Statement of Dallas L. Salisbury, 
President and Chief Executive Officer, Employee Benefit Research Institute*

Testimony Before the Subcommittee on Oversight
of the House Committee on Ways and Means

Hearing on Retirement Security and Defined Benefit Pension Plans

June 20, 2002

Mr. Chairman and members of the Committee, it is a pleasure to appear before you today to discuss retirement security and defined benefit pension plans. I am Dallas Salisbury, President and CEO of the Employee Benefit Research Institute.  EBRI [1] has been undertaking research and education on employee benefit issues since its founding in 1978.  EBRI does not lobby for or against specific proposals, instead our mission is to provide data that will assist others in assessing trends and in making policy decisions.

Since my full submission will be included in the hearing record, I will provide a brief summary of points for your consideration:

1. Since I joined the U.S. Department of Labor in 1975 to assist in the implementation of the Employee Retirement Income Security Act of 1974 (ERISA), defined benefit pension plans have changed a great deal.
2. Then, nearly all paid benefits in the form of annuities for most individuals when they reached normal retirement age.  Essentially all of the nation’s largest employers had a defined benefit plan and a thrift-saving or profit-sharing plan, and multi-employer trusts and public employers had defined benefit plans.
3. Today, largely as a result of decisions made by government, defined benefit pension plans pay more individuals lump-sum distributions than annuities, supplemented by defined contribution plans to which the employer contributes.  Many of the largest new-economy employers that never had a defined benefit plan, and are now among our largest employers, rely exclusively on defined contribution plans.  Most multi-employer trusts and public employers sponsor both defined benefit and defined contribution plans.
4. Data from the Federal Reserve Survey of Consumer Finance documents the trend toward plan change.^[i]  Of all families reporting at least one worker with some type of pension coverage, the portion of those families with at least one worker participating in a defined contribution plan only was 57% in 1998, compared with 38% in 1992, while families with at least one worker participating declined from 40% to 21% between 1992 and 1998, while workers with both stayed steady at 22% in both 1992 and 1998.  The best available estimates suggest that the vast majority of defined benefit plan participants who leave an employer with less than 10 years of service take a lump-sum distribution; that over half of all defined benefit plans now offer a lump-sum distribution at retirement; and that nearly all of the over 500 individual account defined benefit plans (“cash-balance” plans) offer lump-sum distributions. 
5.  I note this trend toward defined benefit plans paying lump-sum distributions because it fundamentally affects the way in which a defined benefit plan contributes to retirement security, yet too many articles and analyses still assume/suggest that all defined benefit plans pay annuities upon retirement, thus shielding retirees from the need to make investment, longevity, rate of spending, and other decisions required of those who are paid lump-sum distributions.  The year 2002 finds far less difference between the amount of retirement security provided by the defined benefit and defined contribution plan systems than existed in 1974. 
6. Public policy change joined with demographics and economics to bring these two plan types closer together.  The primary difference between defined benefit and defined contribution plans to this day is the fact that private employers make the funding contributions to defined benefit plans, and in the event of adverse investment performance must contribute more in order to pay the promised accrued benefit, while both employers and workers generally contribute to private defined contribution plans, and the worker alone bears the burden, or gains the fruits, of bad or good investment performance.   
7. The federal government was one of the first major employers to drastically reduce the generosity of its defined benefit pension plan, while adding a defined contribution plan (1984), but many others in both the public and private sectors have followed suit.  The primary reasons it was done:  a desire to reduce cost and future funding liabilities; a desire to reduce the golden handcuffs that make it difficult for a worker to change jobs; a desire to allow greater fund accumulation for shorter service workers; a desire to provide a program that workers would better understand and be more likely to appreciate. 
8. Rules and regulations related to defined benefit plans are extensive and complex, as is the administration of the plans, as indicated by the recent Department of Labor report.  That report underlined the shift of plans to the payment of lump-sum distributions, and the complexity of making the benefit payment calculations.   The worker tradeoff for this complexity, and the potential for errors, is that the employer typically makes all contributions to the plan and the participant is protected (up to the PBGC guaranty limit) against investment “losses” as well as an entire array of potential deviations from actuarial assumptions.
9. Defined benefit plans (with the exception of a few contributory plans) are full participation plans, as workers do not make a choice on whether or not to participate.  For workers that may not be inclined to contribute to a 401(k) plan (particularly the lowest paid workers), this may make a defined benefit plan preferable to a 401(k) plan (although some 401(k) plans may provide nonelective contributions whether or not the employee contributes).  Were that worker to stay for a full career (a low probability), the benefit value/account balance would grow to a level amounting to a meaningful contribution to retirement security.  Were the worker to leave after a few years of employment, either a defined benefit plan or a defined contribution plan would provide a small lump-sum distribution that likely would be spent.
10. Large employers that had defined benefit plans in 1974, and are still in business, in most cases still have them.  The design changes to lump sums and cash balance have allowed them to compete with defined contribution plans for worker understanding and appreciation.   Proposals such as allowing pre-tax worker contributions to defined benefit plans would further erase the differences between the plan types, and might lead to an increase in the sponsorship of defined benefit plans.   Were benefits paid in lump-sum form, however, this would likely have no favorable impact on retirement security (once the worker retires) relative to a defined contribution plan.
11. My father’s defined benefit pension plan began paying him a monthly annuity in 1978.  Today, that check represents a very important contribution to my parents’ retirement security.  Why?  Largely because he and my mother have lived years longer than they planned or expected to, and they have spent all the money they saved.  The greatest virtue of an annuity is this protection against unexpected longevity.  That is the only true form of retirement income security: a check that does not stop until one dies.   It is no longer the case that all defined benefit plan retirees choose to be paid in annuity form, and few defined contribution participants do so.  Future retirement security should no longer be based on a debate about defined benefit or defined contribution, as that is no longer the central issue when both plan types paying lump-sum distributions at job change and retirement.  The future debate must be about worker education on savings, investing, longevity, retiree health, long-term care, and what choices individuals can make to avoid running out of money before they die. 

