February 27, 2024

The Honorable Bernie Sanders, Chair

The Honorable Bill Cassidy, Ranking Member

Senate Committee on Health, Education, Labor, and Pensions

428 Dirksen Senate Office Building

Washington, DC 20510

Re: Hearing of the Senate Health, Education, Labor, and Pensions Committee to consider “Taking a Serious Look at the Retirement Crisis in America: What can we do to expand defined benefit pension plans for workers?”

Dear Chairman Sanders, Ranking Member Cassidy, and members of the committee:

On behalf of Economic Policy Institute (EPI), I am pleased to submit this statement for the record for the February 28, 2024, Senate HELP Committee hearing on the retirement income crisis. EPI is a nonprofit, nonpartisan think tank created in 1986 to include the needs of low- and middle-income workers in economic policy discussions. EPI believes every working person deserves a good job with fair pay, affordable health care, and retirement security. We welcome this opportunity to comment on the causes of the retirement crisis and ways to alleviate it.

The 401(k) revolution has failed working families

Steady contributions, affordability, and lifetime income are the building blocks of an effective retirement system. Social Security and traditional defined benefit pensions check all three boxes, 401(k)-style defined contribution plans check none. 

The roots of the retirement crisis can be traced to the early 1980s, when employers began shifting from secure pensions to defined contribution plans around the same time that Congress enacted cuts to Social Security to avert an imminent shortfall.

The shift from defined benefit pensions to defined contribution plans is an experiment that failed. 401(k)s were initially intended as a perk for bankers, not a substitute for pensions.1 However, 401(k)s took off quickly and now outnumber defined benefit pensions among private-sector workers by a factor of nearly eight to one (Figure A).

401(k)s do a poor job of building savings

401(k)s made it easy for employers to offer a retirement plan since much of the cost and all of the risk fall on workers. Despite making it easier on employers, the shift from traditional pensions to 401(k)s did not increase the share of workers participating in a retirement plan, which has long hovered around 50%.2 It did, however, widen the gap between retirement haves and have-nots.3

Though defined contribution plans have outnumbered defined benefit pensions in the private sector for three decades, they have done a poor job of growing retirement savings. There are fewer assets in defined contribution plans ($10.0 trillion) than in defined benefit plans ($15.4 trillion). This does not take into account the $13.0 trillion in Individual Retirement Accounts (IRAs), perhaps half of which were rolled over from 401(k) accounts.4 However, a rough equivalence between assets held in defined benefit pensions and retirement account plans is unimpressive given the much larger number of workers participating in defined contribution plans.

Retirement account balances are highly unequal

Most households have nothing or next to nothing saved in retirement accounts. Even among households nearing or starting retirement (ages 60–64), 44% have nothing saved in these accounts, and the median household has only $10,400. The top 10%, meanwhile, have over $1 million saved in retirement accounts (Figure B).5

Racial and ethnic disparities are stark.6 Most Black (65%) and Hispanic (70%) households have nothing saved in retirement accounts.7 Even focusing on households approaching or entering retirement (ages 60–64), 65% of Black and 82% of Hispanic households lack retirement account savings. As shown in Figure C, mean account balances also show large and persistent disparities by race and ethnicity.

Many workers get no help from their employer or the government

There are many reasons why most households have little or nothing saved in retirement accounts, including savings diminished by high fees and pre-retirement withdrawals. But the most important factor is that nearly half of workers (47%) do not participate in a plan, in most cases because their employer does not provide one or because they are not eligible.8

Participation rates vary considerably by industry, occupation, and other factors.9 Union membership is an important factor, especially among blue-collar workers, since 94% of private-sector union members have access to a retirement plan and 84% participate.10 Retirement plan participation is higher among construction (50%), manufacturing (69%), and trade, transportation, and utilities (51%) workers, where unionization rates are relatively high, than among leisure and hospitality workers (16%), where unionization rates are low.11 Unions influence retirement plan access not only among unionized employers but also among competing employers.

