The degree of risk borne by commercial real estate lenders
depends on local market conditions, which in turn depend on local
supply and demand factors. Following the experience of the 1980s,
when overdevelopment led to declining property values in various
markets across the country, the threat of oversupply is watched
with keen interest by lenders and regulators alike. This paper
highlights various metropolitan markets that may be vulnerable to
overbuilding based on the rapid pace of development occurring
within those markets. This analysis is further supported by the
opinions and research of credible industry experts. The ranking
schemes presented are intended to serve as a basis for prioritizing
more in-depth analysis of depository institution risk exposures to
individual markets and to specific market segments. Although this
study does not predict an imminent downturn in those markets
highlighted, it does raise the degree of concern over rapid
development and a related increase in bank construction lending
within these markets.
This paper focuses on identifying markets vulnerable to
overbuilding. The identification process involves comparisons or
rankings of current market activity indicators across markets and
across commercial property segments. The main objective of this
analysis is to prioritize research and analysis efforts conducted
by the FDIC's Division of Insurance going forward. Examples of
these efforts include in-depth market studies, analysis of
geographic commercial real estate credit exposures, and surveys of
specific underwriting practices within riskier markets.
Revisiting the Risks of Overbuilding
History clearly demonstrates the importance of oversupply in
commercial real estate markets. During the 1980s and early 1990s,
banks and thrifts experienced heavy losses related in large part to
overbuilding in markets such as New England, Texas, and Southern
California. As shown in the FDIC's History of the
Eighties project, banks that failed during that period had
higher relative levels of commercial real estate loan exposures
than banks that survived.1
The economic consequences of too much supply of commercial property
are straightforward. When supply outpaces demand, upward pressure
is placed on vacancy rates. Growing volumes of unoccupied space
lead in turn to lower rents for newly developed properties and rent
concessions for existing properties. Since revenue streams
generated by commercial properties serve as the basis of value,
lower rents translate into reduced property prices. As revenues
fall below levels sufficient to service operating expenses and
outstanding debts, loan defaults begin to increase. Finally,
lenders are forced to recognize losses either through foreclosure
and sale of distressed properties or through foregone interest
revenues if loans are renegotiated to cure defaults.
Why Do Markets Become Oversupplied?
The fundamental task developers face is to make accurate
projections of demand for a given property to determine whether the
market will both support development and provide a reasonable
return on invested capital. With commercial real estate
development, however, there is a lag between a project's conception
and its completion. In the case of office buildings and hotels,
this lag can be substantial.
Developers nevertheless make decisions based on projections
weighted heavily toward recent sales and rental trends. If the
market in question is experiencing unsustainable growth (a
"boom" period for example), then such projections are
likely to lead to an overly optimistic outlook for future demand.
The degree of error in forecasted demand may very well be a
function of current growth in demand--the greater the growth rate,
the larger the margin of forecast error due to optimistic
expectations.2 Projections also may
fail to account for the planned development activities of
competitors.
If a developer's demand projections fail to materialize, the result
is an overhang of commercial property beyond what the market can
absorb during a reasonable time frame. Easy access to investment
capital can exacerbate overproduction of space by reducing or
eliminating borrower or developer incentives to make reasoned and
prudent investment decisions.3
Excessive leverage, where the developer has little personal capital
at risk, is a familiar example often associated with the excesses
of the 1980s.
Assessing a Market's Vulnerability to Overbuilding
This study employed a three-step process to rank the vulnerability
of markets to possible overbuilding. First, major metropolitan
markets were ranked in terms of current construction activity4 relative to existing space for each
of five property types: office, industrial, retail, hotel, and
apartment. Second, relative construction activity was compared to
current vacancy rates to assess the competitive pressures faced by
newly developed projects. Third, market-related research was
reviewed to determine which markets analysts consider candidates
for possible supply/demand imbalances.