Introduction

A review of the state of defined benefit pensions must begin with a clear understanding of what a ”pension plan” is.   While this sounds simple, it is done because the “legal”  meaning has clearly changed over the past 28 years.^[ii] Today, the term is used to describe any employer or government-sponsored capital accumulation program that has a stated purpose of providing funds for retirement.  Defined benefit, defined contribution, annuity payment or lump-sum distribution form, all are within the new definition.

ERISA expansion of the definition of pension plan to include capital accumulation plans with lump-sum distributions at “termination of covered employment,” as opposed to “at or near retirement,” actually serves to clearly highlight the “State of Pensions” in the United States.  Both the public and private sector have moved in the direction of sponsoring fewer plans that only pay benefits   “at or near retirement”, and have created more and more plans which pay at “termination of covered employment”.  The result has been dramatic changes in defined benefit pension plans  -- those that promise a fixed accrual and a determinable benefit without worker investment risk – including the development of defined benefit individual account plans (“cash-balance” plans) and growth in the number of defined contribution plans – those that promise payment of funds contributed (once the employee is vested), adjusted for investment earnings, but promise no fixed benefit, as the worker holds investment risk.

I do not provide a normative assessment of whether these trends are good or bad for employers, unions, individuals, or public policy.  They are what they are. 

1996 data from the U.S. Bureau of the Census combined all plan types under the single heading of  “pension,” as do the data from the Federal Reserve.  The data show the impact of a maturing pension system, with the divergence of net flows and net contributions.  Net flows are a measure of new contributions, plus all investment earnings, less benefit payments.  Net contributions are a measure of benefit payments less new contributions.  The fact that net contributions are negative, while net flows are positive, underlines the primary virtue of advance funding, compound interest, and investment earnings.

For the individual worker, the move to more lump-sum distributions from defined benefit and defined contribution plans suggests a number of needs:

• A need for basic financial literacy education.
• A need for understanding saving represents a tradeoff in lifestyle today in order to have money to live on tomorrow.
• A need for understanding investing, fees, returns net of fees, etc.
• A need for evaluation of how important the job-related benefits are, and the degree to which they may determine happiness for a lifetime.

What is the Pension Landscape Today?

Congress acted in 1983 to change the pension system for federal civilian employees.  Prior to 1984, the only federal retirement plan was a final pay defined benefit plan.  For those hired after the 1983 act, a new reduced defined benefit plan was accompanied by a generous 401(k)-type plan.  Those already working had the option of remaining in the old plan or shifting to the new plans. Congress had also acted in 1978 to add two new sections to the Internal Revenue Code, 125 and 401(k).  Proposed regulations in 1981 eventually led to a massive transition of traditional profit-sharing plans into 401(k) plans, which meant that the employee could contribute pre-tax dollars assuming the employer incorporated a 401(k) feature in their profit sharing plan.  State and local governments, and non-profit organizations, had this type of opportunity in 457 and 403(b) plans.  Legislation since 1986 has moved all these so-called “salary-reduction” plans closer together in design and rules, with nearly all employers now able to establish 401(k) plans.  Recent years have seen debates in a number of states over proposals to either introduce expanded supplemental “salary reduction” plans, or to replace defined benefit plans with defined contribution plans.  Demographic change, and economic competition, makes it likely that these debates, and trends, will continue.

