University of Notre Dame

Addressing the Retirement Crisis with Shadow 401(K)s

November 28, 2016

Addressing the Retirement Crisis With Shadow 401(k)s

Deepa Das Acevedo*


The United States has been juggling a handful of socio-economic crises lately. The subprime mortgage crisis, the auto industry crisis, the education crisis, the obesity crisis—the list isn’t short and shows no signs of becoming so. Within this group of economically and socially disruptive developments, the “retirement crisis”—the idea that most Americans will lack the financial resources to be secure and relatively satisfied in their golden years—seems somewhat banal because, for the most part, it has yet to hit. Even though baby boomers first started to age out of the workforce in 2011,[1] the real cost of underfunded retirement is far less palpable than are the evictions, layoffs, and stock market fluctuations accompanying other difficult changes.

But the retirement crisis may prove to be one of the most damaging developments facing contemporary America. It invites procrastination, it’s remarkably immune to class and industry distinctions, and it’s actually two exceedingly complex problems.[2] On the one hand, most Americans aren’t saving smartly or aggressively enough while they work, so the pot of money available to them at retirement is markedly insufficient. On the other hand, and regardless of whether they’ve saved enough during their working years, most Americans aren’t properly managing or capable of properly managing their savings after they retire.

These problems are challenging enough in themselves, but they become even more complicated if we focus on solutions that are politically feasible in the United States. For better or for worse, when it comes to welfare state policies, we aren’t Denmark.[3] This doesn’t mean that we shouldn’t reach a little in our efforts to address the retirement crisis—after all, a “crisis” demands “crisis management.” But we need to work towards viable solutions, and we have to keep in mind that no solution is perfect. This Essay is a contribution toward that effort.

    I.     The “Savings Problem”

This Part explores three reasons why most Americans do not have enough savings accumulated by the time they retire: bad worker defaults, bad employer incentives, and low income. The Part also introduces “Shadow 401(k)s” as a means of addressing the “savings problem.”

A.   Bad Worker Defaults

A host of default practices virtually ensure that workers do not save enough during their earning years for reasons other than a lack of money.[4] Workers may simply choose to allocate resources to things other than a retirement plan. Workers may also fail to undertake crucial trigger actions, like enrolling in a defined contribution plan, choosing a fund into which to invest, or contributing enough to receive matching benefits from their employers. This general problem—now a staple of cocktail conversations everywhere, thanks to Cass Sunstein and Richard Thaler[5]—was significantly, if belatedly, addressed by the Pension Protection Act of 2006 (PPA).[6]

The PPA grants a safe harbor to 401(k) plan sponsors who adopt automatic enrollments, employer contributions, default investments into age-appropriate funds, and automatic increases in worker-contributions.[7] In return for establishing these defaults, the plan sponsor gets a safe harbor from undertaking expensive and complicated non-discrimination compliance testing respecting their plans and matching programs.[8] In effect, the PPA has meant that millennials have almost always lived in a universe where contributions to retirement plans happen as a matter of course.[9]

There’s a lot to like about the PPA, but it’s not all jam. For one, the PPA does not supplant previously available safe harbor provisions, so plan sponsors don’t have to adopt PPA-compliant plans in order to gain its benefits.[10] Secondly, the PPA actually lengthens the vesting period for employers’ matching contributions from “immediate[ly]” to “two years of service.”[11] While we don’t know whether workers opt out more when matching benefits don’t vest immediately, this is just one more reason why we shouldn’t view the PPA as a cure-all for “trigger action” defaults.

The biggest problem with relying on the PPA to increase savings rates is, of course, that employers need not establish any kind of pension plan let alone one that takes advantage of PPA safe harbor standards.[12] As of March 2015, the Bureau of Labor Statistics (BLS) reports that “[e]mployer-provided retirement benefits were available to 31 percent of private industry workers in the lowest wage category.”[13] If we want to take the retirement crisis seriously, it simply won’t do to gloat (or sigh) over the PPA while ignoring the fact that workers at the low end of the earning spectrum often lack any job-related infrastructure to save for retirement.

So what are our options? This Essay suggests authorizing the Social Security Administration (SSA) to create federally administered (though not federally guaranteed) defined contribution plans with PPA-like defaults regarding enrollment and age-appropriate fund selection, as well as a default annuity purchase at retirement. The plans would allow individuals who are interested in more actively managing their savings to select from a limited range of age-appropriate portfolios at predetermined intervals, but the options and the opportunities to act would be purposefully minimal. Let’s call these “Shadow 401(k)s”—shadow because they’re meant to follow workers of all types throughout their professional lives, and also because they aren’t the main event or a one-size-fits-all solution to the retirement crisis.

