Abstract
This article explains how the Social Security Retirement Earnings Test works and why it is particularly relevant for lower-income workers and retirees. It presents original evidence on how misunderstood this important but complex feature of the Social Security Old Age retirement rules is among lower-income retirees and discusses strategies to help those populations most affected by it understand it better.
Key Words
Social Security Retirement Earnings Test, RET, work, retirement, benefits, income, financial literacy
The Social Security Retirement Earnings Test (RET) is an important but complex rule of Social Security’s Old Age (SSOA) retirement program. This rule affects people who continue to work after claiming Social Security benefits but prior to their Full Retirement Age (FRA; currently age 67 for those born after 1959). The key mechanism of the RET is that some or all benefits are withheld when earnings exceed the RET limits, resulting in higher monthly benefits upon reaching FRA. But to what extent do people understand this mechanism? Do they think the benefits held back are lost forever? Individuals’ financial literacy regarding RET can impact their incentive to work after claiming benefits prior to FRA.
In this article, we explain how the RET works and the role it plays for monthly benefits and working and claiming behavior, focusing on lower income/pension wealth individuals. We draw on evidence from prior research, as well as new evidence from our ongoing qualitative research on Social Security literacy and retirement decision-making in underserved communities. Lower income/pension wealth individuals are particularly affected by the RET, and their misunderstandings may cause them to restrict their labor supply and miss out on opportunities to earn higher incomes and lifetime benefits. In our concluding section, we discuss efforts to improve its understanding, with an emphasis on informed decision-making and equity.
How RET Works and Why it Is Relevant for Lower Income Workers and Beneficiaries
If a person claims retirement benefits when they’ve reached their Full Retirement Age (FRA), their benefits represent what is known as the Primary Insurance Amount (PIA)—a person’s monthly benefit at FRA based on their 35 highest earnings years. But many people do not wait until FRA to claim retirement benefits; for one reason or another, as described below, many claim benefits prior to reaching their FRA. When that happens, monthly benefits are reduced from the PIA in accordance with the Actuarial Reduction Factor (ARF). The ARF is the ratio of early to full benefits, reflecting the person’s age when claiming relative to their FRA, and adjusts monthly benefits accordingly. For example, if a person claims Social Security benefits at age 62, they receive only 70% of the benefits to which they are entitled at FRA; this is an ARF of 0.70.
The RET imposes limits on the amount these early claimants can earn without having subsequent benefits withheld (Congressional Research Service [CRS], 2023). Between age 62 and January of their FRA year, an individual’s benefits are reduced by $1 for every $2 they earn above $23,400 (the lower RET limit in 2025). Then, from that January until the month they reach their FRA, their benefits are reduced by $1 for every $3 of earnings above $62,160 (upper RET limit in 2025; CRS, 2023).
Any benefits withheld due to the RET increase the person’s ARF when they reach their full retirement age and monthly benefits are recalculated by the Social Security Administration (CRS, 2023). Benefits are unaffected by earnings after reaching the FRA.
In the first year after claiming (“grace year”), the Social Security Administration (SSA) checks earnings monthly to determine whether a person should receive their monthly benefits (CRS, 2023). In subsequent years, SSA uses estimated annual earnings to assess RET withholding. For example, if estimated annual earnings exceed the RET limit by $14,000, annual benefits will be reduced by $7,000, beginning with the first month of the year until the full $7,000 is withheld. As a result, a person who receives a monthly benefit of $1,400 will have the first 7 months’ benefits completely withheld and then receive their benefits from August to December. Any deviation of earnings from the estimate, including loss of a job, can lead to months with no benefits and no earnings.
‘Beneficiaries frequently fail to understand that withheld benefits are not permanently lost but credited back to them as a permanent monthly benefit increase.’
Table 1 provides benefit scenarios for a hypothetical person born on January 1, 1963 (FRA 67), who claims at age 62 (January 1, 2025) and has a PIA of $2,000, so that their monthly benefits at age 62 equal $1,400 (ARF*PIA=0.70*$2,000=$1,400).
Column 1 shows the case of a relatively high earner ($57,000 per year) working full-time for 5 years. Given $1 withholding per $2 of earnings above the RET limit, their annual excess earnings of $33,600 result in $16,800 benefits withheld ($1,400 per month)—they receive $0 benefits in all 5 years. However, the withheld amounts are not lost: When this person turns age 67 (FRA), the ARF increases from 0.7 (70%) at 62 to 1.0 (100%), and they start receiving benefits of $2,000 per month (ARF*PIA). Because no benefits were received, the RET withholding completely reverses the early claiming penalty.