The following table presents data from the U.S. Department of Labor on private employer pension plans in terms of number of plans.  The trend lines are clear:  defined benefit plans are on the decline and salary reduction plans are becoming the primary “pension” plans in the nation.  The numbers on multi-employer plans reinforce the trend line of increasing use of supplemental and primary defined contribution programs.iii, iv  Finally, the data hide the use of lump-sum distributions in defined benefit plans.

Number of Qualified Private Pension Plans 1975-2002

Sources:  U.S. Department of Labor and author estimates.

As the number of plans has changed, so have the numbers of participants.  Data from the Federal Reserve Survey of Consumer Finance document this trend through 1998.^[iii] Considering all families reporting at least one worker with some type of pension coverage, the number of those families with at least one worker participating in a defined contribution plan only was 57% in 1998, compared with 38% in 1992, while the portion of families with at least one worker participating in defined benefit plans declined from 40% to 21%, while workers with both stayed steady at 22%.

Employer preferences for pensions now focus more on economic performance than retirement income security. Pensions are viewed favorably if they serve to:

• Improve corporate efficiency.^[iv]
• Enhance morale.
• Keep channels for promotion open.
• Facilitate work- force reduction.
• Enhance employee identification with profit.
• Offer a most cost-effective and least administratively intense form of capital accumulation.
• Attract and hold capable employees.^[v]

A senior corporate executive noted in 1998 that “not having benefits at some threshold level will repulse employees, but the mere presence of a more generous benefits package will not attract and retain employees.”[vi]  This view is explanatory of the movement in recent years to flexibility, and an effort to respond to environmental factors with program design. This includes:

• Respond to favorable tax laws that provide an incentive to provide a pension program.
• Respond to demands in labor negotiations.
• Respond to social and indirect government pressures.
• Respond to inherent advantages of group purchase/provision.
• Respond to shareholder desires and competition.

Developments in the retirement plan market represent a response to work-force patterns.  There is now a large body of literature that uses government data to show that the workforce has always had high turnover and that few have spent 25 years or more with one employer.    Not only is this true of the private sector, but it has been so for the public sector as well.  Defined contribution plans and individual account defined benefit plans provide a career-average benefit, as noted above, which may serve to deliver more to most workers (due to relatively short service), than traditional defined benefit plans.  For the employer, they provide a more certain cost, which can be more easily budgeted.  A growing number of all plans provide lump-sum distributions, which are more popular with workers.  They are portable, and once a lump-sum distribution is taken upon job termination they eliminate any employer-specific risks.  Data from the Pension Benefit Guaranty Corporation underline the number of workers for whom this is a consideration.

Can We Return to  ‘The Way We Were’?

Writing prior to the enactment of ERISA, one leading actuary noted:  “A defined benefit final-pay pension plan may be selected precisely because it is the only type of plan which permits the employer to design a pension formula that takes both sources of retirement income – Social Security and company benefits – into account.  By doing so, a firm can provide higher paid employees a proportionately greater company pension.  This compensates for the fact that these individuals receive a lower percentage of final earnings from Social Security.”^[vii]  ERISA and subsequent legislation has limited the degree to which a plan sponsor can integrate a pension plan with Social Security (how much defined benefit can be offset), and funding and benefit limits have shifted much of what is done for high-income workers outside the qualified plan. 

He continued:  “Such a plan may also be necessary to reward an employee whose salary has increased rapidly or whose service was relatively short.  Additionally, only a pension can reward past as well as future service and base the total benefit on final average pay.  Finally, some companies believe that they are better able to assume investment risk…” Taking these in order, new funding and liability rules tied to plan termination insurance have all but ended the consideration of past service due to the liabilities it creates, and the difficulties the new funding limits place of setting aside funding.  Employers and unions that believe they can better absorb risk have continued defined benefit plans, or moved to hybrid plans like the cash balance plan, rather than moving totally to defined contribution.  The combination of the PBGC and tax-funding limits, however, make it unlikely that new defined benefit pension plans will be formed by either single employers or multi-employer groups.  Whether this is good or bad, right or wrong, matters little in light of the overwhelming public policies that make it so. 

