Absence of conflict of interest.
Citation
Friedberg, L., & Webb, A. (2009). New evidence on the labor supply effects of the Social Security earnings test. Tax Policy and the Economy, 23(1), 1-36.
Highlights
- The study examined the impact of changes to the Social Security earnings test in 1996 and 2000 on employment outcomes.
- The authors used a nonexperimental design and the data from the Current Population Survey (CPS) and the Health and Retirement Study (HRS) to compare outcomes before and after the changes in earning test rules for those between the ages of 62 and 74 who faced different earnings test thresholds at different ages and years.
- The study found that upcoming anticipated earnings test between ages 62 to 69 was associated with a significantly lower likelihood of having a job last week. A higher present value of upcoming anticipated earnings threshold between ages 62 to 69 was associated with a significantly higher likelihood of having a job last week.
- The quality of causal evidence presented in this study is low because the study is a nonexperimental analysis that did not demonstrate that the groups being compared were similar and did not account for possible differences in the analysis. This means we are not confident that the estimated effects are attributable to the changes in the earnings test decision. Other factors are likely to have contributed.
Intervention Examined
Social Security Earnings Test
Features of the Intervention
The Social Security program’s earning test reduces the benefits of Social Security beneficiaries who have earnings above an earnings test threshold. Social Security beneficiaries with earnings above the threshold receive lower benefits while their earnings remain above the threshold, but they receive higher benefits after their earnings go below the threshold. The 1996 policy change raised the threshold at ages 65 to 69 in stages. It increased from $11,280 in 1995 to $12,500 in 1996 and rose in scheduled annual increments of $1,000 until reaching $15,500 in 1999. The threshold was set to hit $17,000 in 2000, $25,000 in 2001, and $30,000 in 2002. But the policy change in 2000 eliminated the earnings test for people between the full retirement age (65) and 70.
Features of the Study
The authors used a regression model to estimate employment and earnings outcomes of people between the ages of 62 to 74 from 1992 to 2005 using data from the CPS and HRS. The CPS analytic sample included 305,962 people ages 55 to 74 between the years of 1992 and 2005, and the HRS sample included a total of 305,962 people ages 58 to 74 every two years from 1992 to 2004. Outcomes included whether an individual had a job last week, whether an individual had positive earnings this year, and whether the individual worked last year. The regression covariates included variables reflecting whether there was an earnings test and the value of the earnings test threshold by age and year based on current Social Security rules in each year, as well as age and year indicators.
Findings
Employment
- An upcoming anticipated earnings test for people ages 62 to 69 was associated with a significantly lower likelihood of having a job last week.
- A higher present value of upcoming anticipated earnings threshold for people ages 62 to 69 was associated with a significantly higher likelihood of having a job last week.
- Findings on working last year were mixed.
Considerations for Interpreting the Findings
The authors did not control for race and account for pre-intervention trends in employment and earnings, which means the comparability of the treatment and comparison groups before the intervention was not established. These existing differences between the groups—and not the policy changes—could explain the observed differences in outcomes.
Causal Evidence Rating
The quality of causal evidence presented in this study is low because the study is a nonexperimental analysis that did not demonstrate that the groups being compared were similar and did not account for possible differences in the analysis. This means we are not confident that the estimated effects are attributable to the changes in the earnings test decision. Other factors are likely to have contributed.