Column 2 shows high earnings for 2 years, followed by 3 years of no earnings, which results in 2 years of complete benefit withholding, followed by 3 years of no benefit withholding. During the latter years, monthly benefits received are at the $1,400 level—reflecting the decision to claim at 62 (ARF of 0.7), but following the recalculation at FRA, monthly benefits are $1,600.
Column 3 shows high earnings from 2027 to 2029 after 2 years of no earnings. This labor force re-entry scenario results in $1,400 monthly benefits for 2 years, followed by 3 years of no benefits until FRA, and then upward-adjusted benefits of $1,733.33 per month as the ARF increases to 0.8667 at recalculation.
Column 4 shows lower earnings ($28,000 per year) in all 5 years, resulting in some withholding ($2,550 per year) and, at FRA, moderately higher monthly benefits. Column 5 shows the no earnings scenario.
Column 5 shows a scenario where no withholding takes place: no earnings in all five years after claiming.
As illustrated in these scenarios, the RET can have dramatic effects on benefits received by early claimants who continue to work or return to work. The RET reductions and subsequent boost to monthly benefits upon reaching the FRA are meant to help beneficiaries by reducing the early claiming penalty. Low-income workers with low-pension wealth are particularly affected, because they are more likely to 1) claim benefits early and thus become subjected to the RET (e.g., Coile et al., 2002; Dushi et al., 2021; Friedberg & Webb, 2009; Sass et al., 2013; Shoven et al., 2018); 2) re-enter the labor force unexpectedly after early retirement; 3) have low earnings that fluctuate around the main RET limit; and 4) lack the financial means to cushion any withholding of their monthly benefit. We will briefly discuss the rationale and evidence related to these claims.
Financial Literacy and Social Security Knowledge Among Lower Income Workers and Beneficiaries
Given the importance of the RET’s incentives for lower-income workers and beneficiaries, it is particularly important for these groups to understand how it works and that it is meant to help them in the long run. However, there is concern that lower-income workers and beneficiaries have inadequate knowledge or understanding of the RET. A review of the literature confirms that these populations lack financial literacy and knowledge of SSOA rules, including the RET.
‘Analysis of the survey and focus group data confirmed that the RET is poorly understood.’
Table 2 summarizes the main conclusions from the literature on knowledge deficits and access barriers faced by the general population and specific subpopulations often characterized by low income and wealth. It visually reinforces the pervasive nature of general financial and SSOA program illiteracy, and the specific misunderstandings regarding the RET.
While there is evidence that beneficiaries generally comprehend that benefits will be withheld if their earnings exceed the RET limits, they frequently fail to understand that withheld benefits are not permanently lost but are credited back to them in the form of a permanent monthly benefit increase upon reaching their FRA. Greenwald et al. (2010) found that only 19% of respondents felt “very knowledgeable” about how working after claiming could affect their benefits. Benitez-Silva and Heiland (2007) used a model-based approach to estimate that 30–40% of individuals are aware of the benefit adjustment rules. Consistent with the findings for financial literacy, individuals with lower levels of education and earnings tend to have a worse understanding of the RET’s rules (Brown et al., 2013; Greenwald et al., 2010). Greenwald et al. (2010) observed that African Americans were more likely than whites to incorrectly answer questions about the tax treatment of Social Security benefits. Brown et al. (2013) noted that even people who are aware of the RET often do not know that benefits are increased after FRA to offset withholdings.
New Data and Analysis
We now turn to original evidence on RET knowledge and its impact on behavior from an analysis of qualitative data of Social Security beneficiaries collected as part of a larger project in Metropolitan Chicago, where African Americans and Latinos each constitute about 30% of the population (U.S. Census, 2024). For the larger project, we had planned to involve 64 study participants of African American and Latino heritage with valid Social Security numbers. Participants would be asked to complete extensive survey questionnaires and to take part in focus group discussions, and a sub-sample of respondents would be invited to participate in follow-up interviews. The sudden termination of the SSA grant supporting this research in early 2025 prevented us from carrying out the project as planned (SSA, 2025). Between February and May 2025, we managed to reach 38% of the sample and organize four focus groups (24 people total). We also conducted additional, extended interviews with four more participants.
