Keywords

Introduction

In some regards, the American experience of pensions and extended work is quite unique (Street and Tompkins, 2017). Job creation and lower rates of unemployment post-recession set the US apart from macroeconomic experiences elsewhere. Its older population is relatively smaller and increasing less quickly than most of the other countries considered in this volume. The unusual US health care system often locks workers into jobs for fear of losing employment-based health insurance , since guaranteed health insurance through Medicare is not available until age 65. A larger proportion of older Americans work to later ages than in all but a handful of affluent countries. Finally, increasing the US normal retirement age (NRA) to receive full Social Security OAS retirement benefits was accomplished by amendments to the Social Security Act in 1983, although implementation was not even begun until twenty years later, in 2003. The NRA age increased gradually from 65 to 67 between 2003 and 2025. Not only was the NRA age increased less abruptly, but the time horizon for planning for retirement at older ages was also much longer in the US than in most European countries. There, most countries increased the state pension age twenty or thirty years later than in the US and increased the full retirement age much more quickly, giving European workers much less time to adapt to working longer than those in the US .

Yet similarities are also apparent. The familiar refrain ‘live longer, work longer’ described in Part 1 Chap. 5 permeate debates about older Americans, yet policy discussions often overlook the feasibility or quality of older workers’ potential employment. Emphasized instead are more concrete plans to delay retirement further. Another similarity is that many older American workers have health and disability impairments that make continued employment to older ages difficult or impossible. Similar to the other countries in this volume, older Americans often lack the particular skills or pay expectations that employers demand when recruiting or retaining older workers . Finally, gender gaps in American pay and pensions reflect systematic disadvantages built into labour markets and retirement income systems. Women need higher retirement incomes than men because, on average, they have longer retirements, spend more time living alone, with higher medical costs and are less likely to have a spouse to help care for them than do their male counterparts. US policies fail to adequately accommodate such differing trajectories of women’s and men’s working lifecourses and women’s disproportionate normative burden of responsibility associated with reproductive work and unpaid caregiving , just as in many European countries. This chapter highlights some key factors affecting older people associated with retirement and employment regimes that shape the US extended working life debate.

The US Retirement Income System

The US retirement income system has often been referred to as a ‘three-legged stool’ (Turner and Watanabe 1995) comprised of Social Security (the state pension), employment-based pensions or savings schemes, and individual assets/personal savings . This idea of the three-legged stool is somewhat different from how retirement income is categorized in OECD /EU ‘three pillar’ parlance (OECD 2005) . Means-tested Supplemental Security Income (SSI) provides social assistance for impoverished elders from general revenues, which would be the first pillar . The second pillar would comprise retirement benefits (hereafter Social Security) under the mandatory Social Security Old Age, Survivors, and Disability Insurance (OASDI) earnings-related program. OASDI is by far the largest social transfer program in the US , with outlays exceeding US$950 billion in 2017, funded in 2018 by matching contributions from employees (6.4% of pay) and employers (6.4%) up to a ceiling of US$128,400 earnings. Benefits go to 43.1 million retired workers, 8.6 million disabled workers, 4.1 million widows and widowers, and 2.5 million spouses (NASI 2018). Analagous to the third pillar would be voluntary employer-provided DB and DC plans covering less than half of American workers, and tax incentivised individual schemes such as Individual Retirement Accounts (IRAs) and Roth IRAs. In the spirit of American exceptionalism, the following discussion uses the more typical US ‘three legged stool’ approach rather than the three pillars to discuss the characteristics of the US retirement income system as it bears on debates about extended work.

Public Pensions

The first leg of the US retirement income stool, the contributory Social Security Old Age and Survivors pension (OAS), provides pension benefits for over 85% of American aged 65 and older, with OAS comprising nearly 40% of American retirees’ income. Among older Americans, 28% of married couples and 43% of unmarried individuals rely on Social Security for at least 90% of their income. Social Security benefits are at least 50% of income for 48% of married couples and 71% of unmarried people (NASI 2018). The average monthly benefit for an individual retired beneficiary in 2019 is $1,461 and for a married couple $2,448 (SSA 2019).

