*Arrows reflect the change from the previous quarter’s data. All data represents the most recent numbers released by the BLS. Jobs lost statistic reflects the number of additional jobs older workers would hold if the pre-pandemic older worker EPOP held for the current 55+ population.
By Owen Davis, Bridget Fisher, Teresa Ghilarducci, and Siavash Radpour
- Retirement Boom: At least 1.7 million more older workers than expected retired due to the pandemic recession.
- Retirement Inequality: At earlier ages, vulnerable older workers retired sooner, while more privileged workers delayed retirement. The share of retired workers among adults aged 55-64 rose 5% for those without a college education but fell 4% for those with a college degree.
- Racial Inequality: Black workers without a college degree experienced the highest increase in the share who are retired before age 65. This rate rose 1.5 percentage points, from 16.4% to 17.9%, between 2019 and 2021.
Since March 2020, the size of the retired population in the U.S. expanded beyond its normal trend by an additional 1.7 million people.1 Older workers faced more health risks and disproportionate job loss during the pandemic.2 Thus many older workers were pushed into unplanned retirement earlier than they would in a “normal” year.
Who retired due to the pandemic? If we group all older adults together, regardless of race or education, the increase in retirement comes predominantly from those 65 and older. The share of adults who are retired rose 1.7 percentage points for those 65 and older, compared with 0.3 percentage points for ages 55 to 64. But grouping all older adults together masks differences between educational levels and racial demographics.
Retirement Increased At All Ages For Those Without A College Degree
For older adults without a college degree, retirement increased for everyone over 55—not just those 65 and older. The retirement rate for non-college adults ages 55-64 rose 0.8 percentage points between 2019 and 2021, a 5% increase in the probability of adults in this group being retired. For non-college workers 65 and older, the retirement rate rose 1.1 percentage points between 2019 and 2021.
Although older adults without college degrees retired in greater numbers, the typical worker in this group was not financially prepared for retirement before the pandemic. Older workers without a college degree had median household retirement savings of only $9,000 in 2019, compared to $167,000 for older working households with a college degree.3 Lack of retirement savings leaves non-college workers facing a higher risk of poverty and near-poverty in retirement, particularly if they are forced into an early, involuntary retirement.
Retirement rose at all ages for workers without a college degree, despite their lack of retirement savings, because workers in this group face worse employment prospects than their college-educated counterparts. They have lower job security, and the jobs available to unemployed non-college workers are more likely to be physically demanding or low-paid.4 At the height of unemployment in April 2020, workers 55 to 64 without a college degree were 67% more likely to experience unemployment than college-educated older workers.5 High relative rates of job loss during the pandemic and poor employment prospects likely forced many older workers without college degrees into early and involuntary retirement.
College-Educated Workers Delayed Retirement at Earlier Ages, Increased Retirement At Older Ages
Retirement patterns for college-educated older adults diverged from those without a college degree. Adults ages 55 to 64 with a college degree were less likely to retire during Covid-19 compared to pre-pandemic, while those over age 65 were more likely to retire. The retirement rate for adults 55 to 64 with a college degree decreased from 16.4% to 15.8%, or a decrease of 4% in the probability of retirement—almost a mirror image of the 5% rise in the probability of retirement for non-college adults 55-64.
Compared to older workers without a college degree, college-educated workers younger than 65 have better employment prospects and faced lower rates of job loss during the pandemic. Still, they may have lost income due to reduced hours or furloughs. This makes college-educated workers 55 to 64 more likely to delay retirement to make up for lost income. Recent polling shows nearly a quarter of workers 55 and older delayed retirement due to Covid-19 or expect to do so, with postponements almost twice as high for those who lost income.6
For college-educated workers 65 and older, the retirement rate increased by a substantial 3.2%, a result of both involuntary retirements driven by job loss and accelerated voluntary retirement. The older the worker, the greater the risk of unplanned retirement spurred by job loss. At the height of pandemic unemployment, college-educated workers 65 and older faced a 34% greater probability of unemployment than those aged 55 to 64.7 The increased retirement rate for educated workers ages 65 and over also reflects workers moving up their retirement dates due to the pandemic—as highlighted in recent news media reports focusing on retirement of professionals and affluent workers.8
Because college-educated workers have more retirement savings than those with less education, they can cope better with unplanned retirement. Yet not all older workers with a college degree had sufficient savings prior to the pandemic, and many will likely face downward mobility in older ages due to unplanned retirement. As a recent analysis in the New York Times by Indeed.com chief economist Jed Kolko documented, retirement during Covid-19 is better explained by job loss than retirement asset values.9
Pandemic Retirement Increased Racial Inequality
Black older workers were more likely to be pushed into retirement than white older workers. The retirement rate rose 1.5 percentage points for Black non-college workers 55-64, compared to 1.3 percentage points for whites. The probability of a Black adult aged 55-64 without a college degree being retired increased 9.2%, compared to an increase of 7.5% for white non-college adults.
