The State of the Construction Sector

As recently as 2006, the construction sector of the economy was as large as $1.2 trillion annually. While the decline in residential construction has had most of the headlines, non-residential building construction and heavy construction (mostly infrastructure related) has has been holding up much better. That appears to be about to end.

An online seminar July 23, sponsored by Georgia-Pacific (here), addressed the current state of the construction industries. The speakers were Jim Haughey, Chief Economist for Reed Construction Data (here), Ken Simonson, Chief Economist for the Association of General Contractors of America (here) and Kermit Baker, Chief Economist for the American Institute of Architects (here). The 70 slides presented are available on-line here. Some of the highlights are covered in this article.

Some of the opinions expressed by Jim Haughey include:

• There will be an initially strong recovery followed by sluggish activity;
• Construction activity will decline through the summer, but will be rising by early 2010;
• Construction in 2012 will return to 2006 levels;
• Space surplus for all types of building will peak within a year, but then remain high;
• Public construction will be constrained by lower tax revenues;
• Non-residential building declines are deepening in 2009, but are not the start of a collapse.

The following graph shows Haughey’s data and projections for total construction activity (nominal, not corrected for inflation):

The annual data and projections by major construction categories are shown in the following table:

The first two categories basically cover the building construction business, while heavy construction is comprised of infrastructure activities related to transportation, energy, utilities, etc. Before you get too excited about a projected 10% increase in residential construction in 2010, look at the following graph from Kermit Baker:

If the estimate of 600,000 starts in 2009 is accurate, that would mean an increase in housing starts of approximately 60,000 units nationally from the estimated very depressed level. There will have to be an increase in residential construction in the remainder of 2009 to move up from the current annualized rate, which is closer to 500,000 than 600,000 (about 530,000 for the first six months of 2009, according to the U.S. Census Bureau here).

I would challenge the Baker estimate. I find it hard to conceive that the annual rate for the second half of 2009 would be 670,000 units, the amount necessary to get a 600,000 number for the entire year. That would require the second half of 2009 to have residential housing starts increase by more than 28% from the first half of the year, and equal the 670,000 rate (implied by Haughey’s projection) for 2010. The annual rate for new building permits is also around 530,000 for the first six months of the year, with June at the rate of 563,000. Building permits are not reflecting a dramatic increase in starts for the second half of the year. Is this projection a leap of faith?

The following graph shows Haughey’s view of non-residential construction expenditures:

I am having trouble reconciling the above graph with the following Haughey table:

The table includes residential numbers, in spite of the title, but that is not my concern. Of the 22 non-residential categories, 17 of the categories show declines greater than 20% (11 were greater than 40%). Three categories show declines between 9% and 18% and two categories are actually positive. The table refers to the value of starts during the period and the graph covers the ongoing expenditures for projects that take much longer time periods than six months. The inference is that the decline in the value of starts is believed to be a very short time event and that the second half of 2009 will show a much higher value for starts and this will continue in 2010. Is this another leap of faith? Or will stimulus spending make the difference?

The following chart, constructed by Ken Simonson, shows the construction related portions of the stimulus plan:

Let’s apply the $135 million over two years, a reasonable period of time to complete and pay for all projects. The spending will be applied uniformly over time for discussion purposes, although the actual spending rates may fluctuate. If we start now, then nearly $35 billion will be spent in the second half of 2009 and the first half of 2011, with nearly $70 billion in 2010. Let’s also add the residential construction estimates discussed earlier (and referring to the graph below), another $35 billion added in the second half of 2009 and $30 billion more in 2010.

The total to be added from all these sources is $70 billion (annual rate increase) in the second half of 2009 and $100 billion for 2010. Adding to the value of $964 billion annual rate for May in the graph above, this gives calculated quarterly expenditures for all construction (annual rates) which can be compared to the U.S. Construction spending output in the first graph. These results are shown in the table below (my calculations):

This indicates that without the Haughey projected increase in housing construction and the stimulus package, the construction sector would be contracting through the first quarter of 2010 and still be $54 billion below the May, 2009 rate in the fourth quarter of 2010. This projects that the construction industry will lag the recovery in other parts of the economy by a few quarters if the recession is ending this summer and a recovery ensues. Without the Haughey projected increase in housing construction and the stimulus spending, the rest of construction could take a couple of years longer than other areas of the economy to get back up to levels of the summer of 2009.

Comments

[Old Trader:]

John,

The data you present and your interpertation, don't bode well for the strength of the "recovery". Anecdotal support was provided by CAT's guidance going forward. I seem to recall something about "continued weakness in the North American markets".

Jul 25 12:13 PM

[fran:]

in order for growth of any nature to occur in the areas discussed here, the nation needs GROWTH in jobs and small/medium size businesses. spend your time/energy to build a case for that to happen[growth in jobs,businesses].

enough political and market rhetorical BS has been heard.

WHERE'S THE BEEF??? and no more green jobs BS, please. only supportable facts accepted.

how are jobs growing in ELKHART, INDIANA?
wasn't that the site of the plea for "FAST" passage of the "STIMULUS" legislative miracle?

Jul 25 03:20 PM

[sleepless_o...:]

Isn't construction activity one of the areas of the economy that tends to most lag in the economic cycle? Lead times on commercial construction and infrastructure are quite long and so construction activity tends to continue into the downturn and new projects not to pick up until the recovery is well underway. Housing tends to respond quicker. That is what appears to be happening in the current cycle, as well.

Jul 25 11:27 PM

[billddrummer:]

I agree with sleepless, that construction won't return as a vibrant economic force for years. With capacity utilization still below 70%, there's no incentive for businesses to expand facilities. Lenders are reluctant to finance new construction because they are grappling with higher nonperforming construction loans than anyone predicted 2 years ago. Even build-to-suit projects are being put on hold, because businesses just don't need the room right now.

Until the broader economy picks up, construction will be a drag.

Jul 27 02:39 AM

[Ken Simonson:]

Some highway contractors have won stimulus projects and have added workers or avoided planned layoffs. Many more highway projects will be awarded in the next few weeks as states hold scheduled "bid lettings." But the nonhighway stimulus money has been coming out much more slowly. In any case, even $70 billion per year of stimulus money may not offset the decline in private and state and local government-funded work. Total (private, federal, state and local) nonresidential construction spending was $716 billion in May at a seasonally adjusted annual rate. It is not hard to imagine a 10% derease in this figure by yearend, given the plunge in starts reported by Reed Construction Data and McGraw-Hill Construction, the rising vacancy rates for office, retail, warehouse and hotel properties, the extraordinarily low factory utilization rate and the repeated cuts in state and local budgets.

Jul 27 01:41 PM