Background

Our national retirement system consists of a variety of programs that have evolved over time. Lawmakers and regulators have at times made changes to individual programs to address narrow issues or concerns, without always considering the longer-term impact on broader retirement security policy as a whole. Other changes to retirement programs were designed to achieve objectives unrelated to retirement security, such as a desire to increase tax revenues to fund infrastructure spending. The piecemeal approach under which the various types of retirement plans currently in use have developed has arguably led to inefficiency and a lack of coordination among the programs.

Over the years, several attempts to develop a cohesive framework for the U.S. retirement system have been initiated. In 1979, the President’s Commission on Pension Policy was established by President Jimmy Carter. The Commission’s report, which was never adopted, made recommendations related to employee pensions, Social Security, individual efforts, and public assistance. In June 2016, a Bipartisan Commission on Retirement Security and Personal Savings issued a detailed report of recommendations. Though not a call for a formal national retirement policy, the report has stimulated thought on the shortcomings of the current retirement system. Most recently, in October 2017, the Government Accountability Office (GAO) published a comprehensive evaluation of retirement security, which recommended that Congress establish an independent commission to examine the U.S. retirement system. Legislation was introduced in the Senate during the spring of 2019 that would create a federal Commission on Retirement Security. However, as of the date of this issue brief, the proposed legislation had not yet been acted upon.

Purpose of a National Retirement Policy

A national retirement policy would ideally articulate a set of guiding principles designed to provide individuals with the opportunity to achieve financially secure retirements. These principles could incorporate and address such elements as: availability of retirement programs, benefit adequacy, allocation of risks, treatment of different income levels, use of incentives, individual choice, costs, and portability and leakage. Note that principles, once adopted, can change over time based on changes in demographics and economic and political circumstances. Well-chosen principles will be sufficiently broad and comprehensive to reflect and address the gender and race gaps in wealth, wages, and savings. An effective policy will also take into account the needs of individuals who are not in the workforce on a long-term basis or those who have significant gaps in their careers (e.g., women who leave the workforce during caregiving years). The rise and ramifications of a high-technology-based economy, in which temporary positions for short-term engagements are common, might also be considered.

Although clearly a primary objective of a successful retirement system is to provide financial security to workers after they stop working, difficult questions remain such as how much retirement income is sufficient, or what are the characteristics of the programs that will provide this income. Is it enough for the system to provide everyone with the tools needed to build a financially secure retirement, while placing the responsibility for using those tools on individuals? What role should employers be expected to play in providing retirement benefits? These are the types of questions that a national retirement policy could address.

A national retirement policy does not need to result in a sweeping overhaul of our retirement system to be successful. Rather, it could serve as a guide for future incremental changes that would promote the principles of the policy. Over time, consistently using a well-developed national retirement policy to evaluate existing retirement programs and proposed changes to those programs could help our retirement system become more efficient and effective, while minimizing unnecessary complexity and overlap.

In 2014, the American Academy of Actuaries released a framework—Retirement for the AGES (Alignment, Governance, Efficiency, and Sustainability)—to assist in formulating public policy to help sustain and improve employer-based retirement programs. The principles governing a national retirement policy must necessarily be broader than the principles for employer-based programs. However, the AGES principles, which are discussed in more detail later in this issue brief, continue to be useful in assessing the effectiveness of the employer-based component of the U.S. retirement system.

Elements of a National Retirement Policy

Below are descriptions of each of the possible elements of a national retirement policy noted above. To provide context, we offer illustrative approaches as to how each element could be incorporated into a national retirement policy. These potential approaches are illustrative only; they do not represent positions of the American Academy of Actuaries’ Pension Practice Council.

Availability

In theory, all U.S. workers can voluntarily save for retirement through private after-tax savings, or through either tax-advantaged employer-sponsored programs or Individual Retirement Accounts (IRAs). However, many low-income workers may not have sufficient resources to save for retirement. With some exceptions (e.g., certain employees of state and local governments), almost all workers participate in Social Security. All public and private employers can offer defined benefit (DB) or defined contribution (DC) plans to their employees, but not all employers do so. A 2014 analysis by the Center for Retirement Research at Boston College estimated that roughly 65 percent of workers are covered by an employer-sponsored retirement plan, with approximately 50 percent of the workforce choosing to participate.¹ (For some employees, “coverage” might only include the ability to defer their wages into a tax-deferred account, with no benefits funded by employer contributions.) In other words, a large segment of the population is not covered by an employer-sponsored plan or does not participate in a plan that is available to them.

