Tuesday, February 12, 2013

More Research on Construction Employment

by Bill McBride on 2/12/2013 12:24:00 PM

A key economic question this year is how many construction jobs will be added. Here are a few excerpts from analysis Kris Dawsey and Hui Shan at Goldman Sachs: Housing Sector Jobs Poised for a Comeback

Although many indicators of housing activity improved during 2012, employment in the sector remains close to post-bubble lows. Looking only at residential construction jobs, employment declined by 1.5 million (-42%) from its peak in 2006 to its recent trough in early 2011 and edged up only a modest 100 thousand since then. However, direct residential construction employment is only a part of all residential investment-related employment. Adding in housing-related employment in manufacturing, wholesale trade, retailing, and finance & real estate, employment dropped by 2.8 million (-31%) from its peak, and gained a bit less than 300 thousand from its trough to the present ...

[R]eal residential investment declined somewhat more sharply than housing-related employment in the downturn, resulting in a decline in real value added per residential investment-related worker, according to our proxy measure, from more than $80,000 in 2006 to a bit less than $60,000 in Q4:2012, in chained 2005 dollars. This pattern of declining productivity during a downturn is called "labor hoarding" by economists (although labor hoarding is probably not what most people think of during a period of sharp job cuts) and reflects businesses' reluctance to fire workers at a rate commensurate with the decline in their sales.

The flip side of this phenomenon is more sluggish employment growth than would otherwise be the case once business activity turns around. On top of the only modest turnaround in activity, this secondary effect also argues for only a modest rebound in residential investment-related employment early on in the recovery. However, this effect may shortly be coming to an end. Hours per worker in the construction industry now exceed pre-crisis highs, suggesting that room to increase output on the "intensive margin" (i.e. more hours per worker) is diminishing, and that pushing on the "extensive margin" (hiring more workers) will likely account for a larger share of future increases in residential investment output.
...
Given that we expect real residential investment to continue growing at a roughly stable 10%-15% rate in 2013 and 2014, and that the effects of labor hoarding should be dissipating, what is our forecast for residential investment-related employment growth over the coming several years? In order to answer this question, we estimated two different econometric models: (1) an error correction model of national-level real residential investment and residential investment-related employment, and (2) a state-level panel analysis of the relationship between construction activity and employment. Both models suggest an increase in the rate of housing-related employment growth in 2013 and 2014 relative to 2012, probably to a rate around 25 to 30k per month.
emphasis added
So their analysis suggests construction companies have been increasing hours worked for current employees, but now they need to hire more workers.

Earlier articles on construction employment:

• From Michelle Meyer at Merrill Lynch: Construction Coming Back

• From Jed Kolko at Trulia: Here are the “Missing” Construction Jobs

• From Professor Tim Duy at EconomistsView: Employment Report Nothing If Not Consistent

Last 10 Posts