Essentially, this approach encourages workers who lack employer-sponsored plans to save more than they pay in Social Security taxes, but spares them from having to select, establish, and maintain a plan themselves. Expanding on the SSA’s existing infrastructure (rather than expanding Supplemental Security Income (SSI) taxes) has the advantage of minimizing administrative expenses without triggering politically charged tax debates. Instituting simple defaults allows us to capitalize on the lessons learned from the PPA but leaves some room for individualization—again, unlike SSI taxes. And structuring the plans as defined contribution funds helps avoid the political baggage that is likely to accompany talk of a “federal annuity”[14] or a “universal savings account.”[15]

Because it does not involve federal contributions or risk-bearing, the Shadow 401(k) differs from similar proposals put forward by, among others, Teresa Ghilarducci[16] and Michael Calabrese.[17] This naturally makes the Shadow 401(k) a more incremental response to the retirement crisis and invites—indeed, likely necessitates—alteration down the road. And forced savings without a federal guarantee, savings credit, or matching feature increases the burden on low-wage earners at a time when they are already hurting.

However, these features may also make the Shadow 401(k) more legislatively viable in a period of considerable political tumult and decreased bipartisan support for similar plans, while simultaneously addressing many of the major problems that scholars like Ghilarducci and Calabrese have identified in current or proposed retirement systems.[18] Indeed, some states have approximated this approach.[19] Starting in 2017, for instance, Illinois “will automatically enroll workers without a retirement account into a portable, state-run individual retirement plan called the Secure Choice Pension.”[20] Programs like Secure Choice have the same advantages and disadvantages as the Shadow 401(k), but with the added weakness that they depend upon the political climate of individual states. The Shadow 401(k) provides a feasible federal foundation on which we can build as the opportunity arises and—while we wait—offers a small step towards ensuring that Americans have more money waiting for them at retirement.[21]

B.   Bad Employer Incentives

Just as the current system doesn’t give workers the best incentives to save, it also doesn’t encourage employers to help workers save. To begin with, employers have little reason to take on the responsibilities and costs of sponsoring a retirement plan in a climate where workers change jobs frequently. The average baby boomer worker has held 11.7 jobs between the ages of eighteen and forty-eight,[22] and the average wage and hour job is now held for around 4.6 years.[23] These figures aren’t significantly higher today than they were before the 2008 Recession, and they’re also not higher for millennials than for older workers (even though media coverage tends to suggest otherwise in both cases).[24]

But statistics on job-hopping do indicate the existence of a generalized and long-term shift in employer-employee relations such that both parties increasingly view their relationship as disposable.[25] For a few decades now, employers—even healthy employers—have been more willing to lay off workers in the interests of maximizing shareholder profits, and workers have become more likely to view their careers in terms of building a personal brand across employers rather than a record of good service within one company.[26] All of this adds up to an environment in which “the long-standing assumption of long-term attachment between an employee and a single firm has broken down.”[27]

Relatedly, some features of the existing legal and regulatory structure surrounding employment strongly incentivize employers to create distance between themselves and workers via reclassification. Reclassification, or the practice of categorizing workers as “independent contractors” instead of as “employees,” relieves an employer from fulfilling certain tax and (non-ERISA) employee benefits obligations, and it also takes the entire issue of ERISA benefits off the table. Employers in the “sharing” or “platform” economy have developed new business models that capitalize on this distinction between employees and independent contractors, while many traditional industries like construction and delivery have long relied on a workforce largely composed of independent contractors.[28] If the move towards independent contracting and gig work continues, even more workers will be ineligible for employer-sponsored retirement plans. Even platform companies are beginning to speak up about this problem: in late 2015, a group of executives, venture capitalists, and policy analysts issued a joint letter calling for “portable benefits” plans that would follow independent contractors across jobs.[29]

Finally, there’s a fair bit working against those employers who do categorize their workers as “employees” and who want to sponsor some sort of retirement plan. Economies of scale prevent most small businesses from being able to afford mainstream plans, while the range and complexity of choices often overwhelm small-business owners. In fact, only 14% of small-business employers sponsor retirement plans of any kind.[30] It’s true that the market is starting to produce some solutions: startups like Honest Dollar and ForUsAll offer plans specifically priced for small businesses with relatively jargon-free descriptions and easy online maintenance.[31] Some of these plans even support independent contractors.[32] But these are relatively niche options and they replicate the existing system’s reliance on employer initiative.

The Shadow 401(k) system can address all of these problems because it is independent of employer intent, resources, and skill. Even under a Shadow 401(k) system many employers will find it necessary to offer sponsored benefit plans to recruit high-skill workers. But workers who would appreciate the ease of an employer-sponsored plan will no longer be obliged to wait for their employers to make the first move, and workers whose employers will never make the first move—or workers who are self-employed or operate in the platform economy—will still have a way to save more.

At the same time, it’s worth remembering that even adequate Shadow 401(k) savings don’t inevitably translate into a comfortable retirement for at least three reasons: those savings are vulnerable to market fluctuations (a challenge that cannot be avoided without some element of federally guaranteed returns or benefits), they still need to be managed and spent wisely (a challenge I discuss later in this Essay), and they are taxed very differently than other capital gains (which I leave for another day).