Social Security is a pay-as-you-go (PAYG) defined benefit (DB) public pension , financed through payroll taxes levied on wages up to US$132,900 in 2019 (SSA 2019). The earnings ceiling on Social Security contributions makes funding Social Security regressive, but there are some progressive elements in benefit payouts. Social Security’s retirement benefit formula replaces a smaller percentage of higher earners’ pre-retirement earnings (around 25% at maximum taxable earnings) and a higher rate for low-wage earners’ pre-retirement earnings (about 52%) (NASI 2018). Workers earn entitlement to Social Security retirement benefit through at least 40 quarters of contributions (10 years of Social Security-covered employment) and then the actual OAS retirement benefit is calculated based on lifetime earnings, using the highest 35 years.

The Social Security amendments of 1983 gradually increased the NRA from age 65. From the prior NRA of 65, the NRA gradually increased by two months from 2003 for each subsequent birth year cohort after 1938, with a hiatus at NRA age 66 for the 1943–1954 birth cohorts, then increasing again by two months per birth cohort from 1955 to 1960. Under current law, the SS NRA will reach age 67 in 2025 for everyone born later than 1960.

At the NRA of 66, retiring up to 48 months early at age 62 is permitted, but it decreases the monthly benefit permanently by 25%. Retiring 60 months early at 62 when the NRA becomes 67 permanently reduces monthly benefits by 30% (SSA 2019). For beneficiaries who work while receiving OAS retirement benefits before reaching the NRA (currently 66 for workers born up to 1954), wages earned above an annual limit (US$17,640 in 2019) receive Social Security retirement benefits decreases by US$1 for every US$2 earned above the limit. Once beneficiaries reach NRA, OAS benefits earnings limits are removed (SSA 2019), incentivizing work for those aged 66 and older. Finally, postponing receipt of OAS benefits past the NRA yields a monthly actuarially determined increase, amounting to 130% of the full NRA retirement benefit if it is not taken up until age 70 (SSA 2019). The continuing accrual of retirement benefits supports an interpretation that the NRA is not actually 66 or 67, but rather that 70 has become the de facto age of ‘full benefits’ (Munnell 2013). Increasing the NRA for all workers, eliminating the earnings test once the NRA is reached, and adding to OAS retirement benefit values for 3 extra years for those working up to age 70 are a series of Social Security that extended the financial viability of Social Security . The changes can also be construed as extended working life policies , since they were also enacted to encourage Americans to work longer. The consequences of continued employment and changes in Social Security benefit accrual at different ages indicate that Social Security is, arguably, the major extended working life policy in the United States .

Private Sector Pensions

The second leg of American retirement income is occupational pensions and retirement savings plans. In 1983, the same year that Social Security amendments raised the NRA, 61% of workers who had occupational retirement benefits were in DB pension plans and only 12% were in DC plans. By 2016, only 17% had access to DB plans only, while 73% had access only to DC plans (CRR 2018). Unlike the pooled risk of secure defined benefit monthly pensions (and survivor benefits they provided), changes in occupational retirement plans have created the individualised risks of DC savings plans (Russell 2014). Most private sector US employers now offer only defined contribution retirement plans, providing the mechanism for individual savings and often making matching contributions (up to a limit) to individual accounts. However, eventual retirement income depends entirely on the accrued value of individual contributions and investment returns, rather than any pooled risks. Payouts can be lump sums, annuity purchases or the withdrawal of savings . Unlike the guaranteed monthly amount of traditional defined benefit pensions, retirees can, and often do, outlive their defined contribution savings . Many others have no employer facilitated retirement savings at all (Russell 2014; Weller 2016). The shrinking numbers of defined benefit plans still operating are mainly in the public sector; even in the public sector DB plans are typically being phased out and replaced with DC plans for new hires.

