Who Will Pay For Your Retirement?

The Looming Crisis

A Statement by the Research and Policy Committee of the Committee for Economic Development Who Will Pay For Your Retirement?

The Looming Crisis Library of Congress Cataloging-in-Publication Data

Committee for Economic Development. Research and Policy Committee. Who will pay for your retirement? : the looming crisis : a statement / by the Research and Policy Committee of the Committee for Economic Development. p. cm. Includes bibliographical references. ISBN 0-87186-119-4 : $20.00 1. Old age pensions — United States. 2. Retirement income — United States. 3. Social security — United States. 4. Civil service — Pensions — United States. 5. Individual retirement accounts — United States. I. Committee for Economic Development. II. Title. HD7105.35.U6C65 1995 331.25'2'0973 — dc20 95-3345 CIP

First printing in bound-book form: 1995 Paperback: $20.00 Printed in the United States of America Design: Rowe & Ballantine

COMMITTEE FOR ECONOMIC DEVELOPMENT 477 Madison Avenue, New York, N.Y. 10022 (212) 688-2063 2000 L Street, N.W., Suite 700, Washington, D.C. 20036 (202) 296-5860 CONTENTS

RESPONSIBILITY FOR CED STATEMENTS ON NATIONAL POLICY vi

PURPOSE OF THIS STATEMENT ix

CHAPTER 1: EXECUTIVE SUMMARY AND POLICY RECOMMENDATIONS ...... 1 The Challenge Posed by an Aging Population ...... 2 The Untimely Decline in Retirement Saving ...... 5 Pension Contributions and Funding ...... 5 Economic Effects of Declining Retirement Saving...... 6 Policy Options to Tackle the Retirement Finance Problem ...... 7 Regulatory Reform for Private Retirement Plans ...... 7 Retirement Information and Individual Responsibility ...... 8 Reforming Government Employee Pensions and the Social Security Program...... 8 Summary of Policy Recommendations ...... 9

CHAPTER 2: RETIREMENT SAVING AND ECONOMIC SECURITY ...... 15 The Economic Security of Future Retirees ...... 15 The Economy and Retirement Saving ...... 16 Private- and Public-Sector Responsibilities for Pensions ...... 21 Actions Required to Meet These Responsibilities...... 22

CHAPTER 3: THE BASIC FACTS ON OUR AGING POPULATION AND RETIREMENT SAVING ...... 24 An Aging Population ...... 24 Private Sector Preparation for Retirement of the Baby Boomers...... 27 Private Pension Contributions ...... 28 Private Pension Coverage...... 29 Decline in Defined Benefit Plans ...... 30 Preretirement Withdrawals of Pension Funds ...... 31 The Status of Pension Funds ...... 32 Private Pension Funds ...... 32 Public Sector Funding of Pensions ...... 33 Expectations of the Baby Boomers for Retirement Income ...... 37 How Much Should Workers Save for Retirement? ...... 38 Undersaving by Baby Boomers ...... 39

CHAPTER 4: DESIGNING TAX INCENTIVES TO RAISE SAVING...... 42 The Tax Base Controversy ...... 42 Tax Incentives and Saving ...... 43 Design of Tax-Preferred Saving Incentives...... 44 Proposals for Compulsory Saving Programs ...... 45 The Budget Deficit ...... 46

84 iii CHAPTER 5: THE REGULATORY TANGLE: PROPOSALS FOR SIMPLIFICATION ...... 48 Favorable Tax Treatment ...... 48 Federal Regulation ...... 48 ERISA and Subsequent Legislation...... 49 Regulations Limiting Private Pension Contributions and Benefits ...... 50 Funding Limit for Defined Benefit Plans ...... 51 Reductions in Contribution and Benefit Limits ...... 53 Regulations to Prevent Discrimination in Pension Plans ...... 59 Top-heavy Rules ...... 60 Social Security Integration...... 60 Coverage and Vesting Rules ...... 60 Preserving Pension Funds for Retirement ...... 61 Preretirement Lump-Sum Withdrawals and Borrowing from Pensions ...... 61 Portability of Pensions ...... 61 Pension Benefit Guaranty Corporation ...... 61 The Premium Structure ...... 62 Moral Hazard ...... 63 Rising Regulatory Costs Discourage Pension Saving ...... 64 Sources of Cost Increases ...... 65 Impact of Cost Increases on Plan Terminations and Formations ...... 65

APPENDIX: A BRIEF HISTORY OF GOVERNMENT REGULATION OF PRIVATE PENSIONS ...... 66

NOTES 72

MEMORANDUM OF COMMENT, RESERVATION, OR DISSENT 82

OBJECTIVES OF THE COMMITTEE 83

iv 85 RESPONSIBILITY FOR CED STATEMENTS ON NATIONAL POLICY

The Committee for Economic Development specific legislative proposals; its purpose is to is an independent research and policy organi- urge careful consideration of the objectives set zation of some 250 business leaders and edu- forth in this statement and of the best means of cators. CED is nonprofit, nonpartisan, and non- accomplishing those objectives. political. Its purpose is to propose policies that Each statement is preceded by extensive bring about steady economic growth at high discussions, meetings, and exchange of memo- employment and reasonably stable prices, in- randa. The research is undertaken by a sub- creased productivity and living standards, committee, assisted by advisors chosen for their greater and more equal opportunity for every competence in the field under study. citizen, and improved quality of life for all. The full Research and Policy Committee All CED policy recommendations must participates in the drafting of recommenda- have the approval of trustees on the Research tions. Likewise, the trustees on the drafting and Policy Committee. This Committee is subcommittee vote to approve or disapprove directed under the bylaws which emphasize a policy statement, and they share with the that “all research is to be thoroughly objective Research and Policy Committee the privilege in character, and the approach in each instance of submitting individual comments for publi- is to be from the standpoint of the general cation. welfare and not from that of any special politi- cal or economic group.” The Committee is Except for the members of the Research and aided by a Research Advisory Board of lead- Policy Committee and the responsible subcommit- ing social scientists and by a small permanent tee, the recommendations presented herein are not professional staff. necessarily endorsed by other trustees or by the The Research and Policy Committee does advisors, contributors, staff members, or others not attempt to pass judgment on any pending associated with CED.

vi CED RESEARCH AND POLICY COMMITTEE

Chairman JOSH S. WESTON LINNET F. DEILY RICHARD J. KRUIZENGA Chairman and Chief Executive Officer Chairman, President and Senior Fellow Automatic Data Processing, Inc. Chief Executive Officer ISEM First Interstate Bank of Texas CHARLES R. LEE Vice Chairmen GEORGE C. EADS Chairman and Chief Executive Officer FRANK P. DOYLE Vice President, Worldwide GTE Corporation Executive Vice President Economic & Market Analysis FRANKLIN A. LINDSAY GE General Motors Corporation Retired Chairman W.D. EBERLE WALTER Y. ELISHA Itek Corporation Chairman Chairman and Chief Executive Officer WILLIAM F. MAY Manchester Associates, Ltd. Springs Industries, Inc. Chairman and Chief Executive Officer WILLIAM S. EDGERLY EDMUND B. FITZGERALD Statue of Liberty - Ellis Island Chairman Managing Director Foundation, Inc. Foundation for Partnerships Woodmont Associates ALONZO L. MCDONALD CHARLES J. ZWICK HARRY L. FREEMAN Chairman and Chief Executive Officer Coral Gables, Florida President Avenir Group, Inc. The Freeman Company HENRY A. MCKINNELL REX D. ADAMS RAYMOND V. GILMARTIN Executive Vice President and Chief Vice President–Administration President and Chief Executive Officer Financial Officer Mobil Corporation Merck & Company, Inc. Pfizer Inc. IAN ARNOF BOYD E. GIVAN JOSEPH NEUBAUER President and Chief Executive Officer Senior Vice President and Chief Financial Chairman and Chief Executive Officer First Commerce Corporation Officer ARAMARK Corp. The Boeing Company ALAN BELZER JOHN D. ONG Retired President and Chief BARBARA B. GROGAN Chairman of the Board, President and Operating Officer President Chief Executive Officer AlliedSignal Inc. Western Industrial Contractors The BFGoodrich Company PETER A. BENOLIEL RICHARD W. HANSELMAN VICTOR A. PELSON Chairman of the Board Retired Chairman Executive Vice President, Chairman- Quaker Chemical Corporation Genesco Inc. Global Operations Team ROY J. BOSTOCK EDWIN J. HESS AT&T Corp. Chairman and Chief Executive Officer Senior Vice President PETER G. PETERSON D’Arcy, Masius, Benton & Bowles, Inc. Exxon Corporation Chairman THEODORE A. BURTIS RODERICK M. HILLS The Blackstone Group Retired Chairman of the Board Partner DEAN P. PHYPERS Sun Company, Inc. Mudge Rose Guthrie Alexander & New Canaan, Connecticut Ferdon OWEN B. BUTLER JAMES J. RENIER Retired Chairman of the Board LEON C. HOLT, JR. Renier & Associates The Procter & Gamble Company Retired Vice Chairman Air Products and Chemicals, Inc. * JAMES Q. RIORDAN FLETCHER L. BYROM Stuart, Florida Chairman MATINA S. HORNER Adience, Inc. Executive Vice President HENRY B. SCHACHT TIAA-CREF Chairman of the Executive Committee PHILIP J. CARROLL Cummins Engine Company, Inc. President and Chief Executive Officer CORDELL W. HULL Shell Oil Company Executive Vice President and Director ROCCO C. SICILIANO Beverly Hills, California JOHN B. CAVE Bechtel Group, Inc. Principal SOL HURWITZ ELMER B. STAATS Avenir Group, Inc. President Former Comptroller General of the United States ROBERT CIZIK Committee for Economic Development Chairman and Chief Executive Officer HARRY P. KAMEN ARNOLD R. WEBER Cooper Industries Inc. Chairman and Chief Executive Officer Chancellor Northwestern University A. W. CLAUSEN Metropolitan Life Insurance Company Retired Chairman and Chief HELENE L. KAPLAN, Esq. LAWRENCE A. WEINBACH Executive Officer Of Counsel Managing Partner – Chief Executive BankAmerica Corporation Skadden Arps Slate Meagher & Flom Arthur Andersen & Co, SC JOHN L. CLENDENIN JOSEPH E. KASPUTYS WILLIAM S. WOODSIDE Chairman and Chief Executive Officer Chairman, President and Chief Vice Chairman BellSouth Corporation Executive Officer LSG Sky Chefs RONALD R. DAVENPORT Primark Corporation MARTIN B. ZIMMERMAN Chairman of the Board ALLEN J. KROWE Executive Director, Governmental Sheridan Broadcasting Corp. Vice Chairman and Chief Financial Relations and Corporate Economics Officer Ford Motor Company Texaco Inc.

*Voted to approve the policy statement but submitted memorandum of comment, reservation, or dissent. See page 82.

vii SUBCOMMITTEE ON PENSIONS AND SAVING

Chairman LAWRENCE A. WEINBACH Managing Partner-Chief Executive Arthur Andersen & Co, SC

REX D. ADAMS HARRY G. HOHN Ex-Officio Trustees Vice President-Administration Chairman & Chief Executive Officer JOHN L. CLENDENIN Mobil Corporation New York Life Insurance Company Chairman & Chief Executive Officer IAN ARNOF ROBERT C. HOLLAND BellSouth Corporation President & Chief Executive Officer Senior Economic Consultant FRANK P. DOYLE First Commerce Corporation Committee for Economic Development Executive Vice President ELI BROAD MATINA S. HORNER GE Chairman, President & Executive Vice President SOL HURWITZ Chief Executive Officer TIAA-CREF President SunAmerica Inc. THOMAS W. JONES Committee for Economic Development STEPHEN L. BROWN President & Chief Operating Officer JOSH S. WESTON Chairman & Chief Executive TIAA-CREF Chairman & Chief Executive Officer Officer HARRY P. KAMEN Automatic Data Processing, Inc. John Hancock Mutual Life Chairman & Chief Executive Officer Insurance Company Metropolitan Life Insurance Company Non-Trustee Members J. GARY BURKHEAD ROBERT W. LUNDEEN President W. GORDON BINNS, JR. Retired Chairman Former Vice President Fidelity Management and Research The Dow Chemical Company Company General Motors Corporation LEIF H. OLSEN PHILIP CALDWELL MARSHALL N. CARTER President Chairman & Chief Executive Officer Senior Managing Director Leif H. Olsen Investments, Inc. Lehman Brothers, Inc. State Street Bank and Trust Company JAMES F. ORR III KATHLEEN COOPER MARVIN GREENE Chairman & Chief Executive Officer Managing Director, Retirement Services Chief Economist UNUM Corporation Exxon Corporation Towers Perrin DEAN P. PHYPERS GARY L. COUNTRYMAN New Canaan, Connecticut Chairman & Chief Executive Officer Guests Liberty Mutual Insurance Company S. LAWRENCE PRENDERGAST WILLIAM CHENEY Vice President & Treasurer Economist W. D. EBERLE AT&T Corp. Chairman John Hancock Financial Services * Manchester Associates, Ltd. JAMES Q. RIORDAN CATHERINE L. HERON Stuart, Florida JAMES D. ERICSON Vice President & Senior Counsel, President & Chief Executive Officer IAN M. ROLLAND Tax, Pension, and International Northwestern Mutual Life Chairman & Chief Executive Officer Investment Company Institute Insurance Company Lincoln National Corporation SANFORD KOEPPEL KATHLEEN FELDSTEIN GEORGE F. RUSSELL, JR. Vice President, Issues Management President Chairman Prudential Asset Management Company Economics Studies, Inc. Frank Russell Company EDWARD V. REGAN RICHARD W. HANSELMAN JOHN L. STEFFENS President Retired Chairman Executive Vice President The Jerome Levy Economics Institute Genesco Inc. Merrill Lynch & Co., Inc. CAROL REUTER RODERICK M. HILLS RICHARD F. SYRON President Partner Chairman & Chief Executive Officer New York Life Foundation Mudge Rose Guthrie Alexander American Stock Exchange DAVID M. WALKER & Ferdon Compensation and Benefits Consulting Arthur Andersen & Co, SC

ADVISORS PROJECT DIRECTOR PROJECT B. DOUGLAS BERNHEIM WILLIAM J. BEEMAN ASSOCIATE Professor, Department of Economics Vice President and Director of Stanford University Economic Studies MICHAEL K. BAKER BERNARD SAFFRAN Committee for Economic Development Economist Franklin & Betty Barr Professor of Committee for Economic Development Economics Swarthmore College JOHN B. SHOVEN Charles R. Schwab Professor of Economics Dean, School of Humanities and Sciences Stanford University

*Voted to approve the policy statement but submitted memorandum of comment, reservation, or dissent. See page 82. viii PURPOSE OF THIS STATEMENT

Major reforms are needed in policies that on retirement policies should place top prior- affect both private retirement plans and Social ity not only on coverage and increased ben- Security if the United States is to avoid a seri- efits but also on national saving, capital ous crisis in retirement funding and living formation, and economic growth. standards in the next century. Our goal in The Committee for Economic Development preparing this report is to show as clearly as has a long-standing interest in retirement poli- possible the consequences of inadequate cies. In 1981, CED issued Reforming Retirement retirement saving for those retiring early in Policies, which outlined a strategy for broad- the next century and thereafter. If policies and ening the reach of private pensions, increas- practices are not changed, many twenty-first– ing individual saving, and strengthening century retirees will experience an unexpected Social Security. Although a number of the and substantially reduced standard of living, recommendations in that report have been due in no small part to a retirement system implemented (e.g., a gradual rise in the retire- that is underfunded, overregulated, and soon ment age), recent changes in federal regula- to be strained as never before with the retire- tion (many made in the name of deficit reduc- ment of the baby-boom cohort. tion or antidiscrimination policy) have Who Will Pay for Your Retirement? issues a curtailed new pension formation and sound strong warning, but it also offers comprehen- funding. sive, specific, and workable recommendations. With this policy statement, we call on Con- These recommendations, if taken together, will gress to reverse government policies that are not only allow the nation to avert disaster but strangling private pensions and threatening will also: their financial soundness. Although this state- • Strengthen and streamline the way we fund ment deals primarily with policy for private and regulate retirement programs. pensions, we also say that Social Security cannot be ignored and that changes are neces- • Educate workers and employers about sary to preserve that system well into the next retirement needs, and provide incentives century. We call on all Americans to educate for people to save adequately for their themselves about their retirement needs and own retirement. options so that they can make sound and sen- • Significantly broaden the scope and cover- sible decisions for themselves and their fami- age of private plans. lies. • Preserve Social Security for generations to come by improving funding and limiting ACKNOWLEDGMENTS the growth of benefits. I want to thank the very able group of We are keenly aware of the enormous CED Trustees and advisors who served on impact that retirement policies have on the CED subcommittee that prepared this re- national saving and the economy. Decisions port (see page viii).

ix Very special thanks are due to the sub- We are deeply grateful to Merrill Lynch committee’s chairman, Lawrence A.Weinbach, and Company, the New York Life Founda- Managing Partner-Chief Executive of Arthur tion, Morgan Stanley Group Inc., John Andersen & Co, for the energy, expertise, Hancock Mutual Life Insurance Company, and wisdom he brought to this project. We Lincoln National Corporation, the Investment are also indebted to Project Director Company Institute, Household International, William J. Beeman, CED’s Vice President and and Towers Perrin for their generous sup- Director of Economic Studies, not only for his port of this most important program. comprehensive knowledge but also for the clarity he has brought to this complex set of issues. Thanks are also due to CED Economist Josh S. Weston Michael Baker for his contributions to this Chairman project and to Carol Alvey for her secretarial CED Research and Policy Committee assistance.

