Policy Solutions for the Retirement Crisis

By Monique Morrissey

The U.S. retirement system is flawed and inadequate. The remedy? Expand Social Security and ensure employers and workers contribute to simple, affordable retirement plans.

To address the looming retirement crisis in the United States, economists, advocates, and reformers need to get over the fact that people are flawed and instead focus on fixing the system. Many would-be reformers blame the victim—fix­ating on people’s individual failings while perpet­uating a system that magnifies wealth inequality and protects Wall Street profits. Instead, Social Security should be expanded, while ensuring that employers contribute to simple, affordable plans like Guaranteed Retirement Accounts (GRA).

The Unknowns of Retirement Preparedness

There are many unknowns when it comes to people’s retirement preparedness. Has participa­tion in employer plans held steady or declined? How much have households saved for retirement and how are these savings distributed? How much do retirees need to maintain their standard 
of living?

The evidence needed to answer these ques­tions is surprisingly shaky. Survey respondents are poorly informed about retirement benefits, savings, and income, while administrative records lack key demographic information. New statistical methods linking the two hold promise, but the evidence so far calls into question much of what had been accepted as fact regarding retirement. For example, U.S. Census research­ers linking tax records to household survey data found that older adults often neglect to report pension benefits in household surveys, under­stating their income. Consequently, older adults’ poverty rate, though undoubtedly high for a wealthy country, may be lower than that of working-age adults (Bee and Mitchell, 2017).

Whether or not current retirees are worse off than working-age adults, future retirees will be worse off unless action is taken. The need for retirement wealth has grown, even as this wealth—which includes savings and the value of future benefits—has grown more unequal.

Egalitarian Social Security benefits have been cut back since 1983 when Congress made changes to the program, while risky and highly unequal 401(k)s have largely replaced secure pensions in the private sector. Workers relying on 401(k)s should be saving more to replace lost Social Security and pension benefits, to make up for low interest rates, and to ensure against the risk of outliving their savings. Instead, savings-to-income ratios have stagnated (Munnell, Hou, and Sanzenbacher, 2018).

While aggregate retirement wealth has kept up with income, this wealth is increasingly con­centrated at the top. As a result, most households are far behind where they need to be to main­tain their standard of living in retirement. Among families approaching retirement (ages 56 to 61), the median family responding to the Federal Reserve’s Survey of Consumer Finances (SCF) in 2016 had accumulated less than a third of their annual income in a retirement account—a lower ratio than in 2007, and far too little to replace shrinking Social Security and pension bene­fits. Even the mean retirement account balance, a measure skewed by large accounts, holds the equivalent of only one-and-a-half years of income (author’s analysis of Federal Reserve Survey of Consumer Finances, 2007-2016, microdata).

People of color are especially at risk. Among working-age families headed by someone ages 32 to 61, 65 percent of Hispanic and 59 percent of Afri­can American respondents in the SCF in 2016 had no retirement account savings at all, compared with 32 percent of white non-Hispanic families. The median account balance for families possess­ing such savings was also highly unequal—$23,000 for Hispanics, $29,200 for African Americans, and $79,000 for white non-Hispanics (author’s anal­ysis of 2016 SCF microdata). These stark differ­ences by race and ethnicity, only partly explained by income inequality, also reflect a lack of access to good employer plans and an aversion to investment risk among families with few financial resources, as opposed to differences in saving behavior among similarly situated families (Sullivan et al., 2019).

The Current Retirement System Is an Accident

The 401(k) plans originated in the late 1970s as an obscure fringe benefit for bankers. Congress never envisioned that they would come to be seen as a substitute for pensions, let alone touted as an improvement (Tong, 2013). Their popu­larity has risen and fallen with stock market values. During the dot-com bubble, proponents promoted 401(k)s as a way to harness the power of individual investors to build a prosperous “ownership society”—an idea that came to a head with President George W. Bush’s doomed plan to replace Social Security with “personal accounts.”

Now, in the wake of the 2008 market crash, 401(k)s remain ubiquitous in the private sector, but are hardly beloved. Even those who argue that public-sector workers who managed to hold onto their pensions should be in 401(k)-style plans frame this as a downward leveling, not an upgrade (see Las Vegas Review Journal, 2008).