Supply Focus
In simplest terms, overbuilding occurs when demand fails to keep
pace with new supply. Although not discounting the importance of
the demand side of this equation, the rankings shown in this paper
focus primarily on supply for the following reasons:
1) Projections of supply are perhaps more reliable than
demand forecasts because market participants know with some degree
of certainty what projects are in process and how long it will take
to complete them. Demand forecasts rely more on a variety of
assumptions including employment growth, the general rate of
economic growth, and forecasted rental rates.
2) For certain property types, market performance may
be more closely linked to long-term supply trends than shorter-term
fluctuations in demand.5 Chart 1 illustrates these trends nationally for
the office sector, in which long-term cyclical patterns in
completion rates appear to be leading indicators of patterns in
office vacancy rates.
3) Knowledge of trends in demand factors such as
employment rates and local business conditions may substantially
alleviate potential overbuilding concerns in some markets. Still,
such projections are subject to error, particularly if economic
conditions deteriorate suddenly (a "boom-bust" scenario,
if you will). For the sake of conservatism, this analysis does not
rely on absorption projections. In essence, this analysis asks how
much space will markets have to absorb in the near term if demand
projections fall short of expectations?
Limitations of the Analysis
The rankings presented in this paper are limited to major
metropolitan markets as defined by two primary data sources:
CB Commercial/Torto Wheaton Research (CB Commercial)
for office and industrial markets, and F.W. Dodge for
all other property types. Markets covered and underlying market
definitions employed are not consistent between these two sources.
Furthermore, the two sources use different methodologies to define
construction activity in progress (see endnote
4). A third source, ERE Yarmouth, was used to
determine aggregate measures of construction activity for hotel and
multifamily or apartment space. No attempt was made to reconcile
these three different approaches.
Another limitation in these rankings is the lack of differentiation
between suburban and downtown markets. In general, suburban markets
have experienced much stronger development during the nationwide
real estate recovery and therefore may experience oversupply
conditions earlier than downtown markets.
Finally, insufficient data exist to differentiate speculative6 from nonspeculative construction
activities. Ranking markets on the basis of the relative levels of
speculative development would likely form a much clearer picture of
where excesses are occurring. In this regard, the author relied on
market analysts to gauge the relative level of speculative
construction taking place in a given market.
Construction Employment Growth Is an Overall Indicator
Various indicators can be used to identify active real estate
markets. Construction employment growth is especially relevant
because it directly measures the change in labor resources
allocated to the construction sector. Chart 2
ranks the top 15 markets in terms of their respective three-year
compounded average construction employment growth. As will be
echoed throughout this paper, Las Vegas stands out with over a 15
percent average annual growth rate. Other markets with rapidly
growing construction sectors include San Jose, Dallas, Fort Worth,
Portland, Salt Lake City, and Austin.
Rankings Based on Construction Activity
Charts 3 through 7 show the level of
construction activity relative to the total stock of space at the
end of 1997 for the top 15 major metropolitan markets in each of
five property categories: office, industrial, retail, hotel, and
apartment. Las Vegas is by far the most active construction market
in hotel, office, and industrial. Orlando is higher in retail but
Las Vegas is a close second. Other markets undergoing rapid
development across multiple property types include Salt Lake City,
Charlotte, Atlanta, Portland, Phoenix, and Orlando.
There are other markets worth mentioning where rapid development is
centered mainly within a single property segment. For example,
Columbus, Nashville, and St. Louis all have office construction
activity well above the national average. In addition, markets such
as Kansas City, Greenville, Norfolk, and Denver are experiencing
rapid development in retail. Many analysts consider retail to be
overbuilt, particularly in the regional and power center formats.7
It is instructive to compare the rates of current development
across property types. Nationwide, the hotel sector has the highest
level of construction activity, at 9.1 percent of existing stock,
followed by retail at 6.6 percent. Apartment development has the
lowest level of activity at 1.9 percent.
Over the past few years, the pace of hotel development has jumped
considerably in markets across the country. Although much of this
construction is justified because minimal development took place
during the early 1990s, some analysts have warned about the
potential for overbuilding, particularly in the limited-service or
low- to mid-priced segment.8
Comparing Construction Activity to Vacancies
Levels of commercial property development should not be viewed in
isolation from existing available space. Accordingly, it is
worthwhile to compare construction activity to current vacancy
rates (as shown in Charts 8 through 10) for
office, industrial, and retail space.9 The main idea behind these charts is
that market segments with high existing vacancy rates raise the
degree of competitive pressures for newly built space; markets with
high vacancies may have less justification for continuing increases
in new stock.
