Special Studies, July 3, 2014
By Paul Emrath, Ph.D.
Economics & Housing Policy
Report available to the public as a courtesy of HousingEconomics.com
Recent survey data from NAHB show that shortages of labor and subcontractors have become substantially more widespread since 2013. The incidence of reported shortages is now surprisingly high relative to the current state of new home construction, which has only very partially recovered from its 2008 downturn. The shortages are also particularly acute for workers with basic skills like carpentry, who are needed in substantial numbers for the construction of any home.
In addition, the survey data show more builders reporting a shortage of subcontractors than of workers they employ directly. Partly as a result, costs of subcontractors are rising faster for builders than costs of directly-employed workers. The implication is that any reporting of construction labor statistics that ignores the effects of subcontracting is likely to understate the magnitude and impact of the shortages.
Shortages by Trade
Periodically, the survey for the NAHB/Wells Fargo Housing Market Index (HMI) has included a set of special questions on the availability of labor and subcontractors. After an absence of several years (availability of inputs was not an issue during the trough of the industry’s recession) these questions returned to the HMI in June 2012. They have reappeared twice since—most recently in June of this year.
The questions ask single-family builders to identify shortages in 12 specific construction trades: carpenters-rough, carpenters-finished, electricians, excavators, framing crews, roofers, plumbers, bricklayers/masons, painters, weatherization workers, HVAC and building maintenance managers. For each trade, the builders indicate if there is “no,” “some,” or a “serious” shortage (see the detailed report available in the “Additional Resources” box for the precise wording in the questionnaire).
In June of 2014, the share of builders reporting a shortage (either serious or some) of labor employed directly by their firms ranged from a low of 25 percent for building maintenance managing to 63 percent for rough carpentry (Figure 1). Shortages are particularly acute in some of the basic skills required to build any home, as well over half of builders reported a shortage of all three categories of carpenters (Figure 1).
Although not all builders are reporting labor shortages (several, in fact, explicitly wrote in comments that there were no shortages in their area), the share reporting a shortage has increased for each trade since 2013. Averaged across all 12 trades, 41 percent of builders reported a shortage (either serious or some) in the labor directly employed by their firms—up sharply from 20 percent in 2012 and 28 percent in 2013.
To get a longer term perspective, we can look at the nine trades that have been covered consistently in the HMI survey since 1996. Three of the 12 categories on the 2014 questionnaire (weatherization workers, HVAC, and building maintenance manager) were first introduced in 2012 at the request of Home Builders Institute (an organization established by NAHB to train construction workers and generally promote careers in the building industry).
Averaged across the nine consistently covered trades, 46 percent of builders reported a shortage in 2014. This is the highest the 9-trade shortage has been since 2000—slightly higher even than at the peak of the boom in 2004 and 2005, when the U.S. was averaging around 2 million housing starts a year, compared to current rates that have mostly remained under 1 million (Figure 2).
The figure shows a sharp contrast between the strength of the surge in shortages and the modest nature of the housing market recovery so far. Even as of May 2014, total housing starts have barely recovered to an annual rate of 1.0 million. In comparison, from 1960 to 2000, housing starts averaged over 1.5 million a year and never dipped below 1 million for a full year. But starts have now been under 1 million consistently since 2008, making the last six years the six worst for new home construction since World War II. Against that backdrop, the 9-trade labor shortage percentage more than doubled from 21 percent in 2012 to 46 percent currently.
In addition to the average shortage shown in Figure 2, the full report (available under “Additional Resources”) contains individual graphs for carpenters and bricklayers. The historical trends are basically the same in each case, although the shortage percentages for the bricklayers and carpenters are consistently higher than the 9-trade average.
Between 2002 and the recent surge in reported shortages, the trend in builders’ labor shortages tended to mirror the movement in housing starts. In the earlier period from 1996 through 2000, however, shortage percentages tended to be much higher, with the 9-trade average shortage above 50, or sometimes even 60, percent.
One relevant factor in the late 1990s is the strength of the U.S. economy at the time, with real GDP consistently growing at a rate of over 4 percent a year. Partly as a consequence, unemployment tended to be relatively low in that period, bottoming out at 4.0 percent in 2000 (Figure 3).
For completeness, Figure 3 also includes the value of non-residential construction, which his has the potential to siphon off labor available to home builders. At the national level, however, this doesn’t appear to be much of an issue; as, in real terms, the value of non-residential construction activity was relatively stable over the period covered in the table.
The different historical periods in Figure 3 can be summarized as follows:
1996-2000—the share of builders reporting labor shortages is well above the current 9-trade average of 46 percent. GDP growth remains consistently above 4 percent, driving the unemployment rate down to 4.0 percent.
2002-2006—the 9-trade average shortage remains slightly-to-substantially below 46 percent, even while unemployment stays at or below 6.0 percent and housing starts peak at 2 million.
2007-2011—NAHB refrains from systematically investigating labor shortages during a period of historic weakness in the home building industry, which included a financial crisis and overall economic recession.
2012-2014—the 9-trade average labor shortage more than doubles from 21 to 46 percent. This is consistent with declines in unemployment and increases in housing starts, although the surge in labor shortages is unusually strong in relation to these other trends.