x Chapter 1 Executive Summary and Policy Recommendations

America’s retirement system is under- • Rapid growth in government spending for funded, overregulated, and soon to be chal- the elderly threatens to get so far out of lenged by unprecedented growth in the control when the baby-boom generation retirement-age population. Consequently, our retires that it cannot be financed by reason- nation will confront a major crisis in financing able burdens on taxpayers. the needs of the elderly at the beginning of the Private saving is only a fraction of that twenty-first century unless policies are needed to enable future retirees to fulfill their reformed to make retirement saving a top pri- economic expectations in retirement. One ority. If steps are taken promptly to imple- recent study showed that the baby-boom gen- ment the reforms recommended by CED in eration needs to triple its rate of accumulation this statement, significant sacrifices will be of assets in order to maintain its preretirement required, but they will be manageable. If living standard during retirement. Moreover, action is postponed, the nation will face the this projection assumes that Social Security very unpleasant choice of a substantial cut in benefits will not be cut.2 But it is now clear the economic status of the elderly or an eco- that the Social Security system has made prom- nomically damaging and unfair tax burden ises to future retirees that cannot be kept with- on future generations of workers. out vastly improving prefunding or imposing The economic well-being of future retirees a harsh burden on future workers. This situa- and, indeed, of workers is seriously at risk tion has been exacerbated by ill-advised because of the interaction of several demo- changes in regulatory and tax policies that graphic, economic, and fiscal trends: have discouraged private saving for retire- • Growth in the elderly population in the ment and by federal budget policies that have United States, which is already quite rapid generated huge deficits. These deficits con- because of increasing life expectancy and tinue to absorb the lion’s share of private sav- declining fertility rates, will accelerate when ing needed to ensure the economic prosperity the baby-boom generation reaches retire- of future citizens. ment age in about a decade. A sharp Clearly, for the vast majority of those cur- decline in the ratio of workers to retirees rently in the labor force, their future economic will result.1 circumstances will depend greatly upon their own saving and participation in retirement • Private saving for retirement is woefully plans. Those who are not making sufficient inadequate, and national saving has preparations for retirement appear to have a declined. misunderstanding about the impact of under- • Underfunded pension promises in both saving on their retirement income. CED be- private and public retirement programs are lieves that this situation calls for a major a growing and often understated problem. education campaign sponsored by both gov-

1 ernment and private employers to improve system. Social Security retirement benefits workers’ understanding about their own re- and other benefits for the elderly should sponsibilities and options for retirement. not be promised if they cannot be financed Although the primary focus of this policy either by advance funding that adds to statement is private retirement programs, we national saving or by an acceptable level of recognize that private saving is only one com- taxes on active workers. ponent of the retirement finance problem. We believe that the nation must act quickly Therefore, our discussion of retirement sav- to reform retirement policies and practices so ing would not be complete without a brief that the future economic security of retirees examination of the problem of underfunded and all other citizens will be protected. Post- promises with respect to Social Security and poning action would simply raise the cost dra- pensions for public employees. Indeed, the matically; in contrast, early action imple- outlook for Social Security makes reform of mented in gradual steps would minimize the private retirement policies all the more impact on the living standards of both work- important. ers and retirees. The major thrust of the reforms recom- mended by CED can be described as follows: THE CHALLENGE POSED BY AN • The federal government should streamline AGING POPULATION and simplify regulation and reverse recent Growth in the nation’s real income and changes in regulatory and tax policies that wealth has made it possible for the United act as barriers to private retirement saving States not only to provide for the basic needs and discourage growth in private pension of rapidly growing numbers of elderly but coverage. also to greatly improve their economic • Businesses and government sponsors of well-being. However, because of the aging of retirement programs should fully fund pen- the baby-boom generation, the retired popu- sion promises made to their workers and lation will begin to rise much more rapidly avoid making promises that cannot be kept. and the workforce more slowly during the Business and government should also first decade of the next century. The ratio of encourage an increase in the average workers to retirees is expected to plunge from retirement age in recognition of the its current level of about 3.4 to 1 to about 2 to 1 improved health status of the elderly and by 2030 (see Figure 1).3 Consequently, provid- the projected need for skilled workers, and ing for the needs of growing numbers of eld- to compensate for added retirement costs erly will become a much greater challenge for arising from increases in life expectancy. the country. The share of the nation’s total output consumed by this group will increase, • In order to encourage individuals to in- placing a heavy burden on workers unless crease their saving and take greater respon- the increased consumption by retirees is sibility for their future, workers must be prefinanced by saving, including funded pen- better educated about the effects of retire- sions. Unfortunately, federal regulations ment saving and the choices they make enacted since the mid-1980s have been inimi- about retirement ages, saving rates, and cal to funding. Moreover, legislation requir- investment strategies on their future eco- ing the delay of a large part of funding until nomic circumstances. the later part of workers’ careers, combined • Government should quickly legislate the with the aging of the workforce (as the baby- gradual introduction of benefit-trimming boomers approach retirement), will raise the changes in Social Security in order to pre- cost of many retirement plans sharply, thereby serve the long-run financial health of the discouraging the expansion of retirement ben-

2 Figure 1 efits. Delay in funding was required by the Omnibus Budget Reconciliation Act of 1987 Ratio of Workers to Social Security (OBRA 1987). See Chapter 5 and Figure 25 for Recipients, 1980 to 2050 a description of the effect on funding. (intermediate projection) The upcoming bulge in the elderly popula- Ratio tion presents a particularly serious problem 4 for Social Security. Some of the increase in spending for Social Security benefits can be financed by drawing down the reserve in the 3 Social Security trust funds. However, this will merely delay a tax increase, and probably not 2 for long. The Social Security Administration’s intermediate projection indicates that the an- nual cash flow will turn negative beginning in 1 2013 and that the reserve fund will be ex- hausted in 2029. Because earlier projections of 0 the balance in the Social Security retirement 1980 1990 2000 2010 2020 2030 2040 2050 trust funds have been revised downward per- Year sistently and dramatically (see Figure 2), many NOTE: “Workers” defined as persons having earnings believe it is more realistic to assume that the creditable for Social Security. fund will run out much sooner, perhaps by SOURCE: Office of the Actuary, Social Security 2012, as in the Social Security Administration’s Administration. more pessimistic projection. When the reserves

Figure 2 Revised Intermediate Projections for Social Security Trust Fund Balances (current dollars) Trillions of Dollars

20

1983 projection 15

10

1994 projection 5 1989 projection

0 1985 1995 2005 2015 2025 2035 2045 2055 2063 Year

SOURCE: Office of the Actuary, Social Security Administration.

3 are exhausted, Social Security benefits must Figure 3 be financed by taxes on current wages (or by increased debt), placing a uniquely heavy bur- Social Security Retirement and Hospital den on future workers. Insurance Benefits and Revenues, The actuarial deficiency in Social Security, 1980 to 2030 taking into account all prospective revenues (percent of taxable payroll) and expenditures for the next seventy-five Percent years, was estimated at $1.9 trillion at the end 30 of 1993.4 A rough indication of the potential future burden of Social Security on active workers can be seen by comparing the cost of retirement benefits with the wages of work- 25 ers. The Social Security Administration’s 1994 Benefits intermediate projection indicates that total expenditures for old age and survivors insur- 20 ance and disability and hospital insurance ben- efits, or OASDIHI,5 which are financed by the payroll tax, will increase by an alarming Revenues 12 percent of taxable payroll (from more than 15 15 percent to nearly 28 percent) during the 1994–2030 period, largely as a consequence of 6 demographic change (see Figure 3). OASDIHI 10 benefits are also projected to rise sharply as a 1980 1990 2000 2010 2020 2030 Year share of the gross domestic product (GDP), from 6.4 percent in 1994 to 10.8 percent in NOTE: Revenues exclude interest on trust funds. 2030. These figures do not reflect the rapidly SOURCE: Office of the Actuary, Social Security growing cost of supplementary medical Administration. insurance (SMI, or Part B of Medicare) that is financed primarily out of general tax revenues, the growth in unfunded retirement benefits generations of workers. Senator Robert Kerrey from government jobs that will also be and former Senator John Danforth, co-chair- financed from general revenues, and other men of the Bipartisan Commission on Entitle- programs for the elderly for which long-term ment and Tax Reform, have responded to the projections are not available.7 Thus, it is not Social Security projections by proposing fun- surprising that many have concluded that damental changes in Social Security and the government-related retirement and health ben- U.S. retirement system that would bring efits for the elderly must be cut back. In CED’s future outlays and revenues to near balance. view, reliance on extremely high taxes paid However, there is substantial political resis- by future workers to finance these benefits tance to changes in Social Security at the for the elderly would be very inequitable and present time, although reform is critically have an extremely adverse effect on the living needed and delaying action will only make standard of those workers born after the the problem worse.8 But given the unavoid- baby-boom generation. able impact of the aging population on The impact of demographic change on pay-as-you-go retirement systems, the most pay-as-you-go retirement financing makes it promising way to protect the economic secu- imperative that pension policy be reformed to rity of future retirees without overburden- encourage increased retirement saving and to ing workers is to increase private retirement reduce unfair shifting of the burden to future saving. Many individuals look upon contri-

4 butions to their pension funds, which are their PENSION CONTRIBUTIONS own assets, as much less of a burden than AND FUNDING increased payroll taxes. If policy changes suc- Total private pension contributions ceed in encouraging individuals to forgo a declined in constant 1987 dollars from about significant amount of consumption before $1,470 per worker in 1985 to about $1,140 per retiring, and if sponsors of retirement pro- worker in 1991. The employer component grams fully fund pension promises, the declined in real terms from about $1,039 in income of retirees can be maintained without 1980 to about $506 per worker in 1991 (see an unacceptable sacrifice by workers. Figure 5). Defined benefit plans, which are more commonly sponsored by large firms, suf- THE UNTIMELY DECLINE IN fered a very sharp decrease in contributions 9 RETIREMENT SAVING (see Figure 6). Overall contributions by Unfortunately, saving for retirement and employers to these plans declined largely as other purposes has declined in the United a consequence of three developments: (1) States at the very time when it should be ris- increased earnings on financial assets, (2) the ing in anticipation of the retirement of the enactment in 1987 of lower ceilings on pen- baby-boom generation. Private contributions sion funding eligible for tax deductions, and to pensions as a percent of disposable income (3) a sharp rise in the cost of complying with have declined in the last decade, though not increasingly complex regulations that discour- as sharply as total personal saving (see Fig- aged the growth of defined benefit plans, par- ure 4). ticularly in small firms (see Figure 7). Total

Figure 4 Figure 5 Personal and Pension Saving, Decline in Contributions to Private 1970 to 1991 Pension Plans, 1975 to 1991 (percent of disposable personal income) (in 1987 dollars) Percent Dollars 10 1,600 Total contributions per private Personal saving sector worker 1,400 8 1,200

1,000 6

800

4 600 Employer contributions per private Pension saving sector worker 400 2

200

0 0 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1975 1977 1979 1981 1983 1985 1987 1989 1991 Year Year

SOURCE: Bureau of Economic Analysis; pension saving series SOURCE: The Brookings Institution, Bureau of Labor provided by staff at The Brookings Institution. Statistics, Bureau of Economic Analysis.

5 Figure 6 eral agency that insures private pensions. It should be noted that this insurance program Employer Contributions to Defined also exposes future taxpayers to a huge poten- Benefit Plans, 1975 to 1991 tial liability.11 (in 1987 dollars) Information on underfunding of govern- ment employee pensions is incomplete, but Billions of Dollars the magnitude of underfunding appears to be 70 greater than for private pensions. (Of course, the government’s taxing authority makes it 60 difficult to compare the risks.) It has been esti- mated that the average funding ratio of all 50 state and local pensions is only 80 percent and that the actuarial deficiency for federal, civil- 40 ian, and military pensions at the end of 1992 was $1.5 trillion, taking into account both pro- 30 spective receipts and outlays.12

20 ECONOMIC EFFECTS OF DECLINING RETIREMENT SAVING 10 The decline in retirement saving not only

0 jeopardizes the economic security of individual 1975 1977 1979 1981 1983 1985 1987 1989 1991 Year SOURCE: Pension and Welfare Benefits Administration, Figure 7 U.S. Department of Labor. Regulatory Costs Have Soared (administrative costs for defined benefit plans contributions to defined contribution plans in 1990 dollars) continue to rise, however, as a consequence of Dollars per Worker rapid growth in plan participation. Underfunding of defined benefit pension 500 plans is a serious problem. Although most 1981 private pensions meet legal funding require- 400 ments, CED believes that the funding limits 1991 enacted in the 1980s (described in detail in Chapter 5) have placed many pension funds at risk in the event of unexpected changes in 300 the value of financial assets and in the finan- cial strength of firms. A decline in interest rates, for example, would reduce the return 200 on pension assets and increase the present value of future benefits, thereby causing some pensions to become underfunded. Moreover, 100 the funding status of private pensions has deteriorated even by current legal funding 0 standards. Underfunding of private retirement 15 75 500 10,000 plans has been increasing rapidly, reaching Plan Size (number of employees) $71 billion in 1993, according to the Pension SOURCE: Hay/Huggins Company, Inc. Benefit Guaranty Corporation (PBGC),10 a fed-

6 workers but also affects the economic health and widespread pension coverage. Unfortu- of the country, because retirement saving is a nately, recent legislation appears to have been very large and growing component of national motivated by other concerns, such as the elimi- saving. The U.S. national saving rate (the sum nation of every potential abuse of pensions, of saving for retirement and all other pur- the impact of pension saving on the distribu- poses by individuals, business, and govern- tion of income, and the budget deficit. All ment) has fallen to record low levels in recent these concerns have merit but frequently can- years. In fact, the national saving rate has not be addressed through pension policies with- averaged less than 2 percent of GDP so far in out creating enormously complex and costly the 1990s, down from 4 percent in the 1980s regulation and undermining the primary and about 8 percent in previous decades. objective of retirement policy. Because national saving is the domestic source CED believes that the looming crisis in of funds for investment, its decline has an retirement finance requires the prompt imple- adverse effect on U.S. productivity growth mentation of three measures: (1) tax incen- and international competitiveness. Moreover, tives and regulatory reform to encourage the rate of saving has been low for some time, individual retirement saving and to achieve and the cumulative effect on the nation’s capi- increased funding of, and coverage by, pri- tal stock cannot be reversed quickly. Of course, vate pensions; (2) education programs that weak growth in productivity and real wages increase worker awareness of the need for does not bode well for the living standard of retirement saving and encourage individual future retirees and workers. responsibility; and (3) changes in government priorities to provide full funding of public employee pensions, increased funding of So- POLICY OPTIONS TO TACKLE THE cial Security, and a reduction in the growth of RETIREMENT FINANCE PROBLEM spending programs and promises for the eld- There are limited options for alleviating erly financed from current income. the burden on future workers arising from the inevitable increase in the elderly population. REGULATORY REFORM FOR PRIVATE The most important is to make retirement sav- RETIREMENT PLANS ing and adequate funding of pensions top pri- The objective of the 1974 Employee Retire- orities for individuals, employers, and gov- ment Income Security Act (ERISA), which ernment. Retiree spending from retirement remains the basic pension law of the United saving (including benefits received from States and deals with all facets of private pen- funded pensions) is merely the recapture of sions, was quite simple: to protect the pen- deferred consumption. Unlike pay-as-you-go sions of the elderly. However, numerous financing, this recapture need not be a burden amendments have since been enacted that are on workers because the saving resulting from extremely complex, in large part because of deferred consumption contributes to the their diverse objectives. These changes have growth of the economy and the growth of placed many barriers in the way of increased workers’ real incomes. Indeed, if the added pension coverage and retirement saving. For saving creates sufficient income to pay for the example, complicated discrimination rules and rise in benefits to retirees, active workers funding limits enacted in the 1980s have would experience no increased burden as the sharply increased administrative costs, retiree population grows. reduced contributions, and placed pensions Therefore, the basic objective of retire- at increased risk. From a national saving per- ment policy should be to ensure the eco- spective, the most serious setback was nomic security of retirees by encouraging OBRA87, which limited funding of defined retirement saving, full funding of pensions, benefit plans to 150 percent of plan termination