Two focus groups for retirees were held at senior centers associated with affordable housing; the third focus group involved participants in a church program on financial education and the fourth focus group engaged independent contractors and was held virtually over Zoom. Three focus group discussions were held in English and one in Spanish and all were audio-recorded and transcribed. The survey questionnaires were distributed in digital or paper formats and were completed by all focus group participants representing workers, small business employers, independent contractors, and retirees of each of the three employment categories. Study participants were compensated with a $50 gift card.
Most respondents were ages 62 to 67, had claimed SSOA benefits early (on their own earnings record), and were low-income/pension wealth (with median annual personal income in the $20,000–$24,999 range). Among those ages 62 and older, about half reported being retired and half were still working. Here, we summarize what we learned from the survey responses, focus group discussions, and in-depth interviews from those ages 62 and older who had claimed retirement benefits.
Most of the participants with working-class backgrounds had retired at their earliest opportunity to collect Social Security benefits, worn out from physically punishing manual labor, including construction, meat processing, food service, and commercial and residential cleaning. Others were suffering from chronic diseases, including diabetes and asthma. Although they cited health as the primary reason for leaving the workforce, they also mentioned wanting to enjoy life or escape from an unpleasant workplace, being forced to retire after having their work hours reduced, or getting laid off. They were aware that by retiring at age 62 and not delaying until their FRA, they would get a reduced lifetime benefit. But most did not know the exact percentage of the reduction.
Analysis of the survey and focus group data confirmed that the RET is poorly understood. Although most participants (9 out of 10) who had already claimed benefits knew that they could work while claiming retirement benefits, they did not understand how the RET works. They appeared to generally know that earnings over a certain amount will be reduced for early claimants (7 out of 10), but almost no one knew the approximate maximum amount of earnings allowed before RET withholding and almost all believed that the money withheld due to earnings above the RET limit is lost for good.
Specifically, only 2 out of 10 survey respondents selected correct ranges when asked, “Approximately what is the maximum amount a person is allowed to earn per year before some of their Social Security benefit is withheld?” Similarly, when asked, “If a person earns more than the allowed limit, some of their monthly benefits may be withheld. Do they ever get that money back?” only one person correctly selected the option “Yes, they get the money back in higher monthly benefits when they reach the Full Retirement Age.” One person selected “Yes, they get the money back in one shot when they reach the Full Retirement Age,” which indicates they understood that benefits are not lost but did not fully understand the RET mechanics of higher monthly benefits after FRA. Out of the remaining eight claimants with valid responses, four selected “No, it is gone forever,” and four answered, “I do not know.”
‘The RET is a prime example of how knowledge disparities exacerbate financial inequality among the most vulnerable populations.’
The focus groups revealed concerning details about misleading information on the RET provided by SSA claims specialists. For example, one respondent, who was trying to understand the implications of working full time after claiming, was told that he would have to pay a fine if he worked and earned too much:
“Social Security will give you all the paperwork, all the projections, and right there, they’ll tell you: if you retire at this age, that’s fine. You can work at a part time job, but you can’t earn more than this much because then you’ll have to pay a fine of so much, so let’s say, you earn $1,000, you’ll pay a fine of $100 for those $1,000.”
Reducing the excess earnings-related benefit withholding and adjustment mechanism of the RET to “a fine” risks giving beneficiaries the false impression that benefits withheld are lost.
The misunderstanding about the RET mechanism played a role in participants’ decision-making about how much to work and how much to earn. This knowledge gap affected middle-class study members as well. For example, one middle-class, white interviewee, who was interviewed outside of a focus group, explained how when she retired and started collecting benefits at age 62, she and her husband calculated the exact dollar amount she could earn to avoid having benefits withheld. She took a new job at a nonprofit agency, where she has been for the past 3 years. Until our conversation in April 2025, she was unaware that she would have gotten back any benefits withheld by Social Security after reaching her FRA in 2026.
Another respondent, age 60 when interviewed, was expecting to retire from her full-time job in two years, and to start receiving benefits. She was planning to continue working part-time at another job and earning just enough to stay below the main RET threshold, under the same misconception that anything she earns above that limit would be lost forever.