Not surprisingly, the United States’ acciden­tal do-it-yourself system has serious problems. The 401(k) plans have one real advantage, which is that employers do not take on long-term lia­bilities. However, these plans pose significant barriers to retirement security for employees, including employer contributions that are con­tingent upon employee participation, high fees, poor investment choices, upside-down tax subsi­dies, and the absence of longevity risk-pooling or investment risk-smoothing (Morrissey, 2016).

Why the Slow System Fix? Blame Economists!

Many retirement experts have fallen under the sway of behavioral economics, diagnosing the problem as a failure to save enough to smooth consumption over a lifetime. While economists long have recognized that “humans are human,” the central insight of behavioral economics is that these deviations from rational behavior are not random, but predictable, with implications for retirement (see, for example, Benartzi and Thaler, 2007). Specifically, retirement savers are found to suffer from myopia (favoring current over future consumption) and inertia (avoiding decisions in the face of complex choices).

Poor decision-making is compounded by financial illiteracy and misinformation. Many retirement savers fail to diversify across asset classes; adopt an all-or-nothing approach to risk; assume “you get what you pay for” when it comes to mutual fund fees; engage in counterproduc­tive market timing; have trouble estimating how much they will need to save; underestimate their life expectancy; and are unsure how to spend down their savings in retirement (see Beshears et al., 2018; Society of Actuaries, 2018).

While it may be useful to recognize how ill-suited most people are to managing their retire­ments, the economics profession also has shaped the solutions offered, narrowing their scope to reflect the profession’s biases. The lessons of behavioral economics could point to a number of tried-and-true solutions, such as expanding Social Security. Instead, while recognizing that retirement insecurity has societal costs, experts versed in the economic concept of “externali­ties,” but biased toward pseudo-market solu­tions, have advocated addressing the problem by increasing tax subsidies, for both savers and employers (see Congdon, Kling, and Mullaina­than, 2016; Portman, 2019). In this worldview, choice is paramount, even if it masks a highly inequitable and inefficient system.

Though welcomed by the financial services industry, an approach that relies on tax subsi­dies comes with serious problems. Current sub­sidies for retirement saving are badly designed. They favor high-income people who can eas­ily steer funds into tax-favored accounts and take advantage of the higher expected returns on risky investments. They are poorly targeted, because their value does not depend upon the amount contributed (as is commonly assumed), but rather on taxes that would otherwise be owed on investment returns. And though touted as “incentives,” tax subsidies for retirement sav­ing function more as tax shelters because any­one with money in the bank can shift savings to tax-favored accounts—no new saving is required. Finally, though incentivizing saving is seen as less heavy-handed than mandating saving, there is nothing voluntary about taxing Peter to pay Paul (Morrissey, 2017).

After empirical research found that tax incentives were largely ineffective, expert focus shifted to redesigning “choice architecture” to nudge people into making smart decisions (Chetty et al., 2013). This approach culminated in the Pension Protection Act of 2006, which encouraged employers to adopt automatic enroll­ment and escalation. Under these provisions, workers opt out if they do not want to partici­pate, rather than opting in if they do, and their contributions increase over time. However, results have been underwhelming, as employ­ers have proved less eager to contribute more toward their workers’ retirement than propo­nents had anticipated, reducing 401(k) match rates and default contribution rates to offset the cost of broader participation (Butrica and Karamcheva, 2015).

How to Fix the Broken System?

It is time to stop wringing hands over human frailty and directly address the problems in America’s do-it-yourself retirement system. This will require political will, because all of the low-hanging fruit have already been picked. One way to look at incremental reforms of recent decades, such as the automatic enrollment provi­sions of the Pension Protection Act, is as a series of well-intentioned but ineffective experiments in applying lessons from behavioral econom­ics, sometimes called “libertarian paternalism.” Another is to view them as distractions from the obvious solution: expanding Social Security.

Solution 1: Expand Social Security
Expanding Social Security directly addresses three main problems with the current system. It ensures steady contributions; it keeps costs low and prevents leakage; and it pools longevity risk. Moreover, Social Security’s pay-as-you-go financing, whereupon current retiree benefits are funded largely by current worker contributions—provides an internal rate of return tied to eco­nomic growth—just as financial assets do over the long term, but without the gyrations (Morrissey, 2018; Munnell, Hou, and Sanzenbacher, 2017).

