PENSIONS: FINANCING AND REGULATION

Except in a hypothetical individualistic and primitive society, where each person produces for his or her own consumption, everyone depends for survival, at least in part, on an allocation of goods and services produced by others. This allocation can be based on trade, perhaps facilitated by money, or made in accordance with a perceived obligation. For persons with reduced productivity resulting from advanced age or disability, the obligation might be based on family loyalty, institutionalized charity, or past meritorious service (e.g., in the case of a military veteran). For an individual entrepreneur, the allocation could come as rent for a farm or shop developed in the past.

Current cost funding

In the industrial age, income to elderly persons often derives from past employment and the payment structure is formalized as a pension plan. The obligation becomes financial and contractual. For self-employed persons retirement income often comes from an individualized financial plan rather than from a family's loyalty or public charity. The income transferred to elderly persons is often paid from financial assets built up over the working lifetime or, as is common in social insurance systems, as an allocation of current income without an intervening fund. For example, in 1861 the federal government established the first major nondisability pension program in the United States, providing for the retirement of military officers after forty years of service. This pension program involved direct payments from the military budget. The first retirement programs of private employers also involved direct payments to retirees and the payments were considered a current business expense.

This type of pension funding or budgeting, in which the plan sponsor pays benefits each year from current general revenue and accounts for the payments as a current business expense, is called current-cost or pay-as-you-go funding. There are two basic objections to current-cost funding. The first is an accounting issue. One of the goals of accounting is to match the costs of production with the revenue generated by the goods and services produced. If costs and revenues are mismatched, the faulty information may result in bad business decisions. Current cost funding charges pension benefits against current revenues when in fact the services on which the pension is based may have been provided many years earlier. This could be a serious mismatching of costs and revenues. This objection to current-cost funding could be mitigated if the operating statement of the pension plan sponsor contained an expense item that is an estimate of the cost of future pensions associated with service in the current year. This is now required by financial accounting standards in the United States.

The second objection to current-cost funding of pension plans is the lack of security of employee pension expectations when benefit payments come directly from the plan sponsor. The payments depend on the continued economic viability of the sponsor. In a dynamic economy the survival and prosperity of a plan sponsor is not assured.

There are two methods for increasing the security of plan members. The first is some sort of insurance program to make the benefit payments if the sponsor cannot. The second is the creation of a pension fund, a pool of investments independent of the sponsor from which benefits will be paid and which will receive pension contributions. The separate pool of investments can also generate investment income that will lead to lower required contributions than those in a current-cost system with identical benefits. There is also a powerful macroeconomic reason for favoring a fund. The allocation of income to retirees comes from the body of goods and services created by the economy. If this body is bigger as a result of pension savings that result in investment in plant, tools, research, and education, the total system for maintaining the income of retirees may increase the economic well-being of all. In fact, in the United States savings made through pension plans have been the principal source of new capital since World War II.

Developing a plan

The development of a pension funding or budgeting plan starts with the economic and demographic characteristics of the group to be served. These characteristics can be observed, but the level of retirement benefits to be provided is an important decision that is influenced by the human resource management philosophy of the sponsor. The benefit level decision may also be influenced by the members, sometimes expressed through collective bargaining. The benefit level is often studied by using replacement ratios. The replacement ratio is (income rate after retirement/income rate before retirement).

After the qualitative decision on benefit level is made, the decision is quantified as a formula for benefits that may depend on age, service, and salary history. The cost and liabilities of a pension plan are derived by calculating actuarial present values (APV) of future possible payments. An APV is the product of an expected payment amount and the probability that the amount will be paid, and a discount factor that recognizes that, because of investment income, a dollar today is worth more than one due to be paid in the future. For example, suppose $1,000 will be paid in ten years to a person now age sixty-five if the person is then alive. The APV of this contingent payment at age sixty-five is: The probability of survival from age sixty-five to seventy-five in this example was taken from the 1983 Group Annuity Mortality Table (Females) and an investment return rate of 8 percent per year was assumed. In valuing the liabilities of a pension plan—the actuarial present value of future promises—many calculations of this type are required.

The next decision in constructing a pension plan is to select one of two broad classes of plans, defined benefit (DB) and defined contribution (DC). In the 1990s hybrid plans with some of the characteristics of both types of plans were introduced.

