As the authors of The 100-Year LifeFootnote ¹ inform us, a child born today in the West has a 50 percent likelihood of living to be a centenarian. This longevity revolution has been described as “the most important phenomenon of our time in the world, more than the bomb, the Pill, or the Internet.”Footnote ² The groundswell of individuals in older-age cohorts – whom I will call the “new old” – could activate a world-historical social movement. To fulfill its potential, however, that movement must resist the temptation to advance age-based civil rights solely within shopworn frameworks established by race or sex. The “new old” should seek instead to fashion an innovative antidiscrimination paradigm that honors distinctive age-related concerns.

In Section 6.1, I describe how legislation and doctrine create the temptation to build “like race” or “like sex” paradigms for age. In Section 6.2, I argue that advocates should resist this temptation for both empirical and strategic reasons. In Section 6.3, I show how unmooring the age discrimination paradigm from traditional civil rights models allows us to apprehend it better by considering the distinctive fears older individuals conjure about our own mortality. In Section 6.4, I consider how such insights might be incorporated into antidiscrimination law and politics.

6.1 Understanding the Analogy

Courts interpreting the Age Discrimination in Employment Act (ADEA) and the Equal Protection Clause have traditionally parsed age discrimination through analogy to race or sex discrimination. When courts or legislatures accept this analogy, they afford age discrimination plaintiffs more relief. When such bodies reject that analogy, they look less kindly on those plaintiffs. This pattern presses advocates of older individuals to cast age as analogous to race and sex.

6.2 Rejecting the Analogy

The problem here is a specific form of a general pathology in antidiscrimination discourse. Two decades ago, Janet Halley wrote an essay on how LGBT rights advocates felt impelled to make analogies between sexual orientation, on the one hand, and race and gender, on the other.Footnote ⁷⁴ “Particularly when they argue to judges, who are formally if not actually constrained by precedent, and even when they make more general political appeals, advocates are opportunists looking for a simile,” she observed. She mimed the advocate addressing the Court: “Your honor, this is just like a race discrimination case; this is just like a sex discrimination case.”Footnote ⁷⁵

Halley advanced an urbane critique of this strategy. She entertained no illusions that this strategy would disappear. She acknowledged that “analogies are probably an inescapable mode of human inquiry and are certainly so deeply ingrained in the logics of American adjudication that any proposal to do without them altogether would be boldly utopian.”Footnote ⁷⁶ Nevertheless, she alerted advocates for LGBT rights to the costs of conformity to the regnant paradigms of civil rights – costs that centrally included the misrepresentation of the minority itself. She argued that the analogy faltered on empirical groundsFootnote ⁷⁷ and, less intuitively, on strategic grounds.Footnote ⁷⁸

Halley’s analysis illuminates age discrimination jurisprudence. Courts insist that age is different because it is (1) universal and (2) creates a real difference. Because of the incentives described earlier, advocates for older individuals seek to elide these differences. Their arguments, however, can be challenged on both empirical and strategic grounds.

6.3 Transcending the Analogy

I have argued that distinguishing age from other classifications opens space for alternative progressive interpretations of age discrimination. Thus far, I have relied on doctrinal examples. I now cut deeper by turning to a distinctive feature of age that has not yet surfaced directly in the age discrimination jurisprudence. Social scientists have analyzed age discrimination under a “terror management theory” that suggests that age discrimination arises in part out of a fear of our own mortality – a fear without direct analogues in other realms of antidiscrimination law.

6.4 Doctrinal and Political Reforms

My primary goal here is not to advance legislative or doctrinal reform. Yet because it is easier to dispose than propose, I make some reconstructive gestures as a token of good faith. To respect the difference that age makes is to transform the law and culture of aging.

6.5 Conclusion

Ageism is a problem that cries out for an urgent solution in a period of radically lengthened lifespans. It is all very well to say that individuals should begin their second careers at age sixty-five, but this will be impossible if employers will not hire them at that age. As the problem grows ever more urgent, and the constituency affected becomes ever more numerous, we seem due for a tipping point. I predict that the movement for older individuals will soon have a major mobilizing “moment,” much as other groups did with Black Lives Matter or #MeToo.

I have attempted to suggest some ways for older individuals to use their power when the revolution comes. The traditional way of recognizing a new social movement is to bring discrimination against the group within the pale of the traditional antidiscrimination canon. I argue that this is a mistake (indeed a particular iteration of a more general mistake). If age is truly to be protected, we should not force the “new old” into old paradigms. The opposite of “old” is not only “young”; it is also “new.”

We live in an age of widening inequality.Footnote ¹ In the mid twentieth century, income and wealth inequality was declining, leading to an era of comparative economic equality and a large middle class. But since the early 1980s, income and wealth inequality has been on the rise. The 100-Year Life describes how increasing life expectancy could transform the course of individual lives in detail – but how does this trend intersect with our age of inequality? And what does it mean for law and policymaking more generally?

The dark side of rising life expectancy is that it is unequally shared. In Section 7.1, I briefly discuss how age and inequality are connected – that is, that not everyone shares in rising life expectancy equally and that those who are living longer share many common attributes. Section 7.2 considers the possible impact of the skew in rising life expectancy on representative government. A wide literature in political science shows how inequality skews politics to favor the wealthy; inequality by age will likely do the same, but with some additional twists. The fundamental problem is that power is held unequally in most societies. Addressing this issue has been one of the central questions of political theory, constitutional law, and institutional design – particularly for those interested in representative democracy. Section 7.3, on age and institutional design, outlines some ways in which political structures might be crafted to account for age-related differences.Footnote ²

7.1 Inequality and Rising Life Expectancy

Rising life expectancy often comes with optimism and celebrations of progress.Footnote ³ But rising life expectancy isn’t experienced by everyone equally in society. In fact, the data show a wide range of disparities in life expectancy. Women live longer than men on average, and that disparity is projected to continue into the future.Footnote ⁴ White people live longer than Black people, on average, but Hispanic and Latino Americans live longer than either.Footnote ⁵

Life expectancy is also tied to a range of social factors like wealth, education, and geography. Perhaps unsurprisingly, the wealthy live longer than the poor.Footnote ⁶ But the difference is striking. The top 1 percent of men in America by income have a life expectancy that is fifteen years longer than the bottom 1 percent.Footnote ⁷ Those with college degrees live longer than those without,Footnote ⁸ and studies show that life expectancy has been declining for those without – even as it increases among those with – college degrees.Footnote ⁹ Where Americans live is also tied to life expectancy. People living in metropolitan areas and on the coasts live longer than those in the middle of the country,Footnote ¹⁰ and people living outside metro areas in the south or in Appalachia have lower life expectancies than much of the rest of the country.Footnote ¹¹ Deaths often come from suicide and drug overdoses,Footnote ¹² and life expectancy is tied to the prevalence of smoking.Footnote ¹³ In some cases, geography matters even more than poverty. The life expectancy of the poorest 5 percent in Detroit, for example, is more than ten years lower than their counterparts in New York.Footnote ¹⁴

It is worth noting that economic inequality persists even within age cohorts as well.Footnote ¹⁵ The Black–white wealth gap, for example, is significant within older age cohorts.Footnote ¹⁶ White Americans in their thirties, on average, have three times as much wealth as Blacks (about $147,000). By their sixties, the gap has expanded to seven times as much (or about $1.1 million).Footnote ¹⁷ This is partly a function of the cumulative effects of compounding economic inequality over a lifetime.Footnote ¹⁸

There may also be significant period effects in inequality by age.Footnote ¹⁹ That is, the period of time in which one grew up might shape an entire cohort’s economic opportunities. The generation that grew up in the comparative equality of the mid to late twentieth century will have different economic outcomes than a generation that grew up in the more recent age of wide inequality. Studies show, for example, that the ratio of wealth for older and younger cohorts has changed over time. A Deloitte analysis notes that in 1992 the ratio of median family net worth for households older than seventy-five compared to those less than thirty-five years was 9.4. That ratio had jumped to 24.1 by 2016.Footnote ²⁰ The median family net worth for those under thirty-five actually declined during this period.Footnote ²¹ Pew’s research has led to similar data. In 2009, the median net worth of households headed by adults over sixty-five was 42 percent higher than their counterparts in 1984. But households headed by adults under thirty-five saw a 68 percent decrease in their median net worth compared to their counterparts a generation earlier.Footnote ²² Over the past few decades, incomes have increased at a greater rate for older people; poverty is declining for those above sixty-five, but rising for those under thirty-five; and younger people suffered more from the Great Recession than older people.Footnote ²³

In sum, what the data show is that rising life expectancy is not experienced equally – and will not be experienced equally in the future. Rather, those who live longer – successive cohorts of older Americans – are more likely to be wealthier, white or Hispanic, better educated, and live in metro areas, particularly on the coasts. Within the cohorts of older Americans, inequalities will persist based on race, geography, period effects, and other factors as well.

7.2 Gerontocracy and Its Discontents

Systematic, persistent disparities in age are a problem for representative democracies. The American constitutional system, for example, was designed on a theory that interests – what James Madison calls factions – would be so fragmented that no single faction would be able to dominate politics.Footnote ²⁴ Instead, the clashes between groups would lead to the election of leaders who would advance the public good, rather than the preferences of any one group.Footnote ²⁵ At the same time, the separation of powers was meant to divide institutions, not factions, unlike earlier mixed-government regimes that explicitly saw the primary divides in society as rich versus poor and built those divisions into the constitutional structure.Footnote ²⁶ In recent years, many constitutional scholars have shown the limitations of Madison’s theory. Some have argued that in an ideologically polarized system, political parties and party identity are the central divide in society, and so a system of checks and balances should focus on the “separation of parties, not powers.”Footnote ²⁷ Others have noted that policy preferences overlap with wealth and class, and that the political system is dominated by the wealthy and interest groups; those cleavages, not institutional separations, are primary drivers in politics.Footnote ²⁸ In both cases, the worry is that the Madisonian system will not function as intended because of acute social cleavages.

Age may offer another example of this problem. According to the Interparliamentary Union, only 2.6 percent of legislators around the world are under the age of thirty, despite that age group comprising nearly 50 percent of the world’s population.Footnote ²⁹ Only about 17.5 percent are under forty, and 30 percent are under forty-five.Footnote ³⁰ Outside of elected officeholding, there are age disparities in voter registration, voting, and political acts from writing letters to making campaign donations.Footnote ³¹ Political scientists have also shown that there are clear life-cycle effects in terms of political participation. Activity in politics is low in youth, increases in middle age, and has a short trail-off at extreme age.Footnote ³² As life expectancy grows, we should expect life-cycle effects on the older end to stretch – with older people remaining active for longer. The result of this data is that there is a massive imbalance in representation of the young and an imbalance that will likely expand with rising life expectancy. If, as political scientists have argued, representative democracy is not a system that aggregates the views of individual voters or responds to deliberation, but rather a process by which groups wield power, then the young have and will continue to have comparatively less power and influence than their elders.Footnote ³³

The overlap of age and economic inequality only magnifies this problem. Older people (and those with rising life expectancies) are wealthier than younger people. Political scientists have demonstrated that the poor and middle class are not well represented in government, either through elected officials who come from a blue-collar backgroundFootnote ³⁴ or through participation in contacting representatives, volunteering, donating, voting, or lobbying (personally or through representatives in Washington).Footnote ³⁵ It is perhaps no surprise, therefore, that study after study has concluded that policy outcomes reflect the preferences of the wealthy.Footnote ³⁶ Indeed, when the vast majority of Americans’ views diverge with those of the wealthy, the wealthy’s views tend to control policy outcomes.Footnote ³⁷

The data on age and wealth inequality, and on rising life expectancy, raises the prospect of gerontocracy – a regime defined by the rule of the elderly. If the elderly participate more in politics, serve more in government, and are systematically wealthier than younger people, the elderly will have disproportionately more political power and influence than younger people. Moreover, political influence, like wealth, can compound over time. Decades of donating to political leaders can build close relationships; decades in office can build political power – in the form of favors done, former staff spread throughout the system, and skills and relationships. It may be, as we approach the unequal 100-year life, that we will have an increasingly unequal system in which wealthy, older people have more and more political power – a gerontocracy.

The specter of gerontocracy might not matter if the elderly’s views are similar to those of younger people. But surveys show that different generations have different attitudes toward policy. According to Pew Research, millennials are the most supportive of same-sex marriage, with older generations – Gen X, Boomers, and Silent – exhibiting descending levels of support (the study omitted Gen Z).Footnote ³⁸ Views on whether racial discrimination is the primary barrier for why Black people in America “can’t get ahead” also diverge by generation, and over time the gap has widened even as attitudes changed. Indeed, younger people’s attitudes have shifted further than other generations on whether racial discrimination is a major problem.Footnote ³⁹ This gap in age cohorts persists even if comparing only white millennials to white members of older generations.Footnote ⁴⁰ On other questions too, there are serious divides based on age: Do immigrants strengthen the country? Younger generations say yes.Footnote ⁴¹ Should the US rely more on diplomacy or the military to ensure peace? Younger generations focus on diplomacy.Footnote ⁴² Should the US government be bigger and offer more services? Younger generations say yes.Footnote ⁴³ Is it the government’s responsibility to provide healthcare coverage to everyone? Younger generations say yes.Footnote ⁴⁴

Age differences in policy preferences and political participation create a variety of problems. We might call the first the entrenchment problem. Longer life expectancy may mean that people remain in positions of power longer. Commentators regularly note that Presidents Trump and Biden are older than their predecessors, that some congressional leaders are older than is conventional, and that Supreme Court justices regularly serve into their eighties. If older generations remain entrenched firmly in power for longer, that keeps younger generations from gaining power until they too are far older. The result could be a systematic phase shift in the age of political leadership – and continued lack of representation among those who are younger and less wealthy.

A second problem is what we might think of as the increasing imbalance between the wisdom of the ancients and the wisdom of the young. One of the primary arguments for a political system that operates slowly, with many veto-gates, checks, and balances, is that it recognizes the wisdom of past generations. Those who came before were not fools, and their considered decisions should be given great weight before they are displaced by new ideas and new policies. Similarly, societies revere the elderly for their accumulated wisdom and knowledge. A lifetime of successes, failures, and lessons learned may lead to a just and wise gerontocracy. But the problem is that an equal and opposite principle is at play too: the wisdom of the young. As technologies, innovations, and society change, the elderly may not update their technical knowledge or social views as swiftly as they should, particularly in comparison to younger generations who grow up natively familiar with a new context. The wisdom of youth welcomes the future, is more willing to experiment, and charts new frontiers. The young may also think with a longer time horizon, because they must live with policy decisions for many decades to come.Footnote ⁴⁵ As life expectancy rises and if older people remain in positions of power longer, we should expect a political system that is already designed to move slowly to move even more slowly, to adapt less well to social and technological changes, and to be less innovative. The balance between the conservativism of age and the experimentalism of youth will likely lean further toward the former.