The transformation in occupational pensions away from DB pensions to DC savings plans contribute to extended work. Defined benefit retirement plans determine predictable lifetime payments according to a fixed formula based on salary, years of service , and age. DC plans determine the value of individual accounts on the basis of contributions and the rate of investment return. DB plans created incentives for older workers to retire early, just as soon as they had maximized the combination of age, years of service and salary into a maximum pension benefit , a feature that is not part of DC plans. DC retirement plans do permit workers to maximize contributions at older ages because tax treatment becomes more generous once a worker hits age 50, likely encouraging workers to stay on in employment to build greater DC savings .

Individual Retirement Savings

The third leg of US retirement income is individual savings and income from assets and investments. Like DC occupational plans, Individual Retirement Accounts (IRAs) receive preferential tax treatment, both in terms of savings and returns on investment. Once money is invested in an IRA, withdrawing it before age 59.5, except due to hardship conditions, carries a 10% penalty from the Internal Revenue Service . Once individuals withdraw from IRAs, they pay taxes on the proceeds. Given that most Americans say they would have trouble coming up with an immediate $400 for an emergency (Federal Reserve 2019), it is no surprise that only 39% of Americans do have IRAs. Among those who do, savings are modest. Currently, about half of all American workers work in jobs with no occupational retirement plans, and about a quarter of working aged adults have no retirement savings or access to pension plans beyond Social Security (Federal Reserve 2019).

As Social Security benefits are based on a long work history (35 years) and lifetime wages, retirement benefits are permanently undercut by time out of paid work or by many years in low-wage jobs. Although the OAS benefit formula partly adjusts by providing higher rates of return to low-income earners, it does not completely remedy the impact of a lifetime of low wages or gaps in paid work . Even when employed full time , women earned only 80.5% of men’s earnings in 2017 (Fontenot et al. 2018). Earlier pay gaps lead to Social Security benefit gaps, where retired women receive about 76 cents on the dollar when compared to similar retired men (Herd 2009). Women carers’ lost wages and retirement income is substantial. Over a lifetime, it adds up to an average of US$142,693 in lost wages, US$131,351 in lost Social Security benefits, and another estimated US$50,000 in lost retirement income from other sources (MetLife 2011). Lower lifetime pay, time out of the labour market or part time work to accommodate caring mean that women have gaps in pay, in pension contributions and coverage, and in later life incomes.

Women’s systematic retirement income disadvantage and heavier dependence on Social Security is demonstrated in Table 39.1 As Table 39.1 shows, women depend more on Social Security than do men , who receive more income and are more likely to have income in the first place from more sources. Among Americans having any income from each of the sources, women receive less in every category. Total income from all sources shows that older women’s median income is just 55% that of men’s. Keep in mind that for the data in this table that, other than Social Security for both women and men , and assets for men only, most Americans have no income from the other sources—no pension income, no earnings, or no other income.

Table 39.1 Median annual income received from each source among women and men aged 65 and older who received any income from each source
Full size table

The increasing individualization of risk represented in occupational DC retirement plans and individual retirement savings is worrisome. The 2008 economic crisis wreaked havoc on the value of older Americans’ individual retirement resources. The value of defined contribution plan balances shrunk as the stock market dropped, equity in housing plummeted, and many had to tap into retirement savings in pre-retirement years to make ends meet. Many older Americans dealt with the aftermath of the Great Recession by postponing retirement. Transformations in the second and third legs of the pension stool have contributed to older Americans’ decisions to retire or to try to remain employed, by-products of economic trends rather than purposive policies to extend working lives. However, the sheer size and scope of Social Security may make it a singular influence on extending working lives.