Conclusions
This article describes how the Social Security Retirement Earnings Test works and why it is so important for low-income workers and retirees to comprehend. We argue that understanding the RET is crucial for older lower income workers and retirees, as it directly impacts their benefits during their working years and beyond. As we document using statistics from nationally representative surveys and early results from our own qualitative research in Chicago, this situation is made worse by the complex nature of the RET that has many retirees confused about the implications of working after claiming benefits early, resulting in poor work and retirement decisions. As low-income workers and beneficiaries are particularly likely to misperceive the RET as a penalty or lack the liquidity to offset the front-end loading of benefit withholding, the RET risks reducing the lifetime resources of those who claim early and reduce their work effort to stay below the RET limits. On the other hand, the RET may reduce the incidence of poverty at older ages if it causes later claiming and, thereby, a higher permanent rate of monthly benefits among those with enough longevity.
Given the importance of the RET for lower income and beneficiary groups and the challenges they face in understanding and maneuvering it, the RET is a prime example of how knowledge disparities exacerbate financial inequality among the most vulnerable populations. Language barriers, cultural differences, and limited access to information contribute to these informational inequities and resulting knowledge gaps. For example, research suggests the “my Social Security” online platform has low awareness and usage among Hispanics with limited English proficiency (Rabinovich & Perez-Arce, 2023b). If the RET is not abolished, targeted financial education programs are essential to improve retirement literacy among older individuals with lower income and education levels. Better communication of the retirement rules governing SSOA is crucial, particularly for those who work in advisory roles, such as Social Security claim specialists and financial advisers serving minority and immigrant communities, as they play a vital role in helping individuals navigate these complex programs.
To that end, Social Security could expand its offerings to include targeted information materials on the RET for all early claimants, an info campaign modality that has been shown to be effective (Liebman & Luttmer, 2015). In addition, claims specialists and advisors should attempt to harvest the power of simplified terminology. The communication strategy of Social Security rules would be well-advised to prioritize clarity and simplicity over technical accuracy to ensure that intended incentives are understood and acted upon (Perez-Arce et al., 2024). Simple information has been shown to be most effective in improving RET knowledge (Rabinovich & Perez-Arce, 2021), suggesting a “less is more” approach to RET communication.
Many workplaces inform workers nearing retirement of the rules governing private pensions, which can be similarly complex to SSOA. They are underutilized as places of education on Social Security retirement rules (Vitt & Smith, 2022). While not recommended as the main solution to the RET literacy challenge, broader information strategies should include workplace-based outreach on Social Security. This can help address a systemic failure in outreach that disproportionately affects low-income workers who may lack access to other high-quality financial planning resources.
We recommend a multi-pronged approach that utilizes a combination of these communication strategies. Ensuring that lower-income workers and beneficiaries are well-informed about the RET will help them make more strategic decisions about how much to work after claiming benefits, avoid unwanted periods of low earnings or benefit withholding, and maximize their financial security in retirement.
We hope to continue this research by interviewing the remaining sample. Gathering a large enough number of observations would allow for a separate analysis by race/ethnicity, education, and immigrant status, and clarify the specific causes of the confusion around RET rules.
Acknowledgement: We are grateful for discussions and feedback from Tony Webb and Teresa Ghilarducci at various stages of this research and to Jesús Macarena Avila and The Resurrection Project for assistance with the qualitative research.
Frank W. Heiland, PhD, is a professor of Public and International Affairs at CUNY-Baruch College and associate director of the CUNY Institute for Demographic Research in New York City. Joelle Saad-Lessler, PhD, is an associate dean of Undergraduates and an Industry professor of Economics in the School of Business at Stevens Institute of Technology in Hoboken, NJ. Karen Richman, PhD, is the director of the Latino Studies Program at the University of Notre Dame in Indiana.