A third problem is in the design of public policies. Political scientist Paul Peterson has observed that federal government policies that help children are systematically different from those helping the elderly across a range of dimensions. Programs for children tend to have state-determined eligibility requirements, rather than federal; they tend to have intergovernmental administration, rather than national; they tend to be complex, rather than simple; and they tend to offer substitutable benefits, rather than supplemental benefits.Footnote ⁴⁶ Peterson argues that the dynamics of program design are a function of the political power that the elderly have, compared to the young. Their power enables them to push for programs that bring more benefits, with greater consistency and simplicity.Footnote ⁴⁷

These three problems may lead to a fourth – an increasing gap between the rulers and the radicalized. If policies systematically lag behind changing trends, if programs are better for those who are older and richer than the overall population, and if political leadership is increasingly entrenched in power, we might expect increasing frustration with politics and increasing radicalization among younger cohorts. Younger cohorts may see themselves shut out of direct political representation and their policy preferences advancing far more slowly than they would like. Overlap with economic inequality means that these dynamics could become acute for many with respect to their basic livelihoods.

Looking globally, some commentators have suggested that a “youth bulge” in a country – a large population of young people without sufficient economic opportunities – can contribute to political radicalization and even destabilize governments.Footnote ⁴⁸ It is possible that rising life expectancy, and the political dynamics it unleashes, might be similar. While some young people may choose to remain in the political system and wait their turn, many others may not. Rising hopelessness and distrust in government’s ability to address their concerns might push them to exit the political process and exercise their voice through protest.Footnote ⁴⁹ It is possible that such frustrations could lead, over time, to a greater radicalization of politics, particularly among younger cohorts. In a gerontocracy, the slow pace of reform may lead to calls for revolution.Footnote ⁵⁰

7.3 Age and Political Representation

What can be done about the possibility of gerontocracy? The first option is simply to do nothing. Many social and identity cleavages are not specifically accounted for in political representation – tall people, blondes, fans of crime novels, runners. None of those groups get representation in politics, and few would think they need representation. That intuition likely stems from the view that nothing about their group identity is essential to public affairs or linked to systematic differences in policy preferences. But as noted earlier, this is not true of age. There are significant divergences in policy preferences by age – even among copartisans. Younger conservatives have significantly different views than their elderly counterparts on a variety of policy issues.

Another possibility is that the young are represented, virtually, through their elders. While there is a long tradition of political thinkers who believed representatives should “be in miniature an exact portrait of the people at large,”Footnote ⁵¹ there is also a long tradition of those who believed representatives would carry the views of their constituents to politics, even if they did not “mirror” the population precisely.Footnote ⁵² For example, older members of legislatures have children and grandchildren and may consider those younger generations’ interests when making political decisions. They may also listen to their constituents and represent those views faithfully, for fear of losing an election. Still, political scientists have shown that actual representation matters. Women in office act differently than men; people of color act differently than whites, and people from blue-collar backgrounds act differently than those from white-collar backgrounds.Footnote ⁵³

A second option, which accounts for policy or other changes, is for the political system to informally consider age as a factor in representation. To some extent, this is currently what happens. Some candidates for office run explicitly on generational change, promising fresh ideas and a new approach compared to the tired politics of the past. Their opponents emphasize their experience and entrenched power after decades in office. For appointed roles, leaders may choose to pick people from different age brackets, in part to ensure age diversity. It is not as obvious that leaders take age diversity into account, especially when compared to gender and racial diversity, but they certainly could. Still, with rising life expectancy in the political class, it is possible that the “younger generation” in politics may not actually be so young.

A third option is to grant specific representation by age. This should not be as outlandish as it might first seem. It is worth noting that constitutional theorists have long considered some groups worthy of specific representation. The mixed-government tradition that persisted from ancient times until the American Revolution divided society into classes, giving specific representation to the wealthy and the ordinary people. In comparative politics, the field of consociationalism outlines strategies to give representation to divided societies – decentralization and regional control; power-sharing agreements; and quotas for ethnic group representation at the cabinet or senior level. In societies deeply divided by religion, language, and ethnicity, many have thought to build representation for each faction directly into the political system.

The question, of course, is whether age fits into the same category. People may come to different conclusions on this point, but as noted earlier, there are not only direct correlations between age and wealth but also systematically different policy preferences by age cohort. As life expectancy increases, it is likely that generational divides will increase as well because older generations will be represented for longer. If age is likely to be a more and more significant cleavage over time, it may be worth considering as a factor in institutional design. Let us, therefore, consider a few thought experiments in how age might be incorporated into institutional design.Footnote ⁵⁴

7.4 Conclusion

In representative democracies, public policy depends on who has political power and who uses it. Political power turns not simply on the institutional structures of government but on background social conditions such as wealth, ideology, geography, and race. With rising life expectancy pointing toward a 100-year life, age may become an increasingly important feature in determining political power. Older political leaders may become entrenched in office, and older people tend to be wealthier – another important aspect of political power. The consequence of gerontocracy could over time be radicalization of the young who feel poorly represented. Addressing this problem will be important to ensuring representative government, and there are a number of formal strategies governments can take, from age qualifications for office to term limits to youth quotas, in addition to informal strategies like candidate recruitment. But whatever the approach, as people live longer – and do so unequally – the policy consequences will be significant.

With the number of older adults increasing worldwideFootnote ¹ – and the ranks of the very old rising especially sharplyFootnote ² – countries are confronted with the question of how to respond to their expanding elderly populations. Historically, countries have responded to the “problem” of old age by creating special policies for older adults, such as pension programs for those of “retirement age.”

This chapter considers the role of such age-based law and legal entitlements in a world in which people routinely live to be 100 years old. It asks: In such a world, what should be the legal significance of advanced chronological age? Should an adult’s chronological age affect his or her legal status or entitlements? It posits that the rise of the 100-year life should prompt a shift away from age-based policy. Specifically, it shows why, as life expectancy increases, chronological age will become an increasingly poor proxy for other characteristics, and that, therefore, using chronological age to establish eligibility for legal entitlements will result in increasingly poorly targeted policy interventions. It also explains how the use of age-based classifications will become increasingly inequitable and regressive.

The chapter calls for a wholesale reconsideration of the use of age-based criteria. It argues that social welfare policies should increasingly reject age-based classifications in favor of approaches that target interventions to populations based on their actual vulnerability to the particular problems the policies aim to address. It explains that although this alternative approach may be more challenging to administer, it is increasingly necessary to ensure efficient and equitable distributions of resources.

8.1 The Role of Age-Based Classifications

Age-based legal classifications are ubiquitous. Across time and across cultures, jurisdictions have used chronological age as a mechanism for establishing rights and responsibilities.

Today, chronological age classifications occur in a variety of legal contexts. Chronological age is used to establish legal duties. In both criminal and civil cases, minors are typically held to a lower standard of care than are adults.Footnote ³ Similarly, certain behaviors may be required of individuals at one age, and not at another. For example, states may require education for minors, or special tests for older adults seeking drivers’ licenses.Footnote ⁴ Chronological age can also determine eligibility for benefits. Although states have generally abandoned age as a basis for imposing guardianship or conservatorship, eligibility for public health insurance in the US largely depends on chronological age. Not only is Medicare coverage largely limited to those over sixty-five years of age, but older adults and children can more readily qualify for Medicaid than can younger adults. In addition, chronological age can determine the level of protection the law provides against actions by others. For example, the US Age Discrimination in Employment Act (ADEA) provides protection against age discrimination in employment to those aged forty and over, but no protection to those under the age of forty.Footnote ⁵

The ubiquity of age-based legal classifications, in part, reflects their utility. In some contexts, chronological age is a meaningful predictor of individuals’ needs or abilities, and, therefore, using age-based classifications can help efficiently target policy interventions and resources. Compulsory education requirements for minors, for example, largely reflect the distinct needs and abilities of youngsters, whose cognitive abilities generally develop in predictable ways corresponding with their chronological age.

The ubiquity of such classifications also, however, reflects their substantial administrative appeal. Chronological age classifications are easy to administer because chronological age can be readily determined without any need for discretionary judgment. Thus, policymakers are understandably likely to use age as a proxy for a wide range of other less-readily-ascertained characteristics. Indeed, age-based legal classifications are frequently used as proxies for more-difficult-to-measure attributes such as maturity, disability, or financial insecurity.

Consistent with these advantages, chronological age-based classifications are particularly prevalent in policies designed to respond to the risks associated with aging. Indeed, age-based classifications are a core feature of US laws and policies designed to address the needs of older adults. The major entitlement programs in the US – Social Security, Medicare, and Medicaid – have rules that preference individuals for benefits based on advanced chronological age. Most Americans will only become eligible for government-provided health insurance because they become older.

Age-based classifications are also common in policies designed to address concerns about the safety of older adults. Indeed, the use of age-based classifications has increased in recent decades as the US has sought to respond to the problem of elder abuse and neglect. States are embracing elder-specific social service interventions. Almost all states have adopted mandatory elder abuse reporting.Footnote ⁶ Many of these require third parties to report otherwise confidential information if the suspected victim is above a certain age. For example, Rhode Island requires “any person who has reasonable cause to believe that any person sixty years of age or older has been abused, neglected, or exploited or is self-neglecting” to “immediately” report it to the state.Footnote ⁷ Others require reporting of suspected abuse when the victim is above a certain age and has a requisite level of disability.Footnote ⁸

States have also created new crimes that exclusively apply to behavior directed at older adults.Footnote ⁹ Some states have criminalized certain types of otherwise consensual sexual relationships where the “victim” is an older adult. For example, Washington prohibits sexual relations between disabled persons aged sixty and over and those who provide them with paid transportation.Footnote ¹⁰ Similarly, some states have adopted statutes that create criminal liability for certain types of financial transactions if those transactions involve a person of advanced age.Footnote ¹¹ And Maine allows for certain transactions to be voided if they were the result of undue influence exerted on a person aged sixty or older.Footnote ¹²

8.2 Impact of Expanded Life Expectancy: Recalibration

Expanded life expectancy, and the growing ranks of the extremely old, should trigger recalibration – or, at least, a consideration of whether recalibration is in order – of age-based legal classifications that differentiate among adults.

Recalibration could help to better allocate scarce resources. As the population ages, entitlement programs for older adults will become substantially more costly without recalibration. This has been well recognized in the context of the Social Security system. Scholars and policymakers have long been in conversation about increasing the Social Security retirement age (either the early retirement age, normal retirement age, or both) to maintain the program’s financial solvency. Anne Alstott, for example, has suggested that raising the retirement age for higher-income workers is appropriate both to address the program’s solvency and to further fairness across the life cycle.Footnote ¹³ Similar suggestions have been made, although with less political support, about raising the Medicare eligibility age in response to changing demographics.Footnote ¹⁴

Recalibration could also improve the fit between age-based criteria and the underlying characteristics being targeted. Age-based criteria are often used in policy as a proxy for disability and frailty, but old age is actually a poor predictor of either, especially when it is defined broadly (such as sixty or sixty-five and older, as is common in US policy). At higher age thresholds, by contrast, there is a stronger correlation between chronological age and disability and frailty. For example, although dementia is often thought of as a common ailment in old age, less than 5 percent of those in their seventies have dementia, but over a third of those age ninety and older do.Footnote ¹⁵

Thus, ratcheting up age-based eligibility criteria is likely to result in more tailored policy interventions because chronological age is a more accurate predictor of incidence of physical and cognitive disabilities for the very oldest of the old than for older adults who are younger (i.e., the “old-old” and “young-old”). The older the age classification used, the more likely the individuals covered by that classification will have predictable physical and cognitive vulnerabilities and thus the more likely the classification will be a reasonable proxy for such traits.

Recalibration may also impact the experience of growing older, potentially in positive ways for some. Age-based legal classifications help shape the social meaning of chronological age. For example, turning sixty-five is perceived as an entrance into old age and a major milestone for Americans because it has been deemed “retirement age” for Medicare (and previously was “normal retirement age” for Social Security benefits). As one blogger wrote: “Turning 65 years old is one of those life passages – like college hazing and infantry basic training – many men would rather avoid contemplating until it’s upon us … Nobody wants to think about becoming an elder.”Footnote ¹⁶ Thus, increasing the minimum age criteria for policies designed to address the needs of older adults may prolong the perceived experience of middle age.

On the other hand, recalibration in the form of simply raising the eligibility age for benefits (e.g., Social Security, Medicare, access to retirement accounts without penalty) will come with substantial costs and often inequitable results. As eligibility ages are raised, we can expect people to remain in the workforce longer, potentially crowding out younger workers and workers with lower wages.

Increased inequities can be expected because although people’s chronological age increases at parallel rates, their biological ages do not. Biological age, which may be estimated based on biomarkers and other indicia of health, occurs at different rates in different people. In addition to genetic contributors, socioeconomic status, lifestyle, and environmental factors, including exposure to pollution, may impact the rate of biological aging.Footnote ¹⁷ Thus, responding to increases in longevity by increasing eligibility thresholds – without accompanying those changes with new supports that meet the needs of those of advanced biological age – will tend to be increasingly regressive: disproportionately disadvantaging already disadvantaged and marginalized groups.

In short, the rise of the 100-year life should not simply cause recalibration of age-based classifications; it should trigger reconsideration of their use in the first place.

8.3 The Impact of Expanded Life Expectancy: Reconsideration

The rise of the 100-year life should revive and reenergize the classic debateFootnote ¹⁸ over what role (if any) chronological age should play in determining rights or establishing eligibility for benefits. That is, it should seriously call into question not only what age-based classifications are used but also whether such classifications should be used at all.

One reason that expanding life expectancy should also prompt reexamination of the use of age-based legal classifications is that chronological age is likely to be an increasingly poor proxy for underlying characteristics that policymakers wish to target (e.g., functional abilities or needs).