7 liability. (Termination liability is the liability a paying taxes. We recognize, however, that plan would incur if it were to cease at any workers are not well informed about retire- given time. 150 percent of termination liabil- ment needs and benefits.13 Indeed, financial ity is generally lower than 100% of projected literacy is low in the United States, and many liability, which was the previous funding workers do not have adequate information limit.) This change forced many firms to dis- about retirement income and retirement continue further funding of pensions for sev- options to plan intelligently for their old age. eral years. The fact that private retirement saving per By discouraging retirement saving, these worker may be only about one-third of what regulatory changes have clearly exacerbated will be needed to maintain accustomed living the burden on future workers arising from standards during retirement does not appear demographic change. At the same time, regu- to be widely understood by workers. There- lators have failed to take adequate action to fore, CED recommends that government and preserve existing retirement funds. Reforms business regularly provide all workers with are needed to reduce preretirement with- information on accrued and prospective drawal of retirement funds for consumption retirement benefits. The Social Security purposes and to encourage improvements in Administration is planning to begin the distri- the portability of pension funds. bution of such information soon. This is CED believes that pension regulatory and a positive development. It is critically tax policies must be streamlined and simpli- important, however, that the information on fied and subjected to cost-benefit evaluation. Social Security be as realistic as possible, We propose a number of changes in the regu- including the possibility that benefits may be lation of private pensions that would greatly cut in order to preserve the system. In this simplify discrimination rules, restore the way, individuals can make informed plans for full-funding limit to its pre-1987 level of 100 their retirement. percent of projected plan liability, preserve pension funds for retirement, permit pension REFORMING GOVERNMENT tax preferences to be based on lifetime income EMPLOYEE PENSIONS AND THE (as opposed to current-year income only), SOCIAL SECURITY PROGRAM encourage accelerated vesting and greater It is also important to reform the funding portability of pension assets, increase the of government employee pensions. Indeed, authority of the PBGC to make sure that spon- CED strongly endorses the view that pen- sors of private pensions meet their funding sion promises made by both private and pub- responsibilities, and increase the limit on pen- lic employers should be fully funded and sion benefits and contributions (see “Summary that no promises should be made that cannot of Policy Recommendations”). All our recom- be funded. mendations are designed to encourage saving The specific character of required reforms and greater support for private retirement pro- in Social Security retirement programs is a grams. complex issue beyond the scope of this state- ment. It is clear, however, that merely passing RETIREMENT INFORMATION AND the burden to future generations is both ineq- INDIVIDUAL RESPONSIBILITY uitable and bad economics. The increase in Although reform of pension regulation is total benefit payments that will occur critical, CED believes that it is incumbent because of the baby-boom bulge should be upon individuals to take greater responsi- prefinanced, and/or benefits should be cut bility for financing their retirement. Individu- to limit the burden on future workers. We als ultimately pay for their own retirement by also believe that the rise in life expectancy saving, participating in retirement plans, and warrants a further increase in the normal

8 retirement age beyond the increase already of preparations for retirement and the poten- legislated and that the earnings limit for tial burden on workers, reduced benefits and recipients of Social Security should be raised delayed retirement are not likely to be matters substantially in order to reduce the disincen- of choice. However, the burden on future tive for the elderly to continue working. Of retirees can be minimized by giving them course, a politically realistic resolution of the adequate warning. CED believes that neces- Social Security financing problem is likely to sary reductions in future Social Security ben- involve both cuts in benefits, especially for efits should be enacted well before they take higher-income individuals, and some addi- effect in order to give workers an opportu- tional taxes to increase prefunding of the nity to make compensating adjustments in retirement of the baby boomers. CED has pre- their saving behavior. viously endorsed a proposal for full taxing Such reforms would permit a larger of Social Security benefits in excess of the buildup in Social Security trust fund reserves beneficiary’s contributions. (Because low- and reduce the potential burden on future income retirees do not pay federal income workers by making it possible to hold down taxes, the taxation of all benefits would have payroll tax rates when the baby-boom genera- an effect somewhat similar to means testing of tion retires. However, the effect on the payroll benefits.) Other reforms, such as a gradual tax may be a misleading indicator of the phase-in of a lower ratio of initial benefits to future burden if the increased reserves do not income and/or limits on the indexing of ben- add to national saving. No additional resources efits (above the minimum subsistence level), are created to pay for the rise in benefits should be considered. unless the reserves add to national saving, The economic status of retirees has productive capital, and economic growth. improved sharply in recent decades, a devel- Unfortunately, it appears that the recent posi- opment that has probably raised workers’ tive cash flow in the Social Security program expectations relating to retirement income. To has been used to pay for current government the degree that workers regard promised services rather than to add to saving. CED Social Security benefits as a right, similar to a believes it is imperative that the positive cash property right, that should not be repealed, flow in the Social Security program be used, there may be strong resistance to these not to support government consumption, but options. Nevertheless, given the present state to add to national saving.

SUMMARY OF POLICY RECOMMENDATIONS The economic well-being of future retirees not be provided without imposing a harsh and their dependents is in jeopardy because burden on future generations of workers. of the continuing failure of our society to make Unfortunately, the regulation of private pen- adequate preparation for an inevitable sharp sions has become incredibly complex and rise in the retirement-age population. This situ- aimed at various social objectives that often ation is reflected in the low priority many conflict with the goal of encouraging retire- individuals give to retirement saving, changes ment saving. CED believes that policy mak- in regulations that have discouraged or lim- ers should focus much greater attention on ited contributions to retirement plans, the basic goal of retirement policy, which is underfunded pensions sponsored by both pub- to ensure the economic security of the eld- lic and private employers, and a looming cri- erly by encouraging saving for retirement. sis in the Social Security retirement system, The adverse effects of government policies which has promised future benefits that can- on retirement saving and work effort have led

9 us to the conclusion that current pension laws, concerns, such as the budget deficit, should regulations, and practices must be streamlined, not be addressed in a manner that serves to simplified and designed to achieve six basic exacerbate our private saving deficiency. In objectives: addition, other important retirement policy • Provide retirement plans for the largest issues should not be addressed in a way that possible number of workers. defeats the basic saving objective of pension policy. • Ensure that pension plans meet appropri- The inadequacy of present-day retirement ate fiduciary standards and are funded to saving may be due in part to confusion about fulfill pension promises. who is responsible for meeting the income • Encourage individual saving by making goals of retirees. Apparently, some individu- retirement saving opportunities available als mistakenly believe that their income in to all workers through deferred taxes on retirement is entirely or principally the contributions and earnings and by pro- responsibility of government and business. In viding all workers with realistic informa- fact, the responsibility for the income of tion on their retirement saving needs, retirees is divided among government, resources, and options. employers, and individuals. However, each individual must bear the ultimate responsi- • Place simple and reasonable limits on tax bility for his or her living standard in retire- preferences received by any individual ment. Social Security, which is only loosely (i.e., a single overall limit on eligible con- linked to income, should continue to provide tributions indexed to inflation and a limit a minimum level of retirement income for all on benefits received from a qualified working members of society and their depen- defined benefit plan). dents. Businesses and other employers should be • Preserve retirement saving and pension encouraged to voluntarily facilitate and assist rights by discouraging preretirement saving for retirement to the degree affordable. withdrawals and by improving the port- Company-sponsored pensions are often use- ability of pension assets. ful as a means of attracting and retaining a competent workforce. Employer plans, com- • Compensate for the rise in life expect- bined with Social Security, are generally ancy by encouraging retirement at a later intended to provide retirees with a reasonable age and by reducing work disincentives minimum standard of living, though with less for the elderly who receive retirement ben- income than received by active workers. efits. Finally, saving by individual workers should be Both business and government should encouraged so that they can achieve their encourage later retirement in order to com- desired living standard in retirement. For pensate for the rise in life expectancy, and many workers, their savings are a critical government should raise the earnings limits source of retirement income. applicable to retirees receiving Social Secu- CED continues to favor strong action to rity benefits. Although government tax and reduce the federal deficit as the most certain regulatory policies should encourage private way to increase national saving needed to pension saving, we believe that tax incen- improve the growth and competitiveness of tives should be carefully designed to ensure the U.S. economy. Some of the recommenda- that they raise national saving as well. Pen- tions presented in this statement would sion saving for retirement represent a large reduce the federal deficit; others would raise and increasingly important component of it. We believe that any net increase in the national saving. Other legitimate fiscal policy federal deficit resulting from reform of retire-

10 ment policies should be offset by reductions tion standards mentioned above should, in real spending. However, the purpose of at a minimum, adopt the coverage rule. reducing the deficit is to increase national sav- • Top-heavy rules, which are already largely ing; therefore, attempts to lower the deficit by redundant in view of the nondiscrimina- taxing private saving are counterproductive tion rules in place for all pension plans, and should be avoided. should be eliminated. With these goals and responsibilities in mind, CED makes the following recommen- • Restrictions on the employer’s ability to dations: integrate private pension benefits with 1. Retirement plan nondiscrimination rules Social Security should be eliminated in governing coverage and contributions should cases where the employer contribution for be streamlined and simplified in a manner all employees equals or exceeds an that reduces administrative costs and in- appropriate threshold. creases saving for retirement and coverage. 2. Federal regulation should encourage full Nondiscrimination rules in general, includ- funding of private pensions. ing coverage rules and nondiscrimination tests One of the most important causes of relating to contributions under Sections 401(k) declining business contributions to pensions and 401(m) of the Internal Revenue Code, is OBRA87, which capped funding for defined are complex and extremely costly to adminis- benefit plans at 150 percent of termination ter. These tests reduce pension saving by liability. As a consequence of this legislation middle-income earners as well as more highly and rising asset values, many firms were compensated employees. With appropriate unable to make eligible contributions for sev- grandfathering of rules where necessary, these eral years. Firms can only ensure the viability tests should be replaced by a vastly more of their defined benefit plans if they are per- simple and administratively inexpensive safe- mitted reasonable flexibility to spread the costs guard against discrimination: of funding plans over time in a manner that • All employees who meet nondiscrimina- realistically reflects both expected plan liabili- tory age and service eligibility require- ties and the firms’ ability to make contribu- ments should (a) be covered and (b) tions. The full-funding limit based on termi- receive the same ratio of employer contri- nation liability denies such flexibility and butions to wages. (In the case of defined places a disproportionate burden on the firm benefit plans, the same benefit formula in the later years of workers’ careers. should be applied.) • The full-funding limit should be restored If such a radical reform cannot be adopted to its pre-1987 level of 100 percent of pro- in its entirety, simplification should proceed jected plan liability. Projected plan liabil- along the following lines: ity in the case of flat-dollar defined ben- • Employers who voluntarily comply with efit plans should be calculated to include the two nondiscrimination standards anticipated increases in the dollar benefit specified in the preceding bullet would level that are negotiated over time. be given safe harbor from all further non- 3. Business and government should provide discrimination tests, such as those under workers with adequate and realistic Sections 401(k) and 401(m) of the Internal information about their pensions and Social Revenue Code. Security benefits. • Employers who do not comply with both Many individuals reportedly undersave the coverage and the uniform contribu- because they are not adequately informed

11 about accumulated benefits, retirement needs, • Regulators should investigate options for and the rate of saving required to meet those accelerating vesting and for improving the needs. Individuals can prepare intelligently portability of vested benefits from defined for their retirement only if they have sufficient benefit plans. Individuals who change information about their retirement needs and employers should be strongly encouraged resources as well as their saving and invest- to roll over preretirement lump-sum ment options. Some employers are reported distributions into alternative retirement to be reluctant to provide information on pro- saving instruments such as individual spective growth of funds because they fear retirement accounts (IRAs) and defined that they could be liable for the outcome. contribution plans maintained by their new employers. • Workers should be given realistic infor- mation about retirement needs, accumu- 5. In order to increase saving for retirement lated and projected benefits, the effect of and to encourage small-business owners and different investment choices on their managers to provide pension plans for all retirement income (if they have control employees, Congress should raise the over such choices), and the funded status annual limit on allowable plan contributions of their plan (if it is a defined benefit and benefits to more reasonable levels. plan). Regulations limiting both contributions and benefits are often redundant and tend to • The Department of Labor should provide reduce pension contributions and saving, lead- guidance designed to encourage employ- ing to the abandonment of some plans and ers to provide information on retirement discouraging the formation of new ones. Lim- saving and investment to their employ- its on benefits also discourage higher-risk in- ees and to protect plan sponsors against vestments. The following reforms should be unreasonable lawsuits. For example, made: information provided by employers about • Retirement saving tax preferences should past market performance should not make be based on accumulated lifetime income employers legally liable if future perfor- rather than current-year income only. This mance is inconsistent with experience. would be a stimulus to saving and far more equitable, particularly for those who 4. Regulation should encourage both the pres- must postpone contributing because of ervation of retirement funds until the worker unusual expenses (such as medical expen- retires and the portability of pension assets. ditures) or because of a temporary Preretirement withdrawals from pension absence from the workforce (such as funds are placing the economic security of parents caring for children). future retirees in jeopardy. Although some of these withdrawals are used for investment • With respect to the redundancy of restric- purposes such as education and housing, a tions on benefits and contributions, lim- large proportion is simply used for current its on tax preferences for contributions consumption. CED recommends: are generally preferable to restrictions on benefits. Therefore, the excise tax on • In-service preretirement withdrawal and annuitized pension distributions above a borrowing of employer contributions to stated threshold (now $148,500 for a single pension plans should be prohibited. employee) should be eliminated, and the Access to voluntary employee contribu- limit on benefits from a qualified defined tions should not be prohibited, but exist- benefit plan (now $118,800) should be ing penalties should be retained. raised.

12 • Given the proposed elimination of excise levels. Amortization periods for unfunded taxes on distributions, a limit on total con- liabilities should be simplified and short- tributions and on considered compensa- ened. tion should be retained. However, the con- The PBGC’s poor financial position has sidered compensation limit (now $150,000) three underlying causes: (a) Premiums for risky should be raised at least to its pre-OBRA93 plans are set too low. (b) Premiums are struc- level14 (and indexed); to prevent contribu- tured in a way that creates an adverse-selection tions for lower-income workers from problem in the insurance pool. (c) The PBGC’s declining, the limit on considered com- compliance authority is weak and there is pensation should apply only to those insufficient incentive for companies with whose incomes actually equal or exceed underfunded pension plans to increase their that limit, not to those whose projected funding. Minimum-funding rules also allow incomes equal or exceed it. The dollar amortization of some kinds of unfunded pen- limit on total contributions should also sion liabilities over an excessively long period. be increased. • The PBGC insurance premium should be • Congress should discontinue the recent restructured so that it more closely practice of reducing considered compen- resembles what would be offered in the sation for more highly compensated private market. Such a redesign should employees as a means of raising federal include a stronger link between the pre- revenue. This is necessary because of its mium level and the actual risk that a pen- sion plan poses to the PBGC. For example, adverse effect on pension saving, through the premium calculation could take into its impact on contributions for lower- account the financial strength of the plan income employees and on the willing- sponsor, the marketability of plan assets, ness and ability of sponsors to create and and the proportion of assets tied up in the maintain qualified plans. firm’s own equities. • Individual contributions to 401(k) plans • The PBGC should be given more power to that are primary pension plans should be influence the behavior of sponsors of subject to the same limit that applies to underfunded plans, including the author- other defined contribution plans, such that ity to prevent plans operating below a par- the combined employee and employer ticular funded ratio from granting benefit contributions do not exceed that limit. increases to employees. This will give workers in smaller firms an opportunity to expand their retirement • The amortization schedules for unfunded savings and provide a greater incentive liabilities should be simplified by reduc- for managers of small firms to offer these ing the number of categories, and the plans to their workers. amortization periods should be shortened to accelerate the funding of liabilities. 6. The Pension Benefit Guaranty Corpora- tion reforms should improve both the sol- • The PBGC’s benefit guarantee structure vency of the PBGC itself and the funded should be revised to correlate more closely status of the pension plans it insures. Thus, with the minimum-funding requirements the PBGC premium structure and benefit attributable to specific benefits, such as guarantee should be aligned more closely shutdown benefits (see Chapter 5). with the actual risk posed by a pension plan • The status of the PBGC’s claims in bank- and the PBGC should have stronger compli- ruptcy should be reviewed and enhanced, ance authority to insure adequate funding as appropriate.

13 7. Changes are required to place the rity federal budget deficit and that they Social Security program on a sound finan- result in an increase in real national sav- cial footing and avoid an unfair burden on ings. future generations of workers. 8. Defined benefit plans sponsored by fed- Based on already-legislated benefits, pro- eral, state, and local governments should be jected demographic changes, and other fac- subject to minimum disclosure and funding tors, the Social Security Administration’s standards. Plans should be fully funded, and long-run projections indicate that the pro- pension promises that cannot be funded gram’s assets will be depleted rapidly when should be avoided. the baby-boom generation retires unless ben- The magnitude of underfunding of public efits are cut and/or payroll taxes are increased employee pensions appears to exceed under- sharply. This policy statement focuses on the funding of private pensions. A substantial regulation of private pensions and does not number of state and local government set out a detailed program for Social Security employee pension plans are underfunded. reform. However, the fiscal principles Federal pensions are also greatly under- espoused by CED lead us to call for the fol- funded. To maintain the solvency of under- lowing changes: funded plans, it will be necessary to • All Social Security benefits that exceed increase taxes on future workers and/or cut past contributions should be subject to benefits. income tax. We also believe that the age 9. Pension funds should not be required to requirements for receiving normal make investments to achieve social objec- retirement benefits should be raised tives other than the objective of protecting beyond the increase already legislated and the economic security of the elderly. that the limits on earnings of retirees Many state and local pension funds and receiving benefits should be raised sub- even private funds are under pressure to stantially. Congress should consider other invest in infrastructure and to make various limitations on Social Security benefits, other social investments. CED does not such as reduced cost-of-living adjustments oppose arrangements that permit individuals for benefits above a basic floor, and/or a to choose such investments, but we strongly gradual phase-in of a lower ratio of initial oppose any mandated use of pension invest- benefits to income. ment funds. Unless the individual chooses oth- • Steps should be taken to make sure that erwise, pension investments should be based the building Social Security reserve funds on sound economics (e.g., risk and return), do not serve to mask the non-Social Secu- not on social considerations.*

*See memorandum by JAMES Q. RIORDAN, (page 82).

