Actually, Raising the Retirement Age Is Not Regressive

Some commentators have suggested that raising the retirement age would be regressive given that lower income recipients tend to have a lower life expectancy than middle and higher income recipients (and therefore spend fewer years collecting benefits.) But a look at the data from the Social Security's Office of Retirement Policy shows that this claim is not true.

As a technical matter, raising the Normal Retirement Age (NRA) leads to a reduction in benefits at any given age. For example, someone retiring at age 62 might see their benefits reduced to 25% lower than if they retired at 65, rather than 20%. As a result, raising the normal retirement age affects people of all income roughly equally (and indeed is slightly progressive).

Here are the Social Security Administration's estimates of the effect:

^{Source: Social Security Office of Retirement Policy}

So what about the argument that wealthier recipients tend to live longer? Doesn't that mean raising the normal retirement age is a bigger cut in lifetime benefits for lower earners? Actually, no. That fact doesn't make raising the retirement age regressive, it makes the Social Security program regressive as is (this regressivity, however, is offset by the progressive benefit formula). Increases in the retirement age would result in a roughly equal reduction in lifetime benefits for all beneficiaries who first collect through the retirement program.

As we explained above, the actual effects are slightly progressive. The reasons is that raising the retirement age is a benefit cut that exempts those who first collect through the disability system -- who tend to be lower income. The chart below shows the people over 60 who see a benefit cut from raising the retirement age:

[chart:6872]

All of this is not to say that raising the retirement age leads to an optimal distributional outcome. Policymakers might want to make the system more progressive than it is today rather than simply maintaining the current distribution of benefits. However, evaluating distribution requires looking at the whole package. Take for example the Simpson-Bowles plan, which raises the retirement age but also increases the payroll tax cap, slows the growth of initial benefits for higher earners, and establishes a robust minimum benefit. This plan is quite progressive on the whole.

Furthermore, there are economic benefits to raising the retirement ages. According to a CBO analysis, raising the normal retirement age by 3 years would increase the size of the economy by 1 percent by 2035 as a result of people working longer. Raising the early retirement age by 2 years would have a similar effect. And this is before accounting for any potential growth from increased savings or from smaller deficits.

Policymakers should pusue comprehensive Social Security reform which makes that program sustainably solvent in a way that promotes growth and protects the parts of the program we care about most. The retirement age can be an important part of such a package.