An interesting report from the Minneapolis Federal Reserve sheds some light on changes in the workforce by age and gender. It has appeared for some time young men in particular have taken to living in their parents’ basements and playing video games. And in many states you are literally paid to just stay home and not work–you can get free shelter, free food, free health care, free phones, free everything. It is even easier if you claim to be disabled. Long CV-19 is a popular method for that. And if you work, well, you have to work, who wants to do that. And you pay lots of taxes, mostly so other people can not work. (Mn. Fed. Rep.)
Anyway, according to the report, for age brackets under age 55 people are less likely to work and those over that age are more likely to work than they were in the year 2000. I think the simplest explanation is that older people retain some of the work ethic that is necessary to sustain any society. Our younger people are a bunch of snowflakes and slackers, and that isn’t just my journalistic exaggeration. Many of them have neither the inclination to work nor the understanding that work is a primary method for creating meaning in life.
The share of working teenagers has declined substantially. This may be unavailability of jobs as minimum wages rose, or that more live in families with higher incomes. Another explanation is that more people in their late teens and early 20s are full-time students. The share of those working from age 65 to 69 has risen, but this may reflect decisions to delay Social Security claiming to age 70. The report is interesting but I am doubtful it captures the reasons for some of these changes and it is apparent that in the prime working age groups we have a decline in employment which is very bad for the economy and the country.
An important supplement to this report is another very technical paper on the decline in response rates to surveys used in various labor and employment statistics. The Federal Reserve paper uses survey data and so is subject to issues in the changing response rate. People are much less likely to respond to these surveys over the last decade. This is a widespread problem across all research disciplines. Sorting out the differences between those who will respond and those who don’t is almost impossible and introduces potential biases to results. When the number of non-responders is changing rapidly, it introduces the possibility of even more bias. The authors of this paper claim to have methods to correct for the bias, but I think they are inadequate. What is apparent is that the non-response has led to common labor statistics like the unemployment rate and who is looking for a job to be inaccurate. The authors find the effect on the unemployment rate to be nominal but to be significant in the labor force participation rate. As I said, I am dubious that they are accurately assessing the inaccuracy caused by response rate issues. (Response Rate Study)