A national retirement policy could include a targeted level of availability for employer-sponsored plans. A relatively low target would imply a greater reliance on personal savings and Social Security. A higher target might require substantial incentives to encourage employers to offer plans. Universal availability could likely only be achieved by mandate.

A national retirement policy could also address whether requiring an employer to merely offer employees access to a structured retirement program without employer contributions is sufficient, or whether a targeted level of employer contribution or benefit is necessary. The targets could be different for different segments of the population. For example, lower-wage earners may need to rely more heavily on employer-provided benefits than higher-wage earners, who might be able to save more of their income independently. 

Potential Approach: A national retirement policy could include a goal that all private-sector companies with a minimum number of employees should provide their workers with access to an employer-sponsored retirement plan.

Adequacy

Although what constitutes an adequate level of retirement income is not universally defined, a commonly cited goal is that people be able to maintain the same standard of living in retirement that they experienced while they were working. The portion of pre-retirement income necessary to meet this goal can vary widely from person to person based on factors such as marital status, medical expenses, homeownership, and the need to support dependents. Another possible objective is to target retirement income that exceeds some predetermined level, such as the poverty level or a minimum dollar amount.

The level of retirement income produced by personal savings and voluntary salary deferrals in employer-sponsored plans depends largely on employees’ behavior and their ability to save. This behavior includes the amount of money an employee contributes to the plans while working, how the employee invests the assets, and how savings are drawn down in retirement. Because in these plans employees get to choose whether to contribute the maximum amount allowed or some amount less than that, a national retirement policy could differentiate between the level of retirement income that these plans have the capacity to produce and the level that is actually produced.

An adequacy target could address only total retirement income from all sources, or it could consist of individual objectives for the various sources of retirement income. A national retirement policy might focus on the level of retirement income that is considered to be sufficient for various cohorts of workers, taking into account the fact that not everyone is in the workforce for all of their working years and that income levels for some segments of the population may be too low to allow for significant savings. Alternatively, a more sophisticated approach might establish statistical goals, such as having at least a specified percentage of retirees with sufficient income to stay above the poverty level or to replace a certain percentage of their pre-retirement earnings.

Potential Approach: A national retirement policy could establish an explicit goal that individuals have retirement income from all sources that is sufficient to avoid old-age poverty.

Allocation of Risks

Retirement systems face investment risk as well as longevity and other demographic risks. At a very high level, when a risk produces adverse experience, it will either increase the cost associated with a retirement program or decrease the benefits that will be paid. Demographic and economic forecasts are inherently uncertain, which causes the cost of providing defined benefits, such as Social Security, to deviate from expected, perhaps significantly. Similarly, the level of personal savings or defined contribution account assets needed to support retirement income could be much different from expected—again, maybe considerably.

Longevity risk, if borne by an individual, can result in the individual outliving his or her retirement savings. Longevity risk can also be borne by the government (and therefore the taxpaying public), employers, and insurers. These entities can pool the risk among large groups of individuals, which can be an effective risk-mitigation technique because the average lifespan across a large population is more predictable than the lifespan of an individual.

An example of the federal government bearing longevity and other risks (e.g., birth rates and the growth of national average wages) is Social Security, although those risks could be ultimately passed on to beneficiaries in the form of lower benefits or to workers and employers through higher payroll taxes. Defined benefit plans are generally structured so that longevity risk is borne by the sponsoring employers, while in defined contribution plans individual participants generally bear the longevity risk. In the case of fixed-income annuity contracts, the longevity risk is borne by the insurance company.

The level of risk in a retirement plan can be managed. Both defined benefit and defined contribution retirement plans, for example, could greatly reduce the level of investment risk by allocating all of the plan assets into high-quality fixed-income securities. The downside is that the level of long-term investment returns would likely be substantially lower under a less risky approach, resulting in a combination of higher plan costs and reduced benefit levels.

