Skip Header

Federal Deposit
Insurance Corporation

Each depositor insured to at least $250,000 per insured bank

Home > Industry Analysis > Research & Analysis > San Francisco Regional Outlook - Third Quarter 1998

San Francisco Regional Outlook - Third Quarter 1998

Regular Features

Signs of a Pause in the Prolonged Residential Housing Boom Raise Concerns about Banks' Exposure to Construction Lending

  • The highly cyclical residential construction sector has experienced a long period of prosperity; however, a few warning signs are emerging.
  • In spite of the 1998 rebound in permit activity for the Region, permits continued to fall in Oregon and Utah, two states where other housing indicators were softening and economic growth has been tapering off.
  • Residential building activity is outpacing new household formation (a situation that can lead to overbuilding) in Oregon and Nevada, the two states where community banks have the highest concentration in construction lending in the nation.

Residential housing markets have enjoyed a long period of prosperity in many states in the San Francisco Region; however, recent trends in building permit activity may signal emerging weakness or overbuilding, or both, in a few states. In Oregon and Utah, softening in housing permits over a period of months, an indicator of a slowdown in future housing activity, suggests that the boom may be coming to an end. While the weakness may reflect the cyclical nature of construction activity, it is occurring at a time when the Region's financial institutions are recording rapid increases in construction lending.

Slower job growth, the Asian crisis, and changing demographics all contribute to concerns about housing indicators and increased construction lending. The Region, with a 3.2 percent job growth rate over the 12-month period ending in June 1998, continues to add jobs at a faster pace than the nation (2.5 percent). However, over the past year, job growth rates in most states in the Region have fallen moderately. Moreover, in part because of the Asian crisis, most state-by-state economic forecasts for 1998 project slower growth rates than in 1997. In addition, aside from California, population growth in the Region has slowed slightly but noticeably over the past two years. Slower population growth outside of California has translated into fewer new households formed and likely has reduced housing demand.

The Construction Sector Is Highly Cyclical

Lenders' assumptions about the outlook for the housing sector are very important, because construction is one of the most highly cyclical sectors of the economy. Construction projects are very sensitive to both the cost and availability of financing. Thus, changes in mortgage interest rates and overall economic activity tend to result in much larger swings in construction employment than in most other types of jobs. For example, over the past 25 years in the San Francisco Region, the annual percent changes in construction employment varied about four times more than changes in nonconstruction payroll jobs. Changing inflation expectations and consumer confidence, as well as wealth effects associated with stock market valuation, also can affect housing activity.

Not only does the construction sector experience large cyclical movements, but the timing of the peaks and troughs is significant. Nationally, declines in housing permits occur months or quarters in advance of changes in the overall economy; hence the sector is designated a leading economic indicator. At the state level, permits may occasionally move independently of national business conditions, but 8 of the Region's 11 states recorded downturns in permits prior to the 1990-91 recession.

The Region's Banks Are Highly Exposed to Construction Lending

The cyclical nature of the construction sector is important to the Region's financial institutions because sharp declines in construction activity are usually followed by deterioration in the quality of banks' construction loan portfolios. This importance has grown as institutions have increased their exposure to construction lending. In the San Francisco Region, real estate lending is the largest single loan category at community banks (non-credit card banks with assets of less than $1 billion), accounting for 39.6 percent of total assets as of March 31, 1998. Construction loans, which include both residential and nonresidential construction and land development lending, account for about 5.2 percent of total assets. As a result, changes in real estate markets may have a significant impact on the health of banks in the Region (see Increased Construction Lending and Changing Residential Lending Practices Pose Risks to San Francisco Region Banks).

The analysis that follows examines several state-level housing markets. In Oregon and Utah, housing permits have shown unusual weakness in 1997 and 1998. Comparing the number of housing permits to new households focuses attention on the potential for overbuilding in Nevada and Oregon.

Continuing Weakness in Housing Permit Activity in Oregon and Utah Is a Concern

The Region as a whole reported moderate growth in total housing permit activity throughout 1997 (4.9 percent) and strong growth in permit activity in the first five months of 1998 (10.6 percent annual rate). However, the performance was not uniform across the Region's states. For example, in 1997 California registered a rapid increase in permits (nearly 20 percent) that masked declines in seven other states.

Unlike the other states that experienced declines in housing permit activity in 1997, Oregon and Utah continued to experience declines in the first five months of 1998 (see Chart 1). Moreover, this weakness is consistent with other housing and construction indicators for these two states; housing price appreciation has slowed and sales activity remains below mid-1990s peaks. The weakness in permits, as well as slower rates of job growth in both states over the past year, is a significant indication that the extended housing booms in these states may be coming to an end.

