Skip to content

9 Oct more graphs

  1. The participation rate for prime age workers was moderately constant during 1994-2006. Not true of 16-19 year olds, not true of those age 60 and above.
  2. So setting the average prior to the Great Recession for each 5-year cohort equal to 100 and starting from the Great Recession makes it easier to see changes. These are not picked up in U, and only some in the measures of "discouraged" workers. It has other defects, but does capture things missed by focusing on unemployment measures. Note it is essential to look at participation within age brackets, because the baby boomer retirement skews participation for the population as a whole.
  3. I can use this approach to use actual / projected population data for each 5-year cohort to create an expected normal level of employment. To keep pace with population growth we have to add jobs each month just to tread water. But boomer retirement means we don't need to add nearly as many as 20 years ago.
  4. Jobs data are noisy. The sample is still large, but there are also seasonal effects. The following plots monthly changes for the CPS and the CES measures of total employment. Over time they tend to be about the same, but they can diverge over short time horizons.
  5. Unemployment remains substantial, even though it is better than in 2010. Much better. Here are graphs of both alternate measures of U (in % of the LF) and total numbers by duration. We still have about 1.7 million people who've been out of work 6 months or more.


  7. Longing for the good old days is in part hearkening back to the era when manufacturing jobs were plentiful, however dirty and dangerous they might have been.
  8. Then there's the story of being take over by government. Not in the numbers, and almost all government is local; the Federal government, those bureaucrats, are a very small slice of the labor force. Now health care, that's another story! But not for this set of graphs.