DB plans specify the benefit formula and it becomes the foundation of the funding plan. The sponsor may have a great deal of freedom in funding the plan. DB plans reduce the risk, the possibility of unfortunate events, for plan participants and allow sponsors some flexibility in making contributions to the pension fund. It is also fairly easy to include benefits for members near retirement age and the higher liability for these older members as part of the general liability of the plan. The members, however, will not have the satisfaction of watching an individual account balance grow, and the fixed nature of benefits causes the sponsor to manage the consequences of adverse experience as well as to benefit from favorable experience.

In a DC plan a set of defined contributions for each participant is determined with the objective of achieving the replacement ratio goal set in the initial planning. Each member can have the satisfaction of watching an individual asset account grow. The management of early withdrawals from the plan is easy because of the individual nature of the funding. From the sponsor's viewpoint, DC plans may reduce the flexibility in making annual contributions, but required risk management is far less because the pension obligation is met once the budgeted annual contribution is made.

During the period of rapid growth of private pension plans in the United States following World War II, DB plans were more common. Since the 1980s, DC plans have increased in importance. The early preference for DB plans can be attributed to pressure from organized labor to include workers near retirement age and to have the risk of deviations from expected experience managed by the sponsor. In addition, plan sponsors valued the ability to make actual contributions within a wide range. The shift to DC plans can be attributed to the satisfaction of members in observing an individual account grow, especially in periods of high investment returns, and the ease of managing the transfer from one pension plan to another in a period of high labor mobility. To sponsors DC plans seem to minimize their risk. Within a few years in the late 1970s and early 1980s, real rates of investment return (investment return rate - price inflation rate) went from negative to historic highs. This roller-coaster experience brought home to sponsors the financial risks of DB plans. The mounting costs of compliance with tax regulations and accounting standards also discouraged sponsors from adopting DB plans.

Regulation

The next steps in designing the financing of a pension plan are heavily influenced by government policy. The security of plan participants is enhanced if the pension fund is large. A plan with assets in excess of estimated liabilities can absorb market variations in the value of assets or the higher costs of possible low mortality without requiring additional contributions by the sponsor. On the other hand, pension funds are increased by the inflow of contributions and, in general, such contributions are recognized as a business expense of the sponsor in the determination of income tax. Thus a government must compromise between encouraging large pension fund contributions, which will increase the security of pension expectations but shrink the base of the income tax, and bounding such contributions to save the tax base. A lower bound on annual pension plan contributions would enhance the security of pension expectations but reduce the flexibility of sponsors in meeting pension obligations.

In the United States the regulation of pension plans before 1974 was in the hands of the Internal Revenue Service. The concern was to place a reasonable upper bound on pension contributions deductible for income tax purposes. The upper bound was stated as (normal cost) + (.10 × Initial accrued liability), where normal cost is an estimate of the cost of the plan attributable to the current year under the funding plan adopted, and initial accrued liability is the amount of liability under the funding plan adopted at the time the plan started. The objective was to prohibit a reduction in the income tax by rapid reduction in the accrued liability recognized when the plan started.

In 1974 the comprehensive Employee Retirement Income Security Act (ERISA) was passed, and upper and lower bounds on pension contributions were enacted. The upper bound was somewhat different from the pre-1974 rule mentioned above, but it incorporated the same basic ideas. What was new, with an acknowledgment that adequate funding was in the public interest, was a minimum funding requirement. The minimum contribution requirement was stated in terms of normal cost plus an amount of the initial accrued liability, depending on the type of plan, and an adjustment for deviations in actual experience from that assumed in earlier estimates of liabilities.

A related set of public policy issues arises about the management of the pension fund. One option is the use of book reserves, pension liability estimates that are included in the balance sheet of the sponsoring organization rather than on the balance sheet of a separate pension plan entity. Under this plan the estimated annual cost of the plan is an expense of the sponsoring organization and the contributions increase the size of the internally held book reserve. This plan is attractive in the absence of an active capital market and can be used to finance the expansion of a sponsoring organization. The disadvantages are that this nondiversified investment reduces the security of the income of beneficiaries. In addition, less of a nation's savings pass through open capital markets, where market forces will tend to allocate them to projects with the highest rate of return. With the absence of the discipline of open markets, macroeconomic growth could be compromised.