Social Security is popular among voters of all political stripes, but faces opposition from pow­erful interests. Surveys have found broad sup­port for expanding Social Security, including among Republican voters (Public Policy Polling, 2018). However, anti-government activists and big-money donors have philosophical and self-interested reasons for opposing expansion, espe­cially if paying for it involves not just raising the payroll tax rate, but also raising the cap on earn­ings subject to the tax (currently set at $132,900; most expansion plans would lift this cap).

In the past, some Democrats were lukewarm in their support for Social Security. As econo­mist Paul Krugman pointed out, it was once viewed as a badge of seriousness inside the Belt­way to warn that Social Security costs were escalating out of control due to rising life expec­tancy (Krugman, 2012). However, life expec­tancy is just one factor in the equation, and though costs are projected to rise from 5 per­cent to 6 percent of GDP with the baby boomers’ retirement, they will then level off (Goss, 2010). A gradual increase in the payroll tax, easily ac­­commodated by wage growth, combined with eliminating the cap on taxable earnings, could handily achieve long-term solvency and pay for expanded benefits (author’s analysis of Goss 2019a, 2019b, 2019c).

Democrats in Congress are now united in support of expansion. After years of persua­sive advocacy by Social Security Works and oth­ers, Social Security expansion was added to the Democratic Party platform in 2016, and a state­ment of support for expansion was signed by vir­tually every Democratic senator (Altman, 2017). Two senators, Bernie Sanders (I–VT) and Eliza­beth Warren (D–MA), co-founded a Social Secu­rity Expansion Caucus in 2018 (Birnbaum, 2018). On the House side, an expansion plan crafted by Rep. John Larson (D–CT), chair of the Social Security Subcommittee, has garnered more than 200 co-sponsors (Marans, 2019).

Solution 2: Implement a GRA plan at the federal level
Social Security expansion is necessary, but may not be sufficient. Expansion bills proposed to date by members of Congress have been designed to reverse the decline in retirement preparedness, but not to replace employer plans. Half of private-sector workers receive no help from their employer in saving for retirement (U.S. Bureau of Labor Statistics, 2018). If Social Security is not expanded sufficiently to make employer plans unnecessary, all workers need to be ensured they will benefit from these plans.
Guaranteed Retirement Accounts (GRA) would fill the gap by ensuring that all workers and employers contribute to a plan. When the Economic Policy Institute (EPI) introduced econ­omist Teresa Ghilarducci’s GRA plan in 2008, it was outside the policy mainstream (Ghilarducci, 2008). At the time, even AARP would not support a plan that mandated contributions.

As with Social Security expansion, the GRA plan directly confronts the failures of the cur­rent system, including coverage gaps, high fees, and longevity risks. Under the plan, workers and employers each are required to contribute a mod­est amount—1.5 percent each in the plan’s current incarnation—into a GRA or equivalent plan. With lifetime coverage and improved risk-adjusted returns from pooled and professionally managed investments, even a 3 percent combined contribu­tion rate can result in substantial income in retire­ment—enough to replace an estimated 14 percent of pre-retirement earnings at age 67 (author’s analysis based on a 3.4 percent real rate of return, assumptions from the 2018 Social Security Trust­ees Report; and Clingman, Burkhalter, and Chap­lain, 2017). For an average earner, this would increase retirement income by roughly a third, compared with Social Security benefits alone.

The GRA plan also takes aim at upside-down tax subsidies, making retirement affordable even for low earners, with a flat, refundable tax credit. The GRA tax credit equals 1.5 percent of pay up to $600, enough to fully offset the employee contribution for workers making $40,000 or less. Workers can choose to deposit the credit into their account instead of receiving a refund, resulting in a total contribution rate of up to 4.5 percent (additional voluntary contributions also are allowed). Workers who already partici­pate in an equal or better 401(k)-style plan can remain in that plan. However, lower contribu­tion limits and reasonable caps on amounts that can accrue in all tax-favored accounts will be phased in for younger workers, because lifetime participation will reduce the need for catch-up contributions. These lower limits and annui­tized benefits—ensuring that all funds are with­drawn in retirement rather than passed down to descendants tax-free—will help to offset the cost of the refundable credit and to limit the use of tax-favored retirement accounts as tax shelters. Because current tax benefits are so skewed—roughly three-fourths going to the top income quintile—the vast majority of people would be better off under the GRA plan than under the current system.