As these charts show, Las Vegas stands out in both office and
industrial as having the highest level of new development combined
with relatively high existing vacancy rates. Although its pace of
development is markedly slower, Atlanta's real estate markets also
appear to be expanding rapidly despite high vacancy rates. With
regard to the retail sector, a number of markets, including
Orlando, Atlanta, Greenville, Norfolk, Phoenix, Denver, and Fort
Worth, exhibit a combination of rapid construction activity and
relatively high existing vacancy rates.
Considering the Opinions of Market Analysts
Except for isolated property segments such as power and regional
retail and limited-service hotel space, analysts are generally
upbeat about overall prospects for real estate markets in the near
term. Vacancy rates continue to decline in most major markets, and
most property segments are commanding good investment returns.
Nevertheless, market-watchers are becoming increasingly cautious
about the longer-term outlook in various markets, particularly in
light of the rapid increase in development for various property
types. Cushman & Wakefield, for example, recently
pointed to "early warning signs" of potential
overbuilding in office space in suburban areas such as Atlanta and
Dallas.10 Analysts recently have
noted a large increase in planned office space across various
metropolitan markets as indicating an eventual rise in vacancy
rates. For instance, a Lehman Brothers study
identifies 11 markets as "danger zones" based on
significant projected increases in vacancy rates over the next two
years.11 In many cases, a
predominant volume of planned space will be constructed on a
speculative basis.
Markets Most Vulnerable to Overbuilding
Based on the preceding information, including analyst opinions
concerning specific metropolitan areas, the following markets are
considered most vulnerable to broad-based overbuilding.
Las Vegas
By almost any measure, Las Vegas is the most active construction
market in the country and perhaps the most vulnerable to
overbuilding. Driven by its gaming sector, Las Vegas dwarfs all
other markets in new development of office, industrial, and hotel
space. At the same time, the city's relatively high office and
industrial vacancy rates place additional competitive burdens on
newly completed space. Despite historically strong employment
growth trends, the rapid pace of construction activity in virtually
all property sectors is highly suggestive of speculative
development. Moreover, the city would be especially hard hit by a
downturn in real estate prices since fully 10 percent of Las Vegas
is employed in the construction sector (twice the national rate).
Early evidence suggests that rapid development may already be
taking a toll as second quarter 1998 office vacancy rates rose
sharply during the quarter from 13.0 percent to 15.3 percent.12 Some analysts expect further
weakening resulting from a recent slowdown in employment growth.
Atlanta
Of the nation's largest metropolitan markets, Atlanta's
construction activity ranks among the top five in office,
industrial, and retail, with ratios of construction activity to
current space of 12 percent, 7 percent, and 11 percent,
respectively. Development, a large proportion of which is
speculative, is very active despite relatively high vacancy rates
in each of these property segments. Real estate investment trusts
control a large percentage of the office and industrial market in
Atlanta and are contributing factors to the rapid pace of
construction. Atlanta's real estate markets have been driven
largely by a rapidly expanding population (trailing only Las Vegas
and Phoenix) and strong employment growth. While analysts expect
these trends to continue, vacancy rates in apartment and office are
projected to rise due to even faster supply growth projections.13
Salt Lake City
Salt Lake City ranks among the top three markets in the nation in
office, industrial, and hotel development, with ratios of
construction activity to current space of 13 percent, 7 percent,
and 29 percent, respectively. Rapid development is being driven by
high-tech corporate expansions, population in-migration, and the
city's preparation for the 2002 Winter Olympic Games. Despite
relatively low current vacancy rates and strong employment sector
growth, analysts have warned of a potential glut of office and
hotel space. Some analysts have projected significant increases in
office vacancy rates over the next two years.14
Charlotte
Charlotte is home to banking powerhouses such as NationsBank and
First Union. The city's rapidly expanding finance sector has
spurred a significant level of development activity. Charlotte
ranks among the top 10 metro areas in office, industrial, retail,
and apartment development, with ratios of construction activity to
current space of 14 percent, 4 percent, 11 percent, and 7 percent,
respectively. CB Commercial forecasts a significant
rise in office vacancy rates over the next two years as newly
completed space outpaces absorption. Due to the employment make-up
of the city, Charlotte's real estate markets are highly dependent
on the health of the financial industry.