C.     Low Income

Legal scholars concerned with the retirement crisis—whether writing from within employee benefits, elder law, or tax—all generally focus on providing the right incentives and the right infrastructure to facilitate savings.[33] This is well and good, since there is much work to be done on both fronts. Nevertheless, it doesn’t help much to incentivize or facilitate saving if workers have no money to save.

This is not just a “minimum wage” issue. Technically, only around three million workers earn below the current federal minimum wage of $7.25 per hour, and these folks definitely do not have the ability to save for retirement.[34] When we ask whether workers are able to save for retirement, we’re really discussing workers who earn at or above the minimum wage and how changes in income affect their behavior. Unfortunately, the BLS Consumer Expenditure Reports do not distinguish between pensions, personal insurance, and payments to Social Security, and there’s very little information available elsewhere on retirement savings relative to income fluctuation.

A few facts suggest that we may want to pay more attention to workers’ ability to forego income in the interests of contributing to any type of retirement scheme. Readers should note that these facts are not meant to suggest linear causation—that is, they don’t tell us that workers aren’t saving more because they can’t afford to. But they do suggest that savings practices (which were likely never ideal for the reasons identified by behavioral economics scholars) have taken an extra hit thanks to wage stagnation, job loss, and financial emergencies associated with the Recession.[35] Consider the following:

  • The National Institute on Retirement Security reports that the average working household has a median retirement account balance of just $2500.[36]
  • Pew Trusts states that about 70% of American households qualify as savings-limited, income-constrained, or debt-challenged,[37] while 32% face two or more of these problems.[38] Moreover, this was the case both before and after the Recession.[39]
  • The Federal Reserve observes that 57% of respondents who had savings before 2008 had, largely because of the Recession, used up “some” to “all” of those savings by 2013.[40]

The first two facts point to the general financial strain felt by working families, while the Federal Reserve’s finding suggests that this strain was exacerbated by the Recession. Again, this is not to say that we can ignore bad savings practices. But the assumption characterizing much of the literature on the retirement crisis—that the problem lies in worker or employer incentives or taxation—replicates the view that expenditures are a proxy for valuation. Before we go full steam ahead in our efforts to change workers’ valuation of saving for retirement or try to capitalize on workers’ tendency to inaccurately express their valuation, we should ask ourselves whether the proxy really applies.[41] If it doesn’t, and if part of the issue is that workers simply don’t have enough to save enough, then scholars and practitioners who are interested in addressing the retirement crisis must join the larger conversation on employment conditions and income inequality that is taking place in America today.

II.    The “Management Problem”

Even workers who manage to set aside enough money during their careers face the difficult task of managing that money before and after retirement. This Part considers three kinds of obstacles to the proper management of retirement savings: poor decisionmaking before retirement, basic capability and financial literacy, and decisionmaking by retirees coping with physical or mental decline. It also shows how “Shadow 401(k)s” can help address aspects of the “management problem.”

A.   The Long Road to Retirement

During their professional lives, workers must make several decisions regarding their retirement besides determining how much they’ll save. What amount should they contribute to an employer-sponsored fund? If there are different funds they could invest into, which ones should they choose? In case of an unexpected financial stress that compels them to draw down on existing resources, how should they treat their retirement savings relative to their other assets? The whole thing bears more than a passing resemblance to an obstacle course.[42]

It’s just not possible to come up with general answers to these questions because they are incredibly dependent on context. At the same time, some choices made by workers are predictable—and predictably flawed. For example, many workers trigger “leakage[s]” from their 401(k) accounts—that is, they draw on their retirement savings before retirement.[43] Workers might take out loans against their 401(k) funds, they might request hardship withdrawals in times of extreme financial need, or they might simply cash out their 401(k)s when shifting to a new job.[44] Hardship withdrawals come with especially heavy consequences since workers not only lose accumulated savings but also lose the income those savings might have earned while invested.[45] Even ostensibly wise choices—like rolling over 401(k) funds from a previous job into a third-party Individual Retirement Account (IRA)—heighten the risk of leakages.[46] And this is hardly a niche problem: Gallup estimates that one in five participants in an employer-sponsored 401(k) plan has tapped into her plan before retirement.[47]

The hassle, of course, is that workers have virtually their entire pre-retirement lives in which to encounter financial stresses and trigger leakages. Moreover, 401(k) leakages are just one type of problem involving one type of retirement savings device. But even though we can’t solve all of the problems that come from having to manage money for multiple pre-retirement decades, we can fix a few key issues. A Shadow 401(k) system could sidestep the entire issue of job change rollovers because the Shadow 401(k)s would naturally follow workers to each new job. The system could also avoid the problem of a worker making unwise and perhaps inconsistent 401(k) investment options with each new job because the selection would only need to be done once (or a handful of times, if the worker takes an active interest).

