By: Drystan Phillips and Teresa Ghilarducci
Download the PDF here.
Our research shows that a majority of retirees begin claiming Social Security before their Normal Retirement Age (NRA)—the age at which their benefit is neither reduced for early retirement nor increased for delayed retirement. Low-income workers are more likely to claim early, possibly to supplement their low wages. There is significant variation in how long older workers continue working based on education, a core differentiator of socioeconomic status. If working longer is driven by lifetime factors like education and benefit claiming is driven by current income, increasing the NRA would be an ineffective mechanism for pushing up the age at which people stop working. What’s more, evidence suggests that raising the NRA to age 70 would reduce average lifetime benefits by nearly 20 percent—not much different from the projected cuts in benefits if Congress does not find a solution to increase Social Security’s revenue before 2034. This reduction would fall hardest on low-income people because they rely on Social Security payments for most, if not all of their total income in retirement.1
Our findings suggest that efforts to encourage older workers to work longer will be ineffective. Raising the NRA is unlikely to change retirement behavior for many workers and will increase income insecurity at older ages for workers across the income distribution.
Given these findings, reforms to federal retirement policy should support policies aimed at creating more good quality jobs for older workers while protecting and strengthening the Social Security system for those workers who are unable or unwilling to work at older ages and risk falling into poverty in retirement.
Many older workers claim Social Security benefits early
Working while claiming retirement benefits is common among older workers in the United States. One-fifth of workers start claiming by age 63. Among a cohort of individuals who were working at age 63 in 2010, 20 percent had already claimed Social Security retirement benefits (see Figure 1). By 2012, when this cohort was at age 65, nearly 40 percent of those still working had already claimed. The Normal Retirement Age for this cohort was 66.
Due to the design of Social Security, starting to claim benefits earlier than the full retirement age reduces monthly benefits for the rest of one’s life. For example, an individual with an NRA of 66 with an expected monthly benefit of $1,000 if they start claiming at age 66 would only receive monthly payments of $750 if they started claiming at age 62.
Figure 1: Many Workers Claim Social Security Retirement Early
Source: SCEPA's analysis of Health and Retirement Study 2010-2018 data.
Note: Normal retirement age for this cohort is 66.
Low earners are most likely to claim early
Low-income workers are more likely than other income groups to claim early. We examine workers in 2018 who received Social Security benefits in the previous year at different ages and by their total income level (we define low income as income at or below $37,500; middle income is greater than $37,500 and less than or equal to $70,000; and high income is greater than $70,000). We find that low-income workers at age 63 are more than three times as likely to have received Social Security retirement benefits in the prior year (at age 62–63) as high-income workers at the same age.
We do not presume that early-claiming workers are making a mistake by reducing their life-long Social Security monthly benefit because they claimed before age 70. Our previous research suggests that older workers who are struggling to maintain their standard of living have very limited financial wealth from which to draw on to supplement low wages, leaving them with no choice but to claim early if they want to stabilize their income.2
Figure 2: Low Income Workers Are More Likely to Claim Early
Source: SCEPA's analysis of Health and Retirement Study 2018 data.
Note: Income tertiles are cut at $37,500 and $70,000.
Not all workers can work longer
Working patterns differ by socioeconomic class. To examine patterns between education and working at later ages, we follow a cohort of individuals aged 63 in 2010 who were working in 2008 (see Figure 3). At all ages, individuals without a high school diploma are less likely to be employed than individuals with higher levels of education. Strikingly, we find that individuals with graduate degrees are three times more likely to have a job than an adult without a high school degree; and those with advanced degrees are more than three times more likely to work at age 67.
Individuals with less education are less likely to work at older ages because older adults with low education have higher mortality, are more likely to be institutionalized, and are more likely to develop a work-limiting disability.3 Many individuals in their 50s have unstable and precarious working conditions, poor health, caregiving responsibilities, and face age discrimination or difficult local labor markets.4
Figure 3: Individuals With Low Education Are Less Likely to Work at Later Ages
Source: SCEPA's analysis of Health and Retirement Study 2010-2016 data.
Most Social Security financing proposals which focus on raising the NRA cite increased average life expectancy between the 1980s and 2010s as the prime motivation. However, these gains were not experienced evenly by all groups.5 While high-income individuals saw large increases in their life expectancy over this period, poorer individuals saw almost no gains.6 Life expectancy gaps have also been growing when measured by education disparities.7
Meanwhile, evidence suggests that individuals stop working in line with their differing life expectancy.8 As changing the NRA will not change life expectancy, we expect that most workers will stop working primarily as a function of their health, caregiving responsibilities, and the quality and availability of jobs. What’s more, since 2019 life expectancy in the United States has started declining, suggesting that older workers are less likely to be able to work until later ages.
Complicating things further, many older workers have physically demanding jobs that limit their ability to work longer.9 Thirty-two percent of workers aged 55–64 and 24 percent of workers aged 65 and older have jobs that require significant physical effort most of the time. Workers without a college degree are also much more likely to have a physically demanding job. The correlation between education and level of physical effort required by one’s job helps explain why workers with less education are more likely to stop working earlier. Faced with demanding jobs, many older workers may be physically unable to continue working beyond certain ages.
Policy recommendations
Any reforms to federal retirement policy—while necessary and long overdue—should be built on the foundation of supporting good quality jobs for older workers and protecting and strengthening the Social Security system, not on reducing benefits by increasing the retirement age.