Though some workers opt not to participate in an employer plan, this is not necessarily a short-sighted decision, as is often assumed. Many low-income and part-time workers are not eligible for an employer contribution and do not gain a tax advantage by participating because they do not owe income tax. These workers nonetheless face a 10% penalty if they need to access their own funds before age 59-1/2 unless they qualify for a special exemption.

Thus, it is not surprising that many low-income and part-time workers, including many retail workers, do not participate in a retirement plan. Only 9% of Dollar General workers, for example, have positive 401(k) balances even though the company considers all its employees “active participants” who are eligible to contribute their own money to the plan. Though the company offers a seemingly generous dollar-for-dollar matching contribution up to 5% of pay, workers are only eligible for the match after a year of service and 1,000 hours, and most retail workers—including, presumably, Dollar General employees—work part-time and do not last a full year.12 Though employers are required to extend eligibility to part-time workers who work at least 500 hours per year for three years (reduced to two years starting in 2025, thanks to the SECURE 2.0 Act of 2022), companies like Dollar General can continue to exclude workers by limiting their hours.

In any case, it would be difficult for most Dollar General workers to contribute to their 401(k) plan since the company has the dubious distinction of employing the highest share of workers (92%) earning under $15 per hour out of the large service-sector employers analyzed in one study.13 Dollar General’s reliance on a low-paid, part-time, and high-turnover workforce kept its 401(k) contributions to a measly $200 per worker in 2022.14

Federal subsidies for retirement accounts are costly and mostly go to high-income households

The U.S. Treasury Department estimates that contributions to tax-favored retirement accounts cost taxpayers $138.5 billion in 2021.15 However, we do not actually know how much these subsidies cost because the federal government does not track untaxed investment earnings and other factors that would be needed to calculate tax losses, and different agencies use different estimation methods.16

In contrast, we closely monitor Social Security’s finances, including long-term projections. The annual release of the Social Security Trustees report elicits concern about projected shortfalls, which the Social Security actuaries last year estimated at 1.3% of GDP over 75 years.17 Meanwhile, Andrew Biggs of the American Enterprise Institute and Alicia Munnell of Boston College’s Center for Retirement Research estimate that the combined cost of tax expenditures related to retirement similarly amounted to 1.3% of GDP annually, yet these subsidies receive far less scrutiny than Social Security’s projected shortfall.18 In addition to receiving less attention, some forgone revenues from tax-favored accounts would improve Social Security’s finances because some contributions to employer retirement plans are not subject to the payroll taxes that fund Social Security and Medicare.

This is not to suggest that all subsidies for retirement should be scrapped. However, those that disproportionately benefit upper-income households and do little to increase saving should be repurposed. The Tax Policy Center estimates that 60% of tax expenditures for retirement saving accounts go to the highest income quintile (20%).19 Since these households tend to save anyway, the subsidies do little to incentivize saving but rather induce households to steer savings to tax-favored accounts.20

Tellingly, most households approaching retirement would have accumulated more in retirement accounts if tax subsidies that were supposed to encourage saving had instead been divided equally among households and invested in Treasury bonds with no employer or employee contributions—a damning assessment of the efficacy of these supposed saving incentives.21

We need to fix 401(k)s before expanding them

There are things we can do to make 401(k)s better, but these have received less support than policies to make them bigger. The financial services industry has embraced automatic enrollment, expanding contribution limits, and pushing back the age that participants are required to start withdrawing funds. In contrast, efforts by the Obama administration and others to improve incentives for low-income savers received little industry support and took years to pass because small accounts are not lucrative. Overdue changes to the low-income Saver’s Credit passed in SECURE 2.0 will not take effect until 2027, whereas provisions benefiting higher-income households, such as raising the age for required distributions, take effect sooner.22 Another indicator of who these accounts really serve: exposés of billionaires shielding millions and even billions in tax-favored retirement accounts have not led to better rules and enforcement around contribution limits or capping account balances, though the Obama and Biden administrations proposed limits on accumulations.23