To be sure, we would need to offer more instruction and information at the time of that initial choice and any subsequent adjustments. And we would also need to strengthen penalties on in-service withdrawals, loans, and cash outs, in order to prevent workers from viewing their 401(k)s as just another savings account. But these steps might also be easier with the kind of centralized governmental management that comes with Shadow 401(k)s.

B.   Capability, Social Capital, and Culture

Of course, all of this decisionmaking assumes a great deal of capability and financial literacy. Just as workers must constantly assess their progress to financial security and adjust their investment practices, retirees must constantly re-assess their rate of expenditure, determine which assets to draw down on first, and decide how to reconcile unexpected financial stresses with a fixed income. These are complex issues even for bright, educated, and financially literate folks, but the ungraceful truth is that many Americans lack the ability to tackle such problems.

Some of this is due to inadequate information and education. For example, one survey asked three basic questions about percentage calculations, simple division, and compound interest.[48] Sixteen percent of respondents couldn’t answer the percentage question, one-third failed the division question, and almost 80% of the respondents “did not understand compound interest.”[49] Even more discouragingly, most people aren’t aware of their own financial illiteracy. The National Foundation for Credit Counseling found that about one in five respondents “say they can resolve their own problems without outside help” while another one in four reach out to their friends and relatives rather than to financial professionals.[50] The real surprise is that investment decisions and 401(k) leakages are only as bad as they are.

The good news is that there are some things we can do in this area. People with more general education are likely to make better decisions.[51] People with some economics or financial literacy education are likely to make better decisions.[52] But it’s not all about education: these decisions demand a not-insignificant level of capability and analytic skill. I do not want to make too much of this last point—not only is it decidedly dispiriting, but our legal system regularly ascribes these positive qualities to citizens.[53] Moreover, some scholars think that we underestimate citizens’ ability to make intelligent decisions given adequate information.[54] But it’s worth keeping in mind that the current system of retirement savings assumes that the average American is either smart enough to make complicated financial decisions, or smart enough and emotionally secure enough to recognize when she needs help. That may not be the case.

Of course, “good” decisionmaking—that is, decision-making that maximizes financial security—also depends heavily on environmental factors like social capital and culture. Jump$tart, a non-profit advocating financial literacy education for high school and college students, found that students’ ability to answer a basic question about smart debt correlated to specific demographic categories like race (whites fared better) and gender (males at the high school level and females at the college level fared better).[55] These differences reflect the fact that some groups inherit positive social capital (like financial sophistication) while others inherit negative social capital (like having poor relatives).[56] On top of all this, financial practices are tied to cultural notions of responsibility: some groups may place a higher value on assisting needy relatives or leaving bequests, while others may prioritize real property ownership.[57]

The current system of employer-sponsored defined contribution plans is structured as if differences in intelligence, education, social capital, and cultural values don’t exist. It demands that people who may lack significant financial literacy nonetheless make complex decisions early on and repeatedly throughout their lives. It also demands that people who aren’t fortunate enough to have an employer-sponsored plan—the same people who are also likely lower on formal education and social capital—display extra initiative and financial acumen by investigating and acquiring IRAs. And finally, the system has no way of coping with the fact that reasonable people can disagree about how to use resources. This last problem is a strange twist on the idea that workers lack incentives rather than income because it essentially assumes that we all have exactly the same incentives (saving as much as possible). We don’t.

A Shadow 401(k) system eliminates the first two problems because it involves fewer decisions during an individual’s working life, a default annuity purchase at retirement, and it covers any participant in the labor force regardless of legal classification or the availability of employer-sponsored plans. It also mitigates the disparity between those who value saving and those who prioritize other things like assisting family or investing in real property. Admittedly, it does so by taking away some decisionmaking power from people who would prefer not to save. But the truth is that this aspect of the Shadow 401(k) system is no more paternalistic than Social Security, and the current system doesn’t offer these individuals much of a choice either: they may get to buy that home or help that relative, but only at the cost of a painful and impoverished old age.

C.   Physical & Mental Fitness

Even if retirees have the money, know-how, and social capital necessary to ensure their own comfort, they must do so at a time when they are confronted by physical and mental deterioration.[58] Age-related decline is an unpleasant and tricky thing to consider, not least because there is an incredible diversity of opinion within the scientific community as to when, how, and at what rate we experience reductions in physical and mental capacity.[59] But since it seems relatively certain that some decline is likely for almost all individuals, it’s worth asking how decline affects outcomes under the current system and whether the Shadow 401(k) offers any improvements.[60]