 A national retirement policy could address the level of risk that is incorporated into various components of the retirement system, how those risks are shared among stakeholders, and the extent to which adverse experience associated with those risks results in lower benefit levels or higher costs.

Potential Approach: A national retirement policy could call for the federal government to encourage the use of longevity pooling mechanisms, such as the payment of benefits as annuities under individual account-type plans or defined benefit plans.

Treatment of Different Income Levels

The traditional three-legged stool of retirement income consists of personal savings, Social Security, and employer-sponsored plans. These components can appropriately take on very different structures and roles based on the level of an individual’s income and years of workforce participation.

A national retirement policy might reflect a belief that Social Security should be less about individual equity (benefits based on wages subject to payroll taxes) and more about social adequacy (income redistribution).² For individuals in the higher income brackets, Social Security could have a negligible role and those individuals will likely accumulate sufficient personal savings to achieve financially secure retirement. Employer-sponsored plans could function the same way for these individuals as they do for lower-paid workers, or a national retirement policy might encourage employers to provide more meaningful benefits to lower-paid employees than to high-wage earners. Conversely, most individuals at lower income levels might be unable to accumulate meaningful personal savings, leaving employer-sponsored plans and/or Social Security as the primary sources of retirement income.

Potential Approach: A national retirement policy could provide that Social Security gradually transition to a structure that focuses even more of the share of benefits toward low-wage earners than it already does.

Use of Incentives

To the extent that a national retirement policy anticipates that employer-sponsored plans will be a source of retirement income, the policy might also address the incentives that employers have to offer such plans. Competition for talent in the labor market, a desire for orderly transitions from one generation of workers to the next, and genuine concern for the well-being of retired employees are all factors that can motivate companies to offer retirement plans. A national retirement policy could recognize that the significance of these factors can change over time, and that regulatory incentives could help ensure that employer-sponsored plans consistently fulfill the role that is anticipated by the policy. 

A national retirement policy could address the role that incentives have in encouraging employers to offer retirement plans. The most significant incentive under current law that encourages companies to sponsor retirement plans is the tax treatment available through qualified plans. A policy could also call for consistency between the time period used to measure the impact of tax incentives on the government budget and the long-term nature of retirement programs.³

Additionally, the nondiscrimination rules for qualified plans provide an incentive for companies to offer plans to lower-wage workers (assuming the companies want to provide benefits to higher-wage workers). Other incentives could also be created, such as tax penalties on companies that do not offer retirement plans or exempting retirement benefits from taxation entirely.

Potential Approach: A national retirement policy could call for the use of tax incentives to raise the proportion of lower-wage workers who are covered by employer-sponsored plans to a specified level.

Individual Choice

Different components of a retirement system can involve different levels of choice. In the current system, individual savings and Social Security are at opposite ends of the spectrum with regard to individual choice. For individuals who have sufficient resources to save for retirement, every aspect of individual savings is subject to the discretion of the individual, including how much to save (if anything), how to invest the money, and how to convert savings into retirement income. Social Security, in contrast, only allows eligible individuals to decide when to begin receiving benefits.

Employer-sponsored plans contain a wide array of individual choice features. Certain defined contribution plans allow employee deferrals and provide only matching contributions at the employer level.⁴ These plans provide no benefits unless employees choose to contribute to the plan. However, defined contribution plans can also provide employer contributions independent of an employee’s decision to contribute. Employees often have wide latitude regarding how to invest their accounts. In some cases, this latitude means making investment decisions that employees do not have the knowledge or skill to make effectively, although this can be mitigated through the use of lifecycle or target date funds that automatically diversify the investments and adjust the allocations as participants age. In some defined contribution plans, the employer makes all of the plan investment decisions. Most employer-sponsored defined benefit plans allow a wide range of retirement ages and allow the employee (and spouse) to choose the type of annuity. Many defined benefit plans also permit employees to receive their benefits as lump sums instead of annuities.