14 Chapter 2 Retirement Saving and Economic Security

In coming years, America will face the ernment tax and regulatory policies affecting major challenge of providing for the economic private pension saving must play a critical security of its elderly population without plac- role in this change. ing unreasonable tax burdens on future gen- erations. Retirement systems are already bur- dened by the rising life expectancy and a trend THE ECONOMIC SECURITY OF toward earlier retirement. More importantly, FUTURE RETIREES the United States, like many advanced indus- The retired elderly in the United States trial countries, will experience a very sharp receive income from a variety of sources, (see rise in its retired population and the retiree- “Sources of Funds for Retirement,” page 17). worker ratio when the baby-boom generation The three primary components of the U.S. begins to retire.1 Unlike citizens in many other retirement system are: (1) the huge Social countries, however, Americans are not setting Security retirement program, which provides aside resources adequate for their retirement most retirees with benefits at least sufficient needs. Indeed, overall saving rates in the to meet their basic needs for subsistence; (2) United States have fallen to record low levels the large and diverse private retirement sys- well below those of other industrial nations. tem that grew rapidly with the encourage- Not only has the private saving rate declined ment of government tax incentives during the precisely at a time when greater efforts should post–World War II period and now covers be made to prepare for the aging of the popu- about 49 million workers;2 and (3) the retire- lation, but the federal budget deficit is also ment programs sponsored by federal, state, absorbing the lion’s share of those savings, and local governments for their own employ- thereby further reducing the amount avail- ees, which have grown in importance largely able for investment needed to spur economic because of increasing employment in the state growth. and local sectors.3 Saving for retirement must become a higher These three components, together with priority of individuals and government policy gains in personal wealth (especially invest- makers if future American retirees and tax- ment in homes), have dramatically improved payers are to avoid unexpected hardships. But the well-being of present-day retirees relative building adequate savings, like turning a large to past generations of retirees. Some studies ship, is a slow process; it will take many years suggest that the overall economic welfare of of higher saving to boost investment to the today’s younger retirees is comparable to that levels necessary to generate sufficient income of present-day full-time workers. However, to meet the expectations of future retirees. there are now serious concerns about whether Thus, it is urgent that saving and retirement the economic well-being of future retirees can income policies be changed promptly. In this be maintained at the level enjoyed by current statement, CED argues that the reform of gov- retirees, let alone be improved, as many have

15 come to expect. The reasons for these con- insufficient saving for retirement. Future cerns (discussed in detail in Chapter 3) can be retirees who have inadequate incomes may summarized as follows: demand increased retirement benefits from Social Security (or at least resist any cutback • Individuals are living longer and retiring in benefits). Those with inadequate savings earlier. Greatly compounding this trend is are more likely to become a burden on the the fact that the ratio of workers to retirees government in other areas, such as long-term will decline sharply in the next several health care. Inadequate saving for retirement decades because of the retirement of the also affects the economic health of the nation baby-boom generation. Consequently, the more broadly. cost to each future active worker of sup- porting retiree benefits will rise sharply. Benefits financed on a pay-as-you-go basis THE ECONOMY AND RETIREMENT may have to be cut back significantly in SAVING order to prevent taxes paid by future work- Although this statement focuses primarily ers from becoming too burdensome. on the economic security of future retirees, we • At present, saving for retirement is not suf- cannot overlook the adverse effect of pension ficient in the United States to meet the policy on the economy and, in turn, the im- income requirements of future retirees. This pact of a weakened economy on the well-being can be seen in the overall decline in contri- of future retirees. The decline in saving for butions to retirement plans, the under- retirement and other purposes has serious ad- funding of many private and public plans, verse implications for the future economic and record-low saving rates. prosperity of the nation; a weak economy with little or no growth in productivity and real • There has been a gradual shift in coverage, wages will, in turn, struggle to provide ad- especially among small employers, from equate support for rising numbers of retirees employer-funded retirement plans to more and to address other social problems. The So- discretionary saving vehicles such as 401(k) cial Security Administration’s underfunded plans. The impact of this trend on the eco- liability for baby-boomer retirement benefits nomic security of future retirees is not yet is a case of too many claims on too little out- known with certainty. put. One way to think about the impact of the Prompt changes in retirement policies can burden of rising benefits for baby boomers is improve the prospects of future retirees. To to ask: How much output will be left for work- compensate for the rise in life expectancy, ers? The implication of the Social Security both business and government should Administration’s intermediate projection is encourage workers to take retirement at a that nearly a quarter of the assumed 1.2 per- later age. Policies should also be instituted cent annual gain in labor productivity (real that encourage saving and full funding of GDP per worker) between 2010 and 2030 will pension promises. Recent changes in federal be absorbed by rising benefits paid by Social tax and regulatory policies governing private Security retirement, disability and hospital pensions have often moved policy in the wrong insurance benefits alone. However, the Social direction. (For a detailed discussion, see Chap- Security Administration’s history of overopti- ter 5.) Changes in pension regulation have mistic projections and the record-low national discouraged retirement saving and contrib- saving rates experienced in recent years raise uted to the underfunding of many pensions. the question of whether Social Security’s pro- The American taxpayer has more than one ductivity assumption is again too optimistic. reason to be concerned about the impact of If so, these programs will absorb an even

16 SOURCES OF FUNDS FOR RETIREMENT

There are six primary sources of retirement according to a formula related to income or income, the funding of which will determine profits) to an employee’s retirement account. whether retirees’ incomes meet expectations. Such plans are often favored by smaller firms To the degree that funding falls short, either because they are less regulated and have lower retirees’ entitlements and aspirations will have administrative costs and more certain contribu- to be curtailed (by delay of retirement or by tion requirements. An important advantage of reduced living standards during retirement defined contribution plans is that there is no years), or the intergenerational transfer of pur- public liability for inadequate funding. How- chasing power from the working generation to ever, these plans often produce a lesser final the retired generation must be increased, prima- benefit than a defined benefit plan. There are rily through higher payroll taxes. five times as many defined contribution plans in the United States as defined benefit plans, but 1. Social Security. About 90 percent of all work- many are small, and only about 26 percent of ers participate in the Social Security system, U.S. workers are currently active participants in whose role is to guarantee at least a minimum a defined contribution plan that is their primary level of retirement income for every eligible retirement plan. retiree. This government program has greatly reduced poverty in America, although 12 per- 4. Personal Savings. For retirees, the median cent of older people are still below the poverty cash income from assets was $2,356 in 1990. But level. In 1994, the median benefit for an indi- personal saving has declined sharply in recent vidual was about $9,972. The maximum annual years, perhaps reflecting misinformation about benefit for a couple is now $20,646. the adequacy of provisions for retirement. The Social Security Administration will soon launch 2. Defined Benefit Plans. The largest accumula- a program to provide workers with more infor- tion of retirement funds is in defined benefit mation; this information, together with better plans, which promise specific benefits and are information from employers, is necessary if typically sponsored by large employers. workers are to understand the need for in- Approximately 15 percent of workers are creased saving. enrolled in such plans, though there has been a tendency for small employers to discontinue 5. Postretirement Employment. For many, jobs the plans because of excessive regulatory bur- (usually part-time) provide meaningful oppor- dens and costs. These plans, in combination tunities and often a necessary retirement with Social Security, generally replace about 60 supplement. About 22 percent of retirees had percent of preretirement earnings for those who income from earnings in 1990. Under present have worked at least thirty years. But workers Social Security rules, for those under 65 benefits who change jobs frequently or work part-time are reduced by $1 for every $2 of annual earn- often fail to qualify or accumulate adequate ings in excess of $8,040; and for those 65 to 69 vested benefits from these plans. (Those who do years old benefits are reduced by $1 for every become vested may receive benefits from more $3 of earnings in excess of $11,160. Social than one plan.) Government workers are gener- Security benefits are not reduced for workers ally covered by defined benefit plans, many of who are 70 or older. which are underfunded and frequently have 6. Private Intergenerational Transfers. The made overly generous promises. transfer of savings from deceased parents to 3. Defined Contribution Plans. These plans children is an important source of income that promise no specific retirement benefit; instead, for many could serve as a partial offset to short- employers make specific contributions (often falls in their own retirement funding.

17 higher share of productivity gains. Moreover, business, and government have all contrib- when other transfers are taken into account, uted to this decline. So far in this decade, net the gain in productivity available to active national saving has averaged less than 2 per- workers may decline sharply (see Figure 8). cent of GDP, down from 4 percent during the Beginning in the mid-1970s, the growth in 1980s and about 8 percent in previous de- labor productivity in the United States slowed cades (see Figure 9). This decline in saving dramatically, from nearly 3 percent to about 1 does not bode well for the future growth of percent a year, producing a long period of the U.S. economy because national saving is near stagnation of real wages. The prospects the only source of domestic funds for invest- for restoring rapid productivity growth are ments that are needed to boost productivity. largely dependent on increases in the rate of In fact, there is concern that the extremely low investment in physical and human capital and domestic saving rates experienced in recent advances in technology. But increased saving years are not sufficient to prevent productiv- is required to provide the funds for such ity growth from falling further.4 investments, and unfortunately the nation has The national saving and investment rates chosen to use a growing share of its income to in the United States have also fallen below finance current consumption rather than sav- those of most other advanced industrial ing and investment. Indeed, the U.S. national nations (see Figure 10). This is bad news for saving rate — the sum of net saving by indi- the international competitiveness of the United viduals, business, and government divided States, which hinges on relative rates of pro- by GDP — has declined precipitously in ductivity growth. Admittedly, investment recent years to record-low levels. Individuals, rates in the United States have not been as

Figure 8 A Large Share of Worker Productivity Growth Will Go to the Elderly Annual Average Percent Growth 1.25

Real GDP per worker 1.20

1.15

Less real Social Security retirement benefits 1.10

1.05

1.00 Less real retirement benefits and hospital insurance 0.95

0 2000 2005 2010 2015 2020 2025 2030

Five-Year Period Ending

SOURCE: Tabulated from intermediate projections contained in the 1994 OASDI Trustees report.

18 Figure 9 weak as national saving rates because foreign capital inflows have financed a substantial U.S. National Saving, Selected Periods amount of domestic spending for investment. (percent of GDP) But reliance on foreign capital inflows has Percent serious disadvantages. These capital inflows 10 increase foreign claims on future U.S. output, a particular concern when the inflows result

8 from a decline in domestic saving rather than from rising investment opportunities in the United States. Foreign capital inflows have 6 brought about a dramatic shift (now approach- ing $1 trillion) in the U.S. net foreign invest- 4 ment position. Once the largest creditor nation in the world, the United States has 2 become the largest debtor, and the rising debt to foreigners represents a future burden for 5 0 the nation. 1950–1959 1960–1969 1970–1979 1980–1989 1990–1993 Why has saving fallen in the United States? Years Huge federal budget deficits made the largest NOTE: National saving is net of depreciation. contribution to the decline in net national sav- SOURCE: Bureau of Economic Analysis. ing (see Figure 11, page 20). In recent years, these deficits have absorbed about 70 percent of private savings. The private sector has also Figure 10 contributed to the decline. The corporate sav- National Saving in G-7 Countries, ing rate (retained earnings) declined sharply in the 1980s, though it now appears to be 1960 to 1991 rising as profits recover. But the personal sav- (percent of GDP) ing rate, which fell from about 8 percent Percent of disposable income in the first half of the 30 1980s, remains very depressed at about 4 per- cent of disposable income.6 Retirement saving, 25 Japan which is a very large (and growing) compo- 20 nent of personal saving, declined from nearly 4 percent of disposable income in 1980 to less Other G-7 (a) 15 than 3 percent in recent years (see Figure 12, Germany page 20). Clearly, any serious effort to increase 10 private saving will require a change in current retirement policies. United States 5 In a 1992 policy statement, Restoring Pros- perity: Budget Choices for Economic Growth, 0 CED said that reversing the collapse in 1960 1965 1970 1975 1980 1985 1990 Year national saving should become an explicit goal of U.S. policy. It recommended that gov- NOTE: National saving is net of depreciation. (a) Other G-7 countries are France, Italy, Canada, and United ernment policies should aim to restore the Kingdom. rate of national saving to the 8 percent norm SOURCE: Organization for Economic Cooperation and that existed prior to the 1980s.7 Two of the Development. most potent policy instruments available for

19 Figure 11 achieving that goal are federal budget policies to decrease the budget deficit and retirement Federal Deficits and National Saving, policies to encourage private retirement sav- Selected Periods ing. (percent of net national product) With regard to the federal budget, CED Percent recommended that the deficit be reduced by 12 about $50 billion annually, with the eventual goal of achieving a surplus. From fiscal 1991 10 to 1993, the federal budget deficit averaged about $272 billion, or about 4.6 percent of GDP. Net 8 nonfederal The 1993 budget act and economic expansion saving have put the deficit on a lower path; it is 6 expected to fall from 3.9 percent of GDP in fiscal 1993 to 2.3 percent of GDP in fiscal 1995.8 4 Federal This decline in the budget deficit will make a deficit significant contribution to national saving, 2 though policy makers have a long way to go Net national to get entitlement spending under control and saving 0 to achieve the target recommended by CED 1950–1959 1960–1969 1970–1979 1980–1989 1990–1993 Years for the federal budget. Moreover, the long-term budget forecast is not so sanguine; later in the SOURCE: Bureau of Economic Analysis. decade, the deficit is expected to begin rising

Figure 12 Personal and Pension Saving, Selected Periods (percent of disposable personal income) Percent 10

Personal saving

8 Private pension contributions

6

4

2

0 1972–1974 1975–1977 1978–1980 1981–1983 1984–1986 1987–1989 1990–1991 Years

SOURCE: Bureau of Economic Analysis; pension saving series provided by staff at The Brookings Institution.

20 again as a percent of GDP.9 CED continues to responsibility is it to provide for their pen- recommend implementation of policies that sions? Observers report that some present- will gradually move the federal budget to- day workers have the mistaken belief that gov- ward balance as a means of increasing na- ernment and business will provide for all their tional saving and investment. retirement needs. Such beliefs may arise be- With respect to retirement policies, it is cause workers do not have adequate informa- clear that measures that succeed in increasing tion about the benefits they will be entitled to private retirement saving by expanding pen- in retirement. sion coverage and increasing contributions of Although CED believes that each family current participants have the potential for sub- should take primary responsibility for its own stantially increasing household saving. How- economic welfare, we agree that all sectors of ever, it may be very difficult to restore pen- society, including business, have a stake in a sion saving rates to earlier levels. Changing healthy private pension system and in the individual priorities would be critical because welfare of retirees. We believe that a reason- many individuals are not saving anything for able arrangement would have Social Security their own retirement.10 Some of the blame for and voluntary business-sponsored pensions the decline in retirement saving must be providing a reasonable minimum level of re- attributed to ill-advised changes in public poli- tirement income for workers, with the indi- cies that discourage business contributions to vidual being responsible for improvements pensions. (We examine this issue in detail in beyond that minimum. Given pension arrange- Chapter 5.) These policies should be reversed.11 ments already in place, including the Social It should also be noted that the public sector Security system, the following division of re- has contributed more directly to the problem sponsibilities is recommended: of inadequate saving by underfunding both • Social Security should provide a mini- public employee retirement programs and mum level of retirement income for all Social Security retirement promises (see Chap- working members of society and their ter 3). dependents. The recent weakness in private pension sav- ing is an important component of the national • Business, nonprofit employers, and gov- saving and investment problem. Clearly, the ernments should be encouraged to pro- United States needs to employ pension poli- vide supplemental retirement assistance. cies to encourage saving not only to improve Employers often provide retirement ben- the economic security of individual retirees efits to attract and retain competent work- but also to boost national economic growth ers. The combination of Social Security and competitiveness. More rapid growth of and employer-sponsored plans is gener- the economy, in turn, will make it easier for ally intended to provide retirees with a this country to support its rising elderly popu- reasonable minimum standard of living, lation. though with less income than that received by active workers. PRIVATE- AND PUBLIC-SECTOR • Individual workers should have ultimate RESPONSIBILITIES FOR PENSIONS responsibility for achieving retirement We have seen that both the nation and income above that provided by Social individual citizens have a strong interest in Security and employer retirement plans, retirement savings accumulated for the according to their desired standard of liv- baby-boom generation. But precisely whose ing.