We can sort cognitive decline issues into two broad categories based on the retirement-related problems they generate: memory loss and cognitive impairment. Memory loss can complicate tasks that retirees must perform under the existing system, like deciding whether to withdraw more than the minimum distribution requirement, picking the funds from which they’ll withdraw assets each year, and determining whether and how they’ll reallocate assets after withdrawals. Retirees suffering memory loss will probably find it more taxing to make these choices and may make some mistakes because they’ll have trouble juggling all the data points that go into an ostensibly simple decision like how much to withdraw from a 401(k). The default annuity purchase feature of Shadow 401(k)s means that retirees with memory loss needn’t go through this process every year unless, at retirement, they are so actively involved in managing their finances that they elect out of the annuity purchase.[61]

Cognitive impairment can vary across a fairly broad spectrum, ranging from brain-based changes that hamper affective processing to Mild Cognitive Impairment to various forms of dementia.[62] Overall, 30% of Americans above age eighty-five are thought to be living with some form of dementia.[63] Men and women aged sixty-five have a 11% and 19% lifetime risk, respectively, of developing Alzheimer’s—probably the most well-known form of dementia as well as one of its more severe variations—and that risk increases exponentially with age.[64]

Depending on the severity of the impairment, retirees suffering from dementia may be incapable of making the annual decisions demanded by traditional 401(k)s. Consequently, they stand to benefit from the default annuity feature of Shadow 401(k)s even more than retirees with simple memory loss. Default annuity purchases may also help address the large but underreported phenomenon of elder financial abuse, at least insofar as they commit a retiree’s Shadow 401(k) balance to small-increment disbursals rather than lump sum withdrawals.

Physical deterioration can also impact a retiree’s ability to manage her money, even when that deterioration is something as run of the mill as sensory decline. A recent study of adults aged fifty-five and above found that about one-third of men classified as “Seeing and Hearing Disabled” required assistance with personal finances, as did around 15% of men classified as “Hearing Disabled but Seeing Abled.”[65] Sensory decline can often also lead to cognitive impairment if an individual doesn’t remain socially engaged—and, unsurprisingly, many elders suffering from sensory decline tend to isolate themselves.[66] Moreover, people experiencing sensory impairment often pretend that they’re unaffected, which makes it difficult for even the most skilled and well-meaning third-party to act in a retiree’s interest.[67] And all of this is aside from any major medical conditions that may directly damage a retiree’s finances or impact her ability to think strategically in her own long-term interest.


No one solution is going to rescue our decidedly imperfect retirement system while satisfying taxpayers, and the Shadow 401(k) is hardly an exception to this rule. Shadow 401(k)s force wealth accumulation, rather than provide replacement income, and they still require a not-insignificant level of financial knowledge on the part of workers who may not even have a high school education. These problems will never truly disappear so long as we structure retirement around defined contribution plans where workers must make decisions affecting their own financial futures and bear the consequences of those decisions. Shadow 401(k)s also still leave retirees vulnerable to fluctuations in the stock market, and consequently without a guaranteed amount of funds for retirement. This problem will never disappear so long as we are unwilling to put the full faith and credit of the federal government behind potential solutions. And Shadow 401(k)s also do not guarantee that workers earn enough to save enough for retirement. This problem will not disappear until we ask ourselves what workers need to earn so that they can cover both present spending and retirement saving in the way our system demands.

But if we’re going to address the “retirement crisis,” we have to be open to multiple solutions that each effect incremental improvements. The Shadow 401(k) does just this. Because it follows workers from job to job, defaults into age-appropriate investment funds with minimal (but non-zero) opportunities for individualization, and because it attaches heightened requirements and penalties to leakages, the Shadow 401(k) reduces the odds that individuals will harm themselves through poor decisionmaking during their working lives. Similarly, because it defaults into an annuity purchase at retirement, the Shadow 401(k) lowers the financial literacy and personal capability needed to manage savings after retirement. And lastly, because it applies regardless of employment classification, the Shadow 401(k) includes all participants in the labor force while functioning independently of employer intent, resources, and skill.

Most importantly, however, the Shadow 401(k) is within the realm of political possibility. To be sure, nothing is easy to achieve in Congress these days. But the Shadow 401(k) requires no new administrative apparatus, no “tax,” no added cost to employers, and no pledging of the United States’ full faith and credit. It doesn’t try to buck the trend of defined contribution plans, and it doesn’t sigh for America to be like Denmark. For these reasons, the Shadow 401(k) isn’t as aggressive a response to the retirement crisis as some other proposals.[68] At the same time—and for these same reasons—it may be a viable (and pliable) first step on the road to resolution.

       ©     2016 Deepa Das Acevedo. Individuals and nonprofit institutions may reproduce and distribute copies of this Essay in any format, at or below cost, for educational purposes, so long as each copy identifies the author, provides a citation to the Notre Dame Law Review Online, and includes this provision in the copyright notice.

*     Sharswood Fellow, University of Pennsylvania Law School. AB, Princeton University (2006); PhD, University of Chicago (2013); JD, University of Chicago (2016). My thanks to Chuck Wolf for comments on an earlier version of this paper and, as always, to John Felipe Acevedo. This Essay won the 2016 Alvin D. Lurie Memorial Award from the American College of Employee Benefits Counsel.