Photo credit: Shutterstock/Berit Kessler
Table 1: Social Security Earnings Test Scenario Calculations
| Description/Scenario | Working Full Time 2025–2029 | Working Full Time 2025–2026 only | Working Full Time 2027–2029 only | Working Part Time 2025–2029 | Not Working |
| Assumed Annual Benefits (Age 62 rate) & RET Limit | $16,800 & $23,400 | $16,800 & $23,400 | $16,800 & $23,400 | $16,800 & $23,400 | $16,800 & $23,400 |
| Assumed Annual Earnings Scenario | $57,000/year for 5 years | $57,000 for 2 yrs., $0 for 3 yrs. | $0 for 2 yrs., $57,000 for 3 yrs. | $28,500/year for 5 years | $0/year for 5 years |
| Earnings in each year | $57,000 (2025), $57,000 (2026), $57,000 (2027), $57,000 (2028), $57,000 (2029) | $57,000 (2025), $57,000 (2026), $0 (2027), $0 (2028), $0 (2029) | $0 (2025), $0 (2026), $57,000 (2027), $57,000 (2028), $57,000 (2029) | $28,500 (2025), $28,500 (2026), $28,500 (2027), $28,500 (2028), $28,500 (2029) | $0 (2025), $0 (2026), $0 (2027), $0 (2028), $0 (2029) |
| Excess Earnings in each year | $33,600 (2025), $33,600 (2026), $33,600 (2027), $33,600 (2028), $33,600 (2029) | $33,600 (2025), $33,600 (2026), $0 (2027), $0 (2028), $0 (2029) | $0 (2025), $0 (2026), $33,600 (2027), $33,600 (2028), $33,600 (2029) | $5,100 (2025), $5,100 (2026), $5,100 (2027), $5,100 (2028), $5,100 (2029) | $0 (2025), $0 (2026), $0 (2027), $0 (2028), $0 (2029) |
| RET-related Benefit Withholding Calculation (Excess/2) in each year | $16,800 (2025), $16,800 (2026), $16,800 (2027), $16,800 (2028), $16,800 (2029) | $16,800 (2025), $16,800 (2026), $0 (2027), $0 (2028), $0 (2029) | $0 (2025), $0 (2026), $16,800 (2027), $16,800 (2028), $16,800 (2029) | $2,550 (2025), $2,550 (2026), $2,550 (2027), $2,550 (2028), $2,550 (2029) | $0 (2025), $0 (2026), $0 (2027), $0 (2028), $0 (2029) |
| Benefits Withheld in each year | $16,800 (2025), $16,800 (2026), $16,800 (2027), $16,800 (2028), $16,800 (2029) | $16,800 (2025), $16,800 (2026), $0 (2027), $0 (2028), $0 (2029) | $0 (2025), $0 (2026), $16,800 (2027), $16,800 (2028), $16,800 (2029) | $2,550 (2025), $2,550 (2026), $2,550 (2027), $2,550 (2028), $2,550 (2029) | $0 (2025), $0 (2026), $0 (2027), $0 (2028), $0 (2029) |
| Benefits Received in each year | $0 (2025), $0 (2026), $0 (2027), $0 (2028), $0 (2029) | $0 (2025), $0 (2026), $16,800 (2027), $16,800 (2028), $16,800 (2029) | $16,800 (2025), $16,800 (2026), $0 (2027), $0 (2028), $0 (2029) | $14,250 (2025), $14,250 (2026), $14,250 (2027), $14,250 (2028), $14,250 (2029) | $16,800 (2025), $16,800 (2026), $16,800 (2027), $16,800 (2028), $16,800 (2029) |
| Actuarial Reduction Factor, ARF, when claiming at 62 (in 2025–2029, % of PIA) | 70.00% (0.7) | 70.00% (0.7) | 70.00% (0.7) | 70.00% (0.7) | 70.00% (0.7) |
| Months of benefits withheld (2025–2029) for ARF adjust. at FRA | 60 months | 24 months | 36 months | 9 months | 0 months |
| Net reduction months at FRA (Initial 60-benefit months withheld) | 0 months | 36 months | 24 months | 51 months | 60 months |
| Adjusted ARF at FRA (in 2030, % of PIA, due to total RET withholdings) | 100.00% (1.0) | 80.00% (0.8) | 86.67% (0.8667) | 73.75% (0.7375) | 70.00% (0.7) |
| Adjusted Monthly Benefit at FRA (PIA*Adj. ARF) | $2,000.00 | $1,600.00 | $1,733.33 | $1,475.00 | $1,400.00 |
Notes: This hypothetical individual was born on January 1, 1963, claims benefits at exactly age 62 (January 1, 2025), and has a pension wealth (PIA at FRA of 67) of $2,000, so that monthly benefits at age 62 equal $1,400. The 2025 RET annual earnings limit (before the year the person attains FRA) of $23,400 is used and considered constant from 2025–2029. The PIA is assumed to be unaffected by any work occurring after age 62.
