Here's an extra wrinkle to the puzzling decline in the U.S. labor-force participation rate: People aged 65 and older left the job market when everyone else was joining, and they have since re-entered the job market en masse just as adults in their prime working years have started leaving.

Labor-force participation rate, age 65 and older.  Source: Bureau of Labor Statistics.
Labor-force participation rate, age 65 and older. Source: Bureau of Labor Statistics.

It's easier to explain the second shift. Companies replaced defined-benefit pensions with defined-contribution plans in the 1980s. That worked well when stock prices were soaring in the 1990s but lackluster returns and low real interest rates in the 2000s probably forced many would-be retirees to keep trudging away for a paycheck. At the same time, Social Security became relatively less generous in the mid-1990s when the Boskin Commission recommended changing how inflation was calculated. Finally, life expectancy at 65 has been going up for a while, although there are significant differences between people in the top and bottom halves of the income distribution.

The steady postwar decline is harder to understand. Social Security was introduced in the 1930s and Medicare in the 1960s. Both programs help people who are worried about having enough money to retire, so you would expect to see large declines in labor participation among the elderly when they are introduced. Yet that isn't what the data show.

One possible implication is that a stronger economy might not lead to much more employment -- the elderly could all retire as younger people re-enter the workforce. (So far, that hasn't been happening.) It's yet another reason to be skeptical of the latest jobs numbers.

(Matthew C. Klein is a writer for Bloomberg View. Follow him on Twitter.)