Meanwhile, efforts to ensure that retirement savers are not steered to high-cost investments by salespeople posing as financial advisors have met with fierce industry opposition.24 Though there has been a welcome trend toward lower-cost passive investment strategies, retirement savers are still being steered to inappropriate high-cost investments through gaps in rules intended to protect savers, since these rules do not apply to rollovers or complex insurance products that even financial professionals have difficulty pricing.25

This is exactly backwards. Before we funnel more money to 401(k)s, we need to ensure that they are an affordable savings vehicle for ordinary workers, not a tax dodge for the wealthy. Even then, our priority should be expanding Social Security and access to secure pensions rather than relying more on 401(k)s.

Social Security is key, but benefits are modest

Social Security replaces only 39% of pre-retirement earnings for a medium earner retiring at 65 though the replacement rate is higher for low earners and lower for high earners. Before benefit cuts were implemented in 1983, including a gradual increase in the normal retirement age, the replacement rate for a medium earner who retired at 65 was 50%.26

Social Security benefits are lower than benefits from similar plans in most peer countries. The average replacement rate for mandatory pensions in the Organisation for Economic Co-operation and Development is 51% for a medium earner—similar to the Social Security replacement rate prior to the 1983 cuts.27

Though modest, the value of Social Security benefits exceeds the value of savings in defined contribution plans for all but the wealthiest Americans. For higher income households, the value of defined benefit pensions exceeds both the value of Social Security and that of defined contribution plan savings (Figure D).

Defined benefit pensions provide financial security for most union members and public-sector workers

Traditional defined benefit pensions are more efficient than 401(k)-style defined contribution plans. Pensions pool risk among workers who retire at different times and have different lifespans. In contrast, defined contribution plans are riskier because individual savers do not benefit from intergenerational risk-sharing or longevity risk pooling. Professionally managed pension funds also earn higher risk-adjusted returns and have lower administrative costs than individual accounts. Due to a lack of risk pooling and lower net investment returns, EPI and others have estimated that contributions to 401(k) plans need to be almost twice as large as contributions to defined benefit pensions to provide similar retirement security.28

In the private sector, most rank-and-file workers who participate in defined benefit pensions are union members. Whereas 58% of union members in the private sector have a defined benefit pension, only 7% of their nonunion counterparts do.29 Defined benefit pensions are the norm in the public sector, where secure pensions partly compensate for lower salaries. Three-quarters (75%) of all state and local government workers—including 82% of full-time workers and 82% of union members—participate in a defined benefit pension.30

Public pensions are especially critical for women and Black workers, who gravitate toward public-sector jobs with secure benefits despite a public-sector pay gap.31 Black (27%) and non-Hispanic white (27%) working households have similar rates of defined benefit pension coverage, as do unmarried or unpartnered working women (15%) and working men (13%).32

While some union members, including many auto workers, participate in single-employer pension plans, others participate in multiemployer plans jointly sponsored by employers and unions. These Taft-Hartley plans are common in sectors where workers often have longer-lasting relationships with unions than with individual employers, such as the trucking and construction industries. Multiple-employer plans such as TIAA are also common among mobile professionals, such as university professors and clergy.

Multiemployer Taft-Hartley plans can survive the demise of individual employers but run into trouble when an entire industry, or union membership in an industry, shrinks rapidly. With a low ratio of active participants to retirees, it becomes difficult for mature plans to adjust contributions to offset volatile investment returns even if the plans were adequately funded to begin with.33 Congress included assistance for troubled multiemployer plans in the American Rescue Plan of 2021, effectively stabilizing them for decades to come. However, additional steps will eventually be needed to ensure that these plans can weather the next storm and serve future generations.

