In October, there were 142.4 million employees on the payroll at “establishments” – businesses, non-profits, governments, etc., but not counting gig workers. That was up by 638,000 employees from the prior month, and still down by 10.1 million employees from February (152.5 million), based on surveys of these establishments by the Census Bureau, released by the Bureau of Labor Statisticsthis morning. At the low point in April, there had been 22.2 million fewer employees at these establishments than there had been in February. In other words, these establishments recovered 12.1 million of the 22.2 million lost jobs.
In percentage terms, employment in October at these establishments was still down 6.1% from a year ago. While this is a big improvement from the 13.4% plunge in April, it remains the steepest year-over-year drop in employment prior to the Pandemic since 1945:
I’m going to ignore the official “labor force” metric here because it is based on surveys of households, and if the respondents answer the questions in a specific way, they get surgically removed from the “labor force” though they still exist and would still work. You can see this as the labor force was down by 4 million people in October from February. These 4 million people still exist and would still work, but they answered in a specific way and got erased from the labor force.
But wait… In the three years before the Pandemic, these establishments increased their staff at an average rate of about 177,000 employees per month. It was enough to absorb the growth in the working-age population that was interested in a job (not the official “labor force”).
And that growth in the working age population that would work has most likely continued at a similar rate during the eight months of the Pandemic as it had before the Pandemic. And the labor market will have to absorb them.
So not only does employment have to go back to where it was before the Pandemic, but it also has to make up for the growth in the working age population. This is a moving target. Note in the chart below, the slope of the line of employment at establishments before the Pandemic and where employment will have to go for it to catch up with the long-term trend.
Given the average monthly increase of 177,000 employees over the three pre-Pandemic years, the gain in jobs in October (+638,000) means that employment approached its long-term trend level (let’s call it the “approach speed”) by 461,000 jobs (638,000 minus 177,000):
And this forms a pattern in the recovery that is being duplicated in many other data sets: After the initial plunge came the bounce, and the recovery continues but at a slowing pace and remains still very far from where activity levels had been. This can be seen in the two charts below: first jobs, and second a broad index of activity. Though the charts are very different, the slopes of their lines are very similar.
In October, the employment levels at establishments was still down 6.9% from January. Note the slowing recovery that is by a wide and historic margin below the January level:
And second, the very similarly shaped weekly index of how many people are going to “places of commerce” in the 40 largest metro areas in the US, based on cellphone GPS data. These places of commerce include offices, stores, malls, restaurants, hotels, movie theaters, airports, hospitals, other places of commerce and other points of interest. The index, released by the American Enterprise Institute, compares the number of visits on a weekly basis to the number of visits in the week ended January 15.
The top bold blue lines are Cincinnati and Kansas City (both at 80% of January level). The bottom bold red line is San Francisco (47% of January level). The bold lines in between represent San Antonio (69%), Baltimore (60%), and Seattle (55%). Click on the chart to enlarge it.