Chart 1

Shifting Demographics May Be a Factor in Future Permit Activity outside of California

Shifting demographics may be having an effect on housing activity across the Region. California experienced significant out-migration during the early 1990s when the state suffered a prolonged recession. However, as the state's economy began to accelerate in the latter half of the 1990s, its population growth rate also has increased, and many of its housing markets have registered strong improvements. While still low by historic standards, California's population growth rate for 1997 had doubled from its 1995 low of 0.6 percent.

However, California's increased population growth rate has been accompanied by slower growth rates elsewhere in the Region. Most states in the Region began reporting slower population growth rates and new household formation rates after 1995. Although the Region, excluding California, still is adding residents at a rapid pace of about 2 percent per year, the growth rate is down from about 2.5 percent earlier in the 1990s (see Chart 2). In Arizona, Idaho, Montana, and Wyoming, population growth rates for 1997 were at least 1 percentage point below 1994 rates, according to Census Bureau figures. Hawaii, Nevada, and Utah reported nearly 1 percentage point declines in their population growth rates over the same period. Furthermore, most states also are reporting declines or slower growth in the 25- to 34-year-old population cohort, which typically accounts for a large share of first-time home buyers.

Chart 2

An Excess of Housing Permits Relative to New Households Formed May Be an Issue in Oregon and Nevada

A comparison of the number of housing permits and net new household formation is an indicator of the potential housing supply in the pipeline and the potential demand from households. Over a long period of years, these indicators tend to roughly balance out, but on a year-to-year basis they may vary widely, depending on the point in the business cycle. During and after recessions, when housing demand softens, the growth in the number of housing permits tends to fall below the rate of new household formation, creating a negative gap, which may indicate a shortage of housing units. During periods of expansion, the number of new housing permits often is larger than the number of new households formed, creating a positive gap. However, a large and persistent gap also may suggest that an excess inventory of housing units is building.

In 1997, the most recent period for which data are available, the Region reported 51,000 more housing permits issued than the number of net new households formed. Oregon and Nevada, which have been among the Region's fastest growing states for several years, reported large excesses of permits over new households relative to their populations. However, the positive gaps for Oregon and Nevada may be occurring for different reasons.

Oregon's 10,400-unit gap between housing permits and new households, shown in Chart 3, is of note because of its size and the recent softening in both the state's economy and housing markets. Employment growth over the 12 months ending in June 1998 slipped to 2.3 percent, near the national average. Fewer sales of existing houses, slower home price appreciation, and weak construction job growth in 1998 all are consistent with slowing in housing permit activity. The slowdown also may be linked to slower net in-migration, which is reflected in the lower number of new households formed in 1997 (see Chart 3). Oregon's sizable positive gap and other indications of slowing are significant because community banks hold 9.2 percent of their assets in construction loans, the second highest share in the nation.

Chart 3

In contrast to Oregon's softening, Nevada's residential real estate markets continued to boom at mid-year 1998. The state reported nearly a 24 percent increase in housing permits for the first five months of 1998 compared with the same period in 1997. Although Nevada's 5,000-unit gap in 1997 is less than the 1996 gap, shown in Chart 4, it is still fairly large relative to the state's small population. Despite this gap, the market remains healthy. Existing housing sales in the first quarter of 1998 were at their highest level in ten years, while housing price appreciation remained moderate.

Chart 4

Still, the potential exists for residential overbuilding in Nevada. Although the state's employment growth rate remains high at 4.2 percent for the 12 months ending in June 1998, it is the slowest job growth since 1992. Furthermore, with population growth moderating slightly and permit activity soaring, the housing gap could widen significantly in 1998. Overbuilding is a concern in Nevada because the state's community bank exposure to construction lending as of March 1998 was 15 percent of assets, the highest in the nation.

Implications: Most states in the San Francisco Region have experienced prolonged residential housing booms in the 1990s; however, recent economic and housing conditions may be an indication of change. In this situation, lenders should not assume that the boom in construction will go on indefinitely, especially given the highly cyclical nature of the residential construction sector.

Changing conditions may be especially relevant to the Region's community banks, because as a group they have heavy exposure to construction lending and they have been recording rapid increases in construction and land development loans. The maturity of the nation's economic recovery, the inherent volatility in construction activity, and increased construction lending by banks all indicate that lenders should carefully monitor housing conditions in their markets for signs of weakness. Such action may be especially timely in states like Oregon and Utah, where banks are heavily exposed to construction lending and housing indicators are showing signs of weakness, or in states like Nevada, where residential construction continues to grow faster than household formation.

Gary C. Zimmerman, Regional Economist

Regional Outlook Information
Return to Regional Outlook main page

Last Updated 7/26/1999

Skip Footer back to content