Our findings are relevant for the debate about raising the Social Security Normal Retirement Age (NRA) to age 68 and beyond. Raising the NRA will not change life expectancy or make it more equally distributed, and it is unlikely to reduce the level of early claiming or increase people’s odds of working longer, especially for workers with low income and less education. Instead, raising the Normal Retirement Age will disproportionately disadvantage poor workers by increasing the penalty for early claiming and reducing their lifetime Social Security benefits.
Additional changes are necessary beyond the policy arena, in pension scholarship, the Social Security Administration literature, and beyond. Ending work and starting claiming should not be assumed to be simultaneous events. Pension research that presumes this pairing will assume that workers compare the income value of working another year and the resulting change in their lifetime Social Security payments and will not account for the fact that workers may choose to start claiming benefits while still working primarily to stabilize their current income, even if it reduces their lifetime Social Security benefits. At the aggregate level, pension simulations that assume this concurrent timing are likely to overstate the effect of changing the NRA on the size of the labor force and on Social Security's liabilities. Social Security Administration pamphlets should refrain from descriptions that use retirement and claiming interchangeably as this will often be two separate decisions for workers. Similarly, financial advisors should understand that individuals may need to claim early retirement even while still working in order to provide a stable stream of income.
Create a federal Older Worker’s Bureau
As older workers make up an increasingly large part of the U.S. labor market, it is long past time to create an Older Worker’s Bureau (OWB) within the Department of Labor to hear from older workers and their employers, investigate their needs, coordinate the vast resources of the U.S. government, and modernize age discrimination laws and worker training. An effective OWB will fulfill three functions: identify and analyze issues of concern for older workers; devise innovative policies to address these issues; and engage in outreach and education. By identifying policies that would ensure the availability of age-accommodating jobs with sufficient rewards for older workers, an OWB would increase the number of older workers by encouraging and rewarding work. Better jobs for older workers with sufficient pay would allow individuals to work longer without needing to claim Social Security benefits early to supplement low wages.
Make work more age-friendly
Improving health and safety standards and providing paid sick leave and time off will make the workplace better for all workers—including older workers.10
Increase the special minimum benefit and expand eligibility
The current funding formula for the Social Security special minimum benefit is obsolete. The minimum benefit level should be set to ensure an income floor above the poverty level for an individual. Increasing benefit levels will expand eligibility to a broader population of workers who have low earnings over the course of their careers and whose regular retirement benefits are insufficient to keep them out of poverty.
Appendix
Throughout our analysis, working is defined as currently doing any work for pay and does not include those looking for work. We define receiving retirement benefits as receiving Social Security retirement, spousal, or widow benefits in the year before our interview (this does not include SSI or SSDI). Income levels are computed using total couple-level income in the year before the interview, divided by two if the individual is coupled. Income terciles are defined across each age-specific subsample of individuals who are working. All estimates were produced using the Health and Retirement Study’s cross-sectional analysis weights.
Endnotes
[1] Krepps, Kathleen. (2023). Raising Social Security’s Retirement Age Would Cut Benefits for All New Retirees. Center on Budget and Policy Priorities.
[2] Ghilarducci, Teresa, Papadopoulos, Michael, & Webb, Anthony. (2022). The Illusory Benefit of Working Longer on Retirement Financial Preparedness: Rethinking Advice That Working Longer Increases Retirement Income. Journal of Retirement 10, No. 1 (Summer): 7–27. https://doi.org/10.3905/jor.2022.1.104.
[3] Quinby, Laura D., & Wettstein, Gal. (2021). Are older workers capable of working longer? Journal of Pension Economics & Finance, 1-18.
[4] Berkman, Lisa F., & Truesdale, Beth C. (2023). Working longer and population aging in the US: Why delayed retirement isn’t a practical solution for many. The Journal of the Economics of Ageing, 24, 100438.
[5] Waldron, Hilary (2007). Trends in mortality differentials and life expectancy for male social security-covered workers, by average relative earnings (No. 108). Social Security Administration, Office of Policy, Office of Research, Evaluation, and Statistics.
[6] Waldron, Hilary. (2007). Trends in Mortality Differentials and Life Expectancy for Male Social Security-Covered Workers, by Average Relative Earnings (No. 108). Social Security Administration, Office of Policy, Office of Research, Evaluation, and Statistics.
[7] Case, Anne, & Deaton, Angus. (2021). Life expectancy in adulthood is falling for those without a BA degree, but as educational gaps have widened, racial gaps have narrowed. Proceedings of the National Academy of Sciences, 118(11), e2024777118.
[8] Case, Anne, & Deaton, Angus. (2021). Life expectancy in adulthood is falling for those without a BA degree, but as educational gaps have widened, racial gaps have narrowed. Proceedings of the National Academy of Sciences, 118(11), e2024777118.
[9] Morrissey, Monique, Radpour, Siavash, and Schuster, Barbara. (2022). The Older Workers and Retirement Chartbook. Economic Policy Institute and Schwartz Center for Economic Policy Analysis.
[10] Truesdale, Beth C. (2020). Better jobs, longer working lives: Proposals to improve the low-wage labor market for older workers. The Brookings Institution. Retrieved from https://www.brookings.edu/wpcontent/uploads/2020/11/ES-11.16.20-Truesdale.pdf
Suggested Citation: Phillips, D and Ghilarducci, T. (2023). “Older Workers Claim Social Security While Working, Upending Beliefs About Raising the Retirement Age” Policy Note Series, Schwartz Center for Economic Policy Analysis at The New School for Social Research. New York, NY.