21 ACTIONS REQUIRED TO MEET as a matter of equity and sound fiscal policy, THESE RESPONSIBILITIES CED has long favored the taxation of all Social Security benefits that exceed past con- It is important that each group — govern- tributions. This change would be a form of ment, private employers, and individuals — means test that would also improve equity in meet their responsibilities without placing an taxation by treating income of workers and unfair burden on future generations. As a prac- 14 tical matter, this means that retirement sav- retirees in the same way. CED also recom- ings must be sufficient to fund promised mends that Congress consider other limita- benefits and the average retirement age must tions on Social Security benefits, such as an be raised to reflect longer life spans. CED increase in the normal and early retirement urges both business and government to ages beyond that already legislated, reduced review retirement programs to ensure that cost-of-living adjustments for benefits above they do not provide ongoing disincentives a basic floor, and a gradual phase-in of lower for productive employment by the elderly, replacement ratios (i.e., the ratio of retire- without reducing the ability of firms to ment benefit to income received during years restructure as needed. of employment). To reduce the disincentive Social Security. With respect to programs for the elderly to continue to work, the limit financed largely on a pay-as-you-go basis, such on earnings of recipients of Social Security as Social Security, government must be care- benefits should be raised substantially. Such ful to resist pressures to promise generous changes could place the Social Security pro- unfunded retirement benefits beyond those gram on a sound footing and avoid an unfair necessary to provide a basic floor of retire- burden on future generations. ment income. Thus far, our political leaders Private employers. Fully funded private and have not been willing to face the demographic public employee retirement programs are criti- challenge and the long-term underfunding of cal for many retirees and also have great promised Social Security benefits. Longer social value, because they add to national sav- retirement spans have greatly increased the ing. We believe all employers should be cost of retirement benefits since the inception encouraged to provide retirement plans or at of Social Security, and the Social Security least to give employees the opportunity Administration’s long-term projections indi- to accumulate retirement savings with cate that the system’s assets will be depleted tax-sheltered contributions from their own rapidly when the baby-boom generation incomes. Public and private employers have retires unless benefits are cut and/or payroll an obligation to adequately fund the pension taxes are increased sharply (see Chapter 3). promises they make to their workers and to An increase in the normal retirement age from maintain high standards of fund management. 65 to 67 years is scheduled to be phased in Individuals are also more likely to achieve during the years 2003 through 2025, but this their retirement-saving objectives if retirement increase does not fully compensate for the benefits can be preserved when a worker increase in life expectancy.12 At the same time, moves from one employer to another. CED earnings limitations applied to those receiv- believes that employers should place a high ing Social Security benefits have reduced the priority on improving vesting and the port- supply of productive workers and exacerbated ability of retirement assets. the decline in the worker/retiree ratio.13 Individual responsibility and the education of Because the primary concern of this policy workers. Individuals, on the other hand, can statement is private retirement saving, we do tailor their retirement-saving decisions to their not investigate the merits of alternative pro- particular circumstances. When deciding on posals for “fixing” Social Security. However, an income-replacement ratio that meets their

22 needs, individuals would, in theory, have the the Department of Labor should provide option of choosing the age at which they will guidance designed to encourage employers retire, whether they will continue to work to provide information on retirement saving after retirement, and the rate at which they and investment to their employees and to will draw from existing assets, including protect plan sponsors against unreasonable equity in a home. However, in a world where lawsuits. For example, information provided job security is uncertain, workers may not by employers about past market performance always have such employment choices, and should not make employers legally liable if many workers will have few assets other than future performance is inconsistent with ex- pensions. Most workers will need to set aside perience. Given the long-term funding prob- substantial savings in addition to pensions to lems of Social Security, it is important to in- achieve their lifestyle expectations. form workers that circumstances may However, American workers are not well necessitate the enactment of measures that re- informed about retirement saving and, indeed, duce the growth of future benefits. about personal finance.15 To prepare intelli- Employer-provided pension benefits are gently for their retirement, individual work- future payments for current work: that is, ers require information about prospective ben- deferred compensation. As such, pension ben- efits from Social Security and private pensions. efits, like wages generally, are primarily a In those instances in which workers have some private-sector issue. However, the U.S. expe- control over the amounts they elect to contrib- rience with some private pensions before the ute to pension plans and the distribution of enactment of ERISA suggests that government those contributions over different kinds of also has a responsibility to provide reasonable investments, they also need information on and stable regulation of private pensions in how their choices could affect their retirement order to make sure that pensions meet appro- income. Unfortunately, workers frequently do priate fiduciary standards and that pension not have such information, though it may be promises are kept. The contribution of private readily available (see Chapter 3) and can be pensions to the economic security of retirees provided at low cost. There is also evidence and to the strength of the national economy that Americans have a poor level of financial indicates that the private pension system has literacy and that this deficiency is a strong an important social value that extends beyond factor in poor saving decisions.16 CED believes the specific benefit to individual retirees. In that business and government have the ad- CED’s view, the social benefits of private ditional responsibility of providing workers pensions justify some government oversight with adequate and realistic information (provided that it is responsible, low cost, and about their pensions, including retirement stable) and appropriate tax preferences to needs, accumulated benefits, and the fund- encourage retirement saving. (These issues ing status of their plan. We also believe that are discussed in detail in Chapters 4 and 5.)

23 Chapter 3 The Basic Facts on Our Aging Population and Retirement Saving

AN AGING POPULATION about 12 percent currently to about 20 percent by 2030, when aged persons will number more One of the most significant social develop- than 68 million. The number of beneficiaries ments in the United States during the twenti- of the Social Security retirement and disability eth century has been the phenomenal growth program (OASDI) is projected to rise from of the elderly population. During the first 41.8 million in 1993 to 80.3 million in 2030.1 ninety years of this century, the number of The social implications of such an accelera- people age 65 years or older in the United tion in the growth of the aged population are States rose by nearly 29 million, from 4.6 per- quite profound. Even with a gradual increase cent of the total population to 12.3 percent. in the elderly population, the growth in gov- This rapid growth reflected the combined ernment spending for the elderly has been influence of three factors: (1) a spectacular phenomenal. In the last three decades, spend- increase in life expectancy (from 47 years in ing for Social Security, Medicare, and other 1900 to 75 years in 1990), (2) a sharp decline in related retirement and disability programs rose the birthrate, and (3) a reduction in immigra- tion. All these factors may contribute to the rela- tive growth of the elderly population in the Figure 13 future, though the precise contribution of each Percentage of U.S. Population Age 65 is uncertain. However, the future growth of and Over, 1950 to 2050 the retired population will also be greatly Percent influenced by an additional factor about which 25 there is no doubt: The baby-boom generation will begin to reach retirement age in about a decade. This group, which includes about 76 20 million individuals born between 1946 and 1964, comprises an unusually large segment 15 of the U.S. population because the birthrate rose during that period and declined there- 10 after. Because of the baby-boom bulge in the age distribution, the growth of the elderly population during the first half of the 5 twenty-first century is expected to accelerate

(see Figure 13), substantially exceeding the 0 very rapid growth in this century. The Social 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Security Administration’s intermediate pro- Year jection indicates that the proportion of the SOURCE: Social Security Administration (intermediate population age 65 and older will rise from projection)

24 from 16.5 percent to 36.5 percent of total fed- Figure 14 eral expenditures. In 1993, 55.5 percent of total federal expenditures for purposes other Ratio of Workers to Social Security than defense and interest on the national debt Recipients, 1980 to 2050 went to the elderly. If government programs Ratio for the elderly are not changed, the retirement 4 of the baby boomers will cause the cost of these programs to rise to unsustainable levels. The potential burden on workers to sup- port Social Security retirement benefits can be 3 seen by examining the ratio of workers to retirees. This ratio has remained fairly stable since 1975 in a range of 3.2 to 3.4; but with the 2 retirement of the baby boomers it will decline sharply. The Social Security Administration’s intermediate projection shows the ratio of cov- ered workers (those who have earnings credit- 1 able for Social Security benefits) to beneficia- ries declining from 3.4 in 1990 to 2.0 in 2030 and below 2.0 by midcentury (see Figure 14). 0 This projection is based on highly uncertain 1980 1990 2000 2010 2020 2030 2040 2050 assumptions about immigration, fertility, and Year death rates that some believe are unrealistic. NOTE: Workers are defined here as persons having earnings Under the Social Security Administration’s creditable for Social Security. more pessimistic assumptions, the ratio of cov- SOURCE: Office of the Actuary, Social Security ered workers to beneficiaries would fall to 1.5 Administration. by 2050.2 Obviously, the financing of prom- ised Social Security retirement and disability benefits on a pay-as-you-go basis would re- young people as well as the number of eld- quire a large increase in the payroll tax bur- erly, is not expected to rise as much as the den falling on future workers and employers. aged dependency ratio (see Table 1, page 26).4 The current intermediate projection suggests However, the total dependency ratio substan- that the tax increase from 1994 to 2030 for tially understates the burden on workers promised retirement benefits (OASDI) alone because the cost of supporting the elderly would probably be in the range of 3 to 7 per- greatly exceeds the cost of supporting the centage points, depending on how long the young, particularly when health care costs are increase is delayed and a number of other included,5 and because the dependency ratio unknown factors.3 Furthermore, any such in- does not take into account the fact that many crease in payroll taxes would be in addition to workers retire before the age of 65. Therefore, tax increases to finance the rising cost of health the projected decline in the ratio of covered care for the aged, whether through Medicare workers to beneficiaries represents a reason- or other public or private programs. able measure of the likely impact of the com- Of course, the ratio of covered workers to ing demographic changes on workers and beneficiaries by itself does not present a com- employers. plete picture of the overall economic burden It is often asserted that policy makers could on future workers. The total dependency exert a mitigating effect on these demographic ratio, which takes into account the number of trends by carefully controlling immigration

25 Table 1 so as to boost the number of young workers and reduce the decline in the ratio of covered Dependency Ratios workers to retirees. But given the magnitude of the shortfall in workers, it is not likely that (a) (b) Year Aged Total immigration could make a significant dent in l990 0.209 0.700 the problem. The number of additional work- 2000 0.210 0.691 ers required to stabilize the ratio at its 1990 level (3.4) is about 13 million in the year 2000 2010 0.216 0.652 (see Table 2). The shortfall would rise to 110 2020 0.279 0.701 million by 2030 and 128 million in 2050, about 36 percent of the projected population in that 2030 0.360 0.791 year. The number of additional immigrants (a) Population 65 and over divided by population age 20 –64. necessary to offset this shortfall of workers (b) Population 65 and over plus population under age 20 would be much larger than these figures divided by population age 20–64. because many of the immigrants would be SOURCE: Board of Trustees. Federal Old Age and Survivors retired by 2030.6 (The baseline projection Insurance and Disability Trust Fund, 1994 Annual Report already includes an assumption that about 8.5 (Washington, D.C.: U.S. Government Printing Office, 1994), p. 144. million immigrants, about 4 percent of whom will already be over the age of 65, come to the

Table 2 Number of Workers Required to Hold the Worker-Beneficiary Ratio at Its 1990 Value of 3.4 to 1 (millions)

Number of Workers Needed to Hold Worker/Beneficiary Social Security Projection Ratio to 1990 Level Baseline Social Ratio of Percent of Population Security Workers to Additional Projected Projection Beneficiaries Workers Beneficiaries Workers Baseline Pop.

1990 259 39 133 3.4 0 0%

2000 285 47 147 3.1 13 5%

2010 306 55 157 2.9 28 9%

2020 325 68 161 2.4 70 22%

2030 340 80 163 2.0 110 32%

2040 349 84 166 2.0 121 35%

2050 354 87 168 1.9 128 36% NOTE: Workers refers to covered workers. SOURCE: Social Security Administration intermediate projection.

26 United States between 1990 and 2000.) Recent ments; government pensions and private pen- research indicates that massive immigration sions accounted for about 19 and 31 percent, would be required to have any significant respectively (see Figure 15). Total private pen- effect.7 sion payments grew very rapidly until the The coming bulge in the elderly popula- mid-1980s. Consequently, the Social Security tion is the inevitable consequence of the aging retirement share of benefit payments declined of the baby boomers. No reasonable public (from 62.1 percent in 1970 to 48.1 percent in policy would change these demographic 1986); since the mid-1980s, it has remained trends sufficiently to halt the decline in the about the same. For most individual retirees, ratio of workers to retirees. From this circum- Social Security is the most important source of stance, we conclude that it is critically impor- income; benefits were received by about tant to increase retirement saving in anticipa- 92 percent of the aged in 1992. In that year, tion of the rise in the aged population. about 66 percent of the elderly had income from assets, and 44 percent received retire- ment benefits other than Social Security. About PRIVATE SECTOR PREPARATION 29 percent of private sector retirees receive a FOR RETIREMENT OF THE BABY private pension. In 1994, the median annual BOOMERS Social Security payment was $9,972, which Total retirement income from all sources, was roughly 80 percent higher than the including Social Security and private pensions, median private pension annuity payment, and now exceeds half a trillion dollars annually in 3 times the median cash income from assets. the United States, roughly 10 percent of per- For retirees who receive private pensions sonal income. In 1992, Social Security after thirty years of service from medium-size accounted for half of total retirement pay- and large private establishments, the combined

Figure 15 Retirement Benefit Payments, by Source, 1970 and 1992 (percent of total payments)

1970 1992 Government Private Government Private pensions 22.0% pensions 16.0% pensions 18.7% pensions 31.4%

Social Security 62.0% Social Security 49.9%

SOURCE: Employee Benefit Research Institute.

27 private pension and primary Social Security Figure 17 benefit replaced 59 percent of wages in 1990. Figures 15 (see page 27) and 16 show retire- Private Pension Assets, 1960 to 1992 ment income from all sources, including gov- (in 1992 dollars and as a percent of national ernment.8 wealth) Percent Billions of Dollars 3,500 PRIVATE PENSION CONTRIBUTIONS 15 In recent decades, private pension assets in 3,000 the United States have grown rapidly in dol- lar terms and as a share of national wealth 2,500 10 (see Figure 17). However, much of the increase 2,000 reflects an appreciation in the market value of Percent of national wealth these assets rather than saving out of current 1,500 9 income. This is seen in Figure 18, which shows 5 that total contributions to private pensions in 1,000 Dollars real terms have declined during the last 500 decade despite the rapid growth of the labor force. Real contributions per worker have 0 0 1960 1965 1970 1975 1980 1985 1990 fallen sharply. Year New funding limits enacted in the 1980s (see Chapter 5), together with a rise in asset SOURCE: Pension and Welfare Benefits Administration, Employee Benefit Research Institute, and Federal Reserve. values, caused the employer component of

Figure 16 Percentage of the Aged with Income from Various Sources, 1992

Percent 100

80

60

40

20

0 Social Security Asset Income Pensions Earnings Public Assistance Veterans’ Benefits Source of Income

SOURCE: Office of Research and Statistics, Social Security Administration.

28 pension contributions to decline. As shown in sion, and coverage has become almost univer- Figure 19, total employer contributions per sal for the largest employers. However, total worker fell sharply in real terms. The change pension coverage has not increased signifi- in funding limits forced many firms to halt or cantly since the early 1970s. But although over- delay contributions in recent years, a particu- all pension coverage appears to have stabi- larly unfortunate development from a national lized, coverage for male workers has declined, saving perspective. Such required reductions and participation of younger workers in pen- in contributions, especially when triggered by sion plans is significantly lower than that of short-term increases in interest rates, are also middle-aged and older workers (see Figure very shortsighted (for an explanation, see 20, page 30).12 “Estimating Liabilities and Funding for Despite the weakness in private pension Defined Benefit Plans,” pages 52 and 53); if contributions and coverage, the proportion of there is a reversal of these rates, many compa- households that will receive a pension may nies would quickly find that they need to make rise because of shorter vesting times and the extremely large contributions so that their pen- increased participation of women in the labor sions will not become underfunded.10 A situa- force. On average, however, women’s pen- tion that requires very large future contribu- sions are significantly smaller than men’s tions may discourage growth in pension because they have had shorter working lives coverage. and lower earnings. Moreover, higher-income households are still much more likely to have PRIVATE PENSION COVERAGE11 private pension coverage of some type than Private pension coverage rose sharply in lower-income households, even though non- the early postwar period. By 1970, about 52 discrimination rules were intended to broaden percent of full-time private-sector workers the coverage provided by individual pension were covered by some type of private pen- plans. In 1993, nearly 80 percent of workers

Figure 18 Figure 19 Total Contributions to Private Pension Employer Contributions to Private Plans, 1975 to 1991 Pension Plans, 1975 to 1991 (in 1987 dollars) (in 1987 dollars) Billions of Dollars Dollars Billions of Dollars Dollars 140 1,500 120 1,100

Total dollars 1,400 100 1,000 120 900 1,300 80 Total employer 100 contributions 800 1,200 60 Dollars per 700 80 private sector Employer worker 1,100 40 contributions per private 600 60 1,000 20 sector worker 500

0 0 0 0 1975 1977 1979 1981 1983 1985 1987 1989 1991 1975 1977 1979 1981 1983 1985 1987 1989 1991 Year Year SOURCE: Brookings Institution. SOURCE: National income and product accounts.

29 with annual incomes of $50,000 or more par- Figure 21 ticipated in a pension plan, whereas only about 9 percent of those with incomes of $10,000 or Pension Participation, by Income less participated (see Figure 21). Group, 1993 Percent DECLINE IN DEFINED BENEFIT PLANS 100 The type of pension coverage available to workers is undergoing a dramatic shift. There 80 has been a sharp decline in the private sector in the number of defined benefit plans.13 These

plans, which provide specified retirement ben- 60 efits, are the type of pension most often pro- vided by older and larger companies. In con- trast, the number of defined contribution plans, 40 particularly 401(k) plans, has grown rapidly, thereby preventing a drop in total coverage (see Figure 22). But the assets of defined con- 20 tribution plans are still only 40 percent of total pension assets (see Figure 23). This is because 0 defined contribution pension plans are most Less than $10,000 $10,000–$24,999 $25,000–$49,999 $50,000+ frequently used as primary pensions by small Income Group firms, the self-employed, and individuals, and SOURCE: Employee Benefit Research Institute. as supplementary (frequently voluntary) pen-

Figure 20 Figure 22 Pension Participation, by Age Private Pension Plan Participation, by Group, 1993 Type of Plan, 1979 to 1993 (percent of workforce) Percent Defined contribution Percent 50 70 Defined benefit

60 40

50 30 40

30 20

20 10

10

0 0 1979 1983 1988 1993 21–30 31–40 41–50 51–60 61–64 Year Age SOURCE: Adapted from Department of Labor and Employee Benefit Research Institute tabulations of Current Population SOURCE: Employee Benefit Research Institute. Survey data.