Employment

Despite remaining employed to later ages and increasing rates of older women’s employment in recent decades, employment prospects for older American workers are not uniformly positive. The boundaries between employment and retirement statuses are fluid, featuring bridge employment, second careers, flexible working and unretirement (Wang and Schultz 2010). The aftermath of the ‘Great Recession’ of 2008–09 featured high levels of unemployment, yet simultaneously accelerated a trend towards (Copeland, 2014). Employed older Americans attempted to weather the reversal in their retirement security by hanging on to the jobs they had during a period when good jobs for displaced older workers were very elusive. Once unemployed, older worker s are the age group at highest risk of remaining long-term unemployed (six months or longer with a comprehensive ‘U-6’ unemployment rate (which includes the aforementioned ‘officially’ unemployed, plus discouraged workers, part-time workers who would prefer full time and workers underemployed in jobs that do not fully use their skills and qualifications) (BLS 2016). Long-term unemployment increases with age; older women’s long term rate more than tripled in the aftermath of the Great Recession (Federal Reserve Bank of St. Louis 2015).

Older individuals who want or need jobs often self-identify as retired, either because they can afford to retire despite wanting employment, or to avoid the stigma of being unemployed. When older jobseekers do find work, it is often with fewer hours of work or lower pay, or both (GAO 2012; Lain 2012). Skills–job mismatches make re-employment especially difficult. When older workers are displaced due to factory or businesses closing or downsizing, they often lack the technical skills needed for other available jobs. Workplace stereotypes and ageism also mean that they have fewer training opportunities to enable them to compete (Charness 2013). Whether sufficient jobs, private or public can be created to enable extended working life , how they can be distributed throughout the US , what skills would be needed and what pay could be offered are all missing pieces in the extended working life puzzle. Demand-side issues receive scant attention in extended working life debates; asserting labour shortages and experiencing low levels of unemployment alone does not mean that employers are eager to retain or recruit older workers .

Work–Family Balance, Health, and Anti-age Discrimination

Paid family leave in the US is rare and usually at the employer’s discretion. The Family and Medical Leave Act (FMLA) has provisions for leave due to parenthood or providing care to immediate family members , covering employees meeting minimum standards of job tenure and hours worked annually and who are employed by the government or large private sector organisations (50+ workers). Job-protected FMLA leave can be taken for up to six months, but there is no requirement that workers on leave be paid. Most American workers have no entitlement to paid or unpaid leave at all. All caregivers face short and long-term costs associated with caregiving—ranging from out of pocket expenses, the need to shift to part time work, lower incomes (and thus less surplus to save for retirement or to contribute to Social Security) . While the 35 highest years of pay used in Social Security benefit formula calculations for retirement benefits may give caregiving individuals who are employed some leeway in maximising public pension benefits despite shorter periods of employment, there is no deliberate policy mechanism within Social Security to compensate employed family carers for essential but unpaid work (Ní Léime and Street 2016).

A unique component of the American political economy that encourages older workers to remain employed is the structure of its health-care system, and the linkage of health insurance to employment. Most Americans younger than 65 depend on employers to be eligible for affordable group health insurance , which is a source of ‘job lock’ that keeps many Americans in full-time employment until they become eligible for state-provided Medicare at age 65. The Patient Protection and Affordable Care Act 2010 (aka Obamacare) might have minimized the perverse effects embedded in the structure of US health insurance , but the Trump Administration has made rolling back those gains a top political priority despite spiking medical bills being a major source of vulnerability for older Americans. The irony is that US health policy does not support older workers with health problems or the onset of disabilities to remain in employment, although it often compels them to try to remain employed to have any health insurance coverage at all.

Workplace policies to improve older workers’ employment prospects are piecemeal at best. In 1986, Congress abolished mandatory retirement by amending the Age Discrimination in Employment Act, except for certain public safety professions. The Workforce Investment Act of 1998 established local ‘one-stop centers’ to provide access to employment and training services for displaced workers under several federal and state programs. ‘One-stop centers’ provided services to terminated or laid-off workers (not just older workers) who had exhausted or were ineligible for unemployment benefits, and who were unlikely to return to work in their previous industry or occupation (GAO 2012). The only federal program specifically targeted at older workers is the Senior Community Service Employment Program (SCSEP). SCSEP provides subsidised, community service-based on-the-job training and internships for individuals age 55 and over with incomes less than 125% of federal poverty guidelines, who are unemployed, have poor prospects for re-employment (GAO 2012). It serves approximately 70,000 workers nationwide each year, with an average age of 64 (DOL 2015). SCSEP jobs and training are part-time with low wages–in many ways, quite the opposite of the good jobs needed for ideal extended working life conditions. Even the fragmented ‘one-stop centers’ and SCSEP programs that do exist are unevenly implemented geographically and variously effective across the states.