The actuary concluded:  “The corporate viewpoint on the defined benefit versus defined contribution issue is formed by various competing factors: (1) whether its financial position can sustain the economic uncertainties posed by a defined benefit plan; (2) the extent to which competitive factors determine benefit levels and types; and (3) the corporation’s perception of its responsibility to provide for employees’ retirement and other financial needs.”  Fewer employers are willing to assume that they can financially sustain a plan as they may well be taken over or spun off tomorrow; the new economy employer creates constant pressure to change benefit programs by turning new hire and retention competition to current cash and short-term incentives, not a great pension 25 years hence; and increasingly employers view their primary obligation to be survival so that they can provide work, leaving post-work planning to the individual.  Many employers and unions will view this last statement as overly harsh, but I view it increasingly as the reality.  Because of these factors, defined benefit pensions are inherently problematic in this new world, as the sponsor issues relate to regulation, funding and liability, not to the simpler issue of portability.

What have changed are the regulatory environment, the workforce, world economics, technology, and feelings of employer and worker security.  Taken together, they suggest that we will not return to the defined benefit design dominance of yesterday, regardless of the consequences for individual retiree economic security, and not even to the dominance of annuity payouts.

How 25 Years Has Changed Demands/Motivations^[viii]

The government does influence action, and ERISA changed design drivers.  The law went from no vesting minimum standard to immediate vesting in some cases; from asset use in a plan for building the firm to arms-length transactions; from clear ”capital accumulation” versus ”retirement plan” distinctions, to limited distinctions; from selective provision of lump-sums allowed to the ‘all or none’ requirement; from less government tax revenue from lump sums to greater government tax revenue from lump sums; from a retirement income focus to a cash portability focus; from a regulatory and tax incentive bias toward defined benefit plans to a strong regulatory and tax incentive bias toward defined contribution plans; from a clear emphasis on employer/union provision advantages to an increased focus on individual self determination and “retail delivery”; from a paternalistic assessment basis of social obligation and corporate identification to one of maximum satisfaction of the largest number of workers.

As one expert has put it, movement from “golden handcuffs” to an employee/employer contract of partnership, personal accountability, and self reliance moved the nation away from traditional defined benefit, employer-pay-all plans with their focus on encouraging an employee to remain with a single employer until ”normal retirement age,” and toward greater financial and psychological independence, and identification with the service firm versus the employer.^[ix]

Plan design and recruitment action has moved from broad-based attraction to key employee attraction; from delivery of fast vested matches in short-term savings programs to vested matches for long-term savings programs; from delivery of final pay annuities to long-term workers to smaller accumulations for all workers and a focus on lump-sum distributions,^[x], and from employers, unions and plans dealing with long-term risks, to avoiding long term risks (investment, inflation, mortality) and placing their burden on individuals and families.

Major employers and unions have always provided the pension coverage available today.  Over 95 percent of participants are in large employer settings.  Most large employers with 401(k) plans now use employer stock in the plans; some of the largest unions have negotiated stock ownership, or outright employee ownership.  As one senior executive put it in 1998: “employee ownership allows the corporation to build partnership and a high performance work culture.”^[xi]  As one executive notes:

“While income security is an issue, it is increasingly being recognized that long-term security can best be achieved through personal development and professional growth.  Ironically, the presence of high-cost ‘1950’s, one size-fits-all benefits’ may, in fact, be a precursor to job insecurity as cost-cutting measures may be necessary for an organization to carry this heavy burden.”  And, he continues:  “There is a general question of whose responsibility it is to provide retirement income.  There is increasing emphasis today on the notion that it is up to individuals to provide a greater portion of their own retirement security.”

For the decades ahead such views are likely to dominate pension decision-making.  Many of these views are now entering the debate over the future of Social Security --- proposals by both the 2000 Republican and Democrat candidates for President for voluntary government sponsored individual accounts to supplement today’s Social Security---and many of the same pressures and attitudes reviewed here can be found in that debate.  In short, whatever one would like the pension world to be from a normative perspective, this descriptive review suggests that it will look more like the pension world of the 1990s that that of the 1950s.  The individual will be king, and economic well being once one is no longer working will increasingly rest on what saving and consumption choices the individual made throughout his or her life.  “Choose to Save” is taking on new meaning, as it will determine whether individuals can retire, or must work forever. 

Once a worker retires, a retirement security debate over defined benefit versus defined contribution plans would only be relevant today if one plan type paid only annuities and the other only lump-sum distributions.  As long as both plan types pay lump-sum distributions to all who have achieved small accruals, and as long as both plan types increasingly pay lump-sum distributions at retirement (retirees generally select a lump-sum when given a choice), the argument that one provides a greater promise of retirement security than the other, when both pay lump-sums, cannot be sustained. 

