Why are older people working longer?
In 1983, Social Security changed legislation concerning retirement age, such as increasing the full retirement age (FRA), changing the way benefit formulas were calculated, and increasing the delayed retirement credit, which increased the financial return to working past the FRA. Before these changes, which motivated the delay of retirement and the accompanying benefit collection, the labor force participation rate of men over 65 was at a record low. Since 1983, however, the employment rates of older workers have risen substantially. The employment rate for men ages 65 to 69 increased roughly 10 percentage points. For women ages 65 to 69, the employment rate more than doubled, rising 15 percentage points, and for women ages 60 to 64, it also rose more than 15 percentage points. Why are people working longer nowadays?
Economists have hypothesized different explanations for the increased labor force participation rate of older workers. In “” (National Bureau of Economic Research, November 2018), Courtney Coile, an expert on the economics of retirement, examines how retirement incentives have changed because of changes in Social Security and the shift from defined benefit (DB) pension plans to defined contribution (DC) plans.
In 1983, Congress and the Reagan administration acted to strengthen the financial status of the Social Security program. Some of the changes made were to encourage workers to delay retirement and benefit collection. After 1983, the FRA for workers to receive Social Security benefits increased from 65 to 67 years old. Nonlinear monthly benefit formulas penalized workers who claimed benefits before the FRA. And a delayed retirement credit increasingly rewarded those who stayed past the FRA.
Since 1980, the private sector has noticeably shifted from DB pension plans to DC pension plans. Unrealistic promises that were prohibitively expensive caused the shift to DC pensions. From 1980 to 2014, the share of workers with only DB plans fell from 28 percent to 2 percent, the share with any combination of plans that included a DB plan fell from 39 percent to 13 percent, and those with only DC plans almost quadrupled from 9 percent to 34 percent. Because DC plans do not have the same age-based incentive to retire early that DB plans do and because the loss of generous DB plans has potentially caused a negative wealth effect, workers are increasingly motivated to work past the FRA.
Coile measures incentive behavior through the implicit tax rate on work (ITAX). She calculates the ITAX under given Social Security rules each year from 1980 to 2016, generating a time series of retirement incentives that show only the variation in Social Security rules. She calculates the ITAX for different earners, splitting them among three groups (high, median, and low earners; male and female; and single and married) into 12 worker types: 3 × 2 × 2 = 12. If the ITAX is negative, then Social Security wealth is rising with additional years of work. If it is a positive ITAX, then Social Security wealth is shrinking. Coile controls for age and estimates that, since 1983, changes in Social Security rules have caused a 15-percentage point drop in the ITAX of workers over the age of 65. However, these same changes do not seem to affect workers under the age of 65.
Coile then controlled for private sector pensions (about 45 percent to 50 percent of workers have one) to calculate the value of the DB pension plan. The presence of a DB pension plan in a worker’s portfolio added 22 percentage points to that worker’s ITAX. Thus, having a DB pension plan greatly motivated workers to retire earlier.
Why are older workers remaining in the labor force longer nowadays? According to Coile, when both the shift to DC pension plans and the changes in Social Security are considered, the ITAX has fallen since 1980 to around 20 percentage points for older workers ages 65–69 (inclusive). In other words, these changes to Social Security and pensions incite older workers to remain in the labor force longer. Once again, workers under 65 years old were found to be unaffected.