<%@ Language=VBScript %> <% Response.Write(cszCSS) %> Detailed Report
[Econoday]
Today's
Calendar
 |  Simply
Economics
 |  International
Perspective
 |  Short
Take
 |  Market
Recap
 |  Resource
Center
Back to Resource Center
[Back To Archive]

Nonresidential investment marches to its own drum
Econoday Short Take - July 31, 2002
By Evelina M. Tainer, Chief Economist, Econoday

Though the housing market has remained strong for several years, nonresidential construction investment has been weak and unsteady. Spending on nonresidential construction headed down through the first half of this year, perhaps hitting a bottom in March and beginning a cyclical climb in April and May. It is still a bit early to tell whether nonresidential construction as a sector will continue to recover in coming months, but we may be able to better determine where separate pieces of the sector are going.


Nonresidential construction includes industrial buildings (factories), office buildings, hotels & motels, commercial buildings, hospitals, educational structures and religious buildings. While there is no question that lower interest rates boost all categories of construction, other factors also play a role. We will look at a couple of the key components separately.

Industrial buildings
In starting to analyze this sector, it is useful to see if industrial buildings have followed the same cyclical pattern over time. Between January 1978 and May 2002 the U.S. economy suffered four recessions (1980, 1981-82, 1990-91, 2001-2002). A quick look shows that growth rates were more positive in the early part of this time frame and more negative in the latter part. It appears that industrial buildings have been on a downward secular trend while all along bouncing up and down in normal cyclical variation.


Industrial buildings are essentially factories. Thus, one would suspect that industrial production growth is a relevant component in determining construction of industrial buildings. If industrial production accelerates, then more factories would be needed; conversely, a decline in industrial production would be associated with less need for additional factory space. Once we overlay the yearly growth in industrial building construction with the yearly growth in industrial production, we see that the two indeed move in tandem. The slower pace of construction growth for industrial buildings in the 1990s is associated with slower industrial production growth during this period.


Outlook for industrial building construction: Industrial production has turned around in the past two quarters. This implies that construction of industrial buildings could turn around as well in the coming year. There are some caveats, however. On the negative side, industrial production is recovering at a moderate pace. Manufacturers are not likely to demand additional factory space soon. In the long term, manufacturing is becoming a smaller portion of the U.S. economy. And the more manufactured goods purchased overseas, the less need there will be to build factories here! On the positive side, as we produce goods more efficiently to maximize productivity in the United States, manufacturers may find that their needs have changed in terms of the size and shape of their physical buildings. This may encourage construction of more efficient building space. Bottom line: Look for this sector to post gains, albeit at a moderate pace, in line with industrial production in the near term. Long-term prospects could follow the path of the past 10 years -- this means less construction spending on industrial buildings.

Office buildings
Office building construction has had its ups and downs in the past few years. For the most part, the pattern of growth is generally in line with the business cycle. That is, growth declined during recessions and rose during expansions. It does appear, though, that decreasing spending on office buildings tends to last longer than the recession period.


In fact, construction spending tends to move with demand for space, which itself is often associated with the business cycle. However, given that office buildings are large structures, it isn't unusual to see builders overshoot demand. The office vacancy rate is inversely related to construction spending for office buildings. When the vacancy rate starts to rise, then builders start moderating their construction. When vacancy rates decline, office building construction will accelerate. We are currently in a down-phase on spending and an up-phase on the vacancies. Data for the last 30 years show that downturns typically last for at least a year and often, like now, two years.


Outlook for office building construction: Office building construction expenditures peaked in the fourth quarter of 2000. Following the pattern of a two-year downward trend, we should expect continued weakness until year-end. Office vacancy rates have risen sharply in the past two years, but are nowhere near the highs of the mid-1980s. We could look for a bottom in office building construction before the end of this year.

Hotels & Motels
At first glance, the pattern exhibited for construction of hotels and motels appears different from the cyclical patterns seen for industrial and office buildings. If you look at the downward trends in construction, they are generally associated with the business cycle. Growth rates for hotel & motel construction are generally more positive than negative, although there are many periods when hotel construction is unchanged.


We find that hotel and motel construction indeed moves in line with the business cycle. Here we use the yearly growth in nonfarm payrolls as a proxy for a business cycle indicator. Typically hotel construction will continue as long as vacancy rates are low. A significant amount of construction takes place during periods of low interest rates. Eventually too many hotels and motels are built, which need then to be absorbed. You'll see that even though the economy is growing, hotel & motel construction is flat after an initial burst of activity. This pattern was evident during the 1980s and 1990s.


Outlook for hotel & motel construction: The hotel industry was already suffering in the early part of 2001, but took an even bigger hit after September 11. Construction has been unchanged for nearly two years. However, this pattern is in line with historical trends. Indeed, construction in this sector was unusually robust during a good chunk of the 1990s. This could mean that the absorption time will be a bit longer this time. But still, if the economy turns around in the coming year and vacancy rates decline rapidly, another round of hotel & motel construction could begin in 2003.

BOTTOM LINE
We like symmetry. During economic recessions, we would prefer to see all indicators move down at the same time. Conversely, during expansions, we prefer to see all economic sectors move up at the same time. This type of behavior encourages the belief among market players that the economy is either in a downturn or an upturn. But in fact, some sectors of the economy are counter-cyclical. That's a good thing, otherwise we'd be using all of our resources at the same time and overheating the economy. Since some sectors of the economy grow at a different pace, we aren't overextended. And on the flip side, it keeps the economy from falling too deeply in recession.

Consumer spending and residential construction have certainly played major roles in economic growth during 2001 and 2002. Growth in the housing market, if not now consistent with a bubble, is very likely to moderate over the coming year, at which time expansion in the nonresidential sector, especially office buildings and hotels & motels, may actually accelerate. It could very well be that 2003 sees investment spending as the impetus behind economic growth.

Evelina M. Tainer, Chief Economist, Econoday

Legal Notices | © 2001-2002 Econoday, Inc. All Rights Reserved.