A national retirement policy could take into account the extent to which individuals have choices with respect to their retirement income vehicles. More choice will tend to provide people with the freedom to customize their retirement programs to meet their specific needs. However, additional freedom brings with it the consequences that result from suboptimal decisions. For example, a retirement system with fewer choices could reliably ensure that all retired workers have adequate income. But this approach might also provide retirement income to some people who do not need it, while providing inadequate income to others. Allowing more choice could address these shortcomings but could also lead to people making poor decisions that ultimately leave them with inadequate retirement income and possibly dependent on social safety net programs. In turn, greater dependence on social safety net programs could result in a need for additional tax revenues or reductions to other government programs to offset the increased social safety net spending. Balancing these competing objectives is difficult, and a national retirement policy could provide a framework for achieving this balance.

Potential Approach: A national retirement policy could specify that individuals are provided sufficient flexibility to customize their retirement programs to their specific needs, while not placing demands on them to make choices outside their area of knowledge or comfort level.

Cost

The value that a retirement system provides must be balanced against the cost of supporting it. The fact that the cost of providing retirement income can vary over time complicates this comparison. The cost of supporting Social Security is higher now than it was in past decades because it is a pay-as-you-go system, and there are more retirees for each worker today than there were in the past. The cost of providing retirement benefits through both employer-sponsored plans and individual savings can vary widely based on interest rates and asset returns.

Adequacy and risk are critical factors when evaluating cost. In general, higher levels of adequacy and lower levels of risk will correspond to higher costs.⁵ The objectives of a national retirement policy are unlikely to be achieved if the associated costs are more than individual savers, employers, and taxpayers are willing and able to bear. For this reason, a national retirement policy might address not only the objectives of the retirement system, but also the costs of supporting the system.

Potential Approach: A national retirement policy could target a minimum level of benefits from all sources while constraining aggregate retirement plan contributions so that they do not exceed a specified percentage of gross domestic product.

Portability and Leakage

When employer-sponsored retirement plans were introduced, individuals commonly worked for a single employer for many years. Now, individuals might have many employers over their careers, resulting in a patchwork of benefits payable from multiple sources that can make retirement planning difficult. Retirees might completely lose track of some of their retirement benefits, potentially resulting in their loss. (Although plans are required to search for missing participants, finding these participants can be a long process, during which no benefits will be paid.) Plans may also incur higher administrative costs if they are required to keep track of a large number of small benefits, while those costs could be reduced if employees are able to consolidate their benefits in a single plan. A related concern is “leakage,” which occurs when money that is set aside for retirement is actually used for other purposes, such as purchasing a residence or satisfying debts that are incurred prior to retirement.

Certain defined benefit plan designs accumulate much of their value in later years because they base all benefit accruals on employees’ final salaries. Under these plans, even if a person switches to a new employer that offers the exact same plan, the benefit earned with the first employer will no longer grow with salary increases, and much of the value of the later years of service will be lost. This potential loss of retirement benefit value due to job changes can be viewed as an additional portability concern.

A national retirement policy could help alleviate these problems by supporting policies that make it easier for participants to transfer benefits—both from defined benefit and defined contribution plans—between employers or into IRAs. Making pensions more portable would greatly reduce the problems created by participants simply losing track of their benefits, because benefits will be paid from fewer sources. A national retirement policy could also provide employers with tax incentives to offer plans that do not penalize employees for switching jobs.

Potential Approach: A national retirement policy could call for a regulatory structure that minimizes the administrative burdens associated with rolling benefits earned with a previous employer into a plan sponsored by a new employer or an Individual Retirement Account.

Retirement for the AGES

The Academy’s Retirement for the AGES (Alignment, Governance, Efficiency, and Sustainability) initiative is intended to provide a framework for assessing employer-based retirement programs, or policy changes that would that would affect them, to understand how well they meet the needs of each of the stakeholders.

The AGES framework initiative is based on four key principles:

• Alignment: Retirement income systems work best when stakeholders’ roles are aligned with their skills. Important tasks, such as financial analysis, investment management, and retirement plan administration, should be the responsibility of those who have the knowledge and experience to perform them well.
• Governance: Making and implementing good decisions are essential for successful retirement plans. Good governance helps balance the complex needs of various stakeholder groups, as well as oversees significant trustee, administrative, and investment functions.
• Efficiency: Risk pooling, accurate pricing, appropriate use of guarantees, and other financial techniques should be adopted or incorporated to ensure that a retirement income system is efficient and maximizes income while avoiding excessive risk to stakeholders.
• Sustainability: Roles and skills, good governance, and financial efficiency should be structured to support a sustainable retirement income system that is able to withstand the financial shocks of recessions or times of extraordinary inflation.