Significance of Subcontracting
Most of the discussion so far has dealt with labor directly employed by single-family home builders. The HMI survey also collected similar information on subcontractors.
In the late 1990s, the reported shortages of labor and subcontractors tracked each other almost perfectly, to the point that NAHB decided it was unnecessary to include questions about both in some of the 2002 and 2003 surveys. Since 2004, gaps of a few percentage points between the 9-trade average for subs and labor have sometimes occurred. After a particularly strong surge over the past year, the shortage percentage for subcontractors rose from 31 to 51 percent, while the percentage for directly-employed labor increased from 34 to 46 (Figure 4). The resulting 5 percentage point gap in June of 2014 is the farthest the subcontractor percentage has even been above its equivalent for labor.
This is important for two reasons. The first is that subcontracting accounts for such a large share of the work in residential construction. According to NAHB’s latest (2012) survey on the topic, single-family builders on average employ 25 different subcontractors when building a home. Over half of builders subcontract out at least 75 percent of the construction work (Figure 5). Over 95 percent of builders always subcontract some jobs, like HVAC (see attached survey report available under “Additional Resources”).
The second reason labor-subcontractor differences are important is because government statistics tend to focus on labor, to the extent that trends in subcontracting (where much of the work is done by the owners of small firms who are not technically counted as labor) often go largely unreported.
For example, according to the latest Census statistics, there are 1.7 million one-person trade contractor firms that are counted neither as employers or employees by the federal government. This number was almost stable while the number of employees in residential construction contracted drastically during the housing downturn (see forthcoming blog in Eye on Housing).
For each of the 12 trades in the 2014 survey, the share of builders reporting a shortage of subcontractors was higher than the share reporting a shortage of their own labor. A spread of about 5 percentage points was typical, although the difference was as high as 9 percentage points for framing crews and excavators (compare Figure 1 to Figure 6 below).
Effect of the Shortages
NAHB’s survey also asked builders if labor shortages have had any effects on their businesses over the past six months. At least three out of five builders said that the shortages caused them to pay higher wages/subcontractor bids (65 percent), raise home prices (62 percent) and created difficulty in completing projects on time (60 percent). These were the three most common effects, by a wide margin. Next came making some projects unprofitable (36 percent), causing builders to turn down some projects (18 percent) and slowing the rate builders accept incoming orders (13 percent). Only 9 percent said shortages had resulted in lost or cancelled sales, and 7 percent checked “other.” About half the builders who checked “other” wrote in that they were not experiencing a shortage.
Because the most common response involves paying more, the logical follow-up question is, “How much?” To get an idea, the HMI survey included questions on how much labor and subcontractor costs have gone up for the same house over the past 6 months. On average, builders said their direct labor costs had increased by 2.9 percent, while subcontractor costs increased 3.8 percent. Both numbers were up sharply from March of last year (the first time this particular question was included in the survey).
These percentages represent cost increases over a 6-month period. If extended over a full year, the average percentages from the June 2014 would translate to a 5.7 percent increase in labor costs and 7.7 percent increase in subcontractor costs—substantially above the roughly 2 percent year-over-year increase in the overall Consumer Price Index. In short, the data show that builders’ labor and especially subcontractor costs are rising faster than general inflation.
Both the 2013 and 2014 installments of the HMI survey show subcontractor costs increasing noticeably higher than the cost of labor directly employed by home builders. This, combined with the tendency of subcontracting to account for the lion’s share of construction costs, again demonstrates how important it is to account for subcontracting when evaluating the availability of labor in the home building industry.
The most distinctive aspect of NAHB’s recent survey data is not simply that they show an uptick in the share of builders reporting labor and subcontractor shortages, but how widespread these shortages are relative to the historical experience of the early-to-mid 2000s, and to the current state of the home building industry, which is still struggling to attain an annual production rate of a million starts.
The survey data also show that labor and subcontractor markets do not always change at the same rate, especially in recent times when many of the historically normal housing market conditions have been disrupted. The most recent evidence shows that subcontractor shortages are more widespread and have a stronger impact on construction costs than labor directly employed by home builders. This imbalance could develop naturally if employees laid off during the downturn formed their own one-person contracting companies at the time, and are now again starting to become employees of other firms as construction activity starts to picks up and slightly larger contracting businesses begin to re-form.
In general, analyses of the home building industry that don’t include the effects of subcontracting and subcontractors are missing a large part of the picture. At the very least, it’s necessary to look at labor employed by residential specialty trade contractors, as well as directly by builders, when viewing the Employment Situation release from the Bureau of Labor Statistics. Even that, however, misses work done by the self-employed—which could be an important component of labor shortages felt by builders in the current economic environment.
Download the full article
HMI Survey July 2012
HMI Survey June 2014
Previous Special Studies
 For simplicity and to increase the response rate, the questionnaire gave builders ranges of cost changes and asked them to check one. An average was calculated from the responses assuming the midpoint of the ranges, with the values bounded at the 10th and 90th percentile of the distribution to eliminate the open-ended ranges (a technique usually called “Winsorising”).