As life expectancy increases, we can expect increased heterogeneity in the experiences and life trajectories of people born in the same year. This is because aging fosters heterogeneity. As a generational cohort grows older, its members become more diverse in their functional abilities and health status. Heterogeneity in health-related characteristics is particularly pronounced in early older age.Footnote ¹⁹ Consistent with this general pattern, a recent study of older adults in Canada, which considered thirty-four health-related characteristics, found that, overall, health heterogeneity peaks around seventy years of ageFootnote ²⁰ but continues to increase for certain characteristics – including functional performance – into at least the eighties.Footnote ²¹

The increased heterogeneity among age peers reflects, at least in part, the effect of cumulative advantage and disadvantage over the life course. In particular, those with lower socioeconomic status and lower levels of education, on average, experience poor health and more functional limitations than those with higher socioeconomic status and levels of education.Footnote ²²

A second more fundamental reason why expanding life expectancy should prompt reexamination of the use of age-based classifications is equity. The expansion of the ranks of centenarians, and the oldest-of-the-old more broadly, is not evenly distributed in society. Although it is often said that everyone grows old, this is simply untrue. Old age is not an inevitable experience. Not everyone will reach older age. Illness, accidents, and violence mean that many will die in their youth or younger adulthood. As of 2021, life expectancy at birth for non-Hispanic Black Americans was six years less than for non-Hispanic white Americans, and more than seven years less than that for Hispanic Americans.Footnote ²³

Very old age is even less certain. As of 2018, less than 10 percent of Americans born alive lived to the age of ninety, and under 2 percent lived to be 100. Incidence of very old age varies by race and ethnicity. For example, as of 2018, Hispanic Americans were nearly twice as likely to live to age 100 as non-Hispanic Black Americans, and nearly 2.5 times as likely to live to age 100 than non-Hispanic white Americans.Footnote ²⁴ The likelihood of reaching very old age also varies substantially by socioeconomic status. In fact, in the US, educational attainment level is one of the strongest predictors of adults’ health and mortality. Notably, the gap in life expectancy by socioeconomic status is widening in many parts of the country. A recent study of disparities in mortality found that the gap between those with higher levels of education and those with lower levels of education has widened in many states (especially those in the South and Midwest) since the mid 1980s.Footnote ²⁵

Even if all Americans had an equal chance of surviving to very old age, however, the rise of the 100-year life would create new equity concerns about policies that disproportionately allocate resources to the very old. This is because such policies may create an inequity from a lifespan perspective. When compared with younger adults, older adults appear vulnerable. However, when compared to others of their own birth cohort, older adults are resilient – they are the survivors and privileged relative to others born at the same time who did not live to old age. Thus, policies that target older adults for special assistance may further entrench inequities by reducing the overall share of resources spent on groups that disproportionately die at younger ages (a group disproportionately composed of those with lower educational attainment, the poor, individuals with disabilities, and Black Americans).

In short, the effect of differences in life expectancy means that the use of age-based classifications is likely to increasingly result in an inequitable distribution of resources. As gaps in longevity increasingly accrue to some socioeconomic groups and not others, policies that differentiate among adults based on chronological age will increasingly serve to compound disadvantage or advantage.

Radical reconsideration of age-based criteria may have some advantages for older adults. Age-based classifications are typically thought of as benign with regard to older adults. Yet, this is increasingly untrue as states adopt legislation that – in the name of protecting older adults from abuse and exploitation – selectively burdens the rights of older adults.Footnote ²⁶

Moreover, reconsideration may benefit older adults even when it involves reconsidering policies that in fact provide for a preferential treatment of older adults. Chronological age criteria that are used to establish eligibility for benefits or to set legal duties, but which are not highly correlated with need or do not reflect meaningful differences in capabilities, may perpetuate stereotypes about age. This in turn can fuel ageism (i.e., stereotyping and discrimination based on age). Policies that provide special benefits to older adults or try to limit or punish certain types of arrangements if they involve older adults (e.g., certain types of sexual relationships or financial transactions) may promote stereotypes about older adults as helpless, frail, or unable to make decisions. Such ageist stereotypes are not merely distasteful; they may increase risk of elder abuse. Although the empirical evidence of the link between ageism and elder abuse is still emerging and more is needed to show that ageism in fact causes elder abuse, there is some suggestive evidence that points to a likely causal connection.Footnote ²⁷ As Edman Palmore has argued, “stereotypes of the old as helpless, worthless, and repulsive” may promote elder abuse by encouraging “people to vent their frustrations on elders or to see them as vulnerable targets for exploitation.”Footnote ²⁸

A move away from age-based criteria has precedent. It would be, for example, consistent with a shift in the law’s approach to court-appointed guardians and conservators. Historically US states have authorized the appointment of a guardian or conservator to make decisions for people whom a court determined had a particular status, such as old age and a disability associated with old age, or a particular medical diagnosis. Influenced by the Uniform Probate Code’s alternative approach, states have largely moved away from such status and age-based criteria. Today most states only permit the appointment of a guardian or conservator – and the corresponding removal of rights from the person for whom one is appointed – if a court finds that individual has functional limitations that result in a need for such an arrangement.

8.4 An Alternative Framework

A reconsideration of age-based classifications should lead to a move away from using chronological age as a proxy for other ascertainable characteristics – such as disability, health status, or financial insecurity – in favor of more direct approaches. In a world where chronological age is likely to be an increasingly poor indicator of functional ability or need, and where socioeconomic gaps in life expectancy are substantial, it will be increasingly indefensible to use chronological age to differentiate among adults.

Chronological age should not be seen as a reasonable criterion for targeting public policy interventions simply because it is convenient or perceived to be a proxy for need, health status, or vulnerability. Instead, polities should seek to find alternative targeting mechanisms. To do so, polities should consider the growing literature on how “vulnerability” can be a basis for the allocation of resources. This school of thought builds on Martha Fineman’s “vulnerability theory,” which proposes that vulnerability is inherent to the human condition, and that governments therefore have a duty to respond by ensuring that all people have equal access to societal institutions that distribute resources.Footnote ²⁹ The approach has captured the imagination of some elder law scholars already.Footnote ³⁰ For example, in the context of old-age policy, Swedish scholars Titti Mattsson and Lottie Giertz look to Fineman to argue that states should move away from allocating social services based on group-based identities like age or diagnosis; instead, they propose, the state should focus on vulnerability – which they describe as allowing for more individualistic interventions.Footnote ³¹

The vulnerability approach is a promising alternative to group identity (e.g., race, gender, poverty) as a basis for targeting social welfare policy. However, as I have discussed at length elsewhere,Footnote ³² policies cannot respond to vulnerability merely as a universal characteristic of human beings. Thus, using a vulnerability approach to target policy interventions requires a contextual approach that defines vulnerability relative to the threat being addressed and targets people based on their vulnerability to a specific problem. This would be consistent with a conceptualization of vulnerability being a product of the relationship between individuals and their environment.

By way of example, imagine a state wished to address the problem of financial exploitation. Adopting a vulnerability approach would suggest that states should resist the urge to create age-specific crimes (like “elder abuse” or “exploitation of an elderly adult”). Instead, the state should identify factors that make people susceptible to such exploitation or its impacts and allocate resources with a view to supporting conditions that increase resilience to financial exploitation. Thus, states might support interventions that target conditions associated with the risk of exploitation, such as social isolation, lack of long-term care supports, limited financial literacy, and lack of access to civil legal services for those who experience exploitation.

While vulnerability will increasingly be a superior basis for distributing resources, it would be a mistake to demand an end to all age-based classifications. Even with demographic shifts, polities could reasonably use age-based criteria where their underlying policy goal is specifically related to chronological age. A polity could decide – as a matter of public policy – that it considers age to be an appropriate and ethical basis for allocating resources (e.g., health care, basic income) or assigning duties (e.g., participation in compulsory education). Similarly, a polity might reasonably choose to allocate resources in a way that explicitly favors older adults to express appreciation for older adults or reward them for the contributions they made during their lifetimes. There are times where policymakers may wish to use age-based legal classifications not as a proxy for other characteristics but because policymakers think chronological age itself matters. For example, schemes that provide pension payments to older adults (such as the US Old Age Survivor’s and Disability Insurance program – commonly referred to as Social Security) may also function as a way for society to express appreciation for older adults or reward them for the contributions they made during their lifetimes. Favoring older adults for public benefits may compensate for disadvantages associated with older age. And the prospect of future benefits may help all people by reducing the fear of aging.

Likewise, increased longevity does not call into question distinctions between minors and adults. It remains reasonable for polities to continue to distinguish between adults and minors in a host of situations. Among children, chronological age remains a reasonably robust proxy for the stage of development. Moreover, a polity will frequently have compelling justifications for treating children differently from adults. For example, it might reasonably decide to provide public education to children but not adults on the grounds that children can make best use of that resource because of their development stage or because they will (on average) have more years of life to benefit from that public investment.

In addition, whether abandoning age-based criteria will actually further efficiency and equity depends on politics. Efforts to replace age-based eligibility criteria in critical social safety net programs, such as Social Security and Medicare, risk unleashing political forces that would dismantle the programs.

Thus, those seeking to replace age-based classifications with targeting based on vulnerability will need to be mindful of the potential political consequences of their actions, as well as the additional administrative costs associated with the use of what are likely to be more complex classification systems. But in a world where age-based classifications are becoming more inefficient and inequitable, the benefits of abandoning age-based classifications will likely increasingly outweigh the costs.

8.5 Conclusion

This chapter has examined how the law should respond to the challenge posed by old age when old age itself is changing. It has suggested that, as life expectancy increases, the use of age-based classifications will become less efficient and more regressive. Accordingly, not only will age-based criteria need to be recalibrated to ensure that scarce resources can be allocated efficiently, but the use of such criteria should be reconsidered altogether. A promising alternative to reliance on age-based criteria is to use a vulnerability-based approach: targeting social welfare interventions to populations based on the population’s vulnerability to the particular concern the intervention aims to address. While this approach may lack the alluring convenience of age-based criteria, it has the potential to more equitably and efficiently allocate resources in an aging world.

The premise of this volume – that human beings will soon regularly live to be well over 100 years old – seemed quite plausible just a few years ago. But in the time since Linda Gratton and Andrew J. Scott published The 100-Year Life in 2016, the unprecedented COVID-19 pandemic ravaged the globe, increased morbidity, and, for a period of time, significantly reduced human lifespans. Even assuming we have emerged from this public health debacle with the 100-year life once again a possibility, the pandemic has exposed the extent to which any promise of extended life is likely to be defined by deep inequities. Working-class people, and workers of color in particular, are the least likely to enjoy extended lifespans. And if they do live longer, they are unlikely to enjoy a leisurely retirement in which they decide how they spend their extra years. Instead, they are likely to face more toil and precarity.

This chapter imagines a different future. It asks: What would the 100-year life look like if working people had a greater hand in shaping it? What would a legal regime look like that gave working people power to affect decisions about how work is structured and how resources are distributed in the era of the 100-year life? How might we reimagine labor law for a more just and equitable 100-year life?

9.1 Working to Death

For many Americans, work is characterized by low wages, long hours, little autonomy, and deep precarity. The pandemic also made work life-threatening. While highly paid professionals retreated to computer screens and home offices, low-wage service workers risked their lives, with few rewards. Those who were in unions were able to negotiate greater protective equipment, safety precautions, hazard pay, and leaves of absence for individuals most at risk. Yet the vast majority of US workers lacked such collective representation. And the vast majority were employed at will, subject to dismissal for any reason or no reason at all (save legally proscribed reasons like discrimination). Against this background, most workers were fearful of reporting violations of workplace safety and health standards that put them, their coworkers, and their communities at risk of infection.Footnote ¹ Indeed, when workers requested protective equipment like masks and gloves, some employers retaliated, even barring workers from speaking out about weak COVID-19 workplace safety standards.Footnote ² For older workers, for whom the risks of COVID-19 were vastly increased, the lack of a collective organization and resulting lack of power to negotiate were particularly consequential. Some quit their jobs; others risked their lives. Racial inequities intertwined with these class and age inequities: Working-class Black and Latino males faced the largest decrease in life expectancy as a result of the pandemic, while workers of color were hit the hardest by rising unemployment. These dynamics meant that millions of workers, already on the brink, faced eviction, food shortages, and mental health crises.Footnote ³

Even after the pandemic began to subside, age, gender, and racial inequalities in the labor market persisted. As the job market recovered, older workers had more difficulty reentering the labor market than their younger counterparts; women, too, failed to reenter the job market at the same rate as similarly educated men.Footnote ⁴ The pandemic highlighted and increased the amount of care work that people, particularly women, performed without compensation or acknowledgment. And although the unemployment rate for people of color bounced back to prepandemic levels, the racial unemployment and wage gaps remained “high and widespread.”Footnote ⁵

In short, the pandemic and its aftermath left little doubt that, if our political economy remains unchanged, the 100-year life, if not elusive, will bring only more years of arduous work for the vast majority of workers, particularly low-income workers, workers of color, and women workers.

9.2 The Need for Worker Power

How, then, to think about reforming workplace law in the event of the 100-year life? Two other chapters in this volume focus on employment law. Professor Kenji Yoshino considers the pressing problem of how the law can effectively protect against age discrimination in an era in which workers live longer. He argues for unmooring the age discrimination paradigm from the traditional civil rights model, and for the law to take into account the distinctive fears that older individuals conjure about their own mortality. Professor Cynthia Estlund’s chapter tackles a related problem: ensuring workers’ retirement security over the course of longer lifespans – a problem made more challenging by the fact that most workers already lack adequate pensions and retirement savings. Estlund concludes that it will be necessary to extend work in order to approach a sustainable ratio of workers to retirees, and of work years to retirement years. She also considers how to address the problems of intergenerational competition and integration at work in a world with transformative new technologies and longer lifespans. Noting that a longer life could increase employers’ tendency to dismiss or refuse to hire aging workers, she renews calls not only to strengthen antidiscrimination protections but also to limit American law’s adherence to employment at will. In an earlier work, Professor Anne Alstott argued for enhancing and revising the Social Security program and delinking benefits from paid work to increase portability and part-time feasibility.Footnote ⁶

Alstott, Estlund, and Yoshino are all correct that finding new ways to fund retirement, delinking social benefits from employment, rethinking age discrimination law, and protecting workers from unjust termination are all critical policy reforms in an era of extended lifespans. But those reforms are unlikely to be achieved or sustained in ways that address the fundamental inequities that will characterize work in the 100-year life without also addressing the problem of collective voice among workers. That is, if workers are to enjoy longer lifespans and a more dignified and remunerated work life during their expanding lifespans, the law must also include a framework that gives them greater power over decisions about the shape of work and the political economy.

9.3 The Decline of Unions – and the Failure of US Labor Law

In the aftermath of the New Deal and World War II, more than a third of US employees belonged to unions. With workers able to exercise collective power in the workplace, the economy, and the halls of government, inequality was at an all-time low. Labor unions not only fought for higher wages and better working conditions but also backed early state efforts to establish social insurance programs, lobbied for a comprehensive federal retirement bill, and supported the eventual Social Security Act of 1935.Footnote ⁷ Social Security, in particular, proved to be a crucially important program, lifting millions of seniors out of poverty.