The GRA plan helped to spark a movement. For the Retirement USA project (www.retire ment-usa.org/), a coalition effort to broaden the retirement discourse launched the year after the GRA plan was unveiled, the plan served as a model for other “universal, secure, and adequate” plan designs. In 2014, Senator Tom Harkin (D–IA) proposed USA Retirement Funds as a companion to an early and influential Social Security expan­sion plan (U.S. Senate Committee on Health, Edu­cation, Labor and Pensions, 2014). The Harkin plan shared many features with the GRA plan—including low fees, pooled and professionally managed funds, and annuitized benefits—though contributions were voluntary (Munnell, 2014).

The GRA plan set the stage for efforts in ten states (and counting) (Georgetown Univer­sity McCourt School of Public Policy Center for Retirement Initiatives, 2019). Initiatives for portable low-cost retirement accounts under­way in California and other states, many dubbed “Secure Choice” plans, resemble the GRA plan in many respects. However, under current fed­eral law, states can only require employers to facilitate employee contributions to state plans, not require them to contribute. States also can­not address problems with federal tax subsidies, and current plans do not allow for portability across states. A veteran of these state efforts, for­mer Maryland Lieutenant Governor Kathleen Kennedy Townsend, frustrated by the legal and geographic constraints facing state plans, is now spearheading an EPI campaign to promote the GRA plan at the federal level.

Change Is in the Air

Legislation requiring a modest employer con­tribution has, for the first time, garnered AARP support (Peters, 2019). A co-sponsor of the plan, Senator Amy Klobuchar (D-MN), is running for president, as are Social Security champions Sanders and Warren, raising the visibility of retirement security. The GRA plan is also arous­ing interest on Capitol Hill. No society has ever demanded so much of individual savers, espe­cially in a time of stagnating wages and rocky investment returns. What people really need—and what they want—are secure lifetime ben­efits. Experts and advocates have tried tweaking the do-it-yourself system—but now they should try something that works.

Monique Morrissey, Ph.D., is an economist at the Economic Policy Institute in Washington, D.C.


Altman, N. 2016. “Expanding Social Security Is the Official Posi­tion of the Democratic Party! Will Republicans Continue to Support Cuts?” Huffington Post, July 10. tinyurl.com/y52r6ltv. Retrieved August 14, 2019.

Bee, A., and Mitchell, J. 2017. “Do Older Americans Have More Income than We Think?” (Work­ing Paper 2017-39). Washington, DC: U.S. Census Bureau Social, Economic, and Housing Statistics Division.

Benartzi, S., and Thaler, R. 2007. “Heuristics and Biases in Retire­ment Savings Behavior.” Jour­­-
nal of Economic Perspectives 21(3): 81–104.

Beshears, J., et al. 2018. “Behav­ioral Household Finance.” In B. D. Bernheim, S. DellaVigna, and D. Laibson, eds., Handbook of Be­­havioral Economics: Foundations and Applications 1. Amsterdam: Elsevier.

Birnbaum, E. 2018. “More than 150 Dems Launch Caucus to Expand Social Security Benefits.” The Hill, September 13.

Butrica, B. A., and Karamcheva, N. 2015. “Automatic Enroll­ment, Employer Match Rates, and Employee Compensation in 401(k) Plans.” Washington, DC: Urban Institute.

Chetty, R., et al. 2013. “Active vs. Passive Decisions and Crowd-out in Retirement Savings Accounts: Evidence from Denmark.” Quar­terly Journal of Economics 129(3): 1141–219.

Clingman, M., Burkhalter, K., and Chaplain, C. 2017. “Replacement Rates for Hypothetical Retired Workers.” Actuarial Note 2017(9). Washington, DC: Social Security Administration.

Congdon, W. J., Kling, J. R., and Mullainathan, S., 2016. Policy and Choice: Public Finance through the Lens of Behavioral Economics. Washington, DC: Brookings Insti­tution Press.

Georgetown University McCourt School of Public Policy Center for Retirement Initiatives. 2019. “State-facilitated Retirement Pro­grams: A Snapshot of Program Design Features.” State Brief 19-03. tinyurl.com/y4sydmj2. Retrieved August 19, 2019.

Ghilarducci, T. 2008. “Guaran­teed Retirement Accounts: Toward Retirement Income Security.” Eco­nomic Policy Institute, March 3.

Goss, S. C. 2010. “The Future Financial Status of the Social Secu­rity Program.” Social Security Bul­letin 70(3): 111–25.

