantly, and residential real estate credit quality remains strong for these institutions as well. The results of a quarterly survey conducted by the Mortgage Bankers Association of America (MBAA), a trade group representing insured and uninsured mortgage lenders, indicate that nationwide 5.00 percent of residential mortgage loans were delinquent by 30 days or more in third-quarter 2001, up from 4.12 percent a year before. Moreover, the survey results state that the delinquency rate for residential mortgages originated in Texas was somewhat worse at 6.65 percent in third-quarter 2001, up from 5.62 percent a year earlier.13 13 Data supplied by Haver Analytics. Although the MBAA data evidence some credit quality deterioration, such weakening has not yet appeared among mortgage portfolios of insured institutions; past-due rates for the nation and Texas were 1.91 and 1.49, respectively, as of third-quarter 2001. The past-due rate for institutions headquartered in Texas declined slightly from a year earlier, suggesting that mortgage underwriting standards of insured institutions may be somewhat more conservative than those of other mortgage loan originators. Slowing Home Price Appreciation May Pressure Collateral ValuesTraditionally, residential mortgages in insured institutions have proven to be a lower-risk type of lending. However, some recent developments affecting collateral values suggest that risk in mortgage lending may be growing. The loan-to-value ratio of many new mortgages is increasing. More than 20 percent of all mortgages originated nationwide in 2001 were for more than 90 percent of the value of the house, almost three times the level in 1990. In addition, the increasing volume of cash-out refinancings has reduced homeowners' equity. According to the results of a Freddie Mac survey referenced in this quarter's In Focus article, "over half of all recent refinancings involved material 'cash out' features," providing less of a buffer for lienholders should mortgages default.14 Furthermore, although home prices in these MSAs have not shown widespread declines, anecdotal reports suggest that high-end home prices are coming under pressure, especially in markets that experienced substantial growth in the high-tech sector in the late 1990s but that have since slowed. And finally, recent rapid growth in mortgage portfolios among insured institutions in these five MSAs suggests that many of these mortgages are not seasoned and are based on relatively high real estate values. Should home prices decline, many of these mortgages would be left with leaner collateral positions. 14 "Homeowners Continuing to Increase Their Loan Balances When Refinancing, According to Latest Freddie Mac Study," Freddie Mac Press Release, October 24, 2001. Construction and Development Loan Concentrations Have RisenConstruction and development (C&D) lending is an important element of residential lending for many Dallas Region insured institutions. While call report data do not separate C&D lending into commercial or residential categories, discussions with bankers suggest that a majority of C&D loans support residential building. C&D lending exposure has increased over the past several years, as evidenced by the rise in C&D lending as a percentage of Tier 1 capital. At year-end 1998, 7 percent of insured institutions headquartered in these MSAs reported concentrations of C&D loans in excess of 200 percent of Tier 1 capital; by September 30, 2001, the level in excess of 200 percent more than doubled to 18 percent. Insured institutions in the Austin MSA reported the greatest increase in construction loans. This loan type represented 35 percent of Tier 1 capital among banks in the Austin MSA as of year-end 1998; by third-quarter 2001, this share had climbed to 117 percent. During a period with a slowing economy and declining employment growth, housing inventories may not sell well. As a result, builders may not remain current on C&D loan payments, pressuring the credit quality of insured institutions. By the Dallas Region Staff
Regional Outlook Information Return to Dallas Edition main page Return to Regional Outlook main page |
| Last Updated 03/25/2002 | insurance@fdic.gov |
| Home Contact Us Search Help SiteMap Forms Freedom of Information Act (FOIA) Service Center Website Policies USA.gov |
| FDIC Office of Inspector General |