But the goal of an enduring collective voice for working-class people in the US was not achieved. From the outset, at the insistence of southern legislators, millions of workers, primarily African Americans, were explicitly excluded from labor law – as domestic workers and agricultural workers – or worked in sectors and regions of the country that were, in practice, largely unorganized. And unions soon came under significant attack. The 1947 Taft-Hartley Act curtailed numerous labor rights and excluded yet more workers from coverage. Moreover, although the left-leaning industrial unions in the 1930s and 1940s pushed for a social democratic system of universal health care and other social benefits delinked from employment, by the 1950s they had lost this fight. Retirement benefits, as well as health insurance, became closely tied to employment, and unions themselves became more bureaucratic and inward-facing.Footnote ⁸

By the 1970s, US companies had adopted a position of intense resistance to unions: They globalized and fissured their employment relationships, eliminating union jobs while ramping up their antiunion practices against new organizing efforts. Meanwhile, the Supreme Court interpreted the National Labor Relations Act (NLRA) in ways that limited the ability of workers to bargain and strike in the changing economy.Footnote ⁹ Union membership declined precipitously. Today, while the rate of unionization is on the rise, it stands at about 6 percent among workers in the private sector and only about 10 percent among workers overall.Footnote ¹⁰

Economic inequality is now at its highest point since the Gilded Age, when unionization rates were similarly low.Footnote ¹¹ And as workers have lost economic power, they have also lost political power. Wealthy individuals and corporations spend vast sums to influence elections and legislative and administrative processes, while workers have largely lost their collective organizations that enable them to engage effectively in politics. Not surprisingly, studies suggest that elected officials are almost entirely unresponsive to the concerns of the poor and working class – at least when their preferences diverge from those of the wealthy.Footnote ¹²

Meanwhile, although state and federal employment laws, which proliferated in the 1960s, provide a host of protections to workers as individuals, many rights go unenforced or underenforced. Wage theft, health and safety violations, and sexual harassment are rampant in many workplaces.Footnote ¹³ Individual workers, lacking bargaining power, are frequently forced to sign away their rights to go to court and even to proceed collectively in arbitration.Footnote ¹⁴ Unlike most advanced industrial democracies, the U.S. does not guarantee workers health insurance; nor do most workers receive compensation for the important labor they perform caring for their children or their elderly relatives. Retirement security is, for many, elusive: Just 15 percent of private sector workers had access to a defined benefit plan in 2020,Footnote ¹⁵ while Social Security benefit cuts are still being implemented from reforms in 1983.Footnote ¹⁶

Meanwhile, millions of the most vulnerable workers, predominantly people of color, remain excluded altogether from both collective labor law and individual employment law as independent contractors, domestic workers, or agricultural laborers. These exclusions are particularly relevant for seniors who are more likely to work as independent contractors than younger workers.Footnote ¹⁷ And one in five workers in their fifties and early sixties work in jobs that lack health or retirement benefits, making it much harder to prepare for retirement.Footnote ¹⁸ In fact, seniors often take on these jobs specifically to supplement inadequate retirement savings.Footnote ¹⁹ And while workers in nontraditional jobs are vulnerable due to these statutory exclusions, the ability to work longer to finance a comfortable retirement is already stratified by income: Higher-income workers tend to retire later than lower-income workers, who are more likely to exit the labor market at a younger age due to health issues, caregiving responsibilities, or job loss.Footnote ²⁰

On the positive side, a host of recent organizing efforts among workers, from Amazon warehouse workers to healthcare workers, from Uber and Lyft drivers to journalists, have drawn attention to these problems. These workers seek tangible improvements in their jobs; they also seek basic dignity at work. As one Amazon worker involved in a unionization campaign observed, “You’re running at a consistent, fast pace … You ain’t got time to look around. You get treated like a number. You don’t get treated like a person. They work you like a robot.”Footnote ²¹

Meanwhile, seniors have remained organized and politically engaged on issues related to retirement. Notwithstanding the national decline in membership in unions and other civic organizations, the AARP – an organization devoted to organizing and advocating for retirees – maintains a membership of nearly 40 million. The AARP has played a key role in opposing various efforts to erode retirement benefits. For example, in 2021, the AARP lobbied against the TRUST Act, which would empower congressional “rescue committees” to make changes to Social Security and Medicare. While Republican legislators have recently floated both the idea of “sunsetting” the Social Security Act and a federal budget plan that would cut benefits, AARP’s political power suggests that these efforts will not go very far.

Though the problems that workers will confront in any future 100-year life are palpable and the interest in organizing to solve such problems is significant, the existing legal frameworks are of little utility. About 70 percent of Americans approve of unions, yet obstacles to achieving unionization among workers even in the traditional labor force are substantial.Footnote ²² The law permits employers to engage in a wide range of efforts to persuade workers not to organize while excluding union organizers from the workplace. Moreover, when employers resort to illegal tactics such as threatening or firing workers for organizing, they incur virtually no penalties. Meanwhile, the law does not only exclude independent contractors, agricultural, and domestic workers, but it altogether lacks a framework for retirees and unemployed workers to organize into unions or for unions to bargain to address many problems facing retirees. That is, the law mandates only that employers negotiate with unionized employees about wages, benefits, and working conditions at a single workplace. It does not set up a system for sectorwide bargaining in order to raise wages and benefits throughout the industry, nor does it create a system for bargaining about broader social programs.

One proposed legislative reform – the Protecting the Right to Organize (PRO) Act – would better protect traditional workers in the process of organizing, bargaining, and striking, including by increasing penalties when employers attempt to coerce or retaliate against employees for unionizing.Footnote ²³ The PRO Act would also extend employee status to many workers now classified as independent contractors, like gig workers, FedEx drivers, and many more. These reforms could prove extremely helpful in enabling workers to exercise more voice at work.

However, even with these important reforms, US labor and employment law would not be well designed to give workers a voice in how to structure work and the broader political economy during the 100-year life. This is because even the improved law would lack a framework for retirees and unemployed workers to organize (while still excluding domestic and agricultural workers), and it would lack a system for sectoral or social bargaining. That is, even if the PRO Act passed, US labor law would continue to channel organizing and bargaining to individual worksites, between traditional employees and single firms, without giving worker organizations power to negotiate sectoral standards or broader social benefits. Yet, in a world in which employees, over the course of a lifetime, work for many different firms, move in and out of the workforce over a longer lifespan, spend more years working a reduced load as they age, or spend more years in retirement or caregiving, a labor law system organized solely around worksite-by-worksite bargaining for private benefits makes little sense.

9.4 Alternative Approaches to Worker Voice and Power

The US approach sharply differs from many European systems that are better positioned to enable worker voice over the course of an extended lifespan, though their systems too face significant limitations. For example, in Germany, the union federations participate in basic decisions concerning national wage policy and policies relating to employment, economic growth, and social insurance.Footnote ²⁴ Meanwhile, collective bargaining occurs on a regional basis to set wage scales that cover all workers, at least in manufacturing sectors; those agreements then provide a floor above which local bargaining may occur.Footnote ²⁵ Many German workers also have a voice at their worksite through statutorily guaranteed works councils.Footnote ²⁶ Such works councils are granted codetermination rights on topics such as health and safety and working hours.Footnote ²⁷ They also are entitled to information and consultation rights on topics such as the introduction of new work methods and technologies – an especially important right in a time of rapid technological change. Additionally, workers at large companies have an automatic voice through representation on corporate boards.Footnote ²⁸

In Denmark, unions have played an even more active role in negotiating social policy. Unions and employers have, for example, collectively negotiated national policies on worker training and parental leave. Moreover, union bargaining in Denmark often builds on statutory entitlements. For example, while older workers and workers with disabilities had statutory protections against age discrimination, social partners expanded these protections through collective bargaining such that workers now have the right to be placed in “light jobs” if they are unable to perform their positions.Footnote ²⁹

And in Norway, unions, employers, and the state have joined together to bargain for policies that improve working conditions so that older workers can remain in the workplace.Footnote ³⁰ This bargaining produced the Inclusive Workplace Agreement, which aims to reduce the use of sick and disability leave, extend workers’ working lives, and recruit more workers with disabilities and other vulnerable groups.Footnote ³¹ The social partners also serve on the board of the government-funded Centre for Senior Policy, which performs research regarding senior workers and works with companies to improve their policies.Footnote ³²

In France, the political organization of unions enables them to mobilize on issues regarding retirement. For example, in 2023, in response to President Emmanuel Macron’s proposal to reform the pension system by raising the legal age of retirement from sixty-two to sixty-four, France’s eight largest labor unions coordinated nationwide strikes across multiple sectors, including transportation, education, and energy.

These systems are by no means perfect. Workers’ power in these countries has also declined in the face of globalization and neoliberalism.Footnote ³³ However, a viable framework exists to set fair standards for all workers and to enable senior workers and retirees to exercise their collective voice.

The US lacks any comparable framework. To be sure, the law does facilitate workers’ political voice in important ways. Section 7 of the NLRA has been interpreted to protect workers’ concerted activity that occurs through political channels – even for nonunion workers, as long as such activity relates to employment issues.Footnote ³⁴ In addition, unions, like other organizations, may engage in electoral politics and lobby government officials. Many unions spend a great deal of energy and money on political activity, with significant effect. They have helped enact legislation that protects all workers, from the Fair Labor Standards Act and the Occupational Safety and Health Act to the Family and Medical Leave Act and the Affordable Care Act. But US law does little to facilitate a voice for working people in the various decisions that affect senior workers or that will be posed by the putative 100-year life.

9.5 A New Labor Law for the 100-Year Life

What might a different system look like, one that builds on the current regime but gives workers – who may soon enjoy longer lifespans and many different stages of work – greater power over decisions about the shape of work and the political economy?

First, as in the industrial democracies discussed earlier, the law should provide for sectoral bargaining so that collective bargaining agreements apply to workers throughout a region or sector, effectively forming the basis for employment policy in those sectors, while at the same time providing for worker voice at the worksite and firm level through works councils, local unions, and corporate board representation.Footnote ³⁵ Such a model would enable workers to move between firms more easily over time and limit risks of leaving and entering the workforce. More generally, research suggests that where sectoral bargaining is combined with worksite organization, racial and gender wage gaps decrease and unions can exercise more economic and political power.Footnote ³⁶

Second, the law should also give organizations of workers a formal role in developing social benefits and policies that will be critical for expanding lifespans, for example, through administrative processes at the local, state, and/or federal levels of government.Footnote ³⁷ Beyond simply permitting group participation, as notice-and-comment rule-making currently does, the law could give worker organizations a formal seat at the table in negotiating retirement, disability, family leave, and childcare benefits, and, perhaps, eventually job guarantees, basic income, and more.

Third, to make such social bargaining effective for workers at every stage in the 100-year life, the law should encourage or facilitate new forms of worker organizations that encompass a broader range of working-class people, including those unemployed, engaged in paid and unpaid caregiving, and retired.Footnote ³⁸ Examples of such nontraditional worker organizations exist within our own system, past and present. For example, following the Great Depression, unemployed workers organized in large numbers.Footnote ³⁹ The movement of the unemployed originated in local communities, with protests and marches demanding – and sometimes winning – public relief funding from local and state governments. Members of unemployed worker groups also banded together and entered relief offices demanding immediate relief for individuals denied aid. A national federation called the Workers Alliance was later formed, consisting of most of the large organizations of unemployed workers. This federation prioritized lobbying for national relief programs over direct action in localities; ultimately, it was unable to win passage of its prioritized legislation for expanded direct relief, higher pay on federal Works Progress Administration projects, and a permanent public works program, and its local organizing diminished as well. Still it provided an important voice for unemployed workers through a critical period.

Welfare organizing of the 1960s provides another example of a nontraditional working-class organizing effort. During this period, welfare recipients, the majority of them Black women, began to organize to change local welfare policies, overturn benefit denials for members, and win special grants for welfare recipients to cover household furnishings and clothing.Footnote ⁴⁰

Numerous examples of organizations engaging workers in nontraditional employment exist in the contemporary landscape. For example, home-care organizing has engaged hundreds of thousands of low-wage, formerly isolated workers in collective organization, even without the existence of a central worksite or single employer, helping raise wages and improve funding for caregiving.Footnote ⁴¹ The Freelancers Union represents independent workers, offering insurance benefits and other resources to various types of freelancers and advocating for legislative protections on their behalf. The Domestic Workers Alliance has succeeded in winning domestic workers’ bills of rights in numerous states; the bills provide employment protections such as paid time off and overtime rights. Many Hollywood writers work independently, but through their membership in the Writers Guild of America, they are guaranteed minimum pay levels, residuals, and retirement and health coverage. And community-based worker centers across the nation have organized to combat wage theft, co-enforce workplace protections with state and local governments, and advocate for increased pay and protections.

In a different vein, several organizations already organize retirees. AARP provides individual benefits to members and advocates on behalf of older and retired Americans. The Alliance for Retired Americans is sponsored by unions and aims to mobilize retired union members, seniors, and community activists to advocate for various political causes such as Social Security expansion and reducing prescription drug prices. Several unions also have their own retiree chapters and organizations.

More examples exist abroad. As Professor Katherine Stone has detailed, Japan has seen an explosion of nontraditional forms of labor organizations: City unions have pressured employers in an entire metropolitan area around work-related concerns; women’s unions have organized to further the interests of women workers; and ethnically specific unions have asserted pressure on behalf of ethnic groups.Footnote ⁴² In Europe’s most progressive social democracies, not only are traditional unions engaged in defining social programs that benefit all workers, but nontraditional worker and retiree organizations exist as well, working to advance policy concerning issues such as social security, pensions, and health treatment among retirees.Footnote ⁴³

These kinds of nontraditional but democratic organizations can be nurtured and expanded with supportive legal frameworks.Footnote ⁴⁴ For example, law can facilitate the growth of organizations by enabling the aggregation of resources, by allowing access to workers, information, and free spaces to organize, and by effectively protecting workers from retaliation for organizing and engaging in concerted action. In addition, by granting organizations a role in negotiating social benefits in administrative processes, as discussed earlier, the law can give the organizations greater legitimacy, aiding their organizing efforts. Government programs can also partner with democratic membership organizations that represent underrepresented constituencies, giving them grants to provide and administer benefits. For example, under the Ghent system, unions operate unemployment insurance funds and provide benefits to unemployed workers.Footnote ⁴⁵ US policymakers could extend this model to provide portable benefits, training, and increased workplace enforcement to workers across the span of the 100-year life.Footnote ⁴⁶

9.6 Conclusion

As Anne Alstott has written, the predicted “leap in longevity is likely to challenge many of our expectations about work” and could “revolutionize” many of our relationships.Footnote ⁴⁷ But without a fundamental change in the way work is organized, the leap in longevity will also augment and extend existing power dynamics that define the lives of many workers in America, making inequalities and indignities even more acute. To avoid that result, workers – of all kinds and backgrounds – must have a voice in the decisions about what work looks like over the course of the 100-year life.

Of course, one might worry that the reforms suggested above are fanciful or impossible to achieve. Indeed, past experience highlights the difficulty of labor law reform, even in times of unified democratic control of government.Footnote ⁴⁸ Yet the obstacles have been overcome in the past and can be overcome again.Footnote ⁴⁹ COVID-19 made the economic plight of poor and working-class Americans politically salient, as illustrated by the massive, redistributive economic stimulus enacted in March of 2021 and subsequent legislation like the Infrastructure Investment and Jobs Act of 2021 and the Inflation Reduction Act of 2022. The resurgence of organizing among workers, tenants, and debtors that has occurred in the wake of the pandemic has only increased this salience. Meanwhile, the federal structure of our government creates openings for legislative change at the state and local levels, even when federal change is impossible. Ultimately, even if the more ambitious reforms discussed earlier are not possible in the near term, incremental, achievable reforms can get us closer to a world in which workers are not working to death but rather have a collective voice in the 100-year life.

