Retiring at 65 may still be the norm, despite Social Security changes

Changes to the U.S.’s official full-retirement age, which is gradually being increased to 67, has not made people any less likely to stop working at 65, according to a study published this month.

Ramping up the retirement age that is associated with full Social Security benefits does make people delay the year at which they begin making those benefit claims, though many apparently stop working sometime before that.

The findings have implications for retirement policy, the authors wrote in the paper, “How Sticky is Retirement Behavior in the U.S.? Responses to Changes in the Full Retirement Age,” published in by the National Bureau of Economic Research.

Specifically, if Congress wants to encourage people to retire later, it needs to find a new way of going about it, the report suggests.

“From a policy standpoint, our results suggest that merely increasing the full retirement age is not likely to efficiently achieve the stated goal of inducing later retirement by Americans,” the authors wrote. “If policy makers want to encourage delayed retirement, they may need to combine the FRA increase with other policy levers.”

Under the 1983 Social Security Act Amendments, people born after 1937 have a full retirement age above 65, with two-month increases in the eligible age per year, until it reaches age 67, for people born in 1960 or later, according to the Social Security Administration. People can still begin taking payments at age 62, though for a person whose full retirement age is 67, the monthly benefit will be reduced by 30%.

Currently, the Social Security program is expected to be solvent until 2037, when its reserves would be exhausted, according to a report from the federal agency.

Lawmakers have considered various strategies to increase the program’s solvency, such as further increasing the full retirement age.

The current COVID-19 crisis could be having at least a temporary effect on Social Security, with a historic volume of workers suddenly jobless.

“While the impact on Social Security in the long term is unclear, policymakers will endeavor to smooth out any dips that may be the result of the COVID-19 downturn,” said Joseph Coughlin, director of the MIT AgeLab, in an email. “What is clear is that the pandemic has accelerated a trend already underway, the need and desire to work in retirement. The impact of COVID-19, working longer will be a financial necessity for many.”

But with major parts of the economy on lockdown, working later is simply not possible for many people, he noted. “Unfortunately, depending on the speed of the economic recovery, many of the jobs older adults may have relied upon to fill in the income gaps in retirement may have vanished, making reliance on Social Security even greater.”

The recent NBER paper was written by Manasi Deshpande at the University of Chicago; Itzik Fadlon of the University of California, San Diego; and Colin Gray, an economist at Wayfair. The authors used data from the Social Security Administration to compare retirement and claiming habits, pre- and post-full-retirement-age change.

Across those cohorts, retirement is more common at 62 and 65 than other ages, the data show.

“We find that while increases in the full retirement age strongly and immediately shift claiming ages, retirement ages exhibit persistent ‘stickiness’ at the old full retirement age of 65,” the authors wrote.

Prior studies have examined the relationship between increases in the full retirement age and changes to benefits claiming and actual retirement ages, but the new report is the first to use a massive data set from the Social Security Administration, according to the paper.

Further, the persistent spike in retirement at age 65 does not appear to be linked to Medicare eligibility, which has remained at 65, the authors found. And it is unlikely that the pattern is due to a delay in people adjusting to the increases in full retirement age, according to the report.

However, employers and geography potentially have a role in actual retirement ages, areas the authors said would benefit from additional research.