The challenge of defined benefit pensions is that while they eliminate individual longevity and investment risks through risk pooling, sponsors still bear some cohort longevity and market risks because future lifespans and long-term investment returns cannot be perfectly predicted. Pension plans deal with this challenge by gradually adjusting contribution rates to offset nontransitory changes in life expectancy or investment returns—a strategy that works well for public-sector employers and multiemployer plans with a stable or growing number of active participants. Though quasi-mandatory participation can be enforced through collective bargaining agreements, extending pension-like benefits to a broader group of workers might require mandates or greater risk sharing with participants in so-called “hybrid” plans that combine elements of defined benefit and defined contribution plans.

It is time for an overhaul of our retirement system that builds on what works

Social Security is the most important source of retirement income for most workers but needs to be strengthened and significantly expanded along the lines proposed by Chairman Sanders and other cosponsors of the Social Security Expansion Act.34 We also need to expand Supplemental Security Insurance and other social insurance programs to ensure that seniors and people with disabilities are lifted above poverty and families are not devastated by medical and caregiving expenses.

Defined benefit pensions are critical to the retirement security of public-sector and unionized workers. We need to protect existing plans while exploring ways to extend secure pension benefits to more workers. Hybrid plans in some states, provinces, and countries that equitably share risk between employers and workers could serve as models for expanding multiple-employer plans in the United States.35

Three decades after 401(k)s became the most common retirement plan in the private sector, most workers approaching retirement have little or nothing in these accounts. We should stop pretending that the problem is workers and not a poorly designed system that has failed them.

We should begin by implementing the Department of Labor’s Retirement Security Rule without delay. We should also tighten and enforce 401(k) contribution rules, set a $5 million or lower limit on account balances, tax inherited balances immediately, and use the savings from these reforms to help low-income savers.

We should take steps to prevent employers from discriminating against part-time workers, including workers who would prefer to work full-time but are limited to part-time work by employers seeking to save on benefit costs. A provision in the Part-Time Worker Bill of Rights that extended retirement benefits to part-time workers who were employed for two years was included in SECURE 2.0, but more could be done to expand access to other part-time workers.

Ultimately, employers who want to avoid providing retirement benefits will find ways to do so in a voluntary system. Some state and local governments have therefore taken the step of requiring employers to at least facilitate worker contributions to state-sponsored IRAs via automatic enrollment and payroll deduction. While Auto IRAs do not allow for employer contributions, these efforts could prompt federal action, possibly including requirements for employer contributions or a government match.36 The idea of an employer mandate has gained currency in recent decades though it should not be viewed as a substitute for Social Security expansion.

Sincerely,

Monique Morrissey, Ph.D.

Senior Economist

Economic Policy Institute

---

Notes

1. Sarah Max, “The Inventor of the 401(k) Thinks It Has Gone Awry, Barron’s, November 16, 2018; Taylor Tepper, “Ted Benna, Father of the 401(k), on the State of Retirement Savings Today, Forbes, December 22, 2021.

2. Siavash Radpour, Michael Papadopoulos, and Teresa Ghilarducci, Trends in Employer-Sponsored Retirement Plan Access and Participation Rates: Reconciling Different Data Sources, Schwartz Center for Economic Policy Analysis Research Note #2021-01, January 2021; Andrew G. Biggs and Alicia H. Munnell, The Case for Using Subsidies for Retirement Plans to Fix Social Security, Center for Retirement Research at Boston College Issue in Brief 24-2, January 16, 2024.

3. John Sabelhaus and Alice Henriques Volz, “Are Disappearing Employer Pensions Contributing to Rising Wealth Inequality?” FEDS Notes, February 1, 2019; Nadia Karamcheva and Victoria Perez-Zetune, “Defined Benefit and Defined Contribution Plans and the Distribution of Family Wealth,” Congressional Budget Office Working Paper Series, February 2023.

4. Federal Reserve Board of Governors, Financial Accounts of the United States, Third Quarter 2023, Table L.117, 2023 Q2; Sarah Holden and Daniel Schrass, The Role of IRAs in US Households’ Saving for Retirement, Investment Company Institute, January 2022, Figure 10.