Walk into any bookstore’s self-help section and you will find a wall of books that were written with one goal: to help Americans figure out how to save money. Saving money is widely considered the cornerstone of how to have a successful life and, perhaps most of all, a relaxing and successful retirement. The titles say it all: “Saving for Retirement Made Easy,”Footnote ¹ “Winning at Retirement,”Footnote ² “How to Retire Debt-Free & Wealthy,”Footnote ³ and so on. The message is clear: It is possible to retire with abundant savings – you just have to learn more and be strategic.

The reality, however, is that retirement savings are an insurmountable aspiration for many Americans, particularly low-wage workers. Indeed, almost one in four Americans has less than $400 in savings.Footnote ⁴ One-third of Black households have zero or negative net worth.Footnote ⁵ But much like the self-help books in the corner of the bookstore, the government’s response to this overall lack of savings has been to emphasize the need for individuals to save.Footnote ⁶ Policy initiatives have focused primarily on consumer financial education and nudging people in the “right” direction.

Yet, despite the emphasis on saving for retirement, as Anne Alstott and others have discussed in depth, elderly Americans whose incomes are at the low end of the income distribution are increasingly worse off than their higher-income counterparts.Footnote ⁷ They enter old age with little or no savings, medical problems without adequate coverage, a necessity to work in order to eat and have a place to live, and, unsurprisingly, increasing rates of bankruptcy filings.Footnote ⁸

In the past, retirement years were limited in duration for most Americans, as a brief jaunt of post-work life was soon met with death.Footnote ⁹ Indeed, between 1650 and 1850, elderly Americans were less than 2 percent of the population.Footnote ¹⁰ Yet, as life expectancies have increased,Footnote ¹¹ we are faced with the prospect of elderly Americans living in poverty for decades on end. In other words, in the past the problem of no savings, and thus elderly poverty, was simply less of a problem since Americans, on average, spent fewer years in retirement. This, in turn, meant the financial struggle of the retirement years would also be relatively short. But with the 100-year life becoming a reality for more Americans, an elderly poverty crisis looms.Footnote ¹²

In this chapter, I unpack the contradiction in the message embedded in our cultural narrative about the moral worth of self-sufficiency and savings against the backdrop of the realities of our existing poverty laws. I show how this is a central and unresolved incongruity in US policy toward the poor and near poor. The government does not provide a solid financial foundation for older Americans who were making poverty- and near-poverty-level wages during their preretirement years, yet the government also does not allow those making low wages (low enough to not even provide for basic necessities such as housing, food, and clothing despite full-time work) to build a nest-egg for retirement. This antisavings policy – enforced through asset limitations for basic safety net programs such as the Supplement Nutrition Assistance Program (SNAP, commonly referred to as “food stamps”), Temporary Assistance for Needy Families (TANF, commonly referred to as “cash welfare”), and Social Security Disability Insurance (SSDI, commonly referred to as “disability”) – essentially guarantees that those who are poor will continue to be poor and without savings throughout their extended old age.Footnote ¹³ Therein lies the savings mirage.

I argue that as a society we have to get our story straight. Are we committed to a country in which we require people to save for themselves in order to avoid an old age of hunger and financial instability? If so, we need to implement policies that make it easier to save, not only for those who are middle or upper class but also for those who are low-income throughout their working years. On the other hand, if we are committed to restricting the savings of low-income workers because of our devotion to the notion that public benefits should only go to those who have otherwise stripped themselves of savings, then we need to develop a plan to provide in retirement for those who struggled through their working years, unable to save in part because of strict government policy. But we can’t have it both ways. If we continue with the crash course we are on, an elderly poverty epidemic is bound to emerge. New safety net policies show some promise, but this central contradiction in message and policy needs to be recognized in order to promote substantial policy change.

10.1 A Savings Society

The history of responsibility for the elderly in the US is an evolving one. Life was short in the mid 1800s, with the life expectancy of most white Americans only thirty-nine years.Footnote ¹⁴ But those who did live into old age and were poor were usually out of luck. The elderly were generally considered a “burden on the local taxes,” and many were either sent to poorhouses, which were “dreary, vermin-infested, and laden with human waste,” or auctioned off as farm labor.Footnote ¹⁵ By the late nineteenth and early twentieth centuries, “the status of seniors had reached its nadir,” a time when old age was thought of as a disease in and of itself.Footnote ¹⁶

After the Great Depression, life expectancies were increasing as were the percentage of elderly Americans living in poverty – by 1935, 50 percent of elderly Americans were poor, and by 1940, almost two-thirds were poor.Footnote ¹⁷ These staggering numbers, along with the realities of the Great Depression, led to President Franklin D. Roosevelt signing the Social Security Act of 1935 into law.Footnote ¹⁸ This was the beginning of a significant shift, where American values transformed and the government began investing in elderly economic security.Footnote ¹⁹ Indeed, in 1965, Lyndon B. Johnson’s administration furthered the goals of Roosevelt by enacting Medicare and Medicaid. At the same time, following World War II, employers began to expand retirement benefits for their full-time employees, providing them with defined benefit pensions and other important benefits for old age. Taking all of these changes together, Americans could in general depend on economic security in their old age.Footnote ²⁰

In the late 1970s and early 1980s, however, the concepts of individualism and self-sufficiency began to pervade US culture and politics.Footnote ²¹ A weakening of the safety net for Americans, including older Americans, followed. Employers began to replace mandatory pension plans with optional higher-risk 401(k) plans that do not guarantee a certain income during retirement, social security payments were reduced, and many more jobs did not provide any avenue at all for retirement savings.Footnote ²² Jacob Hacker refers to this as the “Great Risk Shift,” one where “a myriad of risks that were once managed and pooled by government and private corporations shifted onto individuals and families.”Footnote ²³

But people are still getting old, and life expectancies are continuing to increase, with the 100-year life in sight for many Americans. In response to increased longevity, there has been a renewed sense of urgency by the government in setting expectations – expectations that Americans must save for their own retirements. We have transformed into a country that says “take care of yourself” – even when you are old. In 2000, then Treasury Secretary Lawrence Summers gave remarks focusing on the need to encourage Americans to save more.Footnote ²⁴ He said that it was critical not only “to our economic health” but also for families “because it influences their capacity to manage what is new in the New Economy: not least, the fact that people are living much longer than before.”Footnote ²⁵ He went on to discuss the (increased) chances of people living to ninety years old, then concluded that “as people are retiring earlier and living longer, retirement spans for many individuals are approaching half or more of their working lives.”Footnote ²⁶ Summers acknowledged the challenges that come with longer lifespans and then summarized his goals for the speech as follows:

1. 1. First, discuss why increasing saving is so crucial both for the American economy as a whole and for the economies of individual American families.

2. 2. Second, discuss our approach to increasing saving based on improving financial literacy and promoting financial access. Recent economic research demonstrates that saving is like any other form of consumer behavior: It is governed by habit, is responsive to social signals, and it can be promoted.

3. 3. Third, discuss the growing problem of borrowing, or negative saving.

4. 4. And fourth, discuss specific legislation that goes beyond existing incentives to further promote savings among Americans who have not had the opportunity to avail themselves of existing incentives.Footnote ²⁷

The government’s recent policies and rhetoric indicate that Summers’s vision lives on – we must teach Americans how to save and provide behavioral nudges and incentives for doing so.Footnote ²⁸ Indeed, in 2015, the Treasury Department released a treasury note entitled “Helping More Americans Save” in part to mark “America Saves Week, an annual opportunity to encourage all Americans to save and a chance for individuals and families to take stock of their financial health and preparedness.”Footnote ²⁹ The notice, released by J. Mark Iwry, then Senior Advisor to the Secretary of Treasury for Retirement and Health Policy and the Deputy Assistant Secretary for Tax Policy, noted that “[s]avings for retirement is one of the best ways to prepare for a secure financial future.”Footnote ³⁰ And further, the Treasury Department was “ensuring that more Americans have the knowledge and tools to manage money and credit responsibility because educated and financially capable consumers can better attain their own financial goals and contribute to the strength of our overall county.”Footnote ³¹

There are moralist and paternalistic overtones to both of these releases. The idea is fairly simple: Some Americans are not financially educated and need help “learning” how to save, so the government will do the work to improve “habits,”Footnote ³² as Secretary Summers put it. The lesson seems to be that if we create opportunities to save, change habits, and increase social pressure, we will see a difference. There was little discussion in either release (or others like them)Footnote ³³ of the structural conditions that make it difficult for Americans to save.

The federal government and federal laws certainly incentivize savings and wealth accumulation. But as others have documented in detail, most of these programs primarily benefit middle- to upper-income white citizens. For instance, our tax code has written into it several savings incentives, but low-income people rarely benefit from these incentives.Footnote ³⁴ There are some other programs that are more inclusive, but in general they are small in scale and come and go. For example, Treasury created the “myRA” retirement savings program, which was designed to be a starter retirement savings program. It was supposed to help low-income workers who didn’t have access to retirement savings plans through their jobs, but the take-up rate was low, and it was eventually shut down after Treasury found it not to be cost-effective.Footnote ³⁵ Other Treasury programs sought to work with employers to create simple retirement plans for workers not covered by existing plans.

Ultimately the social construct of savings as moral, trainable, and the right thing to do pervades the culture in the US. And existing research shows that low-income Americans have adopted this ethic and in fact desperately want to save – they even work the existing savings-unfriendly systems to try to force themselves to save.Footnote ³⁶ However, their efforts are often in vain, in part because of the underappreciated aspects of our laws and policies that make it almost impossible for the working poor to save. In the next section, I discuss these restrictions.

10.2 Restrictions on Savings

There are a plethora of reasons why it is difficult for Americans to save. Wage stagnation and rising expenses are significant factors. A 2019 study found that 53 million Americans, accounting for 44 percent of American workers between the ages of eighteen and sixty-four, qualified as “low-wage” earners.Footnote ³⁷ The median hourly wage of those low-wage workers was $10.22, with median annual earnings of about $18,000.Footnote ³⁸ Further, over half of low-wage earners were either the only workers in their family or lived in families where all workers were low-wage.Footnote ³⁹ While wages went up a bit in the midst of the pandemic, a recent study found that in 2022 more than 51.9 million workers, or almost 32 percent of the labor force, made less than $15 per hour.Footnote ⁴⁰

Such low wages leave families vulnerable to economic insecurity – just one small shock, such as missing work due to an illness, a funeral of a family member, or a needed repair for a car used for transportation to work, can send these workers into financial turmoil and create a need to depend on the scant public safety net that is available.Footnote ⁴¹ In fact, many low-wage workers even working full time and without experiencing economic shocks must depend on Medicaid and food stamps to get by. Numerous studies and reports confirm that large companies pay their workers such low wages that these workers need public assistance, even when working full time.Footnote ⁴²

Even at times when low-wage workers have the opportunity to save – when perhaps they receive an unexpectedly large tax return, they are given extra hours at work, they receive a stimulus payment/credit, or they sell a prized possession for cash – our poverty laws often prevent these workers from accumulating any meaningful savings. How is this possible in a society that so values saving? It happens through asset limitation laws that affect every major public assistance/antipoverty program (outside of the tax system) that provides money or money-like relief to low-income Americans. These limitations impose rigid restrictions on the amount of cash and assets aid recipients can accumulate. Thus, recipients are in a bind. If they attempt to put away meaningful savings when they receive a lump sum in the form of, for instance, a stimulus check or tax credit, they cannot continue to access the programs they need to survive month to month.

Before Ronald Reagan was elected, states set welfare programs’ eligibility requirements. President Reagan, however, campaigned in the early 1980s on a promise to reform welfare by cracking down on “welfare queens” (a term he popularized) who, he claimed, abused the public welfare system. During a 1976 stump speech, Reagan described the welfare queen as follows: “She has eighty names, thirty addresses, twelve Social Security cards and is collecting veterans’ benefits on four non-existing deceased husbands. And she’s collecting Social Security on her cards. She’s got Medicaid, getting food stamps and she is collecting welfare under each of her names. Her tax-free cash income is over $150,000.”Footnote ⁴³ Welfare recipients were portrayed as primarily Black, and the media and other politicians perpetuated this (inaccurate) image.Footnote ⁴⁴

While Reagan did not succeed in the massive welfare reform he promised, he was able to push through asset restrictions on major welfare programs, which largely remain in place today (though states have regained some control). Reagan did this via the Omnibus Reconciliation Act of 1981, which set asset limits at $1,000 for most welfare programs, a large decrease compared to then-existing state limits. And the federal limits had teeth. A study conducted ten years after Reagan left office showed that low-income single mothers reduced their personal savings by an average of $1,250.Footnote ⁴⁵

Following Reagan’s presidency, the poverty laws in our country moved even further in the direction of building expectations for families to develop and sustain their own safety nets, both during their working years and into their old age. In the mid 1990s, Bill Clinton ran for president on a campaign promise to reform cash welfare (then Aid to Families with Dependent Children – AFDC) into a program that would promote “personal responsibility” and “economic self-sufficiency.”Footnote ⁴⁶ When Clinton was elected, welfare reform was indeed passed, transforming AFDC into TANF, stripping away the existing safety net for families, and turning welfare into a bare-bones program with much state control.Footnote ⁴⁷ Further restrictions for eligibility were put on cash welfare, such as more work requirements and a five-year lifetime limit.Footnote ⁴⁸ As part of the changes, states regained control to set asset limits, but only some states made changes.Footnote ⁴⁹

Asset limitations may seem like small potatoes, but a large percentage of the population is potentially subject to them. Indeed, roughly 20 percent of the US population participates in means-tested government aid programs each month.Footnote ⁵⁰ Generally people move in and out of public assistance programs. Of those who participated in public assistance programs between January 2009 and December 2012, the largest share, 43 percent, stayed in the programs between thirty-seven and forty-eight months. Another 31.2 percent of participants stayed in programs between one and twelve months.Footnote ⁵¹

Asset limits vary significantly program by program and even state by state. For instance, the Supplemental Security Income (SSI) asset limit, which is controlled at the federal level, is $2,000 for an individual and $3,000 for a couple.Footnote ⁵² Asset limits for SNAP are set by the federal government at $2,750, and $4,250 for households containing a member who is disabled or over sixty.Footnote ⁵³ However, states can modify or even eliminate SNAP asset limits through categorical eligibility. TANF limits, now set by states, vary from $1,000 to $10,000 or even no limit, with the average around $1,000–3,000.Footnote ⁵⁴ And Medicaid asset limits again are set by states and most have not increased the dollar limit since 1989.Footnote ⁵⁵ State limitations vary significantly – there are usually none for children, but in about half the states, there are limits for families, usually between $1,000 and $6,000.Footnote ⁵⁶

For some of these programs, one vehicle (or a certain amount of vehicle value) is excluded from the asset limits, but no more than that. And treatment of retirement savings is “confusing and seemingly arbitrary.”Footnote ⁵⁷ When asset tests were first enacted in the 1970s, defined benefit plans such as traditional pensions were common, and asset limitations for some programs (SNAP, for example) did not count such accounts in the asset calculation. But today, most employees, particularly low-income ones, do not have access to traditional pension plans. Instead, if they are part of any plan at all, they tend to be part of defined contribution plans, such as 401(k)s.Footnote ⁵⁸ And if a worker is laid off from a job that includes a 401(k), that 401(k) will likely be rolled into an IRA, which then is counted. Those accounts are not exempt from most asset limitation tests for public benefits.Footnote ⁵⁹

Asset limitations help ensure that the poor remain poor. Indeed, studies have shown that asset limits create disincentives for low-income families to save. Multiple studies have confirmed that families save more when asset limitations are lower or nonexistent, and save less when asset limits are instituted.Footnote ⁶⁰ Further, to the extent that low-income families may find a work-around by saving money “under their mattress” at home, such practices are counter to the fairly well-accepted goal of trying to move more low-income people to be banked rather than unbanked.Footnote ⁶¹ But if there is a sense among low-income families that their bank accounts will be surveilled and used against them via asset limit investigations, then they are acting rationally by remaining unbanked and outside mainstream financial institutions.