30 Figure 23 sion saving by firms whose primary retire- ment program is a defined benefit plan. Private Pension Plan Assets, by Type Employees of large firms are still much more of Plan, 1975 to 1991 likely to be covered by a pension than Billions of Dollars employees of small firms (see Figure 24). 2,500 Many smaller firms do not offer their employ- Non-401(K) defined contribution ees a pension plan. Evidently, economic fac- tors such as the uncertainty of future income 2,000 401(K) have much to do with this. Smaller firms also Defined benefit tend to have a greater percentage of employ- 1,500 ees who are young, part-time, and/or not likely to stay in their job for a long period.14 1,000 Formation of new plans has declined in recent years, and most of the pensions that have been introduced by small firms are of the defined 500 contribution type.15 The decline in defined benefit plans, which 0 is at least partly the result of government regu- 1975 1977 1979 1981 1983 1985 1987 1989 1991 Year lations (see Chapter 5), has also raised fears about the economic security of future retirees. NOTE: Funds held by life insurers under allocated insurance Critics of defined contribution plans believe contracts are excluded. that these plans will not provide retirees with SOURCE: Department of Labor. adequate income. This is not a settled issue, and some researchers have reached the con- clusion that defined contribution plans Figure 24 actually provide greater economic security, 16 Pension Plan Participation, by Size in part because of their portability. But it is possible to improve vesting and portability of of Firm, 1993 defined benefit plans. As we noted in Chapter Percent 2, CED regards improved portability of pen- 70 sion assets as a worthy objective and urges regulators to investigate the various options 60 for achieving this goal.

50 PRERETIREMENT WITHDRAWALS OF PENSION FUNDS 40 Pension contributions and coverage rates do not tell the whole story about the adequacy of 30 ongoing preparations for the retirement of the baby-boom generation. Many plans (includ- 20 ing both defined benefit and defined contri- bution plans) permit borrowing from pension 10 assets and lump-sum distributions at retire- ment or when changing jobs. The increasing 0 1–24 25–99 100–499 500–999 1,000+ number of preretirement distributions received Number of Employees when changing jobs appears to be a serious

SOURCE: Employee Benefit Research Institute. threat to retirement saving. Although the tax code strongly encourages the rollover of

31 lump-sum distributions into IRAs or other Federal regulatory and tax policies pertain- qualified retirement instruments, one survey ing to funding have been schizophrenic. They suggests that only a small fraction of recipi- require minimum-funding levels to protect ents do so; 40 percent of recipients use some benefit promises but limit maximum funding of the funds for consumption.17 Preretire- so stringently that the health of even well- ment lump-sum distributions not rolled over funded plans is threatened. Ironically, the amounted to about $42 billion in 1990.18 maximum-funding requirements, which were Many observers have become concerned enacted to limit federal revenue losses, actu- that lump-sum withdrawals and borrowing ally heighten the exposure of taxpayers to the from pension assets, which are now quite com- possibility of increased Treasury expenditures mon, will undermine saving for pensions and to bail out the Pension Benefit Guaranty Cor- the adequacy of retirement income. On the poration (PBGC), the federal government other hand, a substantial proportion of funds agency that insures defined benefit plans.20 not invested in qualified instruments is placed The PBGC guarantees that a specific por- in other forms of investment, such as housing tion of the promised benefit will be paid and education, and these investments may regardless of the financial condition of the improve living standards in retirement. sponsor. Because of termination of under- Although the net impact on the well-being of funded plans, the PBGC has taken over plans future retirees of preretirement lump-sum with billions of dollars of promised benefits. withdrawals and borrowing from pension The PBGC’s financial condition has raised funds is not certain, there certainly appears to questions about its viability and the security be a risk that such practices will undermine of pensions it guarantees, and this has led the adequacy of future retirement income. to proposals for reform. (Further discussion of Moreover, the depletion of retirement funds the operation of the PBGC is provided in may ultimately put pressure on policy makers Chapter 5.) to raise Social Security benefits. (CED’s rec- The 66,000 defined benefit plans insured ommendations pertaining to preretirement by the PBGC have promised about $900 bil- withdrawal of pension assets are described in lion in benefits. The PBGC believes that most Chapter 5). of these pensions are well funded, but it notes a continuing deterioration in coverage and overfunding THE STATUS OF PENSION FUNDS funding. According to the PBGC, has declined, and there has been a substantial PRIVATE PENSION FUNDS rise in underfunding. The decline in funding There is also evidence of inadequate pen- status reflects the combined effects of falling sion saving in the deterioration of pension interest rates (which increase the present value funding for some defined benefit plans. This of plan liabilities) and funding limits enacted deterioration jeopardizes not only the eco- by Congress in 1987 in an attempt to raise nomic security of future retirees but also the federal revenues. future financial strength of the economy and The total underfunding of all plans insured the employers that have promised pensions. by the PBGC amounted to $71 billion at the Pensions are a major source of national saving end of 1993, an increase of $18 billion from and capital for productive investment. Pen- the previous year. A relatively small number sion plans own about one-fourth of all corpo- of firms account for a large share of the rate equities and one-third of the fixed obliga- underfunding; in 1993, just fifty companies tions of business; and of course, pensions are accounted for 56 percent of the underfunding a major holder of government obligations.19 in single-employer plans. The typical under-

32 funded plans, according to the PBGC, are cent or less for higher-income workers. Social “collectively-bargained flat benefit plans that Security is credited with having sharply do not anticipate future salary and wage reduced poverty among the elderly and, increases in their funding operations.”21 Many together with Medicare and private pensions, believe that the PBGC figures understate the with boosting the income of many retirees funding problem because of the measures of nearly to the level enjoyed by workers (see funding adequacy that the PBGC is required “The Economic Status of the Elderly and to use.22 Nonelderly Compared,” page 34). Because they had paid so little in taxes, the PUBLIC SECTOR FUNDING early recipients of Social Security retirement OF PENSIONS benefits received a very high return on their The primary concern of this statement is contributions. In fact, for almost all individu- private retirement policies. However, a dis- als who currently receive benefits, the annuity cussion of the status of retirement saving value of their benefits exceeds their contribu- would not be complete without some mention tions paid (plus interest). The return on con- of the underfunding of pensions sponsored tributions is generally higher for low-income by government, which is a very serious prob- recipients, though in absolute terms the lem. excess of benefits over contributions has gen- Social Security Funding.23 The Social Se- erally favored higher-income recipients. How- curity retirement system was designed to be ever, the return for all new retirees is fall- a universal retirement program that provides ing rapidly and will become negative for a basic retirement income through a transfer higher-income single individuals (particularly from workers to retirees, including some re- men) later in this decade. Single individuals distribution to lower-income retirees, financed retiring early in the next century face deeply largely on a pay-as-you-go basis by payroll negative returns.24 This situation has led many taxes. In effect, it is a compact between gen- to question the long-term political viability of erations whereby current workers agree to the Social Security program if reform is not support retirees in return for similar consider- forthcoming. ation in their retirement from future workers. Social Security retirement and disability Social Security is the most important source of benefits are financed primarily on a pay- income for the elderly and is received by as-you-go basis. At present, a 6.2 percent pay- 92 percent of retirees. The maximum annual roll tax is paid by both employers and retirement benefit for a worker who retired at employees (a total of 12.4 percent) on earn- age 65 in 1994 was $13,797. It is estimated that ings up to $60,000.25 The system currently has the average worker qualified for about $9,972, a temporary positive cash flow. However, which replaced about 43 percent of earnings. long-term projections of program expenditures Recent budget legislation raised the tax on and revenues show that on a net-present-value Social Security benefits received by higher- basis, Social Security is massively underfunded. income retirees (see Chapter 2). At present, receipts in excess of benefit pay- Social Security is particularly important for ments and other expenses accumulate in the low-income retirees because it replaces a OASDI trust fund, which had an estimated higher share of their income. In 1994, the balance of $427 billion at the end of 1994, replacement ratio (Social Security benefits as about 130 percent of one year’s expenditures. a share of preretirement income) averaged One of the problems with a retirement sys- about 58 percent for workers whose income tem financed on a pay-as-you-go basis is that was below average, compared with 24 per- the tax burden on workers increases as the

33 ratio of retirees to workers increases. This ments also reduced benefits for future retirees effect can be minimized by building trust fund by gradually raising the age requirement for balances that are later spent down during a full retirement benefits from 65 to 67 years. bulge in the retirement-age population. The These changes were projected to generate a 1983 Social Security Amendments raised pay- large annual positive cash flow that would roll tax rates not only to ensure the current boost the OASDI trust funds to nearly $21 solvency of Social Security retirement funds trillion by 2045. The 1983 projections indicated but also to prefund the bulge in benefit pay- that if tax rates and benefits were not changed, ments that will be required by the retirement the trust fund would maintain a positive bal- of the baby-boom generation. The amend- ance until 2063 (under intermediate assump-

THE ECONOMIC STATUS OF THE ELDERLY AND NONELDERLY COMPARED

The income of the elderly has clearly grown are controversial, in part, because the value of more rapidly than the income of the non- Medicare and other noncash income is difficult elderly in the postwar period. Nevertheless, to determine. it is not easy to develop meaningful compari- Poverty rates among the elderly have declined sons of the current economic status of the sharply; government tabulations show a de- elderly and nonelderly. On a before-tax cash cline from 27.9 percent in 1967 to 12.2 percent income basis, the median family unit income in 1990.c In 1990, the poverty rate for the eld- of the elderly was about 70 percent of all non- erly would have been 36 percent higher if no elderly in 1990 and about 50 percent of the Social Security benefits had been received.d income of 45- to 55-year-olds, the age group with the highest earnings. But such figures It has been widely reported that by 1990, the may give a misleading picture of the eco- poverty rate for the elderly was actually below nomic status of the elderly.a To begin with, the the rate for all nonelderly (13.7 percent). How- distribution of incomes of the elderly appears ever, the overall nonelderly rate was boosted to be quite variable; the more senior elderly, by very high poverty rates for children. Pov- for example, generally have much lower erty rates for the elderly are still above the incomes than younger elderly. However, com- poverty rates for nonelderly adults. Moreover, pared with the nonelderly, the elderly often these poverty measures do not take into con- face lower tax rates, have more wealth, and sideration wealth and other factors that influ- receive significantly more noncash income. ence economic well-being. Compared with Indeed, some studies have found that the real younger people, for example, the elderly face median income of the elderly is comparable to a much greater risk that their economic status the median income for all nonelderly if such will be adversely affected by large medical factors are taken into account.b Such findings bills.

a. For income data by age group, see Daniel B. Radner, An Assessment of the Economic Status of the Aged (Washington, D.C.: Social Security Administration, Office of Research and Statistics, May 1993), p. 32. b. Michael D. Hurd, “The Adequacy of Retirement Resources and the Role of Pensions” (October 1993). Hurd adjusted cash incomes for such things as family size, tax liabilities, noncash income from assets (e.g., imputed rent on owner- occupied homes), and Medicare benefits and concluded that the real income of the elderly was 101 percent of the median for nonelderly. c. Radner, An Assessment of the Economic Status of the Aged, p. 43. d. U.S. Department of Health and Human Services, Social Security Administration, Office of Research and Statistics, Income of the Aged Chartbook, 1990, Supplement 1992 (Washington, D.C.: U.S. Government Printing Office, 1992), p . 10.

34 tions), sufficient to finance the bulge in retire- has arisen concerning Social Security funding ment benefits caused by the baby-boom gen- because the temporary positive cash flow that eration. However, actuarial projections of the is now being realized (about $69 billion for solvency of the Social Security trust funds have OASDI in fiscal 1995) is used to purchase debt continued to deteriorate since 1983 and were issues of the U.S. Treasury, which appear to revised down sharply again in 1994. Benefit merely finance government consumption. Of outflows are now projected to exceed tax in- course, the total government deficit measures come beginning in 2013; and the OASDI bal- the impact of the government budget balance ance is expected to be exhausted in 2029, on national saving; thus, the impact of the thirty-four years earlier than envisioned in positive cash flow in Social Security depends 1983.26 Another way to look at the financial on how it affects decisions on non–Social status of the Social Security program is to com- Security spending and taxing. Social Security pare the present value of benefits expected to adds to national saving only if the annual accrue to all past, present, and future workers excess of revenues over expenditures reduces over a specific time horizon with the present the overall budget deficit. Many argue that value of the assets (including future contribu- the Social Security surpluses finance spend- tions) expected to accrue over the same time ing and therefore are merely an accounting period. This provides a measure of the system’s device with little significance because they do actuarial deficiency. The actuarial deficiency not add to saving or encourage growth in the of the Social Security program by this mea- nation’s capital stock.28 In any case, the Trea- sure is estimated by the Office of Manage- sury will have to increase taxes, borrow fur- ment and Budget to be about $1.9 trillion. ther, or cut spending in order to redeem these The frequent revisions in the projections of funds when expenditures begin to exceed Social Security funding highlight the great revenues. uncertainty about the future status of the Funding for Government Employee Pen- Social Security trust funds. Nevertheless, if sions. Nearly 10 percent of U.S. workers are Social Security benefits and taxes are not public employees covered by federal, state, or changed, the most likely outcome is that the local pension programs, mostly of the defined annual positive cash flow will end early in the benefit variety. Unlike private employers, the next century and the trust funds will be quickly public employer has no tax incentive to make exhausted. If action is taken promptly, the pension contributions, and public pensions are measures necessary to fix Social Security will not subject to the funding requirements of be less severe. Congress has already ERISA or backed by the PBGC. Years ago, enacted a gradual increase in the normal many public employee retirement benefits, retirement age, and there are several propos- like other current expenditures, were paid als to reduce Social Security benefits and raise from general funds. But such pay-as-you-go tax rates. Senator Robert Kerrey and former funding raised serious questions about Senator John Danforth, cochairmen of the Bi- intergenerational equity and the true cost to partisan Commission on Entitlement and Tax taxpayers of growing employment in the pub- Reform, proposed fundamental changes in the lic sector. Today, prefinancing of public Social Security retirement system that would employee retirement benefits is generally a bring future outlays and revenues to near bal- requirement, though regulatory details regard- ance. However, there is substantial political ing public pensions differ substantially from resistance to change in Social Security at the state to state. present time, and it now appears that no Government expenditures for pension change will be implemented for some years.27 funding compete with other public-sector It should also be noted that a controversy needs and are therefore subject to budgetary

35 politics. There are many recent examples in report found that 75 of 184 pension plans stud- state and local governments of politicians ied contributed less than the actuarially attempting to address a budget problem by required amount in 1988. Others have esti- skipping the annual pension contribution or mated an average funding ratio of 80 percent by altering the actuarial assumptions under- across all state and local pension plans. Unfor- lying the funding. State and local pension fund- tunately, data problems make it difficult to be ing has also been adversely affected by pres- more precise about this issue. Nevertheless, it sures to use pension funds for some special is clear that underfunding is a serious prob- interest, such as the financing of public infra- lem for many government pensions and that structure or environmental programs. Often, future tax burdens will rise sharply to over- it seems that concern about the credit rating of come this situation if funding is not increased the government’s debt is the only real con- (or benefits cut). CED recommends that straint on abuses of public pensions. defined benefit plans sponsored by state and Although there seems to be great variabil- local governments be subject to minimum ity, a substantial number of state and local disclosure and funding standards. Plans pensions are reported to be underfunded. A should be fully funded, and promises that recent General Accounting Office (GAO) cannot be funded should be avoided.

Table 3 Comparison of Initial Retirement Benefits and Contributions Made by Public- and Private-Sector Employees for Defined Benefit Plans and Social Security

Public Sector

Without With Private Social Social Item Sector Security Security

Combined Social Security and pension benefits as a percent of final earnings Final earnings, $35,000 66.4% 65.4% 87.0% Final earnings, $65,000 57.6% 65.4% 73.3%

Total contributions to Social Security and pensions as percent of final earnings Final earnings, $35,000 6.2% 7.6% 11.3% Final earnings, $65,000 5.5% 7.6% 10.6%

Ratio of initial benefit to contribution in final year Final earnings, $35,000 11 to 1% 9 to 1% 8 to 1% Final earnings, $65,000 10 to 1% 9 to 1% 7 to 1%

SOURCE: William J. Wiatrowski, “On the Disparity Between Private and Public Pensions,” Monthly Labor Review (April 1994): 3–9.

