Fact or Fiction: Is Social Security Regressive?

Is Social Security a distributionally regressive system overall? Some people may have this conception due to a few features of the program: higher-income people retire later and therefore qualify for larger annual benefits than if they retired earlier, they live longer and so collect more Social Security checks, and some of their income is exempt from the payroll tax, which is capped at $113,700. When considering only this evidence, there might appear to be some credibility to the claim that Social Security is a regressive program. However, a report from Third Way by Jim Kessler and David Brown dispels this notion, demonstrating through the Brady Bunch that the return on investment in the system is inversely correlated with income. 

Third Way's report tackles the debate by telling the story of three siblings as they work through four different scenarios to figure out whether or not Social Security is an inherently regressive system. To determine this, they approach it from a pure investment standpoint by calculating and comparing the real return on investment each individual would require if they invested their Social Security contributions in the market as opposed to receiving the benefits through the system. For comparison purposes, the characters in the report hail from three different income classes: Marcia earns the taxable maximum of $113,700, Greg makes half the tax max, while Peter makes one-fourth the tax max.

The authors evaluate four scenarios:

  • Scenario one takes into account only the contributions employees make to Social Security, which are equal to the 6.2 percent payroll tax all employees pay.
  • Scenario two adds the employer portion of the tax into the equation, which many economists believe is ultimately borne by employees. When incorporating this assumption, the rate of return for all three siblings decreases.
  • Scenario three involves the employer contribution and different retirement ages. Because lower-income jobs are often more physically demanding, it is expected that they will often have to retire early. Conversely, those in upper-income groups have more access to healthcare and less physically strenuous jobs, which result in a longer work history. Thus, it is assumed in this scenario that Peter retires at age 62, Marcia at 70 and Greg retires at the same age as the other scenarios at 67. 
  • Scenario four includes all of the above in addition to different life expectancies where Marcia's is above average and Peter's is below average.

The results of Third Way's work show that the less a worker earns over the course of their lifetime, the better a return they can expect from Social Security. Even when differences in retirement age and life expectancy are brought into the debate, the progressive nature of the program is still evident in the real return on investment each income group received. There are many reasons for this ranging from the progressive nature of the benefit formula to the fact that wealthier people have to pay taxes on some of their benefits.

Real Return on Investment for Different Earners
  Low Earner (Peter) Middle Earner (Greg) High Earner (Marcia)
Scenario 1
(Employee Contributions Only)
5.2% 4.5% 3.3%
Scenario 2
(Employer and Employee)
2.9% 2.2% 0.8%
Scenario 3
(Scenario 2 Plus Retirement Ages)
3.0% 2.2% 0.6%
Scenario 4
(Scenario 3 Plus Life Expectancies)
2.5% 2.2% 1.1%

Source: Third Way

Nonetheless, just because Social Security is a progressive system doesn't mean it has the distributionally proper outcome -- this in itself is a question of values. Some plans to reform the program, such as those submitted by Simpson-Bowles and Domenici-Rivlin, would in fact make the program more progressive than it is today by making the benefit formula more progressive and raising the payroll tax cap. There are countless ways to reform the system that either keep or increase its current level of progressivity while also ensuring its long-term sustainability.