It is important to understand asset limitations for what they are: a significant, though often overlooked, part of America’s structural racism that perpetuates the racial wealth gap and disproportionately leaves poor, Black, elderly Americans struggling. While in absolute numbers whites far outnumber Blacks in their receipt of public benefits, rates of participation in public assistance programs are higher among Blacks. According to the US Census Bureau, in 2012, 13 percent of white Americans participated in a public assistance program, compared to 42 percent of Black Americans.Footnote ⁶² Thus, asset limitations perpetuate the racial wealth gap by disproportionately preventing Black Americans from accumulating savings or wealth and, thus, also leave Black Americans particularly vulnerable to experiencing poverty in their old age.

10.3 Where Are We, and Where Can We Go from Here?

So where does this leave us? We are at a crossroads, no doubt. While there is a widespread belief that we do in fact take care of our elderly, that Medicare and Social Security cover people in old age, the data show otherwise. The failure of these programs to fully support the poor in old age has been well detailed by others, particularly Anne Alstott in her important book A New Deal for Old Age.Footnote ⁶³ The data are consistent with Alstott’s findings. Being old and poor is not uncommon.Footnote ⁶⁴ One study found that roughly one in three elderly Americans is economically insecure, living at or below 200 percent of the Federal Poverty Line.Footnote ⁶⁵ The data for Black and Latinx elderly Americans are even worse, with half living at or below 200 percent of the Poverty Line.Footnote ⁶⁶ Another study found that 49 percent of older workers and their spouses will experience downward mobility in retirement.Footnote ⁶⁷

Ultimately, a key problem is that our savings rhetoric largely does not mesh with our law. We pretend that we live in a society that provides structures that enable retirement savings for all, and thus, a policy of “encouraging more low-income Americans to save,”Footnote ⁶⁸ as former Treasury Secretary Lawrence Summers suggested, is a way forward. Encouraging the lowest-paid Americans to save will not cut it, in part because we do not provide realistic structural opportunities for them to do so. Yet we use moralistic overtones around savings to shift responsibility from structures to individuals, all while picking up on Reagan-era tropes about race, potential fraud, and exploitation of government programs in order to support the need for these asset limitation laws.

We must pull out of this false narrative that behavioral interventions are the most promising way forward because savings opportunities are available for low-income workers but often squandered. Once we do so there is much that can be done. Fortunately, the good ideas are abundant, but more attention to this issue is needed. During the height of the COVID pandemic, the Child Tax Credit was improved in multiple ways – increasing the amount per child, broadening the credit’s scope, and making it fully refundable. Allocation of the credit was based entirely on income, not assets. The expansion of the Child Tax Credit was not extended, but perhaps as advocates continue to work toward expansion, the ability to save for the long term should be centered in the conversation. The key problem now is that when poor families gain the opportunity to save, through an expanded social safety net, a promotion at work, or something else, this is a positive change that can help secure their future into old age. But when and if they need to access safety net programs such as SNAP or TANF (or continue receiving help from such programs), the savings will become a burden – something they will have to spend down in order to qualify or continue to qualify.

We can attack the problem in one of two broad categories of change, though the best plan is likely one that pursues both of these avenues. If we want to depend on people to create their own nest-eggs for retirement, then we need to repeal asset limitations and provide benefits and structures, early and often, that allow even our lowest-wage workers to save, and save at a rate that would support them in retirement. Given the very low wages of many Americans, the government would likely have to provide, at a minimum, matching funds, and probably even supplemental funds, in order for such a program to work and make a meaningful difference in retirement savings.

Alternatively, or perhaps in conjunction with an aggressive savings program, reforming social security to truly be progressive can eliminate elderly poverty for many Americans. Such a reform would need to take into account the reality that, at least without change, our structures prohibit many seniors from saving for retirement, and the program would need to provide extensive benefits to these seniors in order to keep them from living an old age of poverty. There are several existing comprehensive plans to reform social security, and an extensive discussion of such plans is not warranted here. However, my broader point is that in order for such plans to gain traction, we need to transform the moralistic narratives around savings that pervade our culture and have been circulating for decades. Indeed, we need to align the national conversation about savings with the realities of the structural systems we have in place – including the laws that prohibit low-income Americans from saving while accessing needed support during their working years.

10.4 Conclusion

Ultimately, only when we shift the cultural narrative, peel back the moralistic and judgment-laced rhetoric around poverty, savings, and retirement, and acknowledge existing structural barriers to savings can we then rebuild these structures and circumvent the looming elderly poverty epidemic. If we want the “golden years” to be golden for everyone, it is imperative to attend to this now. As plans and nudges to help Americans save pervade, asset limitations must be addressed in order for these programs to make a meaningful difference in the ability of low-income Americans to save for retirement.

As human lives grow longer, what will happen to the law that governs death? How will a person’s property be divided at death, and will we see any changes in the way people make wills, trusts, and plans for the distribution of their estates after they pass away? In this chapter, I argue that the answers to these questions are surprising: As lives grow longer, the law of property distribution at death will become less important. People who expect to live into their second centuries will have less property to pass at death and will feel a greater need to hand it over to their descendants and beneficiaries before they die. The law of trusts and estates will remain a bedrock feature of the legal system, but it will look more and more like a backstop and less and less like a preferred mode for dividing up the bulk of accumulated estates.

11.1 Background

The declining importance of the law of decedents’ estates promised by the lengthening of human lives will intensify a trend that has already been growing for decades. As John Langbein explained in a seminal law review article in 1988, the law of estates has been diminishing in importance since at least the middle of the twentieth century, and possibly long before.Footnote ¹ The causes are many. One is the transformation in the character of wealth. Whereas the estates of the old aristocracies prototypically consisted of land, today wealth tends to consist primarily of financial assets, such as securities, savings accounts, and insurance contracts. These forms of financial assets are much easier to transfer than land and, therefore, require less attention from lawyers and judges when they are transferred at death. Similarly, much of the wealth that passes from generation to generation in 2024 consists of investments in human capital that tend not even to register as forms of wealth transfer at all. When a professional family devotes hundreds of thousands of dollars to the private education of its children, no intervention by a probate court is needed.

Still another cause of decline in the law of probate is the growth and now dominance of forms of wealth transfer that self-consciously avoid the law of probate. Life insurance proceeds, bank accounts, retirement accounts, and brokerage accounts can all be crafted to name beneficiaries to take assets in the event of an owner’s death. When the day comes, the assets move solely by the actions of these financial institutions, with no need for a will or probate court. Even a family home can be owned jointly with a right of survivorship, so that a spouse’s death automatically transfers their share to the spouse who survives, again with no need for probate.

Estate-planning practitioners have cut further into the domain of the probate courts by mastering the so-called “revocable trust.” A revocable trust is a trust that its creator – known as a “settlor” – can take back at any time. A revocable trust can work as an excellent substitute for a will because, like a will, it does not become permanent until the settlor’s death. The settlor can add and remove assets from it as the settlor pleases during their lifetime, making no final decisions until moments before they pass away. The revocable trust offers many advantages, one of which is to avoid probate. Since, at the time of a settlor’s death, the property in a revocable trust formally belongs to the trustee rather than the settlor, the property is not part of the settlor’s estate for purposes of probate. The dispositive terms of the trust need not be filed with a probate court, and lifetime transfers to the trustee generally cannot be challenged as part of probate proceedings.

The intentional attempt to avoid probate reflects the reality that probate is often expensive and time-consuming, especially for modest estates that have little need of the benefits probate ostensibly offers. If an estate of modest size can be divided up by a series of financial institutions that merely carry out the orders embedded in financial contracts, the estate can often avoid the costly intervention of lawyers, the potential for dispute, and the occasions for taxation and fee extraction that tend to follow entry into probate.

11.2 Less Wealth at Death

The lengthening of human lives will intensify this downward trend in the importance of probate for two reasons. The first is that decedents will have less wealth to pass. For the same reasons that lengthening lives will strain systems of public retirement support, they will also strain private savings. As working years take up a smaller share of human lives, workers will have to spend more years living off of their savings. To support a period of extended retirement and prolonged illness, family homes will have to be liquidated, 401(k) accounts used up, and savings accounts eaten away. All of this will leave fewer assets to be transferred when a person passes away.

Some portion of the burden on private savings might be lightened by lengthening years of work. If people spend some of their extra years working and saving, the financial pressure of those extra years will not be as intense as if people spent all of their extra lifespans on the golf course. But there remains a real possibility that the extra years in income-earning labor will not be enough to compensate for the extra years in retirement and illness. If that happens, the size of probate estates and even nonprobate transfers will decline. A person will have to use their savings to support themselves rather than to provide for the next generation.

11.3 Greater Urgency of LifeTime Transfer

For the fortunate few who have a surplus of wealth even after their extended years of retirement, the lengthening of life will intensify the downward trend in death-time transfers for a second reason: People will feel a greater need to pass their wealth before they die. The trouble is that the age at which people conceive and bear children is not growing as quickly as the age at which people die. Many people can now expect to live past 100, but relatively few will bear children past forty. As a result, the gaps between generations, though growing, are not growing as quickly as human lives, so that as parents die at older and older ages, their children are receiving transfers of wealth at later and later stages in their lives. Think of the difference between Queen Elizabeth II of the United Kingdom, who ascended to the throne at age twenty-five after her father passed away at the age of fifty-six, and Elizabeth’s son Charles, who remained a prince into his seventies as his mother continued to occupy the throne well into her nineties. Few families have thrones to pass on to their children. But what about homes, cars, and 401(k) accounts?

This postponement of wealth transfer presents a problem. Money is often most useful when a person is young. This is partly a result of the cycles of human lives and earnings. Income tends to grow over the first five or six decades of life, so that an educated person in their twenties will tend to earn less than they will in their forties or fifties. Help with a down payment for a house is thus more useful when the person is young and has a lifetime of mortgage payments and future earnings still ahead than when they are middle-aged and have a padded savings account and a mortgage that is already half paid off. A related concern about life cycles is that a person’s propensity to consume tends to drop later in life. Extra money has little appeal for an infirm elderly person who no longer needs fashionable clothes, comfortable vacations, or extra square footage. A parent with extra resources to spare might therefore feel a greater urgency to subsidize children and grandchildren while they are still relatively young than older children who have established incomes.

Money also tends to be useful when a person is young because money has a time value. The foundation of finance is the conviction that money today is worth more than money tomorrow. All else equal, therefore, a child and his family will prefer to start consuming today than wait for another thirty years while the old family home sits underutilized by parents. Like all rational people, Junior might prefer to spend money now than later. Even if we set aside the concerns about Junior’s uneven income over the course of his life, we might still be concerned about the simple time value cost of forcing Junior and his offspring to wait to enjoy the fruits of grandparents’ savings. Some of these concerns about time value might be mitigated by the possibility of investment growth. That is, Junior might be rewarded for his patience with compounding returns on the savings and a larger ultimate inheritance upon the death of his parents. But savings may not grow at a rate commensurate with the discount factors of younger generations, especially if the savings consist of assets that are costly to maintain or that might decline in value, such as homes and cars.

A final reason to hasten the transfer of wealth is that wealth sometimes requires active management or even active consumption. Think of a family business in which a founder must transfer control to their children. A son might take over the family restaurant, or a daughter might assume the corner suite of the family investment office. In an age when retirements were short and death came early, the transfer of control over a family business often coincided with death. Junior would take over the day Father was laid to rest. As retirements and lives lengthen, death-time succession will no longer suffice. Later generations have to be given control well before earlier generations pass on. We cannot expect an elderly founder to effectively manage a business into their eighties and nineties, and we cannot expect their children to wait for them to retire when they are in their fifties and sixties.

The need for active management often mixes with a need for active consumption. Though remaining in the family home may be satisfying and comforting to an elderly person who has spent most of their adult life there, the prolonged decades of occupancy may come at a significant financial cost if the home is large enough to be used by (and sold to) a family with children and if it decays through the neglect of maintenance and renovation that often accompanies aging.

The need to shift transfers from death to life will intensify a problem that estate planners have always known and have always struggled with, which is how to balance the need for income in retirement with the need for timely transfer to the next generation. Mother might recognize that it would be good to pass on assets to Junior sooner rather than later, but at eighty-five years old, Mother might not know how many of these assets she will need to support herself for the rest of her life. If she has a million dollars in savings, she might need all of it if she expects to live another twenty years, or she might need just half of it if she expects to be gone in just another ten years. If she knows that Junior and his children could use the money now, how is she supposed to predict how much to give now and how much to hold on to? This is an old problem, and it long predates the growth of the 100-year life. But the greater strain on personal finances and the larger bands of uncertainty introduced by longer lives and retirements will make this needle even harder to thread.

The risk of passing too much too early makes it clear that what we need is not simply to smooth the making of transfers during life but also to allow such transfers to be made flexibly, so that gifts can be made and then taken back if the life and needs of the donor become unexpectedly long and great.

11.4 Challenges

What, then, stands in the way of such a system? What do we need to do to make it possible for long-lived people to pass their property before they die even as they retain enough control to take some of it back to meet the needs of an unexpectedly long life?