36 Federal employee pension programs EXPECTATIONS OF THE BABY underwent substantial reform in the 1980s, BOOMERS FOR RETIREMENT providing less generous defined benefit pack- INCOME ages that are expected to limit the growth of future liabilities. Moreover, stability in fed- Past experience has encouraged each gen- eral civilian employment and downsizing by eration of Americans to anticipate an improve- the military should reduce the growth in ment in its living standards. The experience of pension benefits. New federal employees and the elderly population in the United States is previously hired workers who joined the new no different. In fact, their economic status has Federal Employee Retirement System are now improved both in absolute terms and relative covered by Social Security, a reduced defined to that of younger workers (see Table 4). The benefit plan, and a defined contribution plan. median real income of the nonelderly is esti- Given present levels of employee and mated to have risen 31.8 percent between 1967 employer contributions as well as statutory and 1990, whereas the income of the elderly funding by agencies, these civilian and mili- (adjusted for family size) is estimated to have tary retirement programs are expected to main- risen 82.8 percent. The rapid rise in the in- tain solvency. But a substantial unfunded come of the elderly relative to that of the liability exists for earlier service under the nonelderly was largely due to increases in older Civil Service Retirement System. Indeed, Social Security benefits, increased coverage by private pensions, and growing income from the actuarial deficiency for civilian and mili- 34 tary pensions combined is estimated to be $1.5 assets. The question remains: How have the trillion. (The actuarial deficiency at the end of gains of current retirees affected the expecta- 1992 was $881 billion for the civilian and $619 tions of workers? Polls seem to present con- billion for the military retirement systems.) tradictory answers. Many younger workers Moreover, federal pension trust funds are in- report that they do not expect to benefit from vested in U.S. Treasury debt. Therefore, as Social Security, but they generally strongly with Social Security, at some future point the oppose its elimination. Treasury will be required to tax or borrow to The recent experience with declining sav- redeem these pension funds.29 ing rates and rising female participation in the It is often pointed out that where there is a “problem” public employee pension program, it may be the case not only that there is Table 4 underfunding but also that too much has been promised. When strapped for cash, some gov- Annual Percent Change in Real Median ernments have substituted pension increases Family Unit Income Adjusted for Size for pay raises in order to satisfy the demands Under 65 or of workers, thereby magnifying funding prob- Years Age 65 Over lems and passing on the burden to future tax- 1967–1972 3.1% 4.9% payers.30 But as a general rule, public sector pensions do not appear to be more generous 1972–1979 1.1% 1.9% in the initial years than private sector pensions 1979–1984 -0.3% 3.4% when individual contributions are taken into account (see Table 3).31 However, the fact that 1984–1990 1.0% 0.9% public sector pensions are more likely to be Average change, 1967–1990 1.2% 2.6% fully indexed for inflation makes such pen- sions more attractive over a longer period.32 Total change, 1967–1990 31.8% 82.8% Of course, indexing for inflation also raises SOURCE: Daniel B. Radner, An Assessment of the Economic the cost of such plans substantially compared Status of the Aged, p. 38. with the cost of plans that are not indexed.33

37 labor force suggests that families will go to possible to exceed this rate of return and great lengths to maintain and improve living achieve the same replacement ratio with lower standards. In the past, retirees have used po- contributions (see “Retirement Benefits and litical power to enhance their living standards, Saving Requirements”).35 Thus, it is important and there is no reason to assume that this will that participants understand the relation be- not happen in the future. Because of the tween risk and return.36 intergenerational compact involved in Social Available data on pension saving suggest Security, many workers view their entitlement that very few workers have an annual private to future Social Security benefits as inviolable, pension saving rate anywhere near that re- like a property right. quired to generate a pension of 50 percent of If future retirees expect their living stan- final pay. In 1991, the combined total contri- dards to improve as much as those of earlier bution to private pensions by employers and generations of elderly, they may be in for a employees appeared to be about 2.9 percent rude awakening unless pension saving is of disposable personal income, down signifi- sharply increased. However, as we indicated cantly from about 3.9 percent in 1980. Part of in Chapter 2, there is evidence that many work- the decline after the mid-1980s was due to ers do not have adequate information about newly enacted full-funding limits. The aver- pension saving and that, consequently, their age business contribution to pension and sav- expectations about their economic circum- ing plans is about 3.6 percent of payroll for all stances in retirement may be unrealistic. Table 5 HOW MUCH SHOULD WORKERS SAVE FOR RETIREMENT? Percent of Earnings that Must Be Information on the savings needed to pur- Saved to Finance a Fully Indexed chase a specific retirement annuity is avail- Pension of 50 Percent of Final Pay able. Given assumptions about investment income, wage growth, age at retirement, and Number other factors, one can also estimate the of Contri- Retirement Age butory required share of income that must be saved Years 55% 60% 65% 70 to achieve a specific retirement annuity. For % example, the data in Table 5 indicate that an 20 34% 31% 28% 25% individual’s annual private pension saving would have to be 13 percent of wages over a 25 25% 23% 21% 18% thirty-five–year period in order to receive at age 65 a fully indexed private pension that 30 19% 18% 16% 14% paid 50 percent of final pay. As Table 5 dem- onstrates, early retirement and a reduction in 35 15% 14% 13% 11% the number of contributory years have a large 40 —% 11% 10% 9% impact on the required saving percentage. Another crucial factor affecting required sav- SOURCE: These estimates were provided by Don Ezra, ing is the assumed rate of return on the managing director of Frank Russell Company. The underly- invested funds. The example in Table 5 is based ing assumptions are as follows: (1) The investment return is 4 percent greater than the rate of inflation. (2) Salary on a real return of 4 percent. By employing a escalation throughout the working career is 1 percent higher more aggressive investment strategy (e.g., a than the rate of inflation. (3) Pensions are paid monthly so portfolio allocation that places a higher per- long as either the retiree or the spouse of identical age is centage of savings in equities and a lower alive. (4) Mortality after retirement is expected to follow the 1983 Group Annuity Mortality Table. percentage in money market funds), it may be

38 industries and probably about 5.5 percent for willing to accept a very modest standard of all industries that have pensions.37 Of course, retirement income. A recent study by the these figures do not include contributions to Congressional Budget Office (CBO) used the Social Security, which are quite large; but for following standard to judge the adequacy of most retirees, Social Security will not provide ongoing preparation for retirement: Will the adequate income in retirement. baby boomers’ real income and wealth in re- Unfortunately, the need for additional pri- tirement exceed that of their parents? The vate saving does not appear to be well under- study answered this question in the affirma- stood, though some pension plans (especially tive for several reasons: (1) The preretirement defined contribution plans) provide workers real income of the baby boomers is expected with adequate information. Many workers are to exceed that of their parents, thereby auto- unaware of the size of their employer-provided matically triggering higher Social Security or pensions and the additional retirement saving pension benefits. (2) Higher workforce par- required to achieve their objectives for retire- ticipation rates by women of the baby-boom ment income. generation will increase their eligibility for UNDERSAVING BY BABY BOOMERS pensions. (3) Changes in pension regulations How much are the baby boomers under- pertaining to vesting and participation make saving on average? The answer depends in it more likely that the baby boomers will part on how much income they expect in receive a pension. (4) The baby boomers are 38 retirement. Perhaps some baby boomers are likely to inherit more wealth.

RETIREMENT BENEFITS AND SAVING REQUIREMENTS Most retirees will require more income than ment and 5 percent after inflation thereafter) is provided by Social Security. Individuals fre- assumed in this example, reflecting a portfolio quently ask: How much retirement saving is allocation that takes on a higher degree of risk. necessary? Fortunately, a reasonably accurate answer can be provided, given a number of assumptions. Once an individual has decided Required Saving as a Percent of Income on a desirable level of retirement saving, the share of earnings that should be set aside for Desired Retiree retirement by the individual and/or an Income employer depends on such factors as whether as a the retirement benefits are adjusted for infla- Percent of Years to Retirement Annual tion, the number of years that pension contri- Salary 10 15 20 25 30 35 butions are made, the return on investments in the retirement fund, and the expected age of 30% 36% 21% 13% 9% 6% 4% retirement. In the example shown here, the required saving rate is lower than in the ex- 40 48 27 18 12 8 6% ample presented in Table 5 even when the 50 60 34 22 15 10 7% retirement income target (50 percent of final pay, indexed for inflation) is identical. The 60 72 41 26 18 12 9% principal reason is that a higher investment return (6 percent after inflation until retire- 70 84 48 31 21 14 10

SOURCE: T. Rowe Price Associates. The underlying assumptions are as follows: (1) A 9 percent annual investment return prior to retirement and 8 percent after retirement, (2) an inflation rate of 3 percent, (3) retirement income lasts thirty years.

39 However, in keeping with the CBO’s prac- nomic well-being as they enter retirement. tice of not forecasting Congressional actions, Given the rise in expectations brought about the evaluation of the future well-being of baby by the prosperity of current retirees, this boomers employed current-policy fiscal as- decline in living standards may be widely sumptions. The study assumed no change in viewed as unacceptable. promised Social Security benefits, even though How much saving would be necessary for the current Social Security benefits cannot be main- baby boomers to maintain their preretirement liv- tained without a very sharp rise in Social ing standard in retirement? An analysis that Security taxes, given demographic trends. A addresses this question was prepared by Pro- realistic projection needs to address the tax fessor B. Douglas Bernheim; it indicates that burden on workers and the political and eco- the baby boomers are very poorly prepared nomic pressures to reduce benefits. In this for their future retirement.42 Employing a respect, the CBO policy assumption begs the model of life-cycle behavior with currently question of whether the retirement income of prescribed Social Security benefits, and using baby boomers will be satisfactory. survey data on household assets and consump- A recent study based on generational tion, Bernheim estimated that households with accounts, which takes into consideration the a head age 35 to 44 were accumulating assets financing of government-projected liabilities at only 34 percent of the required rate. Thus, for retirement and other programs, concludes the baby-boom generation needs to triple rates of that the majority of baby boomers will actu- asset accumulation. He also found that saving ally have a lower standard of living in retire- adequacy is lower for single individuals and ment, in absolute terms, than their parents tends to decline as income rises, reflecting the now enjoy. This is because currently legis- progressivity of Social Security benefits and lated government policies (e.g., for retirement inadequate private pension saving. Moreover, and health care) have passed forward a mas- if future Social Security benefits are cut even sive burden to future generations. Conse- moderately, the required increase in saving quently, a sustainable fiscal policy will require will rise substantially. substantial spending cuts and tax increases.39 The Bernheim study did not take into It has also been pointed out that current account the fact that many retirees have sub- retirees have enjoyed fortunate circumstances stantial equity in their homes. The increase in that are not expected to benefit baby boomers. saving needed to maintain preretirement liv- For example, many current retirees had large ing standards could be less if the baby boomers windfalls from the rise in housing prices and were willing to cash in on their equity in hous- the decline in the real value of liabilities (mort- ing when they retire. Home equity is esti- gages) resulting from inflation. In contrast, mated to account for about 70 percent of the recent economic research suggests that the total wealth of the elderly; therefore, home baby boomers will experience a decline in the equity conversion, by sale or reverse mort- real value of their investment in housing, gage, is a possible means of significantly which represents a very large portion of their wealth.40 The widespread use of adjustable-rate increasing consumption by retirees. But as we mortgages also makes it less likely that they noted earlier, recent projections indicate that will benefit as their parents did from increased the baby-boom generation is not likely to ben- inflation.41 efit from a rise in the real value of housing. Recognizing the great uncertainty about Moreover, studies show that the housing turn- long-term economic trends, it seems that the over rate among the elderly is very low, even most likely outcome is that the baby boomers though many retirees are fully aware of the 43 will experience a very large decline in eco- potential for equity conversion. Some eld-

40 erly become emotionally attached to their and that increased emphasis on private sav- homes and familiar surroundings. Others wish ing is necessary both to ensure the economic to retain home equity as a reserve for meeting security of future retirees and to avoid catastrophic medical expenses. Therefore, unreasonable burdens on future workers. although housing equity should be taken into Although precise estimates of individual sav- consideration when measuring the well-being ing needs are inherently controversial, in part of retirees, the existence of home equity may because of the difficulty of identifying rea- not greatly affect retirees’ perceptions about sonable retirement needs, CED believes that the adequacy of their income. a massive increase in private retirement sav- These facts on demographic trends, pri- ing over the next decade is necessary if we vate retirement saving, the funding of are to avoid a crisis in retirement finance in private and public pensions, and projected the future. Clearly, Congress needs to take retirement income strongly support the con- steps to remove current disincentives for clusions that retirement saving is inadequate retirement saving (see Chapter 5).

41 Chapter 4 Designing Tax Incentives to Raise Saving

Government tax policy is the major tool for the present income tax system in order to encouraging retirement saving. For many achieve particular objectives. Thus, for years, federal income tax laws have included example, interest payments on mortgages are provisions to encourage business and indi- deductible in order to encourage home own- viduals to make contributions to retirement ership, and pension contributions are deduct- funds. However, the deferral of taxes on pen- ible in order to encourage saving. Some of sion saving is very controversial and has these practices have changed the character of changed radically over time. Recent changes the income tax, moving it in the direction of a that have sharply limited tax benefits for pen- consumption tax. Of course, if the income tax sions were motivated by rising political pres- were replaced by a consumption tax, special sures to reduce the huge federal budget defi- tax provisions for pension saving would not cits and by the belief that tax incentives are a be needed because the exclusion of saving windfall because they have little effect on sav- would be the normal tax treatment. ing. The debate concerning the appropriate tax base has again entered the political THE TAX BASE CONTROVERSY arena because of growing support for the The treatment of pensions in the federal consumed-income tax, which would exempt income tax system is central to the long- saving from taxation. The issue is whether the standing debate about the economic impact United States should place greater reliance on and equity of a system that includes saving in taxing consumed income, rather than com- the tax base. Some favor a consumption-based prehensive accrual income that includes all system that exempts saving from taxation consumption and increases in net worth. There either because of its desirable economic is no consensus among scholars or in the busi- effects (i.e., because it encourages saving and ness community on this issue. long-term economic growth) or because they We do not propose to address the debate 1 believe that the exemption will improve tax about the tax base in this policy statement. equity. Proponents of the basic philosophy This statement is deliberately focused on underlying an income tax system generally retirement saving. Of course, we recognize espouse the traditional view that all incre- that the general debate is relevant to the dis- ments to available resources should be taxed cussion of retirement saving and fiscal policy (i.e., that taxable income is the sum of con- more generally. With respect to retirement sav- sumption plus the change in net worth). ing, tax preferences are seen as a subsidy or Although this approach holds that all income, incentive by those who view income as the including saving, should be taxable, in prac- appropriate tax base, whereas the absence of tice, capital gains are generally not taxed until tax-qualified retirement savings is seen as a realized; and numerous tax deductions, penalty or disincentive by those who view 2 exemptions, and credits have been added to consumed income as the proper tax base. The

42 policy position that we take on retirement tax (as happened), but many feared that IRAs preferences in this statement is that with the would not produce a significant increase in existing income tax, the case is compelling for total personal saving. This is because the tax broadening and simplifying the rules relating benefit could be received by households that to tax-qualified retirement savings. Even if merely shift existing savings into IRAs and by this constitutes a subsidy or incentive, we households that borrowed to finance IRA con- nevertheless endorse it as being in the public tributions. In either case, taxpayers would interest. receive a windfall in the form of lower taxes Without taking a position on the tax base with no net increase in aggregate household controversy, we assume in this report that the saving. Many empirical studies have been un- federal government will continue to rely on dertaken in an attempt to determine how much the present income tax system. of IRA saving was new saving. Several, though not all, of these studies have concluded that TAX INCENTIVES AND SAVING IRAs and other tax incentives were powerful 4 A critical question about the current prac- incentives for the creation of personal saving. tice of employing income tax preferences to A far more demanding and relevant test of encourage pension saving is: How effective the success of tax-preferred saving incentives are these incentives? The impact of tax incen- such as IRAs is their effect on national saving. tives on pension saving is part of the broader National saving is the sum of private saving issue of the determinants of saving, which has (by individuals and business) and government confounded scholars for generations. Econo- saving (or dissaving). The impact on national mists have approached the issue by asking: saving is important because, together with What is the effect of changes in the after-tax inflows of foreign capital, national saving rate of return on private saving? Theory alone provides the funds for investment. Whether a does not provide the answer. An increase in tax incentive generates an increase in national the rate of after-tax return lowers the price of saving depends on the amount of additional future consumption, thereby increasing the private saving, the amount of revenue lost by incentive to save. But an increase in the return the government, and whether government on saving also reduces the amount of saving spending is affected by the loss in revenues. If necessary to achieve a given level of future the new private saving does not exceed the consumption, thereby reducing the saving government’s revenue loss, national saving incentive for target savers.3 Empirical studies will not increase and might actually fall, pro- intended to determine which effect predomi- viding that the deficit rises by the amount of nates tend to support the view that net saving the revenue loss. Of course, unlike the response does increase when the after-tax return rises; of private saving to tax incentives, the use of however, this finding is not unanimous and federal revenues (for spending or for deficit nearly all studies find that the effect is quite reduction) is a politically determined issue modest. that depends on the current priorities. The recent experience with IRAs, which Putting aside the political issue concerning grew rapidly when introduced but much less the government’s use of revenues, the impact so after eligibility for a tax deduction was of tax preferences on national saving depends restricted to lower-income households, has on the taxpayers’ marginal tax rate and on the presented researchers with an opportunity fraction of contributions to IRAs and other to study the effect of targeted tax incentives incentives that is new saving (i.e., that does on saving. Before its implementation, most not represent a shift into IRAs of other economists expected that the tax advantages non-tax-deferred saving). If, for example, 50 would generate a sharp rise in IRA balances percent of the contribution to pension funds is

43 new saving generated by the tax incentive Therefore, it is reasonable to conclude that and the relevant effective marginal tax rate is certain existing tax incentives for retirement 30 percent of income, a $100 contribution to saving, as well as employer contributions to the pension fund could cause national saving pension plans, are likely to increase national to rise by $20 because the tax incentive would saving. generate more dollars of new private saving than the government’s revenue loss. DESIGN OF TAX-PREFERRED SAVING Because it is difficult to determine what INCENTIVES proportion of contributions is new saving, it is It is important that future saving incentives be not surprising that the results of empirical carefully designed to maximize their net ben- studies on the net impact of IRAs and other efit. The shifting of funds from existing saving saving incentives on national saving are not to qualified saving and borrowing to finance unanimous. However, it is significant that a contributions should be discouraged as much recent study by Poterba, Venti, and Wise, based as possible. Some have suggested, for example, on the U.S. experience with 401(k) plans, con- that tax-preferred saving incentives could be cluded that these tax-deferred saving incen- made more effective by making deductions tives succeeded in generating substantial new available only for increases in saving or only saving sufficient to significantly increase not for saving above a threshold that rises with only personal saving but also national sav- income.7 The idea is that a tax incentive should ing.5 Moreover, researchers who have found reward greater efforts to save. that tax incentives created little or no new It is also important that the tax incentive be saving, because of the induced shift from designed so that those who have discretion to existing assets, generally agree that national increase saving qualify for the incentive. saving should rise in the long run. For Using data on educational attainment (which example, recent model simulations of long- is correlated with income), one study found term outcomes by Engen, Gale, and Scholz that college graduates and those who already strongly support the finding that over time, have pensions respond positively to saving the impact on national saving will be posi- incentives, whereas those without college tive.6 degrees may reduce saving in response to Thus far, we have discussed the effects of tax incentives. Apparently, lower-income tax incentives, such as IRAs, that are designed individuals are much more likely to behave as to encourage retirement saving by individu- target savers. The study concluded that “a als. However, tax deductions for employer policy that provides tax incentives for saving contributions made from current-year earn- exclusively for lower-income households ings may have a larger positive impact on excludes those households that are most likely national saving. A high proportion of to increase saving in response to this policy; employer contributions to retirement plans are indeed, it is conceivable that such policies likely to be new savings, that exceed the could actually reduce aggregate personal sav- government’s revenue loss, given the usual ing.”8 Of course, if an incentive does not gen- assumption that wages would be higher in erate a significant increase in personal saving, the absence of retirement benefits. Employees there is little hope that national saving will are not likely to change their saving behavior rise. Thus, pension-related tax incentives in response to a change in employer contribu- that decline as income levels rise may be tions because they frequently have very little self-defeating if the goal is to encourage information about those contributions; aggregate pension saving. employees are more concerned about the size In some instances, the motive for tax of the benefit than about how it is financed. changes that have reduced the tax benefits

