The Bureau of Labor Statistics (BLS) put out its monthly employment situation report earlier today. The primary headline from the report is that nonfarm payrolls increased by 266,000 on a seasonally-adjusted basis. This number has disappointed many people, which is understandable. But to really understand what is going on in the labor market, we need to look at more than the seasonally-adjusted net payroll figures from the establishment survey.

Coverage of the employment situation report generally focuses on the seasonally-adjusted figures from the establishment survey first and foremost. From there, some secondary attention is paid to the seasonally-adjusted figures from the household survey.

The main problem with these figures is that they are reported as net amounts. If 5 million people lose their job while 5.5 million people gain a job, that gets reported as 500,000 new jobs. If, in the next period, 6 million people lose their job, while 6.2 million people gain a job, this is reported as 200,000 new jobs. Saying that the number of jobs created decreased from 500,000 to 200,000 is not inaccurate per se, but can lead people to think that hiring declined by 300,000, when in fact, in this scenario, it increased by 700,000 (6.2 million minus 5.5 million).

Another lesser problem with these kinds of figures is the seasonal adjustment, which has not been terribly reliable in the exceptional circumstances of the pandemic.

So for this post, I look at the non-seasonally-adjusted (NSA) gross labor force flows data, which is released by the BLS alongside the employment situation report.

In the below graph, we see how many people newly found employment in March and April (bar one), how many people went from employment to unemployment in March and April (bar two), and how many people went from employment to not being in the labor force (NILF) in March and April (bar three).

On the employment side, we see that just under 6.5 million people newly found employment in March. In April, that number shot up to 6.9 million people who newly found employment. More jobseekers are finding jobs than before, which is at odds with the idea that people who are out of work have become especially reluctant to come into work due to unemployment benefits or similar.

This increase in new employment was, however, completely overwhelmed by the number of people who went from employed to unemployed and people who went from employed to not in labor force (NILF). The number of people moving into NILF status grew by an especially large amount. Importantly, NILFs are ineligible for unemployment benefits, and so this figure is also at odds with the idea that unemployment benefits are a major driver of recent labor market dynamics.

Without access to the CPS microdata, it is hard to say what is going on with the NILFs with much certainty. Generally, NILFs are people who are disabled, retired, in school, or taking care of children. It is hard to see how disability and school would have ticked up, which leaves taking care of children and retirement as the likely culprits for these labor market exits.

We can produce the same graph as above, but this time for each gender.

Looking at these graphs, we see that, for men and women, the number of people newly moving into employment increased, though far more so for men than women. The number of people moving from employment to unemployment was pretty much the same for both months for both genders. The main thing that stands out is that the number of women moving out of employment and into NILF status increased by a large amount (from 1.8 million to over 2.5 million) while that same number actually decreased for men (from 2 million to 1.8 million).

Insofar as dropping out of work to take care of kids is more typical of women than men, these figures suggest that taking care of kids is the primary reason why NILF status increased last month and the primary reason why net employment growth declined last month.