5. Though households with defined benefit pensions do not have to rely as much or at all on retirement account savings, excluding these households makes account savings look worse, not better.

6. Monique Morrissey, “Gaps in Retirement Savings Based on Race, Ethnicity and Gender,” Statement before the U.S. Department of Labor Advisory Council on Employee Welfare and Pension Benefit Plans, June 25, 2021.

7. Author’s analysis of Federal Reserve Survey of Consumer Finances microdata, 2022.

8. U.S. Bureau of Labor Statistics (BLS), Employee Benefits in the United States, March 2023 (Excel tables).

9. Monique Morrissey, Siavash Radpour, and Barbara Schuster, The Older Workers and Retirement Chartbook, Chapter 2, Economic Policy Institute and Schwartz Center for Economic Policy Analysis, November 16, 2022.

10. U.S. Bureau of Labor Statistics (BLS), Employee Benefits, 2023.

11. U.S. Bureau of Labor Statistics (BLS), Employee Benefits, 2023; U.S. Bureau of Labor Statistics, Union Member—2023 (News Release), January 23, 2024.

12. U.S. Bureau of Labor Statistics (BLS), “Retail Trade: NAICS 44-45,” Industries at a Glance (web page), February 21, 2024. Annual separations in the retail trade industry in 2023, derived by summing monthly rates, was 55% in 2023 based on BLS Job Openings and Labor Turnover Survey (JOLTS) data accessed on the BLS JOLTS website February 27, 2024.

13. Annette Gailliot, Kristen Harknett, Daniel Schneider, and Ben Zipperer, “The Company Wage Tracker: Estimates of Wages at 66 Large Service Sector Employers,” The Shift Project and the Economic Policy Institute Research Brief, April 2022.

14. Author’s estimate based on Dollar General 401(k) Form 5500 for 2022, accessed through https://www.efast.dol.gov/5500Search/ on February 27, 2024; and Dollar General Form 10-K annual report for 2022, accessed through https://www.sec.gov/edgar/searchedgar/companysearch on February 27, 2024.

15. U.S. Department of the Treasury, Tax Expenditures, FY 2024, Table 4, March 2023, accessed through https://home.treasury.gov/policy-issues/tax-policy/tax-expenditures on February 27, 2024.

16. Benjamin H. Harris, Eugene Steuerle, and Caleb Quakenbush, Evaluating Tax Expenditures: Introducing Oversight into Spending Through the Tax Code, Tax Policy Center, July 10, 2018.

17. 2023 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds.

18. Andrew G. Biggs and Alicia H. Munnell, The Case for Using Subsidies for Retirement Plans to Fix Social Security, 2024.

19. Tax Policy Center, “T22-0264 – Tax Benefit of Retirement Savings Incentives (Present Value Approach), Baseline: Current Law, Distribution of Federal Tax Change by ECI Percentile, 2022,” December 29, 2022, accessed via taxpolicycenter.org on February 27, 2024.

20. Research summarized in Andrew G. Biggs and Alicia H. Munnell, The Case for Using Subsidies for Retirement Plans to Fix Social Security, 2024.

21. Author’s analysis of U.S. Treasury Department, “Present Value of Tax Expenditures” tables, 2001-2022; Federal Reserve Board of Governors Survey of Consumer Finances microdata, 2001-2022; and Federal Reserve Board of Governors, “Market Yield on U.S. Treasury Securities at 10-Year Constant Maturity, Quoted on an Investment Basis,” (accessed through FRED Economic Data, Federal Reserve Bank of St. Louis on February 27, 2024).

22. American Society of Pension Professionals and Actuaries (ASPPA), Key SECURE 2.0 Act Provisions and Effective Dates, January 11, 2023.

