Mandatory Spending

Function 650 - Social Security

Make Social Security’s Benefit Structure More Progressive

CBO periodically issues a compendium of policy options (called Options for Reducing the Deficit) covering a broad range of issues, as well as separate reports that include options for changing federal tax and spending policies in particular areas. This option appears in one of those publications. The options are derived from many sources and reflect a range of possibilities. For each option, CBO presents an estimate of its effects on the budget but makes no recommendations. Inclusion or exclusion of any particular option does not imply an endorsement or rejection by CBO.

Billions of Dollars 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2019-
Change in Outlays  
  Use 90/32/5 PIA factors 0 0 * -0.2 -0.3 -0.5 -0.9 -1.4 -1.8 -2.3 -0.5 -7.4
  Use 100/25/5 PIA factors 0 0 -0.3 -0.7 -1.5 -2.5 -4.2 -6.3 -8.8 -11.2 -2.5 -35.5

This option would take effect in January 2020.
PIA = primary insurance amount; * = between -$50 million and zero.


The amount of the Social Security benefit paid to a disabled worker or to a retired worker who claims benefits at the full retirement age is called the primary insurance amount (PIA). The Social Security Administration (SSA) calculates that amount using a formula applied to a worker's average indexed monthly earnings (AIME), a measure of average taxable earnings over that worker's lifetime. The benefit formula is progressive, meaning that the benefit is larger as a share of lifetime earnings for someone with a lower AIME than it is for a person with a higher AIME. To compute the PIA, the SSA separates AIME into three brackets by using two bend points (or dollar threshold amounts). In calendar year 2018, the first bend point is $895, and the second bend point is $5,397. Average indexed earnings in each of the three brackets are multiplied by three corresponding factors to determine the PIA: 90 percent, 32 percent, and 15 percent. (Bend points rise each year with average wages, whereas the factors remain constant.)

For example, a worker with an AIME of $1,000 would have a PIA of $839 because the 90 percent PIA factor would apply to the first $895, and the 32 percent factor would apply to the remaining $105. A worker with an AIME of $6,000 would have a PIA of $2,337 because the 90 percent factor would apply to the first $895, the 32 percent factor would apply to the next $4,502 ($5,397 minus $895), and the 15 percent factor would apply to the remaining $603 ($6,000 minus $5,397). Because the formula is progressive, for an AIME of $1,000, the PIA amounts to 84 percent of the AIME; for $6,000, the PIA amounts to 39 percent of the AIME.


This option would make the Social Security benefit structure more progressive by cutting benefits for people with higher average earnings while either preserving or expanding benefits for people with lower earnings. Starting with people newly eligible in 2020, the first alternative in this option would affect only beneficiaries with an AIME above the second bend point. That approach would reduce the 15 percent PIA factor by 1 percentage point per year until it reached 5 percent in 2029.

The more progressive second alternative in this option would reduce benefits for a larger fraction of beneficiaries with higher lifetime earnings while increasing benefits for people with lower lifetime earnings. The second approach would lower both the 15 percent and 32 percent factors and would increase the 90 percent factor. The factors would change gradually over 10 years until they reached 5 percent, 25 percent, and 100 percent, respectively. (The 15 percent and 90 percent factors would change by 1 percentage point per year, whereas the 32 percent factor would change by 0.7 percentage points per year.)

Effects on the Budget

The first alternative would reduce total federal outlays for Social Security over the 10-year period by about $7 billion, the Congressional Budget Office estimates. That estimate is based on CBO's projections of the share of newly eligible beneficiaries who would be affected by that approach and the average reduction in their benefits. By 2028, based on data provided by the Social Security Administration, CBO estimates that about 2.5 million people, or 13 percent of all newly eligible beneficiaries, would be affected. For people who become eligible in 2028, the average decline in monthly benefits for those affected would amount to 4 percent, or about $150 dollars, relative to amounts under current law.

The second alternative would achieve total federal savings of $36 billion over the 10-year period. CBO estimates that about 45 percent of new beneficiaries would receive benefits that are higher than under current law, while 55 percent of new beneficiaries would receive benefits that are lower. People who become eligible in 2028 and would get increased benefits would, on average, receive 6 percent, or about $70 dollars per month, more than under current law; the average decrease for people whose benefits would be reduced would amount to about 8 percent, or $220 dollars per month.

Annual savings from both alternatives would grow over time as the new benefit structure applied to more beneficiaries. In 2048, the first and second alternatives would reduce Social Security outlays from what would occur under current law by 2 percent and 6 percent, respectively. When measured as a percentage of total economic output, the reduction in Social Security outlays under the two alternatives would be 0.2 percentage points and 0.4 percentage points, as the outlays fell from 6.3 percent of gross domestic product to 6.1 percent and to 5.9 percent, respectively.

To achieve greater budgetary savings, larger reductions in the 15 percent and the 32 percent PIA factors could be implemented. (Conversely, smaller reductions would result in less savings.) In addition, to target benefit reductions more narrowly, one or more additional bend points could be added to the formula.

The overall savings from the alternatives in this option could be higher or lower than shown because the projected distribution of earnings and the resulting benefits are uncertain. For example, if earnings were more equally distributed than CBO has projected, resulting in more people with an AIME above the second bend point, the savings from both approaches would be slightly higher than shown because the reduction in benefits would apply to more people.

Other Effects

An argument in favor of this option is that it would better target Social Security benefits toward people who need them more—protecting or expanding benefits for people with low average earnings while reducing payments to people with higher average earnings. This option would help make the Social Security system more progressive at a time when growing disparities in life expectancy by income level are making the system less progressive. (Beneficiaries with higher income typically live longer and experience larger improvements in their life expectancy than lower-income beneficiaries. As a result, higher-income groups receive benefits for more years, on average, than lower-income beneficiaries.) The second approach in this option would increase progressivity more than the first approach by boosting benefits to lower-income people.

An argument against this option is that it would weaken the Social Security system's link between earnings and benefits. In addition, the second approach would reduce benefits for beneficiaries with an AIME above the 45th percentile, some of whom do not have high lifetime earnings. In particular, CBO projects that in 2028 the second approach would reduce benefits for people with an AIME higher than about $3,100, or approximately $37,000 in annual indexed earnings.