Portland
Portland ranks among the top 10 metro areas in office, retail,
hotel, and apartment development, ratios of with construction
activity to current space of 8 percent, 9 percent, 22 percent, and
6 percent, respectively. Portland's expanding markets have been
driven to a large extent by in-migration and a rapidly growing
technology sector. The market reportedly has a large number of
speculative office and industrial projects in the planning or
initial construction phases. Moreover, although the city's
apartment market may already be saturated, apartment development
continues at a brisk pace.15
Phoenix
After Las Vegas, Phoenix has the second largest population growth
rate in the nation. It ranks among the top 10 metro markets in
office, industrial, hotel, and apartment development, with ratios
of construction activity to existing space of 6 percent, 7 percent,
16 percent, and 6 percent, respectively. Retail development ranks
eleventh in the nation despite a relatively high retail vacancy
rate. Phoenix's strong employment growth is encouraging increased
speculative office development, and apartment construction has
saturated the market in some submarkets.16
Markets Exhibiting Rapid Development in a Specific Sector
In addition to the above markets whose development activities are
fairly broad-based, the markets discussed below are vulnerable to
overbuilding within a specific property sector:
Retail. Most analysts feel that the nation's retail markets
as a whole are overbuilt. Still, a number of markets are undergoing
rapid development despite high existing retail vacancy rates.
- Orlando's retail market is developing more rapidly than
any other metro area, with construction activity at 17 percent of
existing stock. According to research by CB
Commercial, the current stock of space in Orlando is 4.1
percent above the level needed to maintain constant real rents.17 Moreover, CB
Commercial projects 1998 through year 2000 retail
completions of 6.8 million square feet, exceeding projected demand
of 4.5 million square feet.
- Kansas City has the third highest rate of retail
development, with construction activity at 12 percent of existing
space. During 1997, the city had retail completions of 1.7 million
square feet (fourth quarter estimated), which exceeded estimated
demand of 513,000 square feet (fourth quarter estimated) by 3.3
times.
- Denver has the ninth highest rate of retail development,
with construction activity at 9 percent of existing space. This
pace of construction is particularly high considering a 17 percent
inventory of available retail space. CB Commercial
projects 1998 through year 2000 retail completions at 5.4 million
square feet, well in excess of projected demand of 400,000 square
feet.
- Fort Worth had the fourteenth highest rate of retail
development, with construction activity at 9 percent of existing
stock. This pace of development is high considering an already high
retail vacancy rate of 15 percent. CB Commercial
calculates that the stock of existing space is 1.1 percent above
equilibrium levels.
- Smaller metro markets experiencing rapid retail development
include Greenville and Norfolk, both of which also
have high vacancy rates.
Office. Unlike the retail sector, which according to many
market observers may be in the latter stages of a development
cycle, growing development rates in the office sector are a fairly
recent phenomenon. Still, even at this early stage, the rapid pace
of office development in some markets has led analysts to warn of
possible oversupply.
- Columbus ranks fifth among all metro markets in office
development, with construction activity at 11 percent of existing
stock. Much of the development under way is speculative, and
CB Commercial projects a sharp rise in vacancy rates
over the next two years, with expected completions of 1.2 million
square feet as compared to expected absorption of only 349,000
square feet.
- Nashville ranks ninth among all metro markets in office
development, with construction activity at 6 percent of existing
stock. It may be difficult for the city's office market to absorb
the space coming on line given a modest increase in vacancy rates
from 6.4 to 7.2 percent in the second quarter of 1998. Moreover,
CB Commercial projects a sharp rise in vacancy rates
over the next two years, with expected completions of 1.1 million
square feet as compared to expected absorption of only 128,000
square feet.