23. Fred Hiatt, “Obama’s Modest Proposal to Cap Retirement Entitlements,” Washington Post, May 5, 2013; Justin Elliott, Patricia Callahan, and James Bandler, “Lord of the Roths: How Tech Mogul Peter Thiel Turned a Retirement Account for the Middle Class Into a $5 Billion Tax-Free Piggy Bank,” ProPublica, June 24, 2021; Ted Godbout, “Biden Budget Resurrects ‘Build Back Better’ Retirement Plan Revenue Raisers,” ASPPA, March 10, 2023.

24. Tobias Salinger, “DOL Hearing Promises Passionate Support and Resistance from Industry—Here’s a Sample,” Financial Planning, December 11, 2023.

25. Monique Morrissey and Heidi Shierholz, “EPI comments on DOL’s Retirement Security Rule—Definition of an Investment Advice Fiduciary,” statement submitted to the U.S. Department of Labor Employee Benefits Security Administration on January 2, 2024.

26. Kyle Burkhalter and Chris Chaplain, “Replacement Rates for Hypothetical Retired Workers,” Social Security Administration Actuarial Note 2023.9, March 2023.

27. Organisation for Economic Co-operation and Development, Pensions at a Glance 2023, Table 4.1.

28. Monique Morrissey, Toward a Universal, Secure, and Adequate Retirement System. Retirement USA Conference Report, October 21, 2009; Dan Doonan and William B. Fornia, A Better Bang for the Buck 3.0, National Institute on Retirement Security, January 2022.

29. U.S. Bureau of Labor Statistics (BLS), Employee Benefits, 2023.

30. U.S. Bureau of Labor Statistics (BLS), Employee Benefits, 2023.

31. Nari Rhee, Closing the Gap: The Role of Public Pensions in Reducing Retirement Inequality, National Institute on Retirement Security, September 2023.

32. Married or partnered households have more than double the rate of pension coverage (31%) as single people (14%), in part because there may be two people employed. The source is the author’s analysis of Federal Reserve Survey of Consumer Finances microdata, 2022.

33. Jason Russell, The Role of Plan Demographics in the American Rescue Plan Act, Keynote at the National Institute on Retirement Security 2023 Annual Retirement Policy Conference, February 28, 2023 (presentation slides).

34. Stephen Goss, Letter to The Honorable Bernie Sanders and the Honorable Peter DeFazio (analysis of the Social Security Expansion Act), Social Security Administration Office of the Chief Actuary, June 9, 2022.

35. Dan Doonan and Elizabeth Wiley, The Hybrid Handbook: Not All Hybrids Are Created Equal, National Institute on Retirement Security, May 2021; J. Mark Iwry, David C. John, Christopher Pulliam, and William G. Gale, Collective Defined Contribution Plans, Economic Studies at Brookings, September 2021; Charles E.F. Millard, David Pitt-Watson, and Angela M. Antonelli, Securing a Reliable Income in Retirement: An Examination of the Benefits and Challenges of Pooled Funding and Risk-Sharing in Collective Defined Contribution (CDC) Plans, Georgetown University Center for Retirement Initiatives, Policy Report 21-03, April 2021; Monique Morrissey, Risk Sharing in a Voluntary Retirement Plan, Economic Policy Institute Technical Paper, August 2022.

36. See, for example, Teresa Ghilarducci, Guaranteed Retirement Accounts: Toward Retirement Income Security, Economic Policy Institute Briefing Paper #204, March 3, 2008; Gary Koenig, Jason J. Fichtner, and William G. Gale. Supplemental Transition Accounts for Retirement. AARP, January 2018; Chris Coons and Amy Klobuchar. “Most Americans Can’t Save for Retirement. We Want to Fix That.” CNN Business, April 25, 2019; John Hickenlooper and Thom Tillis, “Retirement Savings for Americans Act: Bipartisan, Bicameral Retirement Savings for Americans Act Would Make Saving for Retirement Reliable, Real, and Attainable for American Workers,” October 18, 2023.