Learn which day of the week to post your jobs

This simple (and free) hack can help you improve your applicants by up to 47%.


Understanding Construction Labor Shortages with Data


If you'd like a copy of the data set explained here, juse send us a note at service@roaddogjobs.com and we'd be happy to send it to you. Some of the references in this construction labor article require the spreadsheet.

Make better decisions than the competition by understanding the data.

The most common explanations of construction labor shortages is anecdotal. People talk about what they think might be happening.

There is a better way and understanding this way will put your ahead of the competition in understanding the labor market.

This report will use real construction employment data to develop a metric called Total Available Construction Labor or TACL.


Introduction to Total Available Construction Labor (TACL).

TACL is a data-driven metric that predicts construction labor shortages and trends the direction of the labor market.

Below is the chart that shows the state of the construction labor force. This chart compares the available construction employees (i.e., the unemployed) against the open jobs in construction. In other words, this shows how many people in the construction workforce are available to fill the open jobs.

Anytime the blue line is above zero, it means there are enough unemployed workers to fill the open jobs.  

When the blue line goes below zero (as indicated by the red dots), it means there are not enough workers in the industry to fill the open positions. In other words, during these times there is a data-proven labor shortage!

TACL is not a perfect metric but as a macro measure of the labor market it’s effective. We have found it to be accurate in identifying when the difficulties of finding labor are at their peak.

We should also note that this look at the labor market is only scratching the surface of what can be done with employment data analytics. We would love to work with others who want to push this analysis further.


Reviewing Total Available Construction Labor from January 2010 – April 2021.

You can see from the chart that in July 2018 and August of 2018 and then again from April 2019 through October 2019 there were not enough people available to fill the open construction positions. For most people hiring the skilled trades during those times, you felt that shortage.

Looking back at the data, though, we could have seen this coming.

There was a clear trend that going back to 2013 and we could have made good predictions on when the labor shortage would hit.

Understanding these trends puts you in a better position to assess labor risk than the competition. You will know when you can bid aggressively to win more work and you will be out in front of coming challenges.


How to put together your own data set for Total Available Construction Labor.

Let’s go through step by step how to put this data together. We will also keep this data updated and available monthly on our website. It will be there for you to download and manipulate as you see fit.


An overview of construction employment data

The Bureau of Labor Statistics puts out labor data sets once a month.

For this task we’ll need to work with the BLS Industry at a Glance data set for construction. Go to this page and scroll down to this section for the summary of the employment data.

Here you can see the three data sets that you need to develop the model.

  1. Total Employment. This is the total construction employment shown as the number of people (in thousands). For the example above, in May 2021 the number of people employed in construction was 7,423,000. This data is seasonally adjusted (a statistical technique) to reduce the impact of predictable seasonal patterns during the year.
  2. Unemployment rate. This is the percentage of the total labor force that is not currently employed.
  3. Job Openings. These are the numbers (in thousands) of job openings in the construction industry.

Step 1. Download Employment Data

The following steps will be the same for each of the three (3) data sets.

To get employment data click on the green chart icon in the “Back Data” column.

This will bring up the employment data detail page.

Here you will be able to select the years you want to look at, and the site will generate the data. You can take this chart all the way back to 1939!

In this chart you will see that construction employment has risen steadily from 2011 through early 2020 before COVID-19 impacted the industry. It is now on a recovery track.

To download this data, scroll down the page to this section where you will see the download link.

Download this data and save it. You will need it later.

Step 2. Download Unemployment Rate Data.

To do this step we will go back to the BLS Industry at a Glance data set for construction. Scroll down to the same table as construction employment data and click the green graph link to open the unemployment rate detail page.

This will bring up the construction unemployment rate detail page. You will be able to select dates and download data here. This data only goes back to the year 2000. For this exercise, we will need a complete set of data (employment numbers, unemployment rate, open jobs) for each year. Since unemployment rate only goes back to the year 2000, that is the earlier year we can start this data set at.

Scroll down the page and click the green graph icon to download this data. Save this data for later use.

Step 3. Download Job Openings Data.

Same as the steps before, will go back to the BLS Industry at a Glance data set for construction and this time we’ll download the Job Openings data.

Download and save this data.

Step 4. Set the Data Model.

This is where the three data sets come together. We need to set up a spreadsheet that allows us to do some simple calculations.

First thing is make sure the data covers the exact same months. For example, if you want to look at January 2011 through April 2021, make sure all three data sets capture those same months.

Next you’ll need to do some copy/pasting of the data. The data comes in a format where each year on separate rows. To work with the data we need each data set to be on one row or column. I prefer to set up the data in individual columns.

To do that I set up a sheet with the individual months for the data set in column A. 

Next, we’ll get the data into the adjacent columns. We will put employment data in column B, unemployment rate in column C and job openings in column D. I use a copy > paste special > transpose to get the horizontal rows into vertical columns.

Make sure the unemployment rate as a percent. The data gives us a number (for example, instead of 6% it will give us 6). To turn that number into a percentage just set up a column where you divide the given rate by 100.

Step 5. The TACL Calculation.

What we want to do is compare the total unemployed force to the open jobs. The idea being if there are more people available than open jobs, in theory there are enough people to fill the jobs (we know it is not this simple in practice).

We find the unemployed construction force by dividing the total employed people (column B) by 1 – the unemployment rate in column C below.

This gives us the implied number of available workers (column E).

For January 2010, we set up the equation like this:  =B3/(1-C3) which gives us 1,705 (in thousands, so this is actually 1.7 million people) who were unemployed. We compare that against 54 (thousand) job openings, and (uh-oh!) once the open jobs are filled there are still 1.65 million people left unemployed.

This is a great example of when there was not a shortage of people in the construction industry. This was obviously on the heels of the housing crisis and since construction impacts typically lags the rest of the market January 2010 was in the peak of bad times for construction work.

Running those same calculations for all rows you will get the data for difference between the unemployed in construction and open jobs (column F).

Looking at that data over time you will see that as early as 2013 there was a clear trend toward declining labor availability. Then, in 2018 and 2019 we had several months where there were more open construction jobs than unemployed people.

This situation is a data-proven shortage of people.

There are lots of ways to interpret this data and we are always looking for new and better ways to do so.

Conclusion on Total Available Construction Labor (TACL)

The benefit in understanding TACL (Total Available Construction Labor) is that you get a macro view of the construction labor market. This is not a perfect metric, and it is not precise for specific trades or specific geographies. Still, you’ll understand which way the market is trending and will be able to make better decisions as a result.