"As much as one-fourth of the decline in labor force participation in the early postwar period can be attributed to the growth in pension coverage. "
Since World War II, workers have been retiring at earlier ages. Between 1950 and 1989, the labor force participation rates of men decreased from 46 percent to 17 percent for those aged 65 and over, and from 87 percent to 67 percent for those aged 55 to 64. In the same period, participation rates for women age 65 and over fell slightly from 9.7 to 8.4 percent.
One explanation for this trend is the growth of the Social Security program. Total receipts of the Old-Age and Survivors Insurance trust fund in 1995 were $326.1 billion, up from $18.5 billion in 1950. In 1995, there were 26.7 million retired workers who collected average annual benefits of $8,640. The corresponding figures for 1950 were 1.8 million and $3,328.
Another explanation is the rapid growth of private pension coverage and other entitlements. In recent years, employer-provided pension coverage has leveled off at around 50 million workers, or about 44 percent of the work force. These plans provided about 18 percent of aggregate income of household headed by someone aged 65 and over in 1994, up from 14 percent in 1958.
In New Evidence on Pensions, Social Security, and the Timing of Retirement (NBER Working Paper No. 6534), Andrew Samwick links the demographic, employment, and wealth data on households gathered in the Surveys of Consumer Finances in 1983 and 1986 with information on those individuals' pension plans from a companion Pension Provider Survey. He finds that the most significant economic determinant of the probability of retirement is the accrual of retirement wealth attributable to continued work, not the level of retirement wealth at a given point in time. In other words, a senior worker may delay retirement if the pension plan enables him to accumulate or accrue a bigger pension by staying on for a while.
Changes in pension coverage also have a substantial effect on the probability of retirement. As much as one-fourth of the decline in labor force participation in the early postwar period can be attributed to the growth in pension coverage. Samwick shows that it is private pensions, not Social Security, that primarily determine the change in retirement wealth.
Samwick then analyzes changes in Social Security that are typical of past and proposed legislation and finds that they have modest impacts, reducing labor force participation by about 1 percentage point. Increasing employer-provided pension coverage by about 50 percent generates a much bigger drop in labor force participation: about 5 percentage points for those between ages 50 and 70. That's roughly 27 percent of the actual reduction that occurred during 1955 to 1975 when pension coverage did grow by 50 percent.
Using these numbers, Samwick estimates that if Social Security benefits were cut 20 percent in 2032 -- as present calculations of the Social Security Trustees indicate might be necessary -- the probability of being retired by age 70 would drop about 1 percentage point. The generosity of Social Security benefits does not alter the relevant economic incentives by enough to generate substantially different retirement behavior, Samwick reckons.
A more historic look at the impact of pensions on retirement is The Effect of Old Age Assistance on Retirement (NBER Working Paper No. 6548) by Leora Friedberg. She examines Old Age Assistance (OAA), a means-tested program established at the same time as Social Security, but which dwarfed Social Security until the 1950s. By 1940, 22 percent of the aged population received OAA benefits -- an extremely high rate by the standards of today's welfare programs. OAA was replaced by the current Supplemental Security Income program in 1974.
Using data from the 1940 and 1950 Censuses and historical records on state benefit levels, Friedberg finds that the increase in benefits during the decade had a strong impact on retirement decisions. In that period, the prospering economy encouraged older workers to stay in the labor force. But OAA benefits were sufficiently generous to counter that trend, leading to a decline in the labor force participation of the aged.
-- David R. Francis