Skip Navigation

Current debate about the costs of the Civil Service Retirement (CSR) system has raised concerns about the financial condition of the system, its costs to the government, and the rates at which federal employees and agencies contribute to it. This study, undertaken at the request of the House Budget Committee, addresses these questions and poses several alternative approaches to financing CSR benefits.

Civil Service Retirement was prepared by Sherri Kaplan of the General Government Management staff of CBO's Office of Intergovernmental Relations, under the supervision of Stanley L. Greigg and Earl A. Armbrust. The author gratefully acknowledges the special assistance given by David DelQuadro, Stacy Sheffrin, and Mark Thurber. Johanna Zacharias edited the paper, and Norma Leake typed the various drafts and prepared the paper for publication. In keeping with CBO's mandate to provide objective analysis, the paper offers no recommendations.
 

Alice M. Rivlin
Director
May 1980
 
 

SUMMARY

CHAPTER I. INTRODUCTION

• CSR Financing
• CSR Compared with Retirement in the Private Sector

CHAPTER II. CSR FINANCING ISSUES--SOLVENCY, ACCOUNTABILITY, AND COSTS TO THE GOVERNMENT

• The Financial Condition of the CSR Trust Fund
• Accountability of CSR Costs in Agency Budgets
• The Reasonableness of CSR Costs

CHAPTER III. ALTERNATIVE CONTRIBUTION RATES AND BENEFIT PROVISIONS

• Option I. Continue Current Law
• Option II. Increase Agency Contribution Rates to Reflect the Full Cost to the Government
• Option III. Increase Agency and Employee Contribution Rates
• Option IV. Retain Current Contribution Rates but Limit Early Retirement Benefits and Cost-of-Living Adjustments

APPENDIX A. CHARACTERISTICS OF CSR ANNUITANTS

APPENDIX B. COMPARISON OF AGE OF RETIREMENT UNDER CSR AND PRIVATE SECTOR PRACTICES

APPENDIX C. INFORMATION ON COST-OF-LIVING ADJUSTMENTS IN THE PRIVATE SECTOR

APPENDIX D. SOURCES OF FUNDING FOR CSR
 

TABLES
1.	CSR TRUST FUND FINANCING, BY SOURCE
2.	PROJECTED ANNUAL CSR COST TO THE GOVERNMENT: SELECTED FISCAL YEARS TO 2030
3.	PROJECTED RATIO OF CSR RESERVES TO OUTLAYS: SELECTED FISCAL YEARS TO 2030
4.	PROJECTED CSR DEPENDENCY ON GENERAL FUND APPROPRIATIONS: SELECTED FISCAL YEARS TO 2030
5.	CSR COMPARED TO PRIVATE SECTOR RETIREMENT PRACTICE--COSTS NOT PAID BY EMPLOYEES
6.	SUMMARY OF MAJOR FINANCING PROVISIONS OF OPTIONS
7.	PROJECTED CSR COST TO THE GOVERNMENT UNDER OPTIONS

 

At present, the Civil Service Retirement (GSR) system provides pension coverage for some 2.7 million active federal civilian employees and 1.8 million annuitants. In each of the coming five years, an average of 92,500 more civil servants will retire and begin to draw GSR pensions.
 

COMPARISON WITH PRIVATE SECTOR RETIREMENT

Retirement under GSR differs markedly from that in the private sector. Benefits under GSR are relatively large when compared to the two-part retirement income of private-sector retirees: an employer-provided pension plan plus Social Security. The two areas in which differences have the most significant cost effects are age of eligibility for retirement and cost-of-living adjustments (COLAs) of benefits. GSR enrollees may draw pensions as early as age 55, and their benefits are kept abreast of inflation by being adjusted twice annually to reflect changes in the Consumer Price Index. These two features are largely responsible for the recent and projected steep growth in GSR outlays.

Under current law, GSR's annual outlays will increase from $14.7 billion to $30.1 between 1980 and 1986. About $10.6 billion (69 percent) of this increase will result from COLAs, and such adjustments become progressively more expensive every year. For example, in October 1980, each one-percentage-point adjustment caused annual outlays to rise by $161 million. By October 1985, each such increase will add some $341 million to annual GSR outlays.
 

HOW THE SYSTEM IS FINANCED

Despite large and growing outlays, the system's statutory financing provisions assure its present and future solvency. Last year, the GSR trust fund's reserves of $73.6 billion were five times annual outlays; although this ratio is expected to decline in the long run, the reserves will still provide adequate resources at a ratio of more than one and a half times annual outlays.

Income from five sources finances the GSR trust fund. By far the largest portion--46.5 percent--comes in the form of annual appropriations from the general fund of the U.S. Treasury. Interest on fund balances furnishes a further 20.7 percent. Another large portion comes from contributions from employing agencies and employees, for most of whom participation is mandatory; the employee and agency shares (each set at 7 percent of payroll) together constitute 26.5 percent to GSR trust fund income. The balance, 6.2 percent, comes from off-budget agencies such as the U.S. Postal Service.