Among college-educated workers, however, the effects of the pandemic on retirement by race are more nuanced. For whites, retirement trends for earlier versus later retirements align with the analysis presented above—workers younger than 65 delayed retirement while those over 65 retired faster compared to the pre-pandemic period.
For Black college-educated workers, however, retirement rates decreased at all ages despite experiencing higher unemployment rates than their white counterparts—they delayed retirement, even over age 65. College-educated Black workers 55-64 have better job prospects than non-college Black workers, allowing them to remain in the labor force longer. But they have substantially less wealth than white workers; Black workers with a college education have median retirement savings of $50,000, less than a quarter of white college-educated households’ $206,000.10 This means Black older workers are more likely to need earnings in old age. The racial wealth gap helps explain why white college-educated workers 65 and over retired in higher numbers during the pandemic, while the retirement rate decreased for Black college-educated workers at older ages.
Prohibit Age Discrimination
Without strict anti-discrimination laws and enforcement, older workers cannot compete for jobs. Several studies document the effectiveness of state and federal anti-discrimination laws in combating age discrimination and increasing employment of older workers.11 Yet the Age Discrimination in Employment Act (ADEA) protecting older workers from age discrimination was weakened by a 2009 U.S. Supreme Court Decision. Congress must strengthen the ADEA and ensure that any discrimination motivated by age is illegal.
Extend and Increase Unemployment Benefits
Laid-off older workers experience longer spells of unemployment, contributing to why many give up looking for work and retire involuntarily.12 Increased unemployment benefits—more than the $300 a week set to expire in September 2021—can help older workers avoid dipping into their retirement savings and claiming Social Security early.
Lower Medicare Eligibility Age to 50 & Make Medicare First Payer
Lowering the Medicare age to 50 would ensure older, laid-off workers get the care they need. Moreover, making Medicare first payer—having it cover medical expenses before private insurance—would lower firms’ costs associated with providing health insurance to older workers. Reducing the health insurance cost of hiring older workers would help prevent involuntary retirements while increasing older workers’ health coverage.13
Expand Social Security
Increasing Social Security and instituting a minimum benefit will soften the blow for workers who are forced to retire before they are ready and prevent many from falling into poverty. Congress should expand Social Security benefits by $200 per month for everyone and increase the Special Minimum Benefit up to 125% of poverty levels.14
Create A Federal Older Workers Bureau
An Older Workers Bureau at the U.S. Department of Labor would formulate standards and policies to promote the welfare of older workers, improve their working conditions, and advance their opportunities for profitable employment.
For a Complete List of Policy Priorities...
For a complete 10-point list of policy priorities for the new presidential administration to consider, please see our November 2020 report, “A Policy Agenda for the Biden Administration: Protecting Older Workers & Strengthening Retirement Security.”
The headline unemployment rate (U-3) for workers ages 55 and older was 4.9% in April, up from 4.8% in March and down from a peak of 13.6% in April 2020. ReLab’s U-7 figure includes everyone in headline unemployment, plus marginally attached and discouraged workers, involuntary part-time workers, and the involuntarily retired (those who say they want a job but have not looked for over a year). U-7 decreased from a high of 23.5% in April to 11.4% in March. Both U-6 and U-7 declined as some older workers working part-time during the pandemic recession went back to their full-time schedules, while others completely detached themselves from the labor market after a year of not being able to find work.
The employment-to-population ratio (EPOP) provides a broader view of the labor market than the basic unemployment rate. At times when large numbers of unemployed workers move frequently in and out of the labor force, the EPOP shifts the focus from unemployment and participation rates to the share of the population earning an income. In April EPOP slid for both older workers (55 and older) and mid-career workers (ages 35 to 54). The older worker EPOP was down 7.2% in April from its pre-pandemic level, compared to -6.9% in March.
Labor Force Participation
The labor force participation rate indicates the share of workers who are either employed or looking for a job. In April the older worker participation recovered slightly after continually declining since August 2020. The slight uptick in older workers’ unemployment rate, as noted above, is due in part to older workers returning to the labor force but failing to find jobs (i.e., moving from out of the labor force to unemployed). The older worker participation rate was down 4.8% relative to its pre-pandemic level in April. This is more than double the decline for mid-career workers, whose participation rate was down 1.9% in April relative to the pre-pandemic level for those workers.