My father’s defined benefit pension plan began paying him a monthly annuity in 1978.  Today, that check represents a very important contribution to my parents’ retirement security.   Why?  Largely because he and my mother have lived years longer than they planned or expected to, and they have spent all the money they saved.  The greatest virtue of an annuity is this protection against unexpected longevity.  That is the only true form of retirement income security: a check that does not stop until one dies.   It is no longer the case that all defined benefit plan retirees choose to be paid in annuity form, and few defined contribution participants do so.  Future retirement security should no longer be a debate about defined benefit or defined contribution, as that is no longer the central issue in an age when both plan types paying lump-sum distributions at job change and retirement.  The future debate must be about worker education on savings, investing, longevity, retiree health, long-term care, and what choices individuals can make to avoid running out of money before they die. 

* EBRI is a private, nonprofit, nonpartisan public policy research organization based in Washington, DC. Founded in 1978, its mission is to contribute to, to encourage, and to enhance the development of sound employee benefit programs and sound public policy through objective research and education. EBRI does not lobby and does not take positions on legislative proposals.

[i] See Craig Copeland and Jack VanDerhei, Personal Account Retirement Plans: An Analysis of the Survey of Consumer Finances, EBRI Issue Brief No.  223 (July 2000).

[ii] The Employee Retirement Income Security Act of 1974 (ERISA) states:

“any plan, fund, or program which was heretofore or is hereafter established or maintained by an employer or by an employee organization, or by both, to the extent that by its express terms or as a result of surrounding circumstances, such plan, fund, or program---“(A) provides retirement income to employees, or results in a deferral of income by employees for periods extending to the termination of covered employment or beyond, regardless of the method of calculating the contributions made to the plan, the method of calculating the benefits under the plan or the method of distributing benefits from the plan” (emphasis added). This represented an expansion in concept from the first full version of the legislative proposal, H.R. 2, which limited plans to those which “for the purpose of providing for its participants or their beneficiaries, by the purchase of insurance or annuity contracts or otherwise, retirement benefits, and includes any deferred profit-sharing plan which provides benefits at or near retirement.”[ii]  (emphasis added)  H.R. 2 was closer to the traditional dictionary definition of a pension: “a retirement or disability allowance” (emphasis added).

[iii] See Craig Copeland and Jack VanDerhei, “Personal Account Retirement Plans: An Analysis of the Survey of Consumer Finances,” EBRI Issue Brief no. 223 (July 2000).

[iv] Jerry S. Rosenbloom and G. Victor Hallman, Employee Benefit Planning, third edition, Prentice-Hall, 1991.

[v] See Emily S. Andrews, Pension Policy and Small Employers: At What Price Coverage?, EBRI-ERF, 1989, chapter  IV for a summary of research; Richard A. Ippolito, Pensions, Economics and Public Policy, Pension Research Council, 1986; and Richard A. Ippolito, Pension Plans and Employee Performance Evidence, Analysis, and Policy, The University of Chicago Press, 1997

[vi] See Charles G. Tharp, “Yes,” in Dallas L. Salisbury, ed, Do Employers/Employees Still Need Employee Benefits?, EBRI-ERF, 1998, pp 11–13

[vii] Robert B. Peters, Defined Benefit and Defined Contribution Plans: A Corporate Perspective, in Dallas L. Salisbury, ed., Economic Survival in Retirement: Which Pension is for You?, EBRI-ERF, 1982, pp 81– 86.

[viii] See Daniel M. Holland, Private Pension Funds: Projected Growth, National Bureau of Economic Research, 1966; Private Pensions and the Public Interest, American Enterprise Institute, 1970; and Norman B. Ture with Barbara A. Fields, The Future of Private Pension Plans, American Enterprise Institute, 1976, and Dallas L. Salisbury and Nora Super Jones, eds, Pension Funding and Taxation: Implications for Tomorrow, EBRI-ERF, 1994: and Dallas L. Salisbury, ed, When Workers Call the Shots: Can They Achieve Retirement Security, EBRI-ERF, 1995.

[ix] See Charles G. Tharp, “Yes,” in Dallas L. Salisbury, ed, Do Employers/Employees Still Need Employee Benefits?, EBRI-ERF, 1998, pp 11—13.

[x] Steven G. Vernon, Employee Benefits: Valuation, Analysis and Strategies, John Wiley & Sons, Inc., 1993.

[xi] See Charles G. Tharp, “Yes”, in Dallas L. Salisbury, ed, Do Employers/Employees Still Need Employee Benefits?, EBRI-ERF, 1998, pp 11–13.