        [1]     See D’vera Cohn & Paul Taylor, Baby Boomers Approach 65—Glumly, Pew Research Ctr. (Dec. 20, 2010),

        [2]     See Edward N. Wolff, Econ. Policy Inst., Retirement Insecurity: The Income Shortfalls Awaiting the Soon-to-Retire 2 (2002).

        [3]     See Hillary Clinton, U.S. Presidential Candidate, The CNN Democratic Presidential Debate in Las Vegas, Nev. (Oct. 13, 2015) (transcript available from The CNN Democratic Debate Transcript, Annotated, Wash. Post: The Fix, (“But we are not Denmark. I love Denmark. We are the United States of America.”).

        [4]     See Amy B. Monahan, Addressing the Problem of Impatients, Impulsives and Other Imperfect Actors in 401(k) Plans, 23 Va. Tax. Rev. 471, 481–83 (2004).

        [5]     Richard H. Thaler & Cass R. Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness 103–17 (2008).

        [6]     Pension Protection Act of 2006, Pub. L. No. 109-280, 120 Stat. 780 (codified in scattered sections of 5, 7, 19, 26, 29, 30 U.S.C. (2012)).

        [7]     See Emp. Benefits Sec. Admin., U.S. Dep’t of Labor, Pub. 3998 (Rev. 10-2015), Choosing a Retirement Solution for Your Small Business 5 (2015),; Jack VanDerhei, The Pension Protection Act and 401(k)s, Wall St. J. (Apr. 22, 2008),

        [8]     See VanDerhei, supra note 7.

        [9]     See Jean A. Young, Vanguard, The Auto Savings Generation: Steering Millennials to Better Retirement Outcomes 15 (2015),

      [10]     See Daniel P. Schwallie & Allen Steinberg, How Safe Is Your ADP/ACP Safe Harbor?, 29 Benefits Q., Oct. 2013, at 35, 35–36.

      [11]     Id. at 36.

      [12]     See Retirement Plans and ERISA FAQs, Emp. Benefits Sec. Admin., U.S. Dep’t of Labor, (last visited Dec. 1, 2015).

      [13]     News Release, Bureau of Labor Statistics, U.S. Dep’t of Labor, Employee Benefits in the United States—March 2015 (July 24, 2015),

      [14]     But cf. Lawrence A. Frolik, Protecting Our Aging Retirees: Converting 401(k) Accounts into Federally Guaranteed Lifetime Annuities, 47 San Diego L. Rev. 277, 320–30 (2010) (advocating for a “government-subsidized annuity”).

      [15]     The “Universal Savings Account” plan President Clinton floated in the late 1990s required the federal government to match any percentage of savings. See George Hager & Amy Goldstein, Clinton Details New Retirement Plan, Wash. Post (Apr. 15, 1999),

      [16]     See Teresa Ghilarducci, Econ. Policy Inst., Guaranteed Retirement Accounts: Toward Retirement Income Security 17 (2007).

      [17]     See Michael Calabrese, New Am. Found., Facing Up to the Retirement Savings Deficit: From 401(k)s to Universal and Automatic Accounts 13 (2011).

      [18]     See, e.g., Automatic IRA Act of 2015, H.R. 506, 114th Cong. The latest version of the congressional Auto-IRA proposal, H.R. 506, still relies on employers to create plans and exempts those employers with fewer than ten employees from even that obligation. See id. § 2.

      [19]     These include Connecticut, Maryland, New Jersey, Oregon, Washington, Illinois, Massachusetts, and California. Look to the States for Innovation, Georgetown Univ.’s Ctr. for Ret. Initiatives, (last visited Nov. 3, 2016).

      [20]     David Rolf et al., Aspen Inst., Portable Benefits in the 21st Century 19 (2016),

      [21]     The myRA program recently established by the Obama Administration does offer a federal avenue for payroll-deduction retirement savings, but enrollment in myRA depends on worker initiative. See myRA: How It Works, U.S. Dep’t of Treasury, (last visited Nov. 4, 2016).

      [22]     News Release, Bureau of Labor Statistics, U.S. Dep’t of Labor, USDL-15-0528, Number of Jobs Held, Labor Market Activity, and Earnings Growth Among the Youngest Baby Boomers: Results From a Longitudinal Study 1 (Mar. 31, 2015).

      [23]     News Release, Bureau of Labor Statistics, U.S. Dep’t of Labor, USDL-14-1714, Employee Tenure in 2014, at 1 (Sept. 18, 2014).