In the United States, ERISA held out income tax advantages only to those pension plans with external pension funds. Pension plans are prohibited from acquiring or holding more than 10 percent of plan assets in the securities or real property of the sponsor. Other nations have made more extensive use of book reserves. For example, Germany permits the use of book reserves to finance DB pension plans. Book reserves facilitated the internal financing of the rebuilding of the German industrial plant following World War II, when the German financial markets were chaotic. The sponsor's risk of insolvency in many nations permitting book reserves is managed by compelling participation in an insolvency insurance plan.

Financial accounting rules in the United States force sponsors providing post-retirement health benefits to account for their value using book reserves. These accounting rules are applicable to publicly held companies and are set by the Financial Accounting Standards Board (FASB). FASB is an independent, nongovernmental agency. In 1990 it issued Standard of Financial Accounting 106 which required employers sponsoring post-retirement health plans to report on their financial statements estimates of the liability of these plans. Post-retirement health benefits have not been the subject of detailed federal regulation.

Although ERISA discouraged the use of book reserves for pension plans, it also created the Pension Benefit Guarantee Corporation to ensure at least the partial payment of pension benefits under DB plans when the sponsor was unable to meet the funding requirements. The insurance system is funded by a premium related to unfunded accrued liability that is paid by all DB plans.

Individual and social plans

This discussion has often used the term "sponsor." The pension plans discussed have involved an arrangement between a sponsor, usually an employer, and members, usually employees, to provide retirement income. There are two large, very different, classes of retirement programs that do not fit into this framework.

Standard advice is to save for old age and the inevitable rainy day. The advice remains good, but it has also become embedded in tax policy. To encourage self-employed individuals and workers whose employers do not provide a formal pension plan to save for retirement, Congress has created several sections of the Internal Revenue Code, for different types of workers, that provide for the creation of special savings accounts for retirement purposes. These accounts are special in that income diverted to the account and the investment earnings of the account are not subject to income taxation until withdrawn. These tax-sheltered retirement savings vehicles have become popular. The question of whether they have generated new savings or diverted existing savings to tax shelters remains open.

Examples of these retirement programs will illustrate their history and variety. Individual retirement plans were authorized by ERISA in 1974. Originally these plans were limited to individuals not already participating in an employer sponsored tax qualified plan or other tax favored arrangement. Later the eligibility was broadened. These individual retirement plans are funded through Individual Retirement Accounts (IRAs). Tax Sheltered Annuities have an even longer history and can be sponsored by tax-exempt entities organized for religious, educational, or research activities. Keogh plans were authorized in 1962 and are designed for self-employed individuals. Section 401(k) plans are named for the section of the Internal Revenue Code that regulates them. They are designed for employees and are typically funded through salary reductions and matching contributions by employers.

Roth retirement accounts, named for their sponsor in Congress, are a recent addition to this list of individual retirement programs. Roth accounts are fundamentally different in that contributions are subject to income taxation but withdrawals are not.

The high rate of poverty among the elderly during the Great Depression was among the forces that moved the Congress to create an almost universal Social Security program in 1935. Social Security includes several types of benefits, but the old-age income part is a DB plan with replacement ratios that decline as career average indexed wages increase. Social Security has been funded primarily by a payroll tax on a modified current-cost basis. The modification has been the existence of a trust fund that stabilizes results across economic cycles.

In the early 1990s there were approximately three taxpaying workers for each beneficiary. By 2020 it is likely that there will be approximately two taxpaying workers for each person receiving benefits. The resulting financial crunch for a current-cost system is forcing political consideration of increasing the normal retirement age, shifting the system partially to a DC program, and actions to increase productivity by the time the crunch occurs.

James C. Hickman

See also Employee Retirement Income Security Act; Individual Retirement Accounts; Taxation.

BIBLIOGRAPHY

Costa, Dora L. The Evolution of Retirement: An American Economic History 1880–1990. Chicago: University of Chicago Press, 1998.

Grubbs, Donald S. "The Public Responsibility of Actuaries in American Pensions." North American Actuarial Journal 3 (1999): 34–41.

McGill, Dan M.; Brown, Kyle N.; Haley, John J.; and Schieber, Sylvester J. Fundamentals of Private Pension, 7th ed. Philadelphia: University of Pennsylvania Press, 1996.

Trowbridge, Charles L. "ABC's of Pension Funding." Harvard Business Review 44, no. 2 (1966): 115–126.