<img height="1" width="1" src="https://www.facebook.com/tr?id=373327176680496&amp;ev=PageView&amp;noscript=1">

By: Alex Carrick on February 6th, 2017

Print/Save as PDF

2 Leading Monitors of U.S. Construction Activity, 1 Public and 1 Private

Economic News

The Census Bureau’s put in place construction statistics, as featured in Table 1, are similar to progress payments as projects proceed.


Table 1 takes the investment or capital spending ‘put-in-place’ numbers and, for each type-of-structure, compares the percentage changes of latest-12-months-over-previous-12-months versus latest-three-months-over-previous-three-months (annualized). The background dollar-volume levels are seasonally adjusted.

If the three-month percent change exceeds the 12-month percent change, then construction activity in that type-of-structure category is considered to be speeding up. A check mark is entered in the far right column.

If a type-of-structure category has a latest 3-month percent-change that is negative, but less negative, than its 12-month percent-change, such a circumstance is also considered to be an instance of ‘speeding up’ and warrants a check mark in the right-hand column.

When Table 1 was compiled a quarter ago, for September 2016 data, there were only two stars in the ‘speeding up’ column.

The ‘slowing down’ column, at that time, captured all three stars among the ‘Total’ categories and 14 of the 16 stars among the type-of-structure subcategories.

In current Table 1, however, there have been significant reversals. Both ‘Total Construction’ and ‘Total Residential’ are now judged to be speeding up and nine of the 16 subcategories (i.e., a majority) have picked up the pace.

In other words, the Census Bureau (i.e., the ‘public’ monitor in the headline) in the fall of 2016 was saying that the construction sector had become mired in a swampy place. Since then, it has broken free and is now walking with some spring in its step.

Weighting of type-of-structure category
(% of total $s)
    Year to date
Year to date
(from monthly averages of SAAR data)
    Latest 12 mons vs. previous 12 mons Latest 3 mons vs. previous 3 mons (annualized) Slowing Down
1 of 3
'Total'categories /
 7 of 16 sub-categories
Speeding Up
2 of 3 'Total'categories / 
9 of 16 sub- categories
Total Construction 100.0%     4.7%     4.7%   5.7%      
Total Residential 40.1%     5.3%     5.3%   8.5%    
Total Nonresidential 59.9%     4.4%     4.4%   3.8%      
Lodging 2.3%     24.6%     24.6%   -5.9%      
Office 6.4%     24.6%     24.6%   15.6%      
Commercial (retail) 6.7%     10.7%     10.7%   22.2%    
Health care 3.5%     1.5%     1.5%   -2.1%      
Educational 7.7%     6.6%     6.6%   14.0%      
Religious 0.3%     2.1%     2.1%   -30.0%      
Public safety 0.7%     -7.3%     -7.3%   14.2%      
Amusement and recreation 1.9%     9.7%     9.7%   17.0%      
Transportation 3.6%     -5.7%     -5.7%   6.3%      
Communication 1.8%     -3.7%     -3.7%   45.4%      
Power (electric; oil & gas) 7.9%     3.0%     3.0%   -15.6%      
Highway and street 8.0%     3.3%     3.3%   27.7%      
Sewage and waste disposal 1.6%     -7.8%     -7.8%   -14.9%      
Water supply 1.0%     -8.9%     -8.9%   8.5%      
Conservation and development 0.7%     -3.5%     -3.5%   36.9%      
Manufacturing 5.8%     -4.3%     -4.3%   -31.6%      

In the final two columns, if there is no check mark in the cell, then the type-of-structure category has stayed within 1.0%, up or down.
For December, the year-to-date figure is the same as the 'latest 12 months' figure.

Data source: Census Bureau / Table: ConstructConnect


Graph 1 shows moving 12-month totals of ConstructConnect’s construction starts figures (i.e., the ‘private’ monitor in the headline).

For example, the residential, nonresidential building and engineering figures for June 2016 are the sums of their monthly numbers for July 2015 through June 2016.

December of each year, being the sum of the previous 12 months, is also the annual total.

A positive feature of this methodology is that a rolling forward 12-month total eliminates the effects of seasonality.

Graph 1 highlights how much more severely residential construction was impacted by the Great
Recession than was nonresidential building work. Engineering/civil construction barely felt a twinge.

The 12-month total for residential starts has been on the comeback trail since early-2012, but its level has still not fully recovered to where it was in 2006.

Nonresidential building starts have recovered and surpassed their previous peak achieved in 2008, but they’ve flattened out over the past six months.

The curve for engineering/civil starts has been on a gradual upward sloping incline over the 11 years covered, but it has tailed off a bit of late.

The ‘starts’ series, which are essentially groundbreakings, are ‘leading indicators’ for the construction investment or capital spending ‘put-in-place’ series.

Graph 1: ConstructConnect's 12-month Moving Totals* of U.S. Construction Starts
December 2016
*For example, June 2016 is the sum of the monthly numbers for July 2015 through June 2016.
December numbers are the same as the annual total.
Data source: ConstructConnect.
Chart: ConstructConnect.
Take Control of Your Construction Bidding Pipeline & Find More Project Leads in Your Backyard
Find. Bid. Win. Repeat!

Subscribe to the Blog!