Goss, S. C. 2019a. Letter to Sena­tors John Larson, Richard Blu­menthal, and Chris Van Hollen in response to their request for esti­mates of the financial effects on Social Security of the Social Secu­rity 2100 Act. Social Security Office of the Chief Actuary, January 30.

Goss, S. C. 2019b. Letter to Senator Bernie Sanders and Representative Peter DeFazio in response to their request for estimates of the finan­cial effects on Social Security of the Social Security Expansion Act. Social Security Office of the Chief Actuary, February 13.

Goss, S. C. 2019c. Letter to Repre­sentative Ted Deutch and Sena­tor Mazie K. Hirono in response to their request for estimates of the financial effects on Social Security of the Protecting and Preserving Social Security Act. Social Secu­rity Office of the Chief Actuary, April 10.

Krugman, P. 2012. “Life Expec­tancy of the Living Dead.” The New York Times, November 13.
Las Vegas Review-Journal. 2008. “Pension Reform.” September 23. tinyurl.com/y6eex8d2. Retrieved August 13, 2019.

Marans, D. 2019. “House Dem­ocrats Unveil Social Security Expansion Bill with Unprece­dented Support; Rep. John Larson Believes His Plan to Increase Ben­efits Will Pass the House.” Huff­ington Post, January 30. tinyurl.com/y6gzx3dy. Retrieved August 14, 2019.

Morrissey, M. 2016. “The State of American Retirement: How 401(k)s Have Failed Most American Workers.” Boston: Economic Policy Institute.

Morrissey, M. 2017. “Should They Stay or Should They Go? Panel­ists Share Their Thoughts in a Blog Series on Retirement Tax Incen­tives.” Aspen Institute. tinyurl.com/y4j4t65m. Retrieved June 27, 2019.

Morrissey, M. 2018. “Social Secu­rity Is Looking Like a Pretty Good Investment These Days.” Eco­nomic Policy Institute, Working Economics Blog, April 23. tinyurl.com/y4mg2hyr. Retrieved June 27, 2019.

Munnell, A. H. 2014. “Sen. Harkin’s New Retirement-Savings Plan.” MarketWatch, February 19.

Munnell, A. H., Hou, W., and San­zenbacher, G. T. 2017. “How to Pay for Social Security’s Missing Trust Fund?” (Working Paper, December 2017-18). Boston: Center for Retire­ment Research at Boston College.

Munnell, A. H., Hou, W., and San­zenbacher, G.T. 2018. “National Retirement Risk Index Shows Modest Improvement in 2016.” Issue in Brief 18–1. Boston: Center for Retirement Research at Boston College.

Peters, A. 2019. “Americans Can’t Afford Retirement. Here Are 8 Ways to Fix It.” Fast Company, May 8. tinyurl.com/y3jtlt75. Retrieved June 27, 2019.

Portman, R. 2019. “Portman, Car­din Introduce Sweeping Reforms to Strengthen Americans’ Retire­ment Security.” tinyurl.com/y6ypodxk. Retrieved June 27, 2019.

Public Policy Polling. 2018. National Survey Results. Raleigh, NC: Social Security Works. tinyurl.com/yxatbrhr. Retrieved June 
27, 2019.

Society of Actuaries. 2018. Retire­ment Survey Report Key Findings and Issues: 2017 Risks and Process of Retirement Survey. Schaumburg, IL: Society of Actuaries.

Sullivan, L., et al. 2019. “Not Only Unequal Paychecks: Occupational Segregation, Benefits, and the Racial Wealth Gap.” Institute on Assets and Social Policy. Waltham, MA: Heller School for Social Pol-icy and Management, Brandeis University.

Tong, S. 2013. “Father of Modern 401(k) Says It Fails Many Ameri­cans.” Marketplace, Minnesota Public Radio, June 13.

U.S. Bureau of Labor Statistics. 2018. “National Compensation Survey: Employee Benefits in the United States, March 2018.” tinyurl.com/yxczo3t3. Retrieved June 27, 2019.

U.S. Senate Committee on Health, Education, Labor and Pensions (HELP). 2014. “HELP Chairman Tom Harkin Introduces the USA Retirement Funds Act [Long Sum­mary].” January 30. tinyurl.com/y2nvrucs. Retrieved August 
19, 2019.