A national retirement policy could incorporate these principles as a foundation for developing effective employer-sponsored retirement programs.

Conclusions

The establishment of a national retirement policy commission could be an important step forward. The overarching goal of such a commission would be to review the nation’s current retirement systems and the state of individual retirement security. Such a commission could be charged with developing recommendations for an articulated, cohesive policy to guide and drive future retirement legislation and regulations.

Formulating a comprehensive retirement policy might not be easy to accomplish. A large and diverse population inherently includes individuals and employers with a wide range of economic circumstances and needs. An aging population introduces challenges that past generations did not face. Different ideological perspectives contain stark differences regarding the proper role of government in the lives of individuals. Economic considerations can pose significant challenges. For example, focusing heavily on saving for the future at the expense of current consumption might have short-term consequences on the U.S. economy. In addition, providing tax incentives can hinder the government’s ability to provide other services.

Our current retirement system is disjointed, relying on a variety of laws administered by several regulatory bodies at different levels of government. A national retirement policy could bring focus and clarity to this complex system, ultimately helping to provide retirement income security more efficiently to as many Americans as possible. Whether reforms are undertaken piecemeal or as part of a comprehensive package, the considerations discussed in a national policy could serve as a roadmap for policymakers as they work to improve the effectiveness of our retirement programs.

APPENDIX—

ADDITIONAL RESOURCES FROM THE AMERICAN ACADEMY OF ACTUARIES

The American Academy of Actuaries has published a great deal of information related to retirement policy. Below are various Academy publications that might be beneficial to policymakers as they seek to introduce a retirement policy to help Americans achieve a more financially secure retirement.

A. Social Security benefits could be raised for certain parts of the citizenry. These cohorts include single women, low-lifetime wage-earners, and the very elderly. The program also might benefit from automatic benefit adjustments that would secure its sustainability. Changes would need to be evaluated in light of both financial and political constraints. Some Academy publications on these topics are:

• An Actuarial Perspective on the 2019 Social Security Trustees Report (May 2019)
• Social Security—Automatic Adjustments examines automatic approaches to maintaining long-term solvency. (May 2018)
• Women and Social Security examines specific challenges faced by women under Social Security. (May 2017)
• Helping the ‘Old-Old’—Possible Changes to Social Security to Address the Concerns of Older Americans addresses possible ideas for assisting the very old who are vulnerable to outliving retirement savings. (June 2016)
• A Guide to Analyzing Social Security Reform explores different approaches to changes to the program. (December 2012)
• Means Testing for Social Security examines how the program can be modified to reduce benefits for employees without as much need. (December 2012) 

B. Retiree Lifetime Income creation has become a much larger challenge with the decline of traditional defined benefit plans and increasing life expectancies. Individuals could need help recognizing and adapting to this challenge. Some Academy publications on this topic are:

• Comments on the Report of the Commission on Retirement Security and Personal Savings—provides comments on the recommendations of the Bipartisan Policy Center in its 2016 Report of the Commission on Retirement Security and Personal Savings (February 22, 2019)
• Retirement Income Options in Employer-Sponsored Defined Contribution Plans—position statement in support of policy and educational initiatives to increase retirement income options within employer-sponsored defined contribution plans. (October 31, 2017)
• Risky Business: Living Longer Without Income for Life—Legislative and Regulatory Issues examines the importance of a secure income that lasts a lifetime. (October 23, 2015)
• Risky Business: Living Longer Without Income for Life—Actuarial Considerations for Financial Advisers provides actuarial insights regarding lifetime income planning. (October 23, 2015)
• Risky Business: Living Longer Without Income for Life—Information for Current and Future Retirees explains how risk-sharing can help manage longevity risk. (October 23, 2015)
• Retiree Lifetime Income: Choices & Considerations explores key decisions and options available in the years leading up to and during retirement. (October 23, 2015)
• Retiree Lifetime Income: Product Comparisons examines general insurance and investment products to create lifetime income. (October 23, 2015)
• Risky Business: Living Longer Without Income for Life—A discussion paper by the Lifetime Income Risk Joint Task Force. (June 19, 2013)