Hotel. The pace of hotel development has picked up
substantially over the past two years to levels not seen since the
late 1980s (see Chart 11). Much of this
development is likely related to pent-up demand given the dearth of
hotel development activity through the early 1990s. Nevertheless,
the hotel sector should be watched with some caution because of the
higher risks generally associated with this property type (longer
construction periods and greater vulnerability to economic
downturns, for example). Some analysts already have sounded
warnings of potential overbuilding in the limited-service segment.
Conclusion
This paper identifies rapidly growing commercial real estate
markets where analysts have warned of potential overbuilding. In
some cases, the concerns encompass all development within a
particular market; in other cases, concerns involve only a specific
property type such as retail or office space. Some property types
in general, such as retail and hotel, also should be monitored
closely due to emerging and existing supply/demand imbalances.
As this paper illustrates, a number of U.S. markets are
experiencing rapid growth in commercial property development. Much
of this development is occurring in small to mid-sized markets,
many of which did not participate in the large-scale real estate
downturn of the late 1980s and early 1990s. Although a
"boom-bust" scenario is not imminent for these areas,
there is a heightened degree of concern over the rapid development
and related growth in construction loan concentrations occurring
within these markets.
Endnotes
1 Freund et al., History of the
Eighties: Lessons for the Future, FDIC 1997, 1: 158-159.
2 Liang, Youguo, and John Kim,
"The Office Market Cycle: There's Hope This Time," The
Real Estate Finance Journal, Fall 1997, p. 39.
3 See fourth quarter 1997 and first
quarter 1998 editions of FDIC's Regional Outlook for more
information on current commercial real estate financing trends.
4 Construction or
"pipeline" activity generally refers to recent
completions plus projects in process of being built. More specific
definitions of "pipeline" must be interpreted within the
context of the data sources referenced. In the case of office and
industrial properties, the source of data is CB Commercial/Torto
Wheaton research in which pipeline is defined as space completed
during 1997 or known to be under construction at the end of 1997.
For retail, the source is F.W. Dodge and pipeline is referred to as
space completed in 1997, construction started in 1997, and
"pending projects." F.W. Dodge defines pending projects
as those that are either in the final planning or bidding stage.
For hotel and apartment, the source is ERE Yarmouth, and pipeline
is referred to as completions, projects in process of being built,
starts, and pending projects.
5 See Introduction to CB
Commercial/Torto Wheaton Research's Office Outlook, Spring
1998, pp. II.1 to II.6, and Wheaton, William, and Lawrence Rossoff,
"The Cyclical Behavior of the U.S. Lodging Industry,"
Real Estate Economics, 1998, 26, 1: 67-82.
6 The term speculative here
indicates projects undertaken absent a meaningful degree of
preleasing or presales.
7 See, for example, CB
Commercial/Torto Wheaton Research's Retail Outlook, Spring
1998, pp. II.1 to II.4, and Urban Land Institute's ULI 1998 Real
Estate Forecast, pp. 40-41.
8 See, for example, Urban Land
Institute's ULI 1998 Real Estate Forecast, pp. 36-37, and
lender comments in the "Lenders in Hotel Finance and
Development" section by Mark Faris in Commercial Property
News, August 16, 1998, pp. 16-20.
9 Insufficient data were available
to produce comparable charts for hotels and apartments.
10 As quoted in the Wall Street
Journal in an article by Mitchell Pacelle, "Suburban
Office Vacancies on the Rise," April 29, 1998.
11 These markets are Salt Lake
City, Columbus, Nashville, Charlotte, St. Louis, Atlanta,
Baltimore, Las Vegas, Indianapolis, Dallas, and Miami. See Lehman
Brothers' Commercial REIT Research: Eye on Office Markets,
by Steve Hash and Michele Martin, May 5, 1998.
12 Based on second quarter 1998
CB Richard Ellis Pre-audit Office Vacancy Index.