44 available for high-wage individuals is to com- such as 2 percent of disposable income, for all pensate for the fact that in any given year, workers — in order to make sure that most pension tax policies provide considerably households contribute to national saving and fewer benefits for low-wage workers. Indeed, make preparations for the future. It is argued some have cited the distribution of pension that this approach would be very effective in tax benefits as reason to increase taxes on pen- raising national saving because those who do sion saving.9 However, in a given year, many not save would be forced to save and would of those who do not benefit are younger work- have little opportunity to shift funds from ers who are not yet covered by a pension but existing accounts to qualified accounts. Those who will ultimately be covered and benefit who already save the required minimum 11 from the tax treatment of pensions. Measured would not be affected by the proposal. in terms of lifetime earnings, rather than earn- A mandatory saving program probably ings at a specific time, pension tax benefits would be integrated with Social Security, appear to be more evenly distributed.10 either as a private supplement to Social Secu- We conclude that pension tax policy is rity or as a substitute for the part of Social not a good instrument for redistributing Security that provides benefits above basic income. Proposals for reducing pension tax needs. Either approach would increase fund- benefits for high-wage workers are clearly not ing for retirement, though the former is likely desirable from the viewpoint of increasing to make a greater contribution to saving. Pro- national saving and may not be effective even ponents of mandatory saving plans also argue from the viewpoint of distributional policy. that the fact that individuals would own and Moreover, if society finds the distributional control the funds would make mandatory con- tributions more attractive than mandatory implication of pension policies worrisome, this taxes paid to Social Security. can be fixed by other changes in taxes (e.g., We do not dispute the idea that mandatory rates or exemptions). saving programs could be more effective in On the issue of the design of tax-deferred increasing private saving than other currently incentives, CED concludes that private pen- available options. But policies that force indi- sion participation and saving should be viduals to take certain actions often have great encouraged, but in a manner that is most costs to society relative to policies that merely likely to raise national saving as well. Other encourage certain behavior. For example, a legitimate concerns, such as income distri- requirement for annual contributions could bution, cannot easily be addressed with pri- cause considerable inconvenience for families vate pension policies without defeating a experiencing temporary economic hardships. basic objective of retirement policies: to Moreover, the American public instinctively increase saving. resists any compulsion; therefore, it is doubt- ful that proponents of compulsory saving pro- PROPOSALS FOR COMPULSORY grams could garner widespread political sup- SAVING PROGRAMS port unless large net social benefits are clearly It has frequently been observed that some demonstrable. Thus, although CED believes households are savers and others, including that proposals for mandatory saving pro- some middle-income households, do little or grams merit further study, we do not no saving. Moreover, it appears that the be- endorse the idea at the present time. The havior of these nonsaving households changes nation should consider a mandatory program little, if at all, in response to incentives to save. only if other options for increasing saving, Consequently, some scholars have proposed such as improved availability of pension a mandatory qualified saving program — a information and tax incentives, prove to be requirement for a minimum rate of saving, ineffective.

45 THE BUDGET DEFICIT deficit-reducing tax increases. (See “Tax Expenditures and Pension Tax Preferences,” As we mentioned earlier in this chapter, a below.) We also believe that the government’s number of tax changes were enacted during measure of the subsidy is exaggerated because the last decade that have discouraged pension the present value of future revenue gains is saving. The primary purpose of these tax not taken into account. changes has been to increase government rev- CED has long advocated a change in spend- enues in order to reduce the federal budget ing and tax priorities in order to eliminate the 12 deficit. In seeking additional revenues, policy federal budget deficit.13 We have said that makers have often turned to the federal among major policy alternatives, a reduction government’s list of tax expenditures, which in the budget deficit is the most certain and includes tax preferences for pension saving. effective means of increasing national saving. However, in our view, pension tax expendi- In taking this position, however, CED has tures are not an appropriate target for argued that the purpose of reducing budget

TAX EXPENDITURES AND PENSION TAX PREFERENCES

In its efforts to reduce the budget deficit dur- rather than the annual cash flow. This is be- ing the last decade, Congress often sought to cause the calculation of annual cash flow is raise revenues by reducing tax expenditures in affected by changes in pension funding levels, the budget. Tax expenditures were defined by trends in coverage and benefits, and the aging the Congressional Budget and Impoundment of the population, factors that have recently Control Act of 1974 as “those revenue losses biased estimates upward. It is also important to attributable to provisions of the federal tax realize that currently reported tax expenditures laws which allow special exclusions, exemp- are not a true measure of forgone revenue (the tions or deductions from gross income, or amount of revenue that would be realized if which provide a special credit, a preferential the tax preference were ended) because the tax rate of tax or a deferral of tax liability.” The expenditure calculation does not take into Joint Tax Committee, which has responsibility account any taxpayer response to the increases for estimating the magnitude of tax expendi- in effective tax rates. If the preferential tax tures, reports that the annual revenue loss as treatment of pensions were eliminated, such the result of the preferred tax treatment of responses would surely lower the revenue gain pensions amounted to $56 billion in fiscal implied by the tax expenditure estimate. 1993, making pension contributions the larg- More important than these measurement est tax expenditure in the budget. The Joint issues, however, is the fact that the economic Tax Committee’s estimate of tax expenditures consequences of pension tax expenditures are for retirement plans is the annual aggregate of different from those of other tax expenditures. revenue losses measured on a cash-flow basis. Retirement saving has very positive social In principle, legislation designed to reduce tax effects as well as important benefits for future loopholes may improve economic efficiency if retirees. Moreover, as explained earlier in this the added revenues are used to lower statu- chapter, the preferential tax treatment of pen- tory tax rates. However, it is clear that the sion savings also makes an important con- method of measuring tax expenditures is mis- tribution to national saving, which in turn leading when applied to pension saving. A improves the growth and competitiveness of proper measure would balance taxes paid on the U.S. economy. This cannot be said about future benefits against the immediate revenue many other tax expenditures that encourage loss; that is, it would use a life-cycle approach consumption rather than growth.

46 deficits is to increase national saving, which and increase current budget deficits (e.g., full is needed to boost investment spending and funding of pensions). The short-term net defi- to stimulate more rapid growth in productiv- cit effect of these changes is unknown, but it is ity. Therefore, how the deficit is reduced is probably quite small. More important, the important. Deficit-reduction policies should reforms recommended here will increase give first priority to cuts in real spending. national saving and investment and thereby Although taxing private pension saving would achieve the objectives sought by a reduction reduce the budget deficit, such taxes are not in the budget deficit. an effective method of raising national saving CED has noted elsewhere that any serious because they reduce private saving. In con- program of long-term spending restraint must trast, taxes on consumption reduce the deficit substantially reduce the growth of entitlement without discouraging private saving and are and transfer programs, which have been the therefore a more effective tax tool for increas- major source of budget growth.14 As a practi- ing national saving. cal political matter, it may also be necessary to CED agrees that further reductions in the raise taxes in order to enact further deficit- budget deficit should be a high national pri- reducing measures. However, we believe that ority. We believe that there is room for more the desire to reduce current-year deficits spending cuts in the budget, certainly suffi- should not lead to a political decision to cient to offset any net increase in the budget increase taxes on retirement saving. Such a deficit resulting from reforms of retirement policy would lose sight of the economic ratio- law recommended here. This statement pro- nale for reducing the deficit: to increase poses legislative changes that both reduce cur- national saving in order to improve the growth rent budget deficits (e.g., in Social Security) and competitiveness of the economy.

47 Chapter 5 The Regulatory Tangle: Proposals for Simplification

Government policy toward private retire- pension plans, from 400 in 1925 to more than ment saving has traditionally consisted of two 370,000 by 19742 (when there was a major over- broad features: First, under the Internal Rev- haul of regulation), and the heightened con- enue Code, savings for retirement have been cern about potential abuses of pension tax accorded favorable tax treatment: deferral of preferences. The principal factors that moti- income taxes on contributions and on the in- vated federal involvement were: vestment earnings derived from them. The • Pension promises made by firms to second general feature has been regulation of employees could be (and occasionally were) plan design and conduct under the Internal revoked. Revenue Code and federal labor law.1 • Plans eligible for preferential tax treatment could be established solely for the benefit FAVORABLE TAX TREATMENT of certain key company employees, thereby Since the 1920s, employer and (under vari- raising issues of tax equity. ous arrangements) employee contributions to pensions have been deductible from • Employees could be excluded from partici- current-year income up to certain limits. How- pation through restrictions on age and ever, taxes are merely deferred; taxes on con- length of service combined with timely ter- tributions and invested earnings become pay- minations. able upon distribution of retirement benefits • Pension benefits distributed to key employ- to the employee. The Internal Revenue Code ees were sometimes viewed as excessive in requires compliance with participation, vest- relation to perceived retirement needs. ing, and funding rules in order to qualify a • Plans were often underfunded relative to plan for tax benefits. This favorable tax treat- their future liabilities. ment of income set aside for pensions is a recognition of the importance of retirement • Plans were overfunded in some cases for income to the economic security of retirees the purpose of tax avoidance. and the contribution of retirement saving to CED agrees that some government inter- the national economy as a source of invest- vention in the private pension system is called ment funds. for. However, the pension regulations that have been enacted place excessive limits on FEDERAL REGULATION the ability of individual plans to be tailored to Labor law sets out mandatory requirements individual needs, are much too complex, have for all private pension plans, whether quali- been far too unstable, and seek to achieve too fied or not. These requirements, which cover many goals, some of which are of little or participation, vesting, funding, reporting, and no benefit to society. Indeed, many regula- fiduciary responsibility, came about largely in tions currently on the books conflict with the response to rapid growth in the number of goals of pension policy as CED sees them.

48 We believe that the basic goal of pension benefit plans were already in place by the policy is to ensure the economic security of time the Employee Retirement Income Secu- the elderly by encouraging saving. In our rity Act of 1974 (ERISA) was passed. This evaluation of present pension regulations and extremely complicated legislation was a in our recommendations for reform, CED takes major overhaul of pension law and, together the view that pension regulation must be with numerous amendments, remains the streamlined, simplified, and designed to basic pension law today. (The Appendix pre- achieve six basic objectives: sents the chronology of pension legislation in 1. Provide retirement plans for the largest- greater detail.) ERISA addressed all aspects of possible number of workers. retirement plan design and funding; its pri- mary goal was to broaden pension participa- 2. Ensure that pension plans meet appropri- tion and ensure the deliverability of pension ate fiduciary standards and are funded to promises. fulfill pension promises. To achieve these goals, ERISA established 3. Encourage individual saving by making age and service rules designed to make retire- retirement saving opportunities available ment plans available to a greater number of to all workers through deferred taxes on rank-and-file employees and vesting rules to contributions and earnings and by pro- increase the probability that benefits were viding all workers with realistic informa- actually received. It established fiduciary stan- tion on their retirement saving needs, dards for all plans, minimum-funding rules resources, and options. for defined benefit plans, and insurance to 4. Place simple and reasonable limits on tax protect participants against the loss of pen- preferences received by any individual sions. ERISA required that this insurance (i.e., a single overall limit on eligible con- be funded by employer premiums and cre- tributions indexed to inflation) and a limit ated a new government insurance agency, the on benefits received from a qualified Pension Benefit Guaranty Corporation (PBGC), defined benefit plan.3 to administer it. ERISA also gave workers 5. Preserve retirement saving and pension whose employers did not offer a retirement rights by discouraging preretirement plan a chance to enjoy tax-favored retirement withdrawals and by improving the port- saving by creating IRAs. ERISA was followed ability of pension assets. by legislation in 1978 and 1981 creating new retirement-saving vehicles, including 401(k) 6. Compensate for the rise in life expect- plans, and broadening access to others (per- ancy by encouraging retirement at a later mitting IRAs for all workers). age and by reducing work disincentives At the same time, other provisions of ERISA for the elderly who receive retirement ben- and subsequent legislation have discouraged efits. the growth of retirement savings. ERISA Simplifying pension regulation to achieve placed limits on contributions to all pension these objectives will also result in much lower plans and penalized excessive individual ben- administrative costs than under the current efits from defined benefit plans. Nevertheless, regulations. until 1982, it could be argued that the inherent tension between limiting immediate federal ERISA AND SUBSEQUENT revenue losses and encouraging retirement LEGISLATION saving had resolved itself in favor of the lat- Beginning in the 1930s, Congress moved ter. However, the Tax Equity and Fiscal gradually to regulate pensions. Nondiscrimi- Responsibility Act of 1982 (TEFRA) marked nation rules and limits on funding of defined the beginning of a change in priority. TEFRA

49 and most subsequent important changes to are now so complex that it may be impossible pension regulations were legislated in an for individual pension plans to be in compli- environment of large federal budget deficits ance with all of ERISA’s provisions. and focused primarily on limiting immediate federal revenue losses. Consequently, TEFRA and successive laws passed almost annually REGULATIONS LIMITING PRIVATE during the last decade have concentrated on PENSION CONTRIBUTIONS AND increasing the stringency of contribution and BENEFITS benefit limits and erecting ever more elabo- Regulations enacted in the last decade or rate rules to safeguard against discrimination. so reduced pension saving in several ways, These funding and benefit restrictions are including lowering the funding limits on backed by tax penalties for noncompliance: a defined benefit plans, limiting annual contri- 10 percent excise tax on excess contributions butions and benefits for any individual, and and a 15 percent tax on excess distributions introducing complicated new discrimination (annual benefits in excess of $148,500 in 1994). tests that sharply raised the administrative Some parts of ERISA and subsequent legis- cost of delivering pensions, thereby discour- lation have had beneficial effects. But pension aging their use. We deal here with only the regulation has also had detrimental conse- most pernicious of these regulations. (Esti- quences: mates of the cost of these regulations are included in the last section of this chapter, • It has sharply raised the cost of administer- “Rising Regulatory Costs Discourage Pension ing retirement plans, with a disproportion- Saving,” beginning on page 64.) ate impact on smaller defined benefit plans. There are two broad classes of pensions: This trend has coincided with a steep defined benefit plans and defined contribution increase in the number of terminations of plans. The essential difference is that the obli- defined benefit plans. gation for a defined benefit plan is • It has significantly limited pension contri- expressed in terms of a benefit (the contribu- butions, undermining the ability of some tions being variable), whereas the obligation plans, even those that are currently healthy, for a defined contribution plan is expressed in to make good on their pension promises in terms of contributions to the pension fund the future. This has contributed to the grow- (the benefit being indeterminate). Conse- ing underfunding problem. quently, problems of underfunded pension The complexity of private pension regula- promises (discussed in Chapter 3) arise only tion in the United States cannot be exagger- in the case of defined benefit plans. ated. For example, the nondiscrimination rules, The primary pension plan for larger and which are over 600 pages long, require com- older firms is generally of the defined benefit plex numerical calculations for every employee variety. The design and funding of these plans each year. To ensure compliance with these (see “Estimating Liabilities and Funding for and other provisions of pension regulation, Defined Benefit Plans,” pages 52-53) are quite plan administrators must hire an army of law- complicated, in part because they are heavily yers, actuaries, and accountants, who often regulated and their benefits are partially guar- seem to be the major beneficiaries of the legis- anteed by the PBGC. Many employees prefer lation. As an industry publication noted: this type of plan because the risk and respon- “There are currently penalties for putting too sibility for providing the benefits fall on the much or too little into a plan, taking too much employer and because the amount of the pen- or too little out of a plan, and receiving ben- sion, usually based on years of service multi- efits from a plan too early or too late.”4 Some plied by a percentage of compensation or a legal experts have suggested that regulations flat dollar amount, is known with greater cer-

50 tainty. Employers often prefer defined benefit contained in the Omnibus Budget Reconcilia- plans because they allow greater discretion in tion Act of 1987 (OBRA87). This law limited the timing of contributions than many defined funding of defined benefit plans to 150 per- contribution plans and in the use of deferred cent of plan termination liability (benefits contributions (in lieu of wages) to attract work- accrued to date). Before OBRA87, firms could ers. Managers also hope to reduce the cost of fund up to 100 percent of their projected liabil- pension benefits by pursuing more aggressive ity, which is based on actuarial calculations investments than may be included in the port- of the benefits plan participants will accrue folio of defined contribution plans. However, over their working lives. The new limit estab- the administrative costs of complying with lished by OBRA87 caused delay in funding of regulations pertaining to defined benefit plans many plans until the later years of a worker’s significantly exceed those for defined contri- career. For many firms future pension costs bution plans of comparable size. will rise sharply because backloaded funding Contributