On September 9, the House Ways and Means Committee took a potentially historic step that could help millions of Americans to build financial security.

Under the leadership of Chairman Richard Neal (D-MA), the Committee included a provision in the budget reconciliation package that would require employers that do not offer a retirement plan to automatically enroll their employees in IRAs or 401(k)-type plans. If passed, the legislation would implement the Automatic IRA, first proposed 15 years ago by the Brookings Retirement Security Project (RSP) and the Heritage Foundation with bipartisan sponsorship and the strong support of AARP and other groups. In addition to the Automatic IRA, the Ways and Means retirement legislation includes other provisions and options that would expand automatic contribution plans and arrangements.

The Automatic IRA would help roughly 55 million people build savings that would supplement their Social Security benefits. Employers with more than 5 employees that do not offer another retirement plan would act as a conduit enabling their employees to save a pre-set amount from each paycheck that would be invested in privately-managed, low-cost financial assets. Self-employed and gig workers could also save this way.

Employees would have complete control and could opt out entirely or change their contributions or investments at any time. Meanwhile, employers would not contribute or have any fiduciary liability or any of the responsibilities of sponsoring a plan. Experience with an Automatic IRA program in Oregon shows that, for employers, implementation is easy, usually involves no out-of-pocket costs, and can also spur the adoption of 401(k) plans.

While waiting for Congress to act, other states have also started Automatic IRA programs. Besides Oregon’s program, Illinois and California have active programs, and six additional states and two major cities are in the process of implementing Automatic IRAs. Their experience shows that Auto IRAs work. In just the three currently operating programs, almost 400,000 people at over 37,000 businesses have so far saved almost $325 million. A national program would vastly increase those numbers and would include tax credits for participating employers.

The Automatic IRA would help roughly 55 million people build savings that would supplement their Social Security benefits.

In addition to helping millions of Americans to improve their retirement security, the Automatic IRA has many other benefits. Because people who currently do not have a workplace retirement program are disproportionately Black and Hispanic, some experts believe that the program could reduce the racial wealth gap. And since the Automatic IRA uses a Roth IRA, savers could withdraw their contributions without penalty in the event of a financial emergency. There is evidence that having and using emergency savings strengthens household finances over a longer-term, which may enable them to save even more for retirement over time.

The Ways and Means legislation also includes an expansion of the saver’s tax credit for lower- and moderate-income individuals who save in employer plans or IRAs.  The existing credit, which is seriously under-utilized, would become up to a 50% match deposited into savers’ retirement accounts and become refundable for savers without income tax liability. These changes, long advocated by RSP, have support in the Senate under the leadership of Senate Finance Committee Chairman Ron Wyden (D-OR).

Both the Automatic IRA and the expanded saver’s credit have a long way to go before they are enacted, but both have the potential to increase retirement security for millions of Americans.

The Brookings Institution is financed through the support of a diverse array of foundations, corporations, governments, individuals, as well as an endowment.  A list of donors can be found in our annual reports published online here. The findings, interpretations, and conclusions in this report are solely those of its author(s) and are not influenced by any donation.

      

Every new retirement saver starts with a small account. Over time, balances can grow with continuing contributions by savers or employers, investment earnings, and tax benefits. Not all accounts, however, grow very much. Some account balances are cashed out early, while others are eaten away by administrative and management fees. In far too many cases, employees lose track of their past accounts. These situations make retirement planning more difficult and endanger retirement security for millions of households.

As we show in a new paper, the existence of accounts with small balances is an inevitable byproduct of retirement systems, like those in the United States and several other countries, where individualized, employer-based accounts and automatic enrollment provisions are widespread. And of course, not all small balances are undesirable. New savers and those with lower incomes will inevitably have smaller balances, and small retirement benefits are preferable to no benefits.

Nevertheless, public policies that enable people to navigate the problems that small accounts create could help millions of households save more adequately for retirement. Increasing people’s retirement income by just $1,000 a year could also help states and the federal government save several billion dollars that they otherwise would have spent for retiree support programs. However, the problems raised by small accounts are not the same as those raised by inadequate saving. For example, consolidating all of one’s small accounts into one larger account may not be sufficient to generate adequate retirement wealth. On the other hand, those with high income replacement rates from Social Security—that is, lifetime low earners—might have adequate retirement income despite having only a small private retirement account. There is also an important equity component to addressing small accounts, as they are especially prevalent among Black and Hispanic/Latino IRA holders.

Our paper addresses issues raised by small accounts and proposes a range of solutions: extensive reform of the rollover and account consolidation rules, including seven specific changes; reform of the saver’s credit to make it a 50 percent, refundable matching contribution deposited directly into the saver’s account; creation of a national dashboard where it would be easier to find lost accounts and/or a default account consolidation standard that could see the automatic combination of certain accounts; and development of a system where each worker could have a single retirement savings account that moves from employer to employer over the course of their career.

All these reforms will face challenges before they can be fully implemented. Some would fit more easily into the existing system than others. But policymakers should seriously consider all five as ways to reduce some of the complexity and confusion that retirement savers face.

      

Over the past four decades, the U.S. retirement system has largely shifted from defined benefit pension (DB) plans to defined contribution (DC) plans—mainly 401(k) plans—and individual retirement accounts (IRAs). A key factor driving this change is employers’ desire to avoid the risks associated with providing guaranteed pension benefits. This guarantee—a defining feature of DB plans—can entail large and unpredictable changes in funding obligations, which can wreak havoc on corporate balance sheets and budgets. But the flight from DB plans to 401(k)s and IRAs did not make financial risks disappear; instead, it transferred the risks to individual workers, many of whom are ill-equipped to handle the resulting contingencies.

In a new paper, we explore the potential for collective defined contribution (CDC) plans to help rethink risk sharing between employers and individuals and among savers and retirees. CDCs and other hybrid retirement plan formats combine DB and DC elements in different ways. CDC plans generally let employers avoid the funding volatility and investment risk of DB plans while reducing the financial risks individuals face in many DC plans. Although CDCs are technically DC plans, they look largely like DB plans to savers and retirees. Compared to 401(k) plans, which feature individual accounts, participant-directed investing, and typically lump-sum payouts, CDCs share investment and longevity risk among employees and retirees, providing DB-style pooling of investments, professional investment management, and lifetime retirement income. They do not, however, provide guaranteed benefits. Instead, benefits can be adjusted or required contributions increased if necessary to reflect plan funding and investment performance.

Similar plans already exist in the Netherlands (where they are called “defined ambition” plans) and Canada, are receiving serious consideration in the United Kingdom, and have counterparts and close parallels in the United States.

In the paper, we examine the opportunities and challenges associated with implementing CDCs in the United States. We highlight several issues that CDC plans must confront, regarding expectations, equity, transition, and trust. We conclude that, under appropriate circumstances and contingent on addressing those issues, adding particular CDC features to either a conventional DB plan or a 401(k) has the potential to improve outcomes for workers, retirees, and employers. More generally, we emphasize that evaluations of CDCs depend greatly on the answers to two questions: “Compared to what?” (e.g., traditional DB plans or 401(k) plans) and “From whose point of view?” (e.g., employees, retirees, or employers).

While they face significant issues, CDCs and similar approaches that transcend a strict adherence to traditional DB or 401(k) plan designs can help improve retirement security. Looking beyond the conventional, traditional DB and DC plan designs to explore a new, richer, and more nuanced array of risk-sharing and pooling strategies is a welcome development that will help identify more optimal allocations of financial risks and retirement benefits.

      

In the recently released book “Wealth After Work,” RSP tackled a wide range of issues concerning how to make 401(k)s and other retirement plans work better for the vast majority of workers. Two new reports drill down on specific topics in retirement policy.

The first report looks at how, in an employer-based system increasingly using individual accounts—such as 401(k)s and IRAs—and automatic enrollment, many savers have been accumulating multiple small retirement accounts. This can lead to high administrative fees relative to balances, early withdrawals, and lost accounts;  the report proposes a number of policies to address these problems.

The long-term shift to these individual account plans imposes significant financial risk on individual workers. The second report explains how collective defined contribution plans offer a way to pool those risks and balance them against employers’ desire to limit volatile funding obligations. While these types of plans face significant challenges, some of them show promise and are gaining traction in several other countries.

Viewers submitted questions for speakers via email to events@brookings.edu and via Twitter with #FutureofRetirement.

      

Some criticisms of the multiemployer pension rescue are a bit far-fetched.

In the $1.9 trillion American Rescue Plan Act, Democrats inserted an estimated $86 billion to prevent the collapse of distressed multiemployer pensions and preserve the retirements of some 1,500,000 workers and their families. As I have written elsewhere, this was the first time that a major pension law wasn’t bipartisan: after years of failed attempts at bipartisan compromise, Democrats acted unilaterally. That, in turn, outraged both Republicans (many of whom oppose “union pension plans”) and those who want reforms in pension funding practices and oversight.

Opponents complain that long-discussed reforms in pension funding and oversight were all dropped from the “Butch Lewis Act” (named after a union leader whose widow’s pension would have been devastated). They omit mentioning the reforms were abandoned in order to get a majority that complied with the Senate’s procrustean budget rules. Anticipating dire consequences from the reform-absent rescue, they warn that—if plans don’t learn from the past and Congress doesn’t step in sometime —they could run into trouble again decades from now when the $86 billion is spent.

Even for pension experts, the possibility that plans might get into trouble decades from now isn’t much of a crisis, so opponents are getting creative, coming up with catastrophic scenarios they claim are both real and more immediate.  These claims don’t hold up very well (though explaining why gets a bit nerdy).

Will the rescue lead to mass withdrawals?  No.

In an article published by the Pension Research Council at the University of Pennsylvania’s Wharton School, Aharon Friedman, a former Republican Treasury official and House staff member, argued that the rescue bill won’t actually result in a rescue; instead, employers will withdraw from these plans, causing them to fail anyway.

Mr. Friedman’s analysis focuses on the fact that the financial penalties for withdrawing from such plans will be dramatically reduced by Butch Lewis. These penalties are based in part on the amount of plan underfunding. The $86 billion will reduce that underfunding, so he believes employers will take advantage of the reduced penalties and flee critical/distressed plans. His analysis, in my view, misses several important points:

• The real factor keeping most employers in plans was and is that a union must agree to withdrawal. Prior to passage of the law, participating employers faced the prospect of steadily increasing underfunding and could make a case for being allowed to withdraw. A major employer, UPS, was able to demand and get agreement from a union to withdraw by making new commitments outside the plan.
• The Butch Lewis Act completely changes this calculus: while it is true that financial assistance will reduce the calculated withdrawal penalty, it also eliminates the near-term risk of plan insolvency. These effects reduce both the employer’s need for withdrawal and the union’s willingness to allow withdrawal.
• It’s also worth noting that the likeliest exit from multiemployer plan—conversion to non-guaranteed alternative plan designs under the proposed GROW act—was excluded from the law as enacted.

Is this a “prelude” to a federal bailout of underfunded state & local plans?  No.

Andrew Biggs, an economist who has both written about pensions and is dealing with them as a member of the board appointed to oversee Puerto Rico’s financial crisis, raised a different objection: He claims in a Wall Street Journal piece that Congress’s decision to spend $86 billion to rescue private multiemployer plans is a “prelude” to a federal bailout of underfunded state and local pension plans. This, too, is exceedingly unlikely, for several reasons. First, private multiemployer pensions are regulated by the U.S. Department of Labor and have been federally guaranteed for 40+ years by the Pension Benefit Guaranty Corporation. Congress is unlikely to walk away from pensions it has guaranteed for decades. Public pensions, by comparison, have been neither regulated nor guaranteed by the federal government. Even if the legal objections to federal intervention were overcome, it’s hard to see how Congress, which took over a decade of negotiation to come up with $86 billion, could ever come up with the $1 to $4 trillion that is the public plan shortfall.

Will Congress let plans get into trouble again?  Probably not.

In his post, Mr. Friedman describes another possible disaster scenario under which, decades from now, plans that continue to use optimistic funding assumptions again face insolvency. The current law prohibits any assistance beyond 10 years, so such a scenario would require both (a) that pliant future congresses make no reforms in the decades ahead and (b) instead pass a new law that ignores history, rewards questionable behavior, and reopens the federal checkbook. History suggests the opposite result: Congress legislated multiemployer reforms in 2006, years before even attempting a bailout.

It’s easy to understand why Republicans who dislike unions, pensions, union pensions, or unilateral action dislike the Butch Lewis Act. It’s also easy to understand why pension reformers who want to stop the use of the optimistic investment return assumptions that promote underfunding are unhappy, too.

However, claims that the Act itself will cause disasters are far-fetched. What the Butch Lewis Act will do, however unartfully, is preserve more than a hundred distressed pension plans—some 1,500,000 current and future retirees and their families will sleep easier because of it.

      

Fifteen years ago, this week, we proposed the Automatic IRA as a way to boost retirement saving among the multitudes of American workers – now numbering roughly 55 million – who have no retirement plan at work. Since that day, this Retirement Security Project initiative has helped over a quarter of a million people start building a more secure future through the three operating state-facilitated Auto IRAs now in existence.

From the start, the Auto IRA has been nonpartisan. It was created as a collaboration between researchers at Brookings and the Heritage Foundation. In short order, Republicans and Democrats on both congressional tax-writing committees co-sponsored the Auto IRA, and in 2008, both the Obama and McCain Presidential campaigns endorsed it.

Auto IRAs are simple. They combine automatic enrollment with payroll deduction IRAs. Using inertia to promote saving, individuals are in total control of whether and how much to save and how to invest; but unless they choose otherwise a pre-set amount from each paycheck is regularly contributed into a pre-determined, privately managed investment.

Employers with more than ten workers and that do not offer a 401(k) or pension would have to allow employees to use their payroll system to save in an IRA but have only minimal other responsibilities or obligations. Simplicity matters because most Auto IRA employers would be smaller firms that are less likely to have a separate human resources department.

Although the Auto IRA was structured to be a uniform, nationwide program, congressional action on Auto IRAs stalled due to broader political divisiveness. Instead, states took up the cause. The resulting state-based Auto IRAs – now operating in Oregon, Illinois, and California – have resolved any questions about the Auto IRA’s feasibility or practicality.  Four other states and the City of Seattle are also preparing to implement similar Auto IRA programs; more states will likely follow.

Already, about 276,000 people, participating through more than 30,000 employers, have contributed nearly $170 million in the three active state Auto IRA programs, two of which are still phasing in.[1] And the programs are growing: even during the pandemic year of 2020, the number of active Auto IRA savers increased 142 percent, contributions nearly tripled, and the number of participating employers increased nearly 50 percent.

About two-thirds of employees offered an account make contributions despite typically having moderate or lower incomes.  The median contribution is about 5 percent of income, and participants often qualify for a tax credit that can further increase their balances. While their accounts start small, they will grow over time and can be a meaningful source of income in retirement.

As more states and municipalities implement Auto IRAs, interest grows in a national Auto IRA. The need is increasingly obvious: Tens of millions of American workers still have no way to save for retirement at work using payroll deduction. A national Auto IRA, building on and incorporating the state-based Auto IRAs, can change that.

[1] OregonSaves, CalSavers, and Illinois Secure Choice

      

In the United States, as in many other nations with aging populations, policymakers have embraced the notion that most individuals can (and should) extend their working years. The large majority of Americans approaching retirement will not have enough income to maintain their preretirement standard of living. During the past 30 years, private workplace pensions have collapsed, and the United States, like other nations, has effectively cut public pension benefits by raising the retirement age. Working longer is widely proposed as the best way for older people to boost their fragile retirement security (e.g., Maestas and Zissimopoulos 2010; Munnell and Sass 2009; Wise 2017).

Conversations about how to promote working longer—in the sense of remaining in paid work beyond traditional retirement ages—often begin with a deceptively simple question: How can older Americans be encouraged to delay retirement? However, there are at least two embedded assumptions when we equate working longer with choosing to delay retirement: first, that older Americans have jobs from which to retire; and second, that older workers choose the timing of their retirement. Both assumptions were problematic even before the COVID-19 pandemic. They are even more problematic now.

Connecting the dots between job quality in middle age and working longer requires a life course perspective. This perspective bridges a common divide in U.S. research on the labor force. A growing body of research examines what is driving older workers’ retirement decisions, including the adequacy or inadequacy of Social Security benefits, individual wealth, health and health insurance, spouses’ retirement decisions, and labor demand (Coile 2015, 2018), but this research typically excludes those who are already out of the labor force in their late 50s or early 60s. A separate, large literature examines changes in labor force participation among men and women in prime ages, conventionally defined as ages 25 to 54 (Binder and Bound 2019; Goldin and Mitchell 2017). Bringing these two lines of research together calls attention to the relationship between employment during the prime years and an individual’s chances of working later in life. As the pandemic throws long-term prospects for workers in their 40s and 50s into jeopardy, a life course perspective on working longer will become even more crucial.

When we look at the scale of labor force nonparticipation among Americans in their 50s, especially those with lower levels of education, it becomes clear that we need to rethink substantially what ages we have in mind when we talk about extending working lives. In the American policy conversation about working longer, age 67 (the Social Security full retirement age for those born in 1960 and later) and age 70 (the age at which maximum Social Security benefits become available) stand out as benchmarks. But the issues with working longer begin much younger for many workers—earlier by a decade or more for many.

Over the long run, policymakers should aim for a labor market in which workers of all ages, including those without college degrees, have access to decent jobs that offer adequate pay, sustainable schedules, and the flexibility for employees to care for themselves and their families. We would expect such a labor market to produce lower rates of job turnover for lower-wage workers in their 50s. Lower turnover would reduce the number of people who retire early because they have lost a job and cannot find another suitable one. Longer and more-stable employment histories, in turn, produce higher Social Security retirement benefits and retirement savings. Better jobs would increase both the likelihood and the value of working longer.

Read the paper here for a complete list of potential solutions.

The author did not receive financial support from any firm or person for this article or from any firm or person with a financial or political interest in this article. She is not currently an officer, director, or board member of any organization with an interest in this article.

      

Observers of older workers’ economic conditions in the United States quickly encounter a paradox. On the one hand, the labor force participation rate among workers aged 55 and older (hereafter “older workers”) is roughly half the rate of prime-age workers (aged 25 to 54).^[1] Their employment rate is also far lower.^[2] On the other hand, many older workers have insufficient savings to continue their preretirement standard of living in retirement.^[3] The paradox is that a large portion of older workers, including many without sufficient retirement savings, are not engaged in the principal activity that facilitates more retirement savings: continued employment.

The consequences of this paradox become more ominous when put in the context of a yawning pension gap that has widened over the past 45 years. Social Security benefits, on average, replace only about 40 percent of preretirement income.^[4] So, older workers seeking to sustain their preretirement lifestyles need more retirement income. Employer-provided retirement plans are one important source of that income; however, these plans have changed in a way that makes it more difficult for older workers to ensure they will have adequate retirement income. Employers have largely migrated from defined-benefit (or pension) plans, which provide regular, reliable payments to retirees in predictable amounts for the rest of their lives, to defined-contribution plans, like 401(k) plans, which do not. Defined contribution plans merely give participants the opportunity to save amounts of money they choose (influenced by capped tax protection) that may be augmented by an employer match. Participants invest their own savings, either actively or by default, and bear the risks, including the risks of outliving their money or precipitous market declines like those associated with the recent coronavirus pandemic. Most important defined-contribution plans do not guarantee sufficient savings or produce a predictable and protected lifetime income stream, like a pension would.

In 1975 about 32 percent of employees in the United States (about 29 percent of the total U.S. civilian labor force) had pension plans. In 2016 slightly more than 9 percent of active employees (9 percent of the total workforce) had pensions.^[5] If American employers had provided pensions to the same percentage of their employees in 2016 as they had in 1975, almost 51 million employees in the United States would have had pensions—roughly 37 million more workers than have pensions today. This is the pension gap.

The need for a reliable lifelong income supplement to Social Security is apparent to anyone planning for retirement. One obvious hedge against the risks created by the lack of protected lifetime retirement income would be for older workers to earn and save more by continuing to work later in life. Yet that has not been their response, at least to scale. Older workers’ labor force participation rate grew from the mid-1990s until the Great Recession,^[6] but not enough to fill the pension gap, much less to help the majority of workers who could not have expected to receive pensions even when they were most common.^[7] For example, from 1998 to 2018 the number of older workers participating in the labor force grew by only 9.6 million^[8]—roughly one-quarter of the number of workers trapped in the pension gap. Intriguingly, the steepest decline in labor force participation among older workers occurs around age 65 when the day-to-day economics of retirement should be apparent.^[9]

We are left with the question begged by our paradox: Why have many millions of older workers not continued to work or returned to work so they might earn more and save adequately for retirement?

This paper will argue that older workers act rationally when they exit employment and the labor force because they are escaping employment discrimination that significantly reduces the economic returns from employment and labor market participation. In this paper, “employment discrimination” means any decision by an employer—for example, hiring, discharge, compensation, promotion, training, and discipline—that disadvantages an older worker, whether intentionally or unintentionally, “because of such individual’s age.”[10]

Section I will describe how statutory protections against discrimination for older workers in the Age Discrimination in Employment Act (ADEA), including as interpreted by the Supreme Court, are weaker than protections in Title VII of the Civil Rights Act against race, sex, and other forms of employment discrimination. It also will discuss how problems with the interactive process for satisfying the Americans with Disabilities Act’s (ADA) reasonable accommodation mandate erect a barrier to the employment and labor force participation of older workers with disabilities. Section II will cite evidence that large majorities of older workers perceive that they face employment discrimination (“perceived discrimination”). This section also will describe evidence that these perceptions are often reality—that is, there is widespread actual discrimination against older workers facilitated, in part, by weak age and disability discrimination laws. Perceived discrimination and actual discrimination combine to pressure older workers to abandon their search for employment and exit the labor market.

Section III proposes four solutions: (1) increase the effectiveness of laws protecting older workers from employment discrimination; (2) prohibit employers from mandating their employees engage in pre-dispute arbitration of age and disability discrimination claims rather than litigating them in state and federal courts; (3) require employers to disclose the number and results of their interactive processes with older workers with disabilities regarding workplace accommodations; and (4) enact comprehensive labor law reform so that more older workers will be protected against discrimination by collective bargaining agreements and union representation.

For a complete list of solutions, read the full report here.

The author did not receive financial support from any firm or person for this article or from any firm or person with a financial or political interest in this article. He is not currently an officer, director, or board member of any organization with an interest in this article.

      

Americans are living longer, are healthier at older ages, and increasingly are working beyond the traditional age of retirement. While many who work until late in life do so to stay active and connected or for other nonfinancial reasons, others work out of financial need. Owing to a variety of factors including changes in the structure of private retirement benefits, an increase in the eligibility age for claiming full Social Security benefits, and stagnant wages in recent decades for those at the bottom and middle of the earnings distribution, a large share of older Americans lack adequate savings for retirement.

At the same time, the U.S. economy has become more reliant on older workers. Reflecting not only the increased labor force participation of older workers but also, and more importantly, the aging of the baby boomer generation, today nearly a quarter of the labor force is age 55 and older, an increase of 12 percentage points since the mid-1990s.

A significant challenge to continued employment at older ages is that workers often must change jobs late in life. Particularly for those in manual jobs, the physical demands of their work may become too great, leading them to need to change the type of work they do. Older workers also may need or want to reduce their work hours, or they may lose their job and need to search for new employment, possibly in a different industry or occupation. The rapid pace of technical change and globalization could make such displacement more common in the future.

Yet, finding new work is especially difficult for older adults. Age discrimination in hiring is widespread and can discourage older Americans from searching for work. Older workers’ skills may be outdated, making them less attractive to employers. And, it may have been many years since they last looked for a job. Because the application and hiring process has changed, they may not know how to go about searching for a new job. During the exceptionally strong economy that prevailed in 2019, on average 1 million Americans age 55 and older were unemployed and another 1.5 million who were counted as out of the labor force nonetheless said they wanted to work. Permanent job losses during the current recession have greatly exacerbated the problem.

The federal-state workforce system is the main policy vehicle for helping the unemployed find new jobs. Yet, this system often fails to meet the needs of older adults. In contrast to the situation for jobless youths, another group with needs that are notably different from those of prime-age adults, no special programs exist to serve older job seekers. Moreover, the measures used by the U.S. Department of Labor to evaluate state workforce agencies’ performance create a disincentive to provide services to older workers. Especially given today’s high level of unemployment, the large share of affected workers who are older, and the inadequacy of many older adults’ retirement savings, the need to address these shortcomings is urgent.

We propose seven relatively low-cost reforms that will improve the workforce services provided to older adults and can be implemented quickly. These proposals include:

1. Having specialized staff at job centers who understand older workers’ needs and who can serve them more effectively,
2. Experimenting with job placement programs specifically for older workers,
3. Promoting self-employment among older adults,
4. Providing targeted skills development for older workers,
5. Adopting separate program performance standards for older adults to eliminate disincentives for the provision of services to this population
6. Restoring funding to the Senior Community Service Employment Program (SCSEP, a program that serves disadvantaged older adults), and
7. Evaluating the new programs and initiatives we recommend to ensure they have the intended effect.

To provide motivation and background for these recommendations, we begin by discussing employment trends among older adults, the difficulties older workers face in transitioning to new employment, and other employment barriers faced by this population.

Read the full paper here.

The authors did not receive financial support from any firm or person for this article or from any firm or person with a financial or political interest in this article. Neither author is currently an officer, director, or board member of any organization with an interest in this article.

      

One key policy imperative in response to the aging of the U.S. population is to increase employment among older workers. There is ample evidence of age discrimination, especially in hiring, that can impede policymakers’ efforts to encourage or induce older people to work longer, and that inhibit working longer more generally.

The United States has federal legislation to prohibit age discrimination—the Age Discrimination in Employment Act (ADEA), passed in 1967. The ADEA has many parallels to antidiscrimination laws based on race, sex, ethnicity, and religion enshrined in Title VII of the Civil Rights Act of 1964, although the ADEA differs slightly in recognizing that sometimes age can play a legitimate role in labor market decisions and outcomes. For example, the ADEA originally allowed mandatory retirement to continue, while pushing the age from 65 to 70, before eventually prohibiting it for most workers. The other major federal antidiscrimination legislation is the Americans with Disabilities Act (ADA), passed in 1990; the ADA extends discrimination prohibitions to people with disabilities, as defined by the federal law. In addition, many states have their own age (and disability) discrimination laws that can strengthen federal laws (see, e.g., Neumark and Stock 1999, 2006). While the ADEA is obviously most relevant to the question of discrimination based on age, the interrelationships between all these antidiscrimination laws are potentially important in protecting older workers from discrimination, for reasons I explain in the paper.

There is evidence that both federal and state age discrimination laws are effective at boosting employment of older workers. More specifically, there is evidence that enacting the ADEA (and earlier state age discrimination laws) improved labor market outcomes for older workers, and more-contemporaneous evidence that stronger age discrimination laws in some states have increased the effectiveness of policies intended to induce workers to continue working until an older age. At the same time, some recent court rulings have weakened the ADEA.

Although age discrimination can potentially affect many margins of the employment relationship, this paper focuses on age discrimination in hiring for a number of reasons. First, most people do not just retire from their career job, but instead take bridge jobs or move to partial retirement, sometimes in part due to declining health (Cahill, Giandrea, and Quinn 2006; Johnson 2014; Johnson, Kawachi, and Lewis 2009; Maestas 2010). Partial retirement will surely become even more common if people work until older ages, since many older workers move from their career jobs to jobs that are less physically demanding. Second, the most rigorous evidence we have establishing age discrimination concerns discrimination in hiring. Third, both empirical evidence (Adams 2004; Neumark and Stock 1999) and the workings of age discrimination enforcement suggest that, at present, the ADEA is more effective at reducing age discrimination in terminations than in hiring (Neumark 2009). And fourth, there appears to be more scope for policy to reduce age discrimination in hiring, especially in light of recent court rulings.

These considerations lead me to propose the following steps to reduce age discrimination in hiring to help confront the challenge of population aging:

1. Increase damages under the ADEA to match the larger damages that some states allow.
2. Amend the ADEA to clarify that disparate impact claims are allowed in cases of hiring discrimination.
3. Amend the ADEA to clarify that the standard for establishing discrimination is not “but for” age.[1]
4. Amend the ADEA to allow intersectional claims, in particular those regarding discrimination against older women.
5. Extend affirmative action for federal contractors to older workers.
6. Consider closer integration of the ADEA and the ADA.

My proposals can be viewed as complementing other recent proposals in the New Approaches to Retirement Security project. Clark and Shoven (2019) focused on strengthening labor supply incentives in Social Security and Medicare; those incentives are usually related to changing policy to increase the effective posttax wage of older workers. Munnell and Walters (2019) suggest stronger labor supply incentives via expanding the Earned Income Tax Credit, as well as raising the retirement age and providing more information about retirement finances. They also suggest trying to increase demand for older workers by means of education, and by restoring mandatory retirement to forestall employer concerns of having to retain newly hired older workers for too long (although mandatory retirement seems likely to have effects in the opposite direction as well).

Read the full paper here.

The author did not receive financial support from any firm or person for this article or from any firm or person with a financial or political interest in this article. He is not currently an officer, director, or board member of any organization with an interest in this article.