Maintaining the GSR trust fund according to current financing provisions involves considerable out-of-pocket cost to the federal government. This sum is calculated as the difference between the annual income from employee contributions and payments from off-budget agencies, less that year's outlays. In 1980, the GSR cost to the government was $9.6 billion (65.3 percent of outlays) , and over the coming five decades, it is projected to grow in 1980 dollars to $20.2 billion (78.0 percent of outlays).
 

PROS AND CONS OF GSR

Proponents of the GSR system and of maintaining it in its present form stress three points in particular. First, they note that the federal government has the obligation, as a model employer, to protect its retired personnel fully against inflation; this goal is served by the COLA provision of GSR. Second, they point to the tax-exempt status of the Social Security income of private-sector retirees; the generosity of GSR pensions is viewed in part as offsetting this relative private-sector advantage. Third, they point to the relatively lower pay the federal government offers for many comparable jobs in the private sector; again, generous pensions are cited as fair recompense for salary inequities.

On the other side of the debate, the cost to the government is one of the causes of concern about the GSR system. Critics of the present system view its expense as excessive and seek ways of reducing it.

The fact that employing agencies pay only a small share of GSR's federal cost is another source of criticism. The full expense of this important fringe benefit is not accurately reflected in operating agency budgets. Increasing the employing agencies' contributions to GSR has been cited as a way to improve cost accounting and, in turn, budgeting and decisionmaking.

A third criticism arises from the wide disparity between GSR and private-sector retirement practices. Opponents of the present system would like to see the costs or benefits of GSR brought more into line with the private-sector combination of company pensions plus Social Security.
 

POSSIBLE SOLUTIONS

In response to the above concerns about GSR, the effects of maintaining current law (Option I) and adopting one of three alternatives are outlined below. The net reductions in costs to the government that could be achieved by the three alternatives (Options II-IV) are summarized in the table below.

Option I: Continue Current Financing

Advocates of continuing the status quo point out that current financing mechanisms will continue to provide adequate income to the GSR fund. Opponents argue that the present employee and agency contribution rates do not provide sufficient income to cover the full cost of benefits for either current annuitants or active employees. They further point out that the cost to the government is excessive in comparison with retirement practices in the private sector.

Option II: Increase Agency Contribution Rates to Reflect the Full Cost to the Government

If the agency contribution were to reflect the full cost to the government as computed by actuaries, the rate would have to be increased from 7 to 29.5 percent of payroll. For most agencies, this increase would be offset by a corresponding adjustment in annual appropriations, and thus it would simply represent an internal transfer from one budget account to another with no effect on federal receipts. For off-budget agencies, however, the increased rate could result in additional GSR fund income that would reduce annual costs to the government by $3.2 billion (23 percent) in 1982 and accumulate to $17.8 billion (20 percent) through 1986.

The General Accounting Office believes that this alternative would improve management decisions about the level and nature of federal programs, because personnel costs of federal operations would be more visible and accurately recorded. This approach would not change GSR benefit provisions or employee withholding rates. It would, however, result in substantial increases in GSR reserve levels, with trust fund balances reaching 12.5 times annual outlays in 1986.

Option III: Increase Agency and Employee Contribution Rates

This alternative would incorporate essentially the same GSR funding provisions as Option II, but it would require federal employees to shoulder some of the rate adjustment. Specifically, over the next four years, employee contributions to the GSR fund would increase from 7 to 9 percent of pay, while employing agency rates (reflecting the remaining actuarial cost) would increase from 7 to 27.5 percent of payroll.

Because of phased-in implementation, this alternative would initially yield a smaller reduction in federal costs than would Option II. But in 1986, it would yield the greatest annual savings to the government among the alternatives considered--$5.2 billion (23 percent) below current financing.

Increasing the employee withholding rate from 7 to 9 percent of pay would bring GSR's long-term cost to the government into line with private-sector standards. Employee organizations would oppose this alternative, because any mandatory increase in employee withholding rates would widen existing disparities in take-home pay between many federal employees and their private-sector counterparts.

Option IV: Retain Current Contribution Rates but Limit Benefits for Early Retirement and Cost-of-Living Adjustments

This approach would reduce the government's cost for GSR by limiting cost-of-living adjustments in benefits to once a year and to 70 percent of changes in the Consumer Price Index. Also, a reduction in initial benefits for persons who choose to retire before age 65 would be phased in over 20 years, eventually reaching 10 percent at age 60 and 20 percent at age 55. The 20 percent maximum reduction would still be less severe than early retirement reductions required by private pension plans and Social Security. The government's cumulative five-year savings, relative to current law, could total $10.5 billion (12 percent).

This option would be supported by critics who view current GSR cost-of-living and early retirement provisions as too generous and costly to the government when compared to private-sector practices. GSR beneficiaries might point to the argument mentioned above concerning the federal government's responsibility as a model employer. Obviously, this plan would have particularly little appeal to retirees during a period of high inflation.

This document is available in its entirety in PDF.