New research suggests a slowing of the trend toward later retirement
As a strong proponent of working longer, I have been delighted to see the increase in the labor force participation of men 60-74 in recent years. I, and other researchers, attribute this pattern to a host of factors, including changes in Social Security (lower replacement rates as the Full Retirement Age increases and the maturation of the delayed retirement credit), the shift from defined benefit plans with strong early retirement incentives to 401(k)s, an improvement in the health and education of older workers, less physically demanding jobs, the desire to postpone retirement until the availability of Medicare, and the joint decisionmaking of dual earner couples. With all these forces at play, my assumption was that we would continue to see gains in the labor force activity of older workers as they responded to declines in the retirement income system by remaining in the labor force longer.
A recent study by Gary Burtless of the Brookings Institution has caused me to worry. Burtless explored the extent to which the increased educational attainment of older workers – both absolutely and relative to the attainment of prime-age workers – could explain their greater labor force participation.
The gains in educational attainment among older men have been dramatic. In 1985, only 15 percent of men age 60-74 had been to college; today that fraction has more than doubled, reaching 32 percent. Similarly, in 1985, more than 40 percent of older men had not finished high school; today only 13 percent lack a high school diploma.
Just as important, the gap in education levels between older and prime-age men has largely disappeared. Men in their 60s are now as likely to have completed college as those in their prime. And, in a few years, the difference in the two groups in terms of those lacking a high school diploma will also be eliminated. As the educational gap between older and prime-age workers has narrowed, so too has the wage gap. Today, older men earn about the same as their prime-age counterparts.
In order to figure out how much of the increase in educational attainment explains the increase in labor force participation, Burtless calculates the change in the participation rate that would have occurred if the only factors that changed between 1985 and 2010 were the age distribution of the older population and the percentages of older Americans who were in the different educational groups. Since better educated men are more likely to participate in the labor force than less educated men, the trend toward more schooling should raise participation rates in the 60-74 age group.
The results of this exercise suggest that more than half of the 8.7-percentage-point rise in the labor force participation of older men can be attributed to their greater educational attainment. (Another 1.4 percentage points can be explained by the fact that, with the retirement of the baby boom, the 60-74 age group has become younger, with a greater share of 60 year olds than in the past.)
The fact that education played a major role in the increased employment of older men would not be a source of concern if improvement in educational attainment were expected to continue. But it is not. Mid-career workers today have the same educational levels as older workers, so it appears that we have reached a steady state. If education has been the main driver – as opposed to the incentives in the retirement income system – and it is running out of steam, then the trend toward later retirement would be expected to slow.