Many of the obstacles involve taxation. The system of unified estate and gift taxation we now have in the United States is a major accomplishment, in that it tries to treat life- and death-time transfers as part of a single pool. The details are complicated, but the basic gist is that if I make a gift during my life, I have to file a gift tax return in the year I make it, but I will owe no tax until the total of my life- and death-time gifts together exceeds the value of my exemption from the estate tax. Any excess over the exemption gets taxed only at the time the threshold of exemption is crossed, whether that happens in life or death. This system largely avoids the need to pay taxes on transfers during lifetime for any but the very wealthiest families.

Some obstacles to lifetime transfers remain, however. One is the so-called step-up in income tax basis at death. If a donor takes a piece of property in life and then holds it as it appreciates in value, the income tax basis for the donor is usually the price at which the donor purchased the property minus any depreciation. Were the donor to sell the property during their lifetime, they would owe income tax on the appreciation. But if the donor waits until their death and then transfers the property by will, the basis of the donees will equal the value of the property at the time of the donor’s death, appreciation and all. Were the donees to sell the property immediately after the donor’s death, they would owe no income tax. Lifetime transfers do not receive the same favor. The donee of a lifetime gift receives the same basis as the donor.

The trouble, of course, is that this system creates profound incentives to hold property until death. If I bought a house in Palo Alto in 1975 and watched it skyrocket in value through 2024, it would be disastrous to sell or donate the house before I passed away. Instead, the tax incentive would be to hang on to the house for all 110 years of my long life and then leave it to my children in my will. A lifetime transfer would be an income tax disaster. A death-time transfer is the only rational strategy.

One can imagine strategies for accelerating the use of the property, but none is fully satisfying. I could rent the house out and make a gift of the income to my children every year. But the income might exceed the annual gift tax exemption. Additionally, I would owe income tax on the rental income, and my children would not have immediate access to the full value of all the capital accumulated in the house.

Another obstacle to flexible and accelerated transferring is the generation-skipping transfer tax. If a 110-year-old donor hesitates to make a gift to their child who is already eighty-five years old, then the donor could achieve greater value for the gift by handing the assets to a grandchild or even great-grandchild who can make better use of the money. If the goal is to support someone who has yet to attend college, a donor can simply select a beneficiary further down the family tree. The donor can wait until death and still see the assets go to a person who is young enough to genuinely need them.

The problem with this strategy is the generation-skipping transfer tax, or GST tax. The tax is designed to ensure that assets subject to the estate tax get taxed at every generation. Congress did not want a person to be able to pass their fortune to their great-grandchildren without making the fortune subject to estate taxation at the deaths of the person’s children and grandchildren. To put it a bit crudely, the GST tax imposes the estate tax multiple times, based on the number of generations the donee sits below the donor on the family tree.

Private estate planners have developed strategies for handling the GST tax. One is the creation of a perpetual trust that is funded with assets up to the full amount of the estate tax exemption and then made perpetual so that it can benefit the donor’s children, grandchildren, great-grandchildren, and other descendants later on. Since the transfer was made only once, the trust can continue to benefit the donor’s descendants across many generations without encountering estate or gift tax.

The perpetual trust has its drawbacks, however, including the obvious governance problems that come with having something exist forever, not to mention the fact that it only works for the estate tax exemption amount, which, though relatively high right now, might easily come down in the future to levels that make life uncomfortable for a greater share of the public.

A further obstacle to lifetime giving is the tax benefits of retirement accounts. A person who contributes to a 401(k), IRA, or similar retirement account owes no income tax on the money contributed at the time of the contribution and continues to owe no income tax as the investments appreciate until the time the person makes a withdrawal. Such accounts can be leveraged to great effect, ensuring that assets appreciate over time without being periodically reduced by capital gains taxation. The effect can be compounded by continuing to hold the account until death and then handing it over to children and letting them continue to enjoy the same benefits until they are forced to make withdrawals of their own. The length of time a child can hold the account was diminished by the tax reforms of 2017, but the possibilities for savings by keeping assets inside the account after a donor’s death remain substantial. The benefits available under these retirement accounts thus encourage death-time transfers, because there is no way to hand these assets to spouses or children during the donor’s lifetime without undoing the tax benefits.

One might argue with respect to all of these tax issues that convenience and efficiency are not the only policies we must consider in taxation. We have to think about the raising of revenue, redistribution, and so on. Retirement savings, for instance, are supposed to be used to fund retirement, not to accumulate dynastic wealth, and so one might not be troubled by the possibility that lengthening lifespans may complicate the accumulation of dynastic wealth. My point, however, is that whatever our other goals for taxation might be, the lengthening of life will put pressure on them to the extent people come to prefer life over death in the timing of wealth transfers. Tax policy has several goals, but one of them is to avoid distorting people’s behavior. And to the extent tax law encourages a person to cling to property they do not need until they pass away at 110 just so their descendants can save money on taxes, it will be shaping people’s actions in ways that are not socially optimal.

In addition to posing problems for taxation, a shift toward lifetime giving will raise other challenges for the system of wealth transfer as well. In particular, one wonders about the future of life insurance as a form of wealth transfer. By its nature, a term life insurance policy does not pay out until the insured person dies. So, what if the great length of a person’s expected lifespan pushes the person to prefer a transfer before their death? Life insurance will not be a good option. People will continue to use life insurance as a protection against the prospect of an untimely death, but it will cease to be an attractive option for transferring wealth when death comes late.

11.5 Conclusion

One might think that as people live longer lives, the system of law that governs the transfer of their property at death would become more important. But the opposite is true. People will have fewer assets to transfer at death and a stronger desire to transfer them during life. People will want flexibility between life and death, and they will want the law to help them achieve it.

Americans have no right to grow old. Or so we are told. In 1985, in a case holding that the rights of people with disabilities were to receive no special constitutional protection, Justice Byron White worried in his majority opinion that people with intellectual impairments constituted so broad a class that there would be no “principled” way to distinguish discrimination against them from burdens on “the aging, the (physically) disabled, the mentally ill, and the infirm.”Footnote ¹

The clear implication was that these other burdens were self-evidently permitted. And so, for the Court, the idea that laws burdening the elderly might be worthy of special attention from judges was not just wrong but also so obviously wrong that it could serve as the lead float, as it were, in a parade of horribles. This chapter is about what US courts find so horrifying about a right to grow old, and whether they are right to be afraid.

It is important first to clear some definitional underbrush. There is, to be sure, a right to grow old in a sense. Deriving from the Fifth and Fourteenth Amendment rights not to be deprived of life without due process of law and the Eighth Amendment right against cruel and unusual punishments, Americans enjoy a right against arbitrary killing at the government’s hands.Footnote ² Should the state choose to end your life before you become elderly, it had better have a good reason for doing so. This is the right to grow old in its negative sense, as a right against certain kinds of state action. This negative form is the one US constitutional rights typically take.

Even this negative right has its limits. If the state is not ending life but is merely burdening the elderly in some other way, courts don’t take much notice. Age discrimination is barely scrutinized under US constitutional law.Footnote ³ The traditional reasons for greater scrutiny, grounded in a group’s relative lack of political power, its historical marginalization, or its vulnerability to overbroad and harmful stereotypes, can apply to the elderly with great force. US courts nonetheless tend to consider discrimination against older people to be obviously rational even when it is sloppy and harmful.Footnote ⁴

The positive version of the right to grow old, such as it is, is still more limited. In this form, it is not so much the right to life as the right to live. It is the right to the support needed to maintain an adequate quality of life up until death. The elderly face predictable challenges that arise less from choices they have made than simply because they are human, and therefore mortal. Our physical dexterity and mental acuity typically lessen as we age, sometimes to a point where they become debilitating in the way of severe physical or mental disabilities. Providing care for older Americans also imposes burdens on the professional lives of family members, and especially women. These challenges require our care, our attention, and our money. The costs of medical care and supervision needed for the elderly to retain their dignity are overwhelming. And even if the morbidity of old age is reduced in the future,Footnote ⁵ we can still expect that as Americans live longer on average, the wages they will have earned in their working lives will be insufficient to cover these costs. They – we, inescapably – will need help.

It might seem as though help is not forthcoming, at least not from constitutional law. Judges accustomed to the American way of thinking seem to resist positive rights. The US Supreme Court has been unusually clear that the Constitution protects individuals against certain state-inflicted harms but generally does not require the state to ensure a minimum quality of life or prevent or deflect the predations of private actors.Footnote ⁶ Of a piece with this narrow rendition of constitutional rights, the Court has also been clear that state actions that reliably burden even protected classes, such as those defined by race, sex, or religion, will be viewed as constitutionally uninteresting unless a victim can show that the state was trying specifically to harm the members of those classes.Footnote ⁷ The US Supreme Court’s theory of the Constitution’s protections appears to view rights as arising only in those cases in which the government is acting pathologically, not when ordinary governance happens (even knowingly) to spoil some lives along the way. This kind of constitutional austerity sweeps the legs from under any putative right to grow old.

An aging population calls for reconsideration of the constitutional status of the elderly. We can expect rational stereotypes about the capacities of older Americans, already troubling, to become ever more outdated and limiting. We can also expect inequality to rise as Americans get older; basic survival skills tend to atrophy with age, and there is reason to be skeptical that future Americans will muster the political will to redistribute wealth to make up for it.

Constitutional law can only do so much for a miserly nation, but the prospects for some contribution are less bleak than they might appear. Positive rights are more prevalent in American constitutional law than US courts typically admit. Although courts are, for the foreseeable future, unlikely to create new substantive positive rights out of the cloth of the US Constitution, positive rights are best understood not as new rights but as the positive sides of older ones. And there are US constitutional rights whose protection requires the government to take on affirmative obligations. The Supreme Court’s rejection of overt protection for positive rights has been deliberate, reflecting less an inherent feature of the US Constitution than a persistent divide in US politics, particularly around issues of race and class. Carving out some constitutional space for a right to grow old would require less a sledgehammer than a scalpel, less a constitutional invention than an exercise of judgment.

12.1 Positive Rights at the Supreme Court Bar

The Supreme Court has said repeatedly that the constitution protects negative rather than positive rights. Most bracingly, in the 1989 case of DeShaney v. Winnebago County Department of Social Services, the Court denied that a boy who had been beaten into a coma by his father had any constitutional rights against the county welfare workers who had negligently kept the boy in the father’s care.Footnote ⁸ “Our cases have recognized,” Chief Justice Rehnquist wrote, “that the Due Process Clauses generally confer no affirmative right to governmental aid, even where such aid may be necessary to secure life, liberty, or property interests of which the government itself may not deprive the individual.”Footnote ⁹

The refusal to understand the Fourteenth Amendment’s rights provisions as conferring a duty on the state to protect citizens from private injuries is consistent with a set of cases that arose in the 1970s, when the constitutional status of the welfare state remained very much in play. Thus, the Court held that a family had no right to have its state welfare benefits increase commensurate with family size,Footnote ¹⁰ that a woman had no right to a publicly funded abortion, even one that was medically necessary,Footnote ¹¹ and that kids in poor neighborhoods had no right to have their public schools funded on par with those in rich neighborhoods.Footnote ¹²

But as hostile as the Burger Court and its sequelae have been to positive rights, the constraint the Court’s body of cases imposes on the recognition of such rights is best understood as one of degree rather than kind. Positive rights are not a question of which rights the Constitution protects but rather about what the government is obligated to do in respect of those rights. Rights to food, shelter, education, health care, and other rights often associated with the positive dimension of rights do exist under US constitutional law. The Constitution does not permit the government to take food off American tables or prevent Americans from educating themselves.Footnote ¹³ When courts and commentators say that Americans lack positive rights, what they typically mean, rather, is that the federal Constitution does not obligate the government to provide these entitlements or facilitate access to them. In the language of international human rights law, the government has a duty to respect these rights, but it does not have a corresponding duty to protect or promote them.

This limitation is strict and marks the US as a global outlier, but it is not categorical. The Supreme Court has recognized some instances in which the respect the government must afford to rights extends to positive duties to facilitate their exercise. For example, the constitutional right to counsel is understood not simply as a “negative” right of criminal defendants to bring their own counsel to court but has been read to include a positive right to have the government provide counsel to indigent defendants.Footnote ¹⁴ The right to vote is not simply a right not to have one’s vote unreasonably denied but also imposes obligations upon states to create an effective infrastructure for receiving and counting ballots. The right to marry is not just a right to hold oneself out as married but includes an obligation on the state to solemnify the marriage and, therefore, make married persons eligible for certain state benefits. Even in the US, then, the fact that a right is recognized does not, ipso facto, tell us the degree to which its positive dimensions are constitutionally resonant.

The text of the US Constitution is unhelpful here. Many courts around the world protect the positive dimensions of rights, but the different approach they take to these questions is not, as many believe, because their written constitutions provide for these rights explicitly. Although most of the world’s constitutions are indeed longer and more explicit than that of the US, including when it comes to rights often characterized in positive terms, the text of the US Constitution determines neither the substance nor the degree of its rights commitments. As David Strauss has observed, the rights the Constitution protects are surprisingly disconnected from its specific language.Footnote ¹⁵ The Constitution contains no explicit rights against race or sex discrimination, no right to vote, and no free speech rights against states. The degree to which these or other rights are respected, protected, or fulfilled by US courts depends on acts of construction that could as easily apply to rights to food, health, education, and so forth.

Indeed, US courts nearly accepted this invitation in the 1960s and 1970s. Positive rights relate closely to what American lawyers call the “state action doctrine” and what lawyers in many other parts of the world describe in the language of “horizontal effect:” To what extent does the Constitution require private actors to respect constitutional rights? A constitutional ban on private forms of discrimination is functionally related to a constitutional requirement that the government protect individuals against abuse by private actors. And by the late 1960s, six justices on the Supreme Court were willing to hold that, in at least some senses, the US Constitution could be read to reach such abuse.Footnote ¹⁶ This posture was manifest in a series of cases that prevented privatization as a means of evading desegregation measures, shielded “sit-in” protesters from state trespass laws, and affirmed Congress’s power to guard against private discrimination.Footnote ¹⁷

During the same period, in cases concerning the state’s obligations toward those it seeks to prosecute criminally, the Court held that the state had to engage in prophylactic acts that lowered the risk of faulty convictions, most famously when it found that the government had to pay for a lawyer for indigent defendants.Footnote ¹⁸ In extending this last right to the appeal stage, the Court relied not on the Sixth Amendment right to counsel but on the Equal Protection Clause, reasoning that, in the government’s refusal to fund appellate counsel for indigent defendants, “an unconstitutional line [had] been drawn between rich and poor.”Footnote ¹⁹ Later, in striking down Virginia’s $1.50 poll tax, the Court again relied on the presumptive impermissibility of government distinctions on the basis of wealth.Footnote ²⁰ For courts to take seriously the right to be free of the disparate burdens of wealth might, in a practical sense, induce courts to require states to provide certain goods and services to the poor; the only alternative would be to take those goods and services away from the wealthy. An equal protection basis for such holdings could extend them beyond the criminal justice or voting contexts and into other important interests not held to have a specific constitutional basis, such as food, shelter, and health care.

