Wages and the sectoral composition of the labor market

A fact often thrown around these days is that wages for the average worker have stagnated, but a really interesting short paper from the Chicago Fed, What Does the Changing Sectoral Composition of the Economy Mean For Workers?, by business economist Isaac Sorkin, reminds us that looking at a single number isn’t fully informative. The  basic idea here is that  we should take into account the changes in the types of jobs available as well as their pay and nonpay aspects when we look at the labor market.

To demonstrate why this is, imagine an economy with two types of jobs: Job 1 pays $50/hr and Job 2 pays $10/hr. At time t=0, Job 1 is 40% of the labor market and decreasing slowly over time via a random normal process with trend.


If wages remain constant, the average wage in the economy will decrease due solely to compositional effects.


However, even if we allow for wages in both jobs to rise via a random normal process, the average wage in the economy can still decrease thanks to the sectoral shift.

Here’s the path of wages:


And here’s the average wage with the same change in the job share path as before:


So it’s possible to tell a story with across the board increases in wages declining/stagnant average wages.

Another aspect to focus on is the nonpay part of the job: this is things like schedules, working conditions, benefits, etc. If we focused solely on wages, and assumed Job 1 and Job 2 were the same in all other aspects, then the previous example would look unequivocally bad for workers. But imagine that the reason Job 1 pays so much more than Job 2 is that it is very dangerous – the high pay is to offset the low nonpay aspects. Could we still then easily say that the ending state of lower wages and safer jobs is is a worse situation than the starting state of higher wages and dangerous jobs?

This is the difficult question Mr. Sorkin begins to address in this paper. In previous work he has created measures of nonpay job aspects by sector and made an attempt to quantify them. By combining with wage data, he comes up with this graphic:


The diagonal line is where the pay and non-pay aspects cancel out. Jobs above the line are more desirable, and those below are less desirable. So for example, we can see that the sector with the lowest nonpay dimension, mining, is more than compensated by the highest pay dimension. With this information, we can start to see how simply looking at wages is not telling the whole story as the economy evolves.


The second graphic shows how this has played out with the sectoral shifts in the labor market over the last 25 years. As the sectoral composition of the economy has changed, shifting away from manufacturing and toward more service related jobs (though not as much as we might have expected), pay has clearly decreased but nonpay value has increased. Overall though, the net trend is still negative.

It’s interesting to note that the trend in nonpay seems to be jumping during recessions and remaining pretty stable otherwise, whereas pay has been on a steady decline. My hunch is that the last two recoveries have been led mostly by those less desirable jobs, rather than a broad based increase in all jobs back to their previous level.

Now obviously quantifying nonpay aspects of work is difficult, and Mr. Sorkin doesn’t pretend to have a full grasp on the issue, but as a starting point this is a very interesting story and I hope to see more research along these lines in the future.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s