Even when jobs are relatively abundant, most evidence points to employers preferring to hire younger rather than older workers . While the US was a world leader in enacting anti-age discrimination legislation with the Age Discrimination in Employment Act (ADEA)—its de facto impact has been limited to only isolated cases, despite the widespread perception that age discrimination is pervasive in the American workplace (GAO 2012; AARP 2014; Ghilarducci 2015). Age discrimination can be a barrier to older Americans seeking employment, and can also be a factor in pushing willing older workers out of employment they would prefer to keep. Another potentially helpful policy to help older workers remain in employment is the American with Disabilities Act (ADA) enacted in 1990. If a (older) worker is deemed to have an eligible disability, workplaces of a certain size must offer “reasonable accommodation” to employees who need adaptations to their conditions of work. On their face, both anti-age discrimination and disability accommodation legislation (if effective) would appear to be potentially helpful to older individuals seeking to remain in work. However, neither the ADEA nor the ADA have shown particular effectiveness in helping older workers extend their working lives (Wegman and McGee 2004).

On the demand side of the extended working life equation , employers seem to have few jobs for which the skills of most older workers in are high demand. Even when age is not an overt part of the job recruitment strategy, language in job ads such as ‘fun workplace’ or targeting job ads only to younger adults on social media reflect some of the subtleties of ageism that feature in the modern American workplace. Although unemployment rates in 2019 skirt record lows, whether the strong demand for labour will translate into demand-side conditions favourable for extending work is unknown.

Discussion

The United States has few social policies that are particularly woman- or family-friendly, and a complex health insurance system that still leaves many Americans uncovered. In fact , older workers without employment-based insurance in the immediate pre-retirement years and before Medicare eligibility at age 65 face serious barriers to arranging health insurance . Routine access to high quality, affordable child and eldercare institutions is rare. With few affordable services available, Americans (often older women ) have to figure out for themselves how to provide or arrange care for children and grandchildren , frail elder parents or partners, or disabled family members . Rates of poverty among older Americans are high, compared to most OECD countries, and are especially so for women.

There are certainly some paradoxes when considering the American case for extended working lives. At first glance, conditions seem favourable for potentially good outcomes. The population is younger, the pension age was raised earlier and over a longer period, and the political turmoil in the US has paralyzed policy debates in ways that, so far, have safeguarded the structure of the redistributive Social Security program. The United States was an early adopter of legislation that aimed to protect workers’ rights, including the ADEA and ADA. As many contemporary older Americans have higher levels of education than previous cohorts, prospects for continued employment in rewarding jobs seems feasible for some, especially for those living in job markets that support it. Individuals who are able to delay retirement until age 70 continue acquiring credits towards Social Security OAS benefits, improving their financial security in retirement. For some working women, extended working life could mean that their public pensions will be calculated based on a full 35 years of employment earnings, with fewer missing quarters than in the past (unless benefit formulas change) . Raising the NRA and leaving the calculation formula intact at 35 years will create a larger cohort of women eligible for retirement benefits in their own right as workers, rather than dependent Social Security benefits derived from marital status. The booming economy means that there are more jobs to go around.