The Court recognized this possibility in later cases placing limits on its incipient wealth discrimination doctrine. Pointedly, in a 1971 decision holding that indigent people can’t be denied marital dissolution based on their inability to pay court fees, the Court relied solely on the Constitution’s due-process provisions rather than equal protection.Footnote ²¹ Two years later, the Court decided that Texas could tie school funding to local property taxes, plainly disadvantaging students living in poor neighborhoods.Footnote ²² This arrangement did not count as unconstitutional wealth discrimination, Justice Lewis Powell wrote for the 5–4 majority, and the students in poor schools did not suffer so much as to implicate any fundamental right to education.Footnote ²³

The San Antonio case was, to be sure, bad news for supporters of positive constitutional rights. Wealth discrimination claims have gotten nowhere since then, and the Court appeared to deny that the Constitution mandated state support for important interests. It is important to linger here, however, because Justice Powell’s rejection of a right to education was not total. Because the state did not absolutely deny poor children’s right to a public education – they got to go to school, after all – the Court did not reach the question of whether a state had an affirmative obligation to provide a minimum level of public education.Footnote ²⁴ And the most significant case to have addressed this question since the San Antonio case, Plyler v. Doe,Footnote ²⁵ suggests that the state might indeed hold such an obligation.

Plyler arose out of a Texas policy that denied free public education to the children of undocumented immigrants. The Court struck it down. While paying lip service to the Rodriguez Court’s suggestion that public education is not a constitutional right, “neither,” wrote Justice Brennan for the majority, “is it merely some governmental ‘benefit’ indistinguishable from other forms of social welfare legislation.”Footnote ²⁶ Justice Brennan singled out “the importance of education in maintaining our basic institutions and the lasting impact of its deprivation on the life of the child.”Footnote ²⁷

Taken in tandem, the Rodriguez and Plyler cases reinforce the idea that judicial recognition of the positive dimensions of rights may be denied at retail, but cannot be denied wholesale. In thinking through how much the crack in the door to positive rights could widen in the case of rights to various forms of support for older Americans, it is necessary to examine the political economy of the right to grow old, especially in relation to rights for which the door has remained closed.

12.2 The Political Economy of Positive Rights

The cases in which US courts stifled the growth of positive rights were not random. They were tied quite directly to race and class and were the products of courts built to produce them. A potential right to grow old involves very different politics.

When Richard Nixon ran for president in 1968 and 1972, his two most significant domestic issue areas were criminal justice and busing.Footnote ²⁸ He used these issues and the racialized frame around them as wedges to try to pry working-class whites away from the Democratic Party. Criticism of the Supreme Court’s decisions on “law and order” and desegregation made regular appearances in Nixon’s stump speeches (as they were, more pointedly, in George Wallace’s).Footnote ²⁹ At the time of the election, President Lyndon Johnson’s nomination of Abe Fortas to replace the retiring Earl Warren was stalled in the Senate, and so the Warren Court was almost literally on the ballot.

The year before, Johnson had made Thurgood Marshall the Court’s first African American justice. Marshall was also, of course, the face of desegregation, and the Nixon and Wallace campaigns and their supporters frequently conflated racial integration in schools with crime. “They are afraid of street disorders and crime,” the New York Times wrote in an article on Wallace’s supporters. “They also hate provocative political demonstrators and they resent incursions of Negroes into predominantly white neighborhoods and schools.”Footnote ³⁰ The militaristic language of infiltration was used to describe violent criminals and Black schoolchildren alike. Nixon was vice president when Brown was decided, and he publicly supported President Eisenhower’s federal interventions in support of integration. Still, especially in light of Wallace’s presence in the race, Nixon suggested that the Warren Court’s desegregation rulings had been too broad and consistently emphasized themes of federalism and local political control.Footnote ³¹

These law-and-order and desegregation themes permeated Nixon’s judicial selection process. Warren Burger’s selection as chief justice and the choice of Harry Blackmun as Nixon’s second Supreme Court pick owed in part to the perception that they were “law-and-order” judges.Footnote ³² The racial undertones of Nixon’s other two selections were undeniable. Lewis Powell had presided over the Richmond School Board during its resistance to desegregation efforts, vowing that he would never favor compulsory integration.Footnote ³³ As a law clerk to Justice Robert Jackson, William Rehnquist had written a memo supporting Plessy v. Ferguson.Footnote ³⁴ (Nixon’s failed nominations of southerners Clement Haynesworth and Harrold Carswell were widely perceived as efforts to appeal to voters who disapproved of civil rights reforms.)

The racial themes of Nixon’s campaign against the Warren Court engulfed his stance on government welfare as well. The common association of “welfare” with irresponsible behavior by racial minorities is well established. Nixon’s 1968 convention speech played on these themes, accusing government welfare programs of “reap[ing] … an ugly harvest of frustration, violence, and failure.”Footnote ³⁵ The case rejecting any constitutional right to welfare, Dandridge v. Williams,Footnote ³⁶ included Chief Justice Warren Burger in its five-justice majority opinion. The cases rejecting the right of Medicaid and state welfare recipients to abortion funding – Maher v. RoeFootnote ³⁷ and Harris v. McRaeFootnote ³⁸ – both included Burger, Powell, and Rehnquist in their narrow majorities. Race was barely below the surface of the abortion issue. As Justice Marshall wrote in dissent in Harris, the abortion rate for nonwhites was about twice what it was for whites, and nonwhites were disproportionately likely to be indigent.Footnote ³⁹

The Court’s opinion in the public school funding case, Rodriguez, decided two months after Roe, had all four of Nixon’s appointees in the 5–4 majority, which Powell authored.Footnote ⁴⁰ Rodriguez was on its face a case about wealth discrimination, but of course it was also about race, on several levels. Rodriguez was a desegregation case. Demetrio Rodriguez brought the lawsuit on behalf of a group of parents of students in schools in Edgewood, a neighborhood whose public schoolchildren, due in large part to housing discrimination, was more than 90 percent Mexican American.Footnote ⁴¹ More broadly, southern states’ massive resistance to the Supreme Court’s injunction in Brown v. Board of Education that they may not racially segregate their public schools was followed by more passive resistance. Minority-to-majority transfer schemes, whereby students could freely transfer out of schools in which they were a racial minority, were struck down by the Supreme Court in 1963.Footnote ⁴² Closing the public school system to avoid desegregation was ruled invalid in 1964.Footnote ⁴³ Freedom of choice plans, which tended to replicate dual school systems, fell in 1968. Most jurisdictions ultimately settled on geographic zoning as the way to assign students to schools. The problem is that geographic zoning is vulnerable to white flight that can reproduce and reinforce racially identified neighborhoods and schools, such as in Edgewood.

The problem of white flight is stubborn enough on its own – as the busing saga demonstrated – but allowing schools to be financed through local property taxes greatly exacerbates the problem by effectively privatizing public schools. Well-financed schools attract more affluent residents, raising property values. The cycle then repeats. In this way, formally separate but formally equal schools were replaced by informally separate and formally unequal schools. This inequality, though naked, was ostensibly produced through private market ordering and was therefore invulnerable to constitutional attack.

In nearly every case in which the Burger Court conspicuously limited the respect that needed to be shown to the positive dimensions of constitutional rights – in Dandridge, in Rodriguez, in Washington v. Davis,Footnote ⁴⁴ in Maher – those rights were bound up in race and class politics.

The right to grow old is different in this respect because “[a]ll children, except one, grow up.”Footnote ⁴⁵ The challenges of aging disproportionately affect the poor and the disadvantaged, of course, but none of us are immune entirely from them. Social Security and Medicare, both of which primarily benefit the elderly, remain two of the most popular government social programs. Research has suggested that media portrayals of these programs tend to be positive and tend to depict white recipients, and indeed that whites with conservative views on race view Social Security more positively than their liberal counterparts.Footnote ⁴⁶ If the positive dimensions of social rights are ever to find favor with US courts, the rights of an aging population are more likely to be recognized than others.

12.3 Putting Age Discrimination in Context

Viewing social obligations toward older Americans as constitutionally enforceable may require a wholesale shift in cultural attitudes about duties toward others, but adopting a different posture toward rights would get some of the way there. American courts faced with rights disputes tend to load the weight of analysis on threshold legal questions, such as whether the Constitution contains particular rights, who is entitled to hold them, and what the standard of review is when those rights are abridged. Identifying rights with particular categories of deference is thought to discipline the inquiry and subject rights to the forms of analysis – interpretation of written texts, historical understandings, and legal precedents – that judges are competent to perform.

These questions are abstractions, deliberately so, and tend toward a style of slippery-slope reasoning familiar to US lawyers. To those socialized into American rights review, declaring a “right to grow old” seems to entail judges attempting to require the government to fund elder care, to make medical and drug payments, to make its buildings and services more accessible to people with physical and intellectual challenges, to impose certain accommodation requirements in workplaces and public-facing businesses, or even to restructure hiring practices or pension guarantees to enable older Americans to remain gainfully employed. A right to grow old might also seem to require a reassessment of the bounds of the category itself. If there is a right to grow old, why not a right to be sick; to be an addict; to be obese, or short, or tall; or to be less intellectually agile? What accommodations, and at what cost, do people in these conditions require of the government? Where does it end? Not knowing the answer to this question makes a right to grow old seem either unwieldy or lawless. There can be no such right. Right?

Many of the world’s courts approach the positive dimensions of rights differently. They do not refuse to recognize them for fear of slippery slopes, chaotic decision-making, or empowering too many litigants. They also, and relatedly, do not assume that recognizing the need to protect or promote constitutional rights requires them to do so fully or immediately.

In thinking through what particularized, contextual decision-making might look like in the area of age-related burdens, consider two roughly contemporaneous Canadian cases decided under the Charter of Rights and Freedoms. In the first, McKinney v. University of Guelph, private universities in Ontario were sued by professors and a librarian over the schools’ mandatory retirement age of sixty-five.Footnote ⁴⁷ In the second, Tetreault-Gadoury v. Canada (Employment and Immigration Commission), a woman who lost her job just before turning sixty-five sued over being denied unemployment benefits, which were unavailable to those eligible for retirement benefits.Footnote ⁴⁸

Either of these cases would be straightforward losses under the US Constitution as interpreted by the Supreme Court. Never mind that the University of Guelph is a private institution and therefore immune to constitutional attack under US state action doctrine (and, it turns out, under Canada’s as well). The key point is that age discrimination receives rational-basis review.Footnote ⁴⁹ Any conceivable rational reason for age discrimination passes constitutional muster, and there will always be conceivable rational reasons for age discrimination. This is particularly so in the context of social welfare programs designed to address problems that typically arise at particular life stages.

The particular challenged provisions in McKinney and Tetreault-Gadoury had perfectly rational justifications. Mandatory retirement requirements cannot be viewed in isolation from other labor and welfare policies. Often they are the products of delicate collective bargaining or implicit trade-offs that include compensating pension or job security guarantees (including, in the case of a university, a tenure system). These requirements can create job opportunities for younger workers in tacit exchange for requiring those workers to help fund the retirement of their predecessors. In certain settings, moreover, the requirement obviates the need to dismiss older employees for performance reasons, which can be more acrimonious and dignity-effacing than mandatory retirement. The restriction of unemployment benefits considered in Tetreault-Gadoury was designed to prevent people over the age of sixty-five from receiving both pension and unemployment benefits at the same time and to prevent manipulation of the Unemployment Insurance Act by those who were planning to retire.

Unlike in the US, age discrimination is specifically prohibited under the Canadian Charter of Rights and Freedoms.Footnote ⁵⁰ But rather than decide either that age discrimination was so constitutionally offensive that only the most compelling of reasons could permit it, or that age discrimination was easily justified on any conceivable rationale, the Canadian courts carefully examined the justifications themselves. Doing so enabled the Court to come out differently in the two cases based on relevant differences between them.

The McKinney Court acknowledged that the mandatory retirement policies at issue constituted age discrimination but held that the discrimination was justified under section 1 of the Charter, which says that the Charter’s rights guarantees are “subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.”Footnote ⁵¹ Under the proportionality test the Canadian Supreme Court applies in rights cases, this inquiry requires the Court to assess the importance of the state’s objectives, the fit between those objectives and the challenged practice, the necessity of the challenged practice in light of available alternatives, and the burden of the practice in relation to its benefits. Here Justice La Forest, writing for the majority, agreed with the universities’ submission that mandatory retirement enables faculty renewal and bolsters tenure protections, which enhance academic freedom, and that these objectives could not easily be achieved without the policy.Footnote ⁵² The Court recognized that the important questions were not about whether there was age discrimination, which there obviously was, but whether it was justified in a way that respected both the rights of individual employees and the needs of an important societal institution.

The Court came out differently in Tetreault-Gadoury, which Justice La Forest also wrote. But the contrast with McKinney was not in whether particular rights were recognized or whether a particular standard of review applied. In both cases, the policy before the Court contravened section 15 of the Charter, which prohibits age discrimination. In both cases, the Court reached that view independent of whether the discrimination was intentional or invidious. Under the Charter, discrimination is actionable even if it is indirect or based solely on the adverse impact on a population.Footnote ⁵³ In both cases, moreover, the objectives behind the policy were legitimate and important. The difference was that, in Tetreault-Gadoury, unlike in McKinney, the Court believed those objectives could be achieved without burdening the rights of sixty-five-year-olds. There was no evidence that sixty-five-year-olds were any more likely than others to abuse the system by seeking unemployment benefits for which, by not actively looking for work, they were ineligible.Footnote ⁵⁴

The Supreme Court of Canada differentiated cases on their facts, on the margins, rather than by putting them into separate legal categories. It was able to both recognize and discipline a right against age discrimination without abstracting away from the context of the cases. The next section generalizes this strategy and discusses others that enable courts around the world to do the same.

12.4 Strategies

There are at least five distinct strategies that courts can deploy to discipline a right to grow old without extinguishing it: (1) proportionality, (2) polycentric constitutionalism, (3) a “minimum core” approach, (4) progressive realization, and (5) remedial flexibility. I briefly discuss each strategy below.

12.5 Conclusion

The right to grow old is a tall order in the US, but it is not an impossible one. The inevitability of aging means that such a right likely faces fewer obstacles than other rights that have significant positive dimensions. It does not require the invention of new substantive rights – indeed, it is already constitutionally protected. It does, however, require a recognition of governmental duties of protection and fulfillment of rights that are not typical (though they do exist) in US constitutional law. The anxiety US courts face about such duties could be soothed by incorporating elements of proportionality analysis, by giving deference to political bodies acting to protect the rights of older Americans, and by availing themselves of some of the tools of incremental or partial rights realization that have become familiar within international human rights law.