Social Security Replacement Rates and Other Benefit Measures: An In-Depth Analysis

In this report, CBO examines whether Social Security benefits enable retired workers to meet their basic needs and the extent to which benefits replace preretirement earnings. Focusing on workers with long careers, who generally have higher average earnings than all workers, CBO finds that those benefits enable most of those workers to cover their essential living expenses as measured by the official federal poverty threshold. However, the extent to which benefits replace preretirement earnings varies substantially, depending on the way benefits and earnings are measured.

To show benefits from different perspectives, CBO presents alternative specifications of basic needs measures and of the Social Security replacement rate—that is, the amount of Social Security benefits received in retirement, expressed as a percentage of preretirement earnings. Among CBO’s findings:

• Retired-worker benefits for most long-career workers born in the 1940s exceed the official federal poverty threshold. For workers born in the 1960s and 1980s, even more are projected to have retired-worker benefits above that threshold.
• Replacement rates that compare benefits with earnings just before retirement show that, across cohorts, benefits replace about two-fifths of substantial late-career earnings, falling short of providing income continuity as workers transition out of the labor force. (Substantial earnings are annual earnings that are at least half of the worker’s average indexed earnings.)
• Replacement rates that are designed to capture overall changes in the standard of living between working years and retirement show that Social Security benefits replace a significantly higher percentage of average earnings over a lifetime, adjusted for changes in prices over time.
• Because the Social Security benefit formula is progressive, meaning that benefits replace a larger share of earnings for lower-income workers, replacement rates for workers in the lowest earnings quintile are about two to three times higher, on average and across cohorts, than replacement rates for workers in the highest quintile.
• If future benefits are limited to the annual revenues credited to Social Security once the program’s combined trust funds are exhausted, which is projected to occur in 2031—that is, payable benefits—the fraction of workers with initial benefits below the poverty threshold is projected to increase slightly between the 1940s cohort and the 1960s cohort, and then to increase substantially more for the 1980s cohort. Replacement rates based on payable benefits would be significantly lower than the replacement rates based on scheduled, or full, benefits.

CBO also compares individual measures of benefits with household-based measures and before-tax measures with after-tax measures. In addition, the agency compares replacement rates that are based on a single year of benefits with rates that are based on the receipt of benefits over multiple years. According to CBO’s projections:

• Household benefits, counting those available to a retired worker and his or her spouse, fall below the official poverty threshold for only a very small percentage of long-career workers.
• Replacement rates that are based on benefits and earnings shared within a household are similar to replacement rates based on individual benefits and earnings.
• Accounting for payroll taxes and federal income taxes results in after-tax replacement rates that are several percentage points higher than before-tax replacement rates.
• Replacement rates that are based on average benefits received throughout the course of retirement are similar to replacement rates that are based on a single (initial) year of benefits. However, based on benefits and earnings shared within a household, married workers experience a sharp reduction in replacement rates after the death of a spouse.

The analysis in this report focuses on Social Security benefits and does not examine other sources of income that retired workers may have. Therefore, it does not provide an assessment of the adequacy of retirement income overall.