To determine the pre-pandemic retirement trend for Figure 1, we first calculate the retirement rate of adults 55 and over from January 2015 through April 2021, with the retirement rate defined as the share of people 55 and over who report their employment status as “retired” in the Current Population Survey (CPS). The trend is calculated using a linear regression of retirement rate on time for January 2015–February 2020. We then calculate the divergence in retirement rates as the difference between the actual retirement rate and the pre-pandemic trend for the entire sample period. We multiply this divergence in retirement rates by monthly Bureau of Labor Statistics 55+ population estimates to arrive at the divergence of age 55+ retirement levels from trend.
To explore how the pandemic changed retirement by education and race (Figures 2 and 3), we calculate the changes from 2019 to 2021 in the retirement rates of different demographic groups. Retirement rates are calculated as the share of each group listing their employment status as “retired,” according to CPS data. The retirement rate for 2019 is calculated as the full-year 2019 average. The 2021 retirement rate is the average of January–April retirement rates.
Calculations of median retirement assets by demographic categories are derived from the Survey of Consumer Finances. Our measure of retirement assets is calculated based on households’ balances in all account-type retirement plans.
1. If the five-year 2015-2019 trend in retirement rates for adults 55 and over had held constant through March 2021, the population of adults 55 and older who are retired would be 1.7 million lower than it is now. An adult is defined as retired if they are not working, not looking for a job, and say they are retired in the Current Population Survey (CPS). See methodology in Technical Appendix.
2. Davis, O., Fisher, B., Ghilarducci, T., and Radpour, S. (2020). “A First in Nearly 50 Years, Older Workers Face Higher Unemployment than Mid-Career Workers.” Status of Older Workers Report Series. New York, NY. Schwartz Center for Economic Policy Analysis at The New School for Social Research.
3. SCEPA’s calculations based on Survey of Consumer Finances (SCF). Older working households are defined as households with at least one worker ages 55 to 65. College educated households have at least one spouse with a college, professional, or graduate degree.
4. Davis, O., Radpour, S., and Ghilarducci, T. (2020) “Chartbook: Retirement Insecurity and Falling Bargaining Power Among Older Workers.” Schwartz Center for Economic Policy Analysis, The New School for Social Research.
5. This statistic reflects the percent difference between age groups in unemployment rates in April 2020, calculated from CPS data. Not seasonally adjusted.
6. Horowitz, J. M., Brown, A., and Minkin, R. (2021). “A Year Into the Pandemic, Long-Term Financial Impact Weighs Heavily on Many Americans.” Pew Research Center. Retrieved from https://www.pewresearch.org/social-trends/2021/03/05/a-year-into-the-pandemic-long-term-financial-impact-weighs-heavily-on-many-americans/
7. SCEPA calculations of CPS data. See note 5.
8. Tanzi, A. and Sasso, M. (2021, April 30). “Affluent Americans Rush to Retire in New ‘Life-Is-Short’ Mindset.” Bloomberg. Retrieved from https://www.bloomberg.com/news/articles/2021-04-30/more-americans-are-considering-retirement-because-of-covid
9. Kolko, Jed. (2021, May 12). “In Reversal, Retirements Increased During the Pandemic.” New York Times. Retrieved from https://www.nytimes.com/2021/05/12/upshot/retirements-increased-pandemic.html
10. SCEPA’s calculations of SCF data. Households’ race is defined based on the race reported by the respondent. See note 3.
11. Neumark, David and Joanne Song. (2013). “Do Stronger Age Discrimination Laws Make Social Security Reforms More Effective?” Journal of Public Economics 108: 1–16. Retrieved from https://www.nber.org/papers/ w17467 13. McLaughlin, Joanne Song. (2019). “Age Discrimination Laws, Physical Challenges, and Work Accommodations for Older Adults.” Generations 43:3: 59–62. Retrieved from https://www.jstor.org/stable/26841733#metadata_info_tab_contents
12. Retirement Equity Lab. (2020). “Older Workers Know They Face An Unfriendly Labor Market.” Status of Older Workers Report Series. New York, NY. Schwartz Center for Economic Policy Analysis at The New School for Social Research.
13. Ghilarducci, T. (2020). The time is now to lower the Medicare age to 50. Forbes. Retrieved from https://www.forbes.com/sites/teresaghilarducci/2020/04/10/now-more-than-ever-is-time-to-lower-medicares-age-to-50
14. Radpour, S., Ghilarducci, T., and Webb, A. (2020). “Expanding Social Security Benefits All Workers.” Schwartz Center for Economic Policy Analysis and Department of Economics, The New School for Social Research, Policy Note Series.
Suggested Citation: Davis, O., Fisher, B., Ghilarducci, T., and Radpour, S. (2021). “The Pandemic Retirement Surge Increased Retirement Inequality.” Status of Older Workers Report Series. New York, NY. Schwartz Center for Economic Policy Analysis at The New School for Social Research.