      [24]     See Steven F. Hipple & Emy Sok, Bureau of Labor Statistics, U.S. Dep’t of Labor, Tenure of American Workers 2–3 (2013),; Ben Casselman, Enough Already About the Job-Hopping Millennials, FiveThirtyEight (May 5, 2015),; Jeanne Meister, Job Hopping Is the ‘New Normal’ for Millennials: Three Ways to Prevent a Human Resource Nightmare, Forbes (Aug. 14, 2012),

      [25]     See Declining Employee Loyalty: A Casualty of the New Workplace, Knowledge@Wharton (May 9, 2012),

      [26]     Wharton professor Adam Cobb observed that “[f]irms have always laid off workers, but in the 1980s, you started to see healthy firms laying off workers, mainly for shareholder value. . . . If I’m an employee, that’s a signal to me that I’m not going to let firms control my career.” Id.

      [27]     Katherine V.W. Stone, From Widgets to Digits: Employment Regulation for the Changing Workplace 3 (2004).

      [28]     See generally Deepa Das Acevedo, Regulating Employment Relationships in the Sharing Economy, 20 Emp. Rts. & Emp. Resp. 1 (2016).

      [29]     Byron Auguste et al., Common Ground for Independent Workers, Medium (Nov. 9, 2015),; see also Nick Hanauer & David Rolf, Shared Security, Shared Growth, 37 Democracy, July 2015, at 6, 14 (describing the Shared Security System “as analogous to Social Security, but encompassing all of the employment benefits traditionally provided by a full-time salaried job”).


      [31]     See Stacy Cowley, Tailoring Retirement Plans to Companies with a Handful of Workers, N.Y. Times (July 15, 2015),

      [32]     See What Types of Companies or Entities Can Use Honest Dollar?, Honest Dollar, (last visited Nov. 5, 2016).

      [33]     See, e.g., Frolik, supra note 14, at 278; Teresa Ghilarducci, Retirement Security Worse on ERISA’s 40th Anniversary, 6 Drexel L. Rev. 453, 459 (2014); Daniel Halperin, Employer-Based Retirement Income—the Ideal, the Possible, and the Reality, 11 Elder L.J. 37, 44 (2003).

      [34]     Bureau of Labor Statistics, U.S. Dep’t of Labor, Report 1054, Characteristics of Minimum Wage Workers, 2014, at 1 (2015). Readers should note that the “three million workers” statistic is highly misleading since state laws require employers to “top off” a tipped worker’s wages at vastly differing rates (or not at all)—and even where topping off requirements are comparatively generous, enforcement often remains weak. See Suzanne Dershowitz et al., Food Labor Research Ctr., Working Below the Line: How the Subminimum Wage for Tipped Restaurant Workers Violates International Human Rights Standards 15 (2015), (discussing wage theft); Minimum Wage for Tipped Employees, U.S. Dep’t of Labor (Aug. 1, 2016), (listing the direct wage, tip credit, and overall minimum wage requirements for each state).

      [35]     See generally Michael Lind, Opinion, Can You Have a Good Life if You Don’t Have a Good Job?, N.Y. Times (Sept. 16, 2016),

      [36]     Nari Rhee & Ilana Boivie, Nat’l Inst. on Ret. Sec., The Continuing Retirement Savings Crisis 1 (2015).

      [37]     Erin Currier et al., Pew Charitable Trs., The Precarious State of Family Balance Sheets 1 (2015).

      [38]     See id. at 13.

      [39]     Id. at 12.

      [40]     Bd. of Governors of the Fed. Reserve Sys., Report on the Economic Well-Being of U.S. Households in 2013, at 2 (2014).

      [41]     The most relevant of these concepts are “willingness to pay” (the amount A is willing to spend for a widget is a reflection of the value A subjectively ascribes to the widget) and “income effect” (as A’s income fluctuates, so will A’s consumption). The problems, of course, are that (1) A is not a rational, self-aware, and self-controlled actor, and (2) valuation itself fluctuates. See generally R.H. Coase, The Problem of Social Cost, 3 J.L. & Econ. 1 (1960); Daniel Kahneman et al., Experimental Tests of the Endowment Effect and the Coase Theorem, 98 J. Pol. Econ. 1325 (1990).

      [42]     See Monahan, supra note 4, at 478–79 (describing 401(k) employee participation requirements).

      [43]     See Thomas Olson, Note, 401(k) Leakage: Crafting a Solution Consistent with the Shift to Employee-Managed Retirement Accounts, 20 Elder L.J. 449, 451 (2012).

      [44]     See id. (observing that “[l]eakage manifests itself in three forms: cash outs, loans, and early withdrawals” (citing Robert Steyer, DC Plan Leakage Problem Alarming, Solutions Evasive, Pensions & Invs. (Apr. 4, 2011),

      [45]     See id. at 461.

      [46]     This is because IRAs are not as strictly regulated as defined contribution plans, so IRA withdrawals are easier to effect and subject to lighter penalties. See Alicia H. Munnell & Anthony Webb, Ctr. for Ret. Research, The Impact of Leakages on 401(k)/IRA Assets 2 (2015).