Table 2: Summary of Financial Literacy, General SSOA and RET Knowledge Gaps by Demographic Group
| Knowledge Area | General Population | Vulnerable Groups | Notable Findings |
| Overall Financial Literacy | Widespread illiteracy, especially on complex topics (Lusardi & Mitchell, 2011, 2014). | Lower levels, particularly among women, certain racial/ethnic minorities, least educated, and foreign-born individuals (Lusardi & Mitchell, 2011, 2014). | Ill-informed individuals respond to own misperceptions (Chan & Stevens, 2003). Linked to poorer retirement planning, less proactive savings, lower confidence (e.g., Lusardi & Mitchell, 2011, 2014). |
| SSOA Program Rules (General) | Know basic existence of programs; little about future benefits, calculation, or survivor benefits (Alattar et al., 2019; Greenwald et al., 2010; Carman & Hung, 2018; Chard et al., 2023). | Lower levels, particularly among certain racial/ethnic minorities, least educated, and foreign-born individuals (Alattar et al., 2019; Carman & Hung, 2018; Chard et al., 2023; Greenwald et al., 2010;). | Many perceive they know more than they do (Carman & Hung, 2019). Non-Hispanic white respondents answered 53.9% of questions correctly, compared with 48.8% for Asian, Hawaiian, and Pacific Islander; 45.3% for non-Hispanic Black; and 43.3% for Hispanic and Latino (Chard et al., 2023). LEP Hispanics have significant knowledge gaps, especially on benefits and claiming age (Rabinovich & Francisco Perez-Arce, 2023b). |
| Retirement Earnings Test (RET) Mechanics | Often misunderstood; generally know benefits are withheld, but not that they are credited back at FRA (Brown et al., 2013; Greenwald et al., 2010). | More likely to misperceive RET as a tax. Less likely to know of RET’s actuarial adjustment (Brown et al., 2013; Greenwald et al., 2010). | Only 19% of respondents felt “very knowledgeable” about how working after claiming could affect their benefits (Greenwald et al., 2010). Even people who are aware of the RET do not know that benefits are increased after FRA to offset withholdings (Brown et al., 2013). African Americans are more likely than whites to incorrectly answer questions about the tax treatment of Social Security benefits (Greenwald et al., 2010). |
| Information Access & Sources | Varying reliance on formal vs. informal sources (Lusardi & Mitchell, 2011). | Lower internet literacy and education are barriers to online platforms (Rabinovich & Perez-Arce, 2023a). LEP Hispanics report few information sources, limited knowledgeable social networks, and less info from employers/ financial institutions (Rabinovich & Perez-Arce, 2023b). | Simple information can improve understanding, but comprehensive info doesn’t add much (Rabinovich & Perez-Arce, 2019). (Liebman & Luttmer, 2015) conducted RCT using brochure on RET and found it increased employment among recipients ages 55–70. Workplaces provide info on pensions but not Social Security (Vitt & Smith, 2022). Viceisza et al. (2023) designed interventions that cater to specific demographic groups. |
References
Alattar, L., Messel, M., Rogofsky, D., & Sarney, M. A. (2019). The use of longitudinal data on Social Security program knowledge. Social Security Bulletin 79(4), 1–9. https://www.ssa.gov/policy/docs/ssb/v79n4/v79n4p1.html
Benitez-Silva, H., & Heiland, F. (2007). The Social Security Earnings Test and work incentives. The Journal of Policy Analysis and Management, 26(3), 527–555. https://doi.org/10.1002/pam.20264
Brown, J. R., Kapteyn, A., Mattox, T., & Mitchell, O. S. (2013). Framing the Social Security Earnings Test [Working paper 112]. Wharton Pension Research Council. https://repository.upenn.edu/prc_papers/112
Carman, K. G., & Hung, A. A. (2019). Social Security household benefits: Measuring program knowledge. Proceedings of the Annual Conference on Taxation and Minutes of the Annual Meeting of the National Tax Association. https://www.jstor.org/stable/27067392
Chan, S., & Stevens, A. H. (2008). What you don’t know can’t help you: Pension knowledge and retirement decision-making. The Review of Economics and Statistics, 90(2), 253–266. https://www.jstor.org/stable/40043144
Chard, R. E., Messel, M. & Rogofsky, D. (2023). Public knowledge about the Social Security retirement program: Differences by race and ethnicity, Social Security Bulletin, 83(3), 1–10. https://www.ssa.gov/policy/docs/ssb/v83n3/v83n3p1.pdf
Coile, C. C., Diamond, P., Gruber, J., & Jousten, A. (2002). Delays in claiming social security benefits. Journal of Public Economics, 84(3), 357–385.