Nonetheless, policies that bear on extended working lives tend to be piecemeal, underfunded and unevenly implemented and experienced across the country. The adversarial US political climate gives little reason to expect that the policy compromises needed to support work–life balance or employment practices that could better support extended working lives will materialize. Although extra quarters of Social Security credit for lost earnings for carers who must quit jobs or work part time to do it would address some of the gendered disadvantages experienced in the US , it would not resolve the intransigent gender pay and gender pension gaps . Better support for flexible work arrangements or mechanisms for phased retirement could help older workers stay in employment longer. Putting teeth into anti-age discrimination regulations for employers could also help Americans extend their working lives. However, none of these policy innovations seem likely to appear on the near horizon.

Politicians, think tanks as diverse as the leftish Economic Policy Institute and the right-leaning Cato Institute, academic research institutes and advocacy organisations ranging from the American Association of Retired Persons, to the Institute for Women’s Policy Research, have weighed in on the merits and risks of extending working life or delaying retirement, but not with a single voice. The fragmented and strident nature of current US policymaking and the non-aligned interests of various political actors stymies integrated discussion of what extending working life means for Americans with different life circumstances.

Lifecourse scholars underscore the accumulation of advantages and disadvantages over entire life courses that feed into outcomes in later life, whether due to gender, lower levels of education, or compromised health status . Future directions for research and policy include reformulating the idea of extended working life , fully comprehending the roles of age, gender and other social statuses intersecting with employment structures, and accommodating the unequal life chances that structure individual lifecourses through carefully refined policies. In the increasingly bifurcated world of employment, the ‘job-rich’ older worker may ‘extend work’ in the later phase of their professional, satisfying, highly-compensated careers and feel empowered while being enriched by working to later ages. However, for ‘job-poor’ workers with a lifetime of low wages and unfulfilling and often precarious or insecure employment, the demand to extend work ignores their persistent disadvantages and barriers to working longer (see Chap. 5 in Part I). Policy innovations could adjust Social Security-credited earnings by adequately taking time out of employment for motherhood and caring into account, compel employers to sustain ‘age-friendly’ workplaces and employment practices, and take into account the kinds of work that actually compel workers in some occupations to retire earlier. Proponents of extending working lives must confront the challenges of employment capacity and adequately remunerative employment given other bifurcations that occur for older workers , what Ghilarducci (2015) characterises as the contrasts between ‘elders who must work, and those who can afford not to.’ Gender and compromised health statuses are heavily implicated in the between ‘must work’ and ‘can afford to stop’, and between ‘need the money’ and ‘love to work’.

The neoliberal turn happened in the US political economy earlier than in many other countries. To the extent that extended work policy initiatives exist in the US , they are less about empowering older individuals’ choices to pursue satisfying employment and more about disempowering older workers by having to settle for poor quality employment and pinching pennies in retirement income systems. That requires discouraging retirement for as many as possible and for as long as possible, regardless of life circumstances. Workers in high-quality jobs may well perceive staying in work as a good personal choice. Disadvantaged and vulnerable individuals compelled to be employed even longer to make ends meet would see the situation much differently. A fair assessment of the United States’ extended working life policies would note that, such as they are, they do little to help older workers stay employed longer, but do plenty to compel them to. ‘Live longer, work longer’ might be a more convincing argument if, as Burgess (2015) observed, “increases in life expectancy were spread evenly across the workforce . They are not.”

The US welfare state is under renewed assault by the current Republican administration that has slashed taxes for high earners and launched expensive trade wars. The ballooning budget deficits those actions have created will starve the government of resources and tie the policy hands of future governments for years, maybe decades, to come. Many pundits regard the combination of political tenor, antipathy to redistributive government programs, and business de-regulation as making it nearly inevitable that pension ages and contribution levels to Social Security will have to rise in the future and that employers will have free rein to hire workers of the ages they prefer. The upshot given current circumstances is that older Americans will need to save even more themselves but have fewer chances for retirement income security , and to work even longer if demand for their work supports that. Raising retirement age is relatively easy to enforce, even if politically unpopular for some, so that seems to be the likeliest catalyst to extending working lives. After all, shifting governmental priorities, a thriving economy , and neoliberal enthusiasms have failed to coalesce to create meaningful debate about what policies it would actually take to extend working lives beyond their current limits.