      [47]     Art Swift, One in Five Investors Have Tapped into 401(k) Prematurely, Gallup (Dec. 22, 2014),

      [48]     Olivia S. Mitchell, Retirement Risk Management in Times of Turmoil, 17 Elder L.J. 439, 444–45 (2010) (citing Annamaria Lusardi & Olivia S. Mitchell, Financial Literacy and Retirement Preparedness: Evidence and Implications for Financial Education, 42 Bus. Econ. 35, 37–38 (2007)).

      [49]     Id. at 445 (citing Lusardi & Mitchell, supra note 48, at 37 tbl.1).

      [50]     Nat’l Found. for Credit Counseling, 2015 Consumer Financial Literacy Survey 4 (2015),

      [51]     Mitchell, supra note 48, at 445.

      [52]     Id. at 446.

      [53]     Think, for example, of how the Brandenburg Paradigm does this in the context of First Amendment law. See Steven G. Gey, The Brandenburg Paradigm and Other First Amendments, 12 U. Pa. J. Const. L. 971, 992 (2010) (stating that “implicit in the Brandenburg paradigm is an image of a proper democratic citizen” possessing “a certain number of characteristics that are essential to effective participation in a political system governed by the First Amendment” including “incredulity, rationality, critical intelligence, insensitivity to political slights, and an understanding of the political protocols that accompany an orderly transfer of power”).

      [54]     See Jennifer Nou, Regulating the Rulemakers: A Proposal for Deliberative Cost-Benefit Analysis, 26 Yale L. & Pol’y Rev. 601, 642 (2008) (“[T]here is little basis for assuming that lay citizens will be unable to appreciate or comprehend technical information.” (citing Stuart Hill, Democratic Values and Technological Choices 55–89 (1992))).

      [55]     Lewis Mandell, Jump$tart Coal. for Pers. Fin. Literacy, The Financial Literacy of Young American Adults: Results of the 2008 National Jump$tart Coalition Survey of High School Seniors and College Students 167 (2009).

      [56]     See Allison Courtney Scott, Ethnic Differences in Reliance on Family Versus Personal Savings in the Face of a Financial Shock 8–9 (2014) (unpublished M.S. thesis, University of Colorado) (on file with author).

      [57]     See id. at 7–10 (arguing that blacks may prioritize assisting needy relatives or leaving bequests, while Hispanics may do the same for real property ownership).

      [58]     See Lawrence A. Frolik, Rethinking ERISA’s Promise of Income Security in a World of 401(k) Plans, 20 Conn. Ins. L.J. 371, 384–87 (2014); Frolik, supra note 14, at 292.

      [59]     See, e.g., Timothy A. Salthouse, When Does Age-Related Cognitive Decline Begin?, 30 Neurobiology of Aging 507, 508–09 (2009).

      [60]     See Ian Stuart-Hamilton, The Psychology of Ageing: An Introduction 111 (5th ed. 2012) (arguing that “short-term memory tasks . . . present compelling evidence of an age-related decline”).

      [61]     On the importance of annuitization, see generally Jonathan Barry Forman, Removing the Legal Impediments to Offering Lifetime Annuities in Pension Plans, 22 Conn. Ins. L.J. (forthcoming 2016).

      [62]     See Aging, Memory Loss and Dementia: What’s the Difference?, Alzheimer’s Ass’n, , (last visited Nov. 5, 2016); Mariann R. Weierich et al., Older and Wiser? An Affective Science Perspective on Age-Related Challenges in Financial Decision Making, 6 Soc. Cognitive & Affective Neuroscience 195, 198 (2011).

      [63]     Margaret Gatz, Genetics, Dementia, and the Elderly, 16 Current Directions Psychol. Sci. 123, 123 (2007).

      [64]     Id.

      [65]     Parminder Raina et al., The Relationship Between Sensory Impairment and Functional Independence Among Elderly, 4 BMC Geriatrics, December 2004, at 6.

      [66]     Nicholas R. Nicholson, A Review of Social Isolation: An Important but Underassessed Condition in Older Adults, 33 J. Primary Prevention 137, 140–42 (2012); Yukari Yamada et al., Joint Associations of Dual Sensory Impairment and No-Activity Involvement With 1-Year Mortality in Nursing Homes: Results From the SHELTER Study, 71 J. Gerontology Biological Sci. & Med. Sci. 643, 644 (2016).

      [67]     See Susan Seliger, Why Won’t They Get Hearing Aids?, N.Y. Times: The New Old Age Blog (Apr. 5, 2012, 11:53 AM),

      [68]     See, e.g., Ghilarducci, supra note 16, at 17; Jonathan Barry Forman & Michael J. Sabin, Tontine Pensions, 163 U. Pa. L. Rev. 755, 759, 774–807 (2015); Frolik, supra note 14, at 320–30.