Congressional Research Service. (2023). Social Security Retirement Earnings Test: How earnings affect benefits [CRS Report R41242]. https://crsreports.congress.gov/product/pdf/R/R41242
Dushi, I., Friedberg, L., & Webb, A. (2021). Is the adjustment of Social Security benefits actuarially fair, and if so, for whom? [Working paper 2021-421]. Michigan Retirement and Disability Research Center. https://dx.doi.org/10.7302/4176
Friedberg, L., & Webb, A. (2009). New evidence on the labor supply effects of the Social Security Earnings Test. In R. A. Moffitt (Ed.), Tax policy and the economy (pp. 1–35). University of Chicago Press. http://www.nber.org/chapters/c10570
Greenwald, M., Kapteyn, A., Mitchell, O. S., & Schneider, L. (2010). What do people know about Social Security? [Working paper no. WR-792-SSA]. Pension Research Council. https://www.rand.org/content/dam/rand/pubs/working_papers/2010/RAND_WR792.pdf
Liebman, J. B., & Luttmer, E. F. P. (2015). Would people behave differently if they better understood Social Security? Evidence from a field experiment. American Economic Journal: Economic Policy, 7(1), 275–299. https://doi.org/10.1257/pol.20120081
Lusardi, A., & Mitchell, O. S. (2011). Financial literacy and planning: Implications for retirement wellbeing [Working paper no. 17078]. National Bureau of Economic Research. http://www.nber.org/papers/w17078
Lusardi, A., & Mitchell, O. S. (2014). The economic importance of financial literacy: Theory and evidence. Journal of Economic Literature, 52(1), 5-44. https://www-jstor-org.remote.baruch.cuny.edu/stable/24433857
Perez-Arce, F., Rabinovich, L., Yoong, J., & Alatta, L. (2024). Three little words? The impact of Social Security terminology on knowledge and claiming intentions. Journal of Pension Economics & Finance, 23, 132–151. https://doi.org/10.1017/S1474747222000269.
Rabinovich, L, & Perez-Arce, F. (2021). Improving understanding of the Retirement Earnings Test. Journal of Pension Economics & Finance, 20(4), 496–503. https://doi.org/10.1017/S1474747219000222
Rabinovich, L., & Perez-Arce, F. (2023a). Mixed-methods study to understand public use of Social Security’s online platform. Social Security Bulletin, 83(4). https://www.ssa.gov/policy/docs/ssb/v83n4/v83n4p1.html
Rabinovich, L., & Perez-Arce, F. (2023b). Understanding the Social Security communication needs of Hispanics with limited English language proficiency [Working paper 2023-476]. University of Michigan Retirement and Disability Research Center. https://mrdrc.isr.umich.edu/publications/papers/pdf/wp476.pdf
Sass, S. A., Sun, W., & Webb, A. 2013. Social Security claiming decision of married men and widow poverty. Economics Letters, 119(1), 20–23.
Shoven, J. B., Slavov, S. N., & Wise, D. A. (2018). Understanding Social Security claiming decisions using survey evidence. Journal of Financial Planning, 31(11), 35–47.
Social Security Administration. (2025, February 21). Social Security slashes cooperative agreements [Press release]. https://www.ssa.gov/news/press/releases/2025/#2025-02-21.
United States Census Bureau. (2024). Quickfacts. U.S. Department of Commerce. https://www.census.gov/quickfacts/fact/table/US/PST045222
Viceisza, A., Calhoun, A., and Lee, G. (2023). Racial and ethnic disparities in retirement outcomes: Impacts of outreach. The Review of Black Political Economy, 52(1), 87–108. https://doi.org/10.1177/00346446231182343
Vitt, L. A., & Smith, B. A. (2022). Where is information about Social Security retirement benefits provided to the public? An initial assessment. Social Security Bulletin, 82(1), 1–9. https://www.ssa.gov/policy/docs/ssb/v82n1/ssb-v82n1.pdf
