While there is broad agreement that it is necessary to boost retirement savings, there is less uniformity of thought regarding how exactly to accomplish that. A panel of experts recently shared some ideas.
Building a Better Retirement System for all Americans, a Brookings Institution webinar, featured the insights of William G. Gale, Senior Fellow and Director of the Retirement Security Project, The Arjay and Frances Fearing Miller Chair in Federal Economic Policy, Economic Studies at Brookings; J. Mark Iwry, Nonresident Senior Fellow, Economic Studies, Brookings and Visiting Scholar at The Wharton School at the University of Pennsylvania; David C. John, Nonresident Senior Fellow and Deputy Director, Retirement Security Project, Brookings and Senior Policy Advisor, AARP Public Policy Institute; Olivia S. Mitchell, International Foundation of Employee Benefit Plans Professor at The Wharton School at the University of Pennsylvania; and Anita Mukherjee, Assistant Professor, Wisconsin School of Business, University of Wisconsin-Madison and Faculty Affiliate, University of Wisconsin Law School.
The panelists shared their views on a variety of proposals for increasing retirement saving and enhancing the effectiveness of current savings vehicles and approaches. Following are some highlights of the discussion.
Participant Behavior
Part of the problem, panelists indicated, lies in participant behavior. For instance, withdrawals from retirement savings accounts can hurt retirement readiness; they observed that in 2014, 20%—or $65.5 billion worth—of withdrawals from defined contribution plans and IRAs went to retirement-age people, most often due to job changes. Mitchell remarked, “The ability to take money out raises the question—are they really retirement accounts, or rainy day accounts?”
And not only action, but also inaction, can be a problem. The panel noted that:
- 3% of 73-year-olds missed required minimum distributions in 2017;
- 40% of accounts remain inactive for 10 or more years; and
- inactivity appears 10 times higher in accounts created by default enrollment.
Panelists advocated increasing financial literacy as a way to help address retirement preparedness. “You shouldn’t have to be a retirement expert to navigate the retirement system,” said Gale. Mitchell suggested that financial literacy should be incorporated into other vehicles for increasing saving; for instance, she expressed the wish that auto-IRAs could also include “a certain amount of financial literacy.”
“We need to balance concern over participants’ worries with increased financial literacy,” said Iwry, remarking that when plans de-risk, it’s “not a de-risking” for participants, who are ill-prepared to handle it. He added that 401(k)s and IRAs are “do-it-yourself” plans. He argued that it also would be helpful to provide income projections to employees, and more useful than showing shorter-term incremental amounts added to accounts since the last statement.
Small Accounts
The panel identified a variety of reasons why small accounts pose problems for retirement preparedness:
- A flat annual fee can be a higher percentage of a small account’s balance than a large account’s balance. For example, an annual fee of $50 represents a full 5% of a $1,000 balance but only 0.05% of a $100,000 balance. They add that in 2017, a study estimated that the median annual DC plan recordkeeping fee was $59.
- Small accounts involuntarily rolled into an IRA are about 10 times more likely to be abandoned than small accounts that are not involuntarily transferred.
- They are most often held by minority, young and lower-income savers.
The panel had many suggestions regarding how to address the problems posed by small accounts:
- Require DC plans to accept rollovers with appropriate protections.
- Permit Roth IRAs to be rolled over to qualified plans.
- Prevent shrinkage and promote growth of small auto-rollover IRA balances by allowing them to be invested in QDIAs.
- Use the entire account balance in determining the $5,000 limit on auto rollovers to IRAs.
- Support automatic portability rollovers between employer plans when employees change jobs.
John and Mukherjee suggested that another answer could be consolidation of accounts; Mukherjee added that increasing rollover time and improving features of auto-rollover accounts could also help. She further argued that creating a national retirement dashboard could help people keep track of and consolidate accounts.
John and Mukherjee also shared a belief that a lifetime provider would help address the problem of small accounts. Doing so would be the most extensive reform that would help, John said; Mukherjee also sees it as a way to raise balances via aggregation.
Collective Defined Contributions
Collective defined contributions (CDCs) are a hybrid of defined benefit and defined contribution plans. With CDCs:
- employers avoid benefit guarantee and associated funding cost and volatility;
- there is pooled, professional investin—it is not participant-directed;
- retirement income is paid, much as it is with DB plans—however, the amount can vary based investment experience and is not guaranteed; and
- there is pooled longevity risk.
But CDCs may not be a panacea, Iwry indicated. He identified some drawbacks with CDCs, remarking that they offer employers less flexibility, are less portable and shift risk to employees.
In addition, the panel said, there are questions concerning CDC implementation:
- Will participants understand and accept the variable nature of benefits over time?
- Can the plan mitigate the transition from a DB plan to a CDC in order to avoid harming classes of participants?
- Can plan management handle broad discretion to reduce and adjust benefits to achieve equity while avoiding misunderstandings, conflict, mistrust and blame?
- Can lack of portability be sufficiently mitigated?
- Will employers have sufficient motivation and incentive to adopt CDCs?
- Sound, wise plan management, governance and regulation are important.
Saver’s Credit
The panel suggested that the Saver’s Credit could help improve the situation, and that it builds balances while also reducing costs for both the saver and the provider. Increasing people’s retirement income by just $1,000 a year could save states and the federal government several billion dollars that they otherwise would have spent for retiree support programs, they argued. They further suggested making the credit a 50% refundable match that is deposited into the retirement account, and making the credit available on all tax forms.
Mukherjee and John further argued that enhancing the Saver’s Credit also could help address the problems occasioned by the proliferation of small accounts. Mukherjee suggested improving the Saver’s Credit and replacing the nonrefundable credit on the federal tax return with a refundable credit given as a savings match.
State-run IRAs
The panel expressed support for state-run IRA programs, but it was qualified support. To wit: the panel said that state-based mandatory savings accounts are helping, but cited statistics showing that in Oregon the participation rate is 34% and the average account balance is $865 after a year. Iwry remarked that state IRAs “illustrate how a single account system could work very well,” but added that at the same time pursuing such programs should not come at the expense of the employer-based system.
Build on the Foundation
Panelists supported building on the current system, not abolishing it. “We have a system that is not working as well as it could, but is working in many ways,” said Iwry. “Many employers have done so much good,” he noted, adding that minorities, women and low-paid employees have done much better when they are in an employer-based plan, especially one that has implemented auto-enrollment. And he added a note of caution that while pursuing state-based programs, “We need to be careful while working down that road not to do harm to the employer-based system.”
Every retirement system is past-dependent—we are where we are because of past developments, said Mitchell, but added that that is not enough, saying, “Forward-thinking forecasts must offer ranges of outcomes.”