13 Refer to FDIC's Bank
Trends June 1988 issue (98-06) entitled "Metropolitan
Atlanta Construction and Development Trends," by Jack Phelps
et al., for an in-depth analysis of the Atlanta metropolitan real
estate market.
14 See, for example, CB
Commercial/Torto Wheaton Research's Office Outlook, Spring
1998, and Lehman Brothers' Commercial REIT Research: Eye on
Office Markets, May 5, 1998.
15 Refer to "Market
Snapshot" comments in ERE Yarmouth's Real Estate Outlook:
1998.
16 See note
15.
17 By way of comparison, the
national retail market is 3.8 percent above equilibrium levels. See
CB Commercial/Torto Wheaton Research's Retail Outlook,
Spring 1998.
References
Burton, Steven. Federal Deposit Insurance Corporation. 1998.
Trends in commercial real estate loan pricing and underwriting.
Regional Outlook, First Quarter.
Burton, Steven, and Ternullo, Gary. Federal Deposit Insurance
Corporation. 1997. Strong demand and financial innovation fuel
rebounding commercial real estate markets. Regional Outlook,
Fourth Quarter.
CB Commercial/Torto Wheaton Research. 1998. Office Outlook,
Spring.
CB Commercial/Torto Wheaton Research. 1998. Industrial Outlook,
Spring.
CB Commercial/Torto Wheaton Research. 1998. Retail Outlook,
Spring.
CB Richard Ellis Pre-Audit Office Vacancy Rate Index, Second
Quarter 1998.
ERE Yarmouth. 1998. Real Estate Outlook: 1998.
Faris, Mark. 1998. Sales count on E Walk's cost, uniqueness to
limit entrance of competing projects. Commercial Property News,
16 August.
Freund, James, et al. Federal Deposit Insurance Corporation. 1997.
History of the Eighties: Lessons for the Future, vol. 1.
F.W. Dodge. 1998. Real Estate Analysis and Planning Service:
4th Quarter 1997.
Hash, Steve, and Martin, Michele. 1998. Lehman Brothers.
Commercial REIT Research: Eye on Office Markets, May.
Liang, Youguo, and Kim, John. 1997. The office market cycle:
There's hope this time. The Real Estate Finance Journal,
Fall.
Pacelle, Mitchell. 1998. Are developers forgetting the lessons of
the real-estate bust of the early 1990s? Wall Street Journal,
April 29.
Phelps, Jack, et al. Federal Deposit Insurance Corporation. 1998.
Metropolitan Atlanta construction and development trends. Bank
Trends, June.
Urban Land Institute. 1998. ULI 1998 Real Estate Forecast.
Wheaton, William, and Rossoff, Lawrence. 1998. The cyclical
behavior of the U.S. lodging industry. Real Estate
Economics, vol. 26.
About the Author
Steven Burton is a senior banking analyst in the Depository
Analysis Section of the Division of Insurance.
About the Division of Insurance
The Division of Insurance (DOI) was created in 1995 to identify,
analyze, and report on existing and emerging risks to the banking
industry and depository insurance funds. Arthur J. Murton is
Director of DOI.
About Bank Trends
Bank Trends is a series of occasional papers published by
the Division of Insurance. The papers address current issues in
banking, economics, and finance as they relate to exposures to the
banking system and deposit insurance funds. These analyses are
available free of charge on the FDIC's website at www.fdic.gov. Requests for
copies also can be made to
FDIC, Public Information Center
801 17th Street, N.W., Room 100
Washington, D.C. 20434
Phone: 800-276-6003 or 202-416-6940
FAX number: 202-416-2076
Other DOI Products
Regional Outlook is published quarterly by each of the
FDIC's eight Regions and explores potential risks and trends
affecting insured depository institutions from a Regional and
national perspective. These publications and other products are
available on the FDIC's website at www.fdic.gov.
Disclaimer
The views expressed in this article are those of the author(s) and
do not necessarily reflect the official position of the Division of
Insurance or the Federal Deposit Insurance Corporation.
