Comparing Presidents, the Civilian Labor Force

This post is on the Civilian Labor Force level and how it changed over time. I will be using data from the BLS, in particular Table LNS11000000. Because the data is produced monthly, I will be looking at the data for January of each year, as January is the month in which a President takes (and leaves) office.

A few comments before I go on. I recognize that the Civilian Labor Force is not especially affected by actions of the President, but I feel its important to look at this series because it may help explain some of the difference in economic performance between Presidents. For instance, one would assume that real GDP per capita might grow more quickly as the Civilian Labor force grows more quickly, in part because presumably that increases the ratio of potential workers in the economy, and in part because of network effects.

I also hasten to add… as pointed out by reader I Kitov on another thread, there are problems with this series. In particular, “the BLS used postcensal estimates before 2000 and intercensal estimates after 2000.” Now, I’m going to use this data anyway because I want to look at the employment to population ratio in a follow-up post, and the employment rate developed by the BLS is based in part on the same population estimates and I want to be able to look at the two series (civilian labor force and employment to population) together. One does what one can…

OK. So what does the seasonally adjusted civilian labor force look like?

Eyeballing, it looks pretty steady. But is it? Here’s a summary…

The slowest growth rate of the civilian force… under GHW. Second fastest under GW. Third under Ike. In terms of growth rate of growth rate of real GDP per capita, these three came out last, third last, and second last of all administrations beginning in 1952. So far, what we expect. But Carter and Reagan at the top of the Civilian Labor Force growth rates? they were 4 and 3 in terms of real GDP per capita.

I Kitov suggests in comments that perhaps the yearly average makes more sense… a one month figure could be heavily skewed by error. Table 2 below looks at that. (Hence, for Carter, the growth rate is the change from 1976, the year before he took office, to 1980, his last year in office.)

Offhand, results are even less suggestive of a relationship between a change in the number of civilian workers and a change in real GDP per capita.

FWIW, the correlation between the % change in real GDP per capita and the % change in the average size of the civilian labor force is only 0.2. And between % change in the average size of the civilian labor force and next year’s % change in real GDP, its -.11. (i.e., negative 011)

Clearly something is not right. My guess… it’s the folks that are employed, and the ratio of the employed to the total population that matter. I’ll try to post on these topics soon.



My Friday post seemed to be riddled with errors. Fortunately, readers caught those errors, and I made changes. Whatever the problem was Friday, there seems to be a residual of it still here today… I Kitov points out the ranking on the summary table as originally posted was misleading, and that led to changes in the verbage as well. Hopefully I’m now over these mistakes, or I have to spend a bit more time editing. Apologies.