I’m piggy-backing off of the FRED blog and off of Jeremy’s post with yet more data. Let’s set the stage.

• FRED blog, using BLS data from the Current Population Survey (CPS), shows that the labor force participation rate (LFPR) fell by about 1.4pp for people 55 years and older between 2017 & 2023. CPS data is released quickly, but the sample sizes are not massive. There are 3.4 million people in the 7 years of monthly data (so, a little over 40k people age 55+ per monthly observation).
• Also using CPS data, Jeremy shows that FRED commits the fallacy of composition because there are very different people who are 55 and older. Specifically, he illustrates that the LFPR for people ages 55-64 have experienced about a 1.3pp *higher* LFPR in 2023 vs 2017. The implication is that something is happening to the people older than 64.
• I use annual CPS instead. Why? Because it can be corroborated with the annual American Community Survey (ACS) data for 2017-2023.

Continue reading →