C. Multiemployer Plan Sustainability threatens the retirement income of over a million current and future retirees. Solutions to this national issue involve no easy decisions. Some Academy documents on this topic are:

• Follow-up Letter to Joint Select Committee on Solvency of Multiemployer Plans summarizes key topics discussed during June 22 meeting regarding loan proposals and other solutions for troubled multiemployer pension plans. (September 26, 2018)
• Multiemployer Pension Plans: Potential Paths Forward explores options to address failing multiemployer plans and ways to strengthen the multiemployer pension system. (June 27, 2017)
• Honoring the PBGC Guarantee for Multiemployer Plans Requires Difficult Choices examines the Pension Benefit Guaranty Corporation’s multiemployer pension program, which could exhaust its assets in less than 10 years. (October 20, 2016)
• The Multiemployer Pension Plan System: Recent Reforms and Current Challenges examines the future of the program and its many challenges. (March 17, 2016)

D. Public Defined Benefit Plan Funding issues have received much attention over the past several years. Although some public plans are well-funded, certain states and municipalities sponsor defined benefit plans that are significantly underfunded, and considerable debate has ensued as to how public sector plans should evaluate their funding levels. Some Academy publications on this topic are:

• Assessing Pension Plan Health: More Than One Right Number Tells the Whole Story explores various methods to measure the financial health of pension plans. (July 13, 2017)
• Objectives and Principles for Funding Public Sector Pension Plans introduces the objectives and principles for funding pension plans for state and local government. (February 19, 2014)
• Measuring Pension Obligations examines the different measurements of the obligations of defined benefit pension plans. (November 21, 2013)

E. Multiple Employer Plans are uncommon, but their expansion could benefit employees of small employers. Some Academy documents relating to this topic are:

• Comments to Senate Aging Committee on Open Multiple Employer Plans (MEPs) discusses changes to funding rules and administrative responsibilities for defined benefit MEPs and a new concept to offer retirees in defined contribution plans the opportunity to roll assets over to a provider specializing in retiree solutions. (September 1, 2016)
• Retirement for the AGES Assessment of Proposal: USA Retirement Funds contains an assessment of Senator Harkin’s proposed USA Retirement Funds, which would be hybrid pension plans available to small employers. (November 13, 2014)

F. Alternative Retirement Plan Designs could offer an opportunity for employers to provide lifetime income to employees, while protecting employers from the financial risks present in the current DB/DC paradigm. The Academy discussed this opportunity in:

• Retirement for the AGES Assessment: New Brunswick Shared Risk Model contains an assessment of a risk-sharing defined benefit structure adopted by certain Canadian plans. The shared-risk model seeks to provide promised benefits with a high degree of probability while utilizing actuarial stress-testing and self-adjusting mechanisms to ensure sustainability. (November 13, 2014)

G. Retirement Ages should be evaluated and potentially updated to reflect current longevity experience. Some Academy publications on this topic are:

• Essential Elements publication “Raising Social Security’s Retirement Age” is a paper on the benefits of raising Social Security’s retirement age to help solve the program’s long-term financial problems. It provides statistics on the demographics of the system from the 2017 Social Security Trustees’ Report, the impact on beneficiaries, and the benefits of raising the retirement age. (July 2017)
• Rethinking Normal Retirement Age for Pension Plans states that raising the maximum allowable normal retirement age in defined benefit retirement plans would align U.S. pension policy more closely with Social Security’s increasing retirement age and could benefit workers by allowing them to amass more retirement savings. (March 7, 2013)
• Retiree Lifetime Income: Choices & Considerations talks about when to retire and provides links to other sources of information. (October 2015)
• Actuaries Advocate Raising Social Security’s Retirement Age: The Academy’s first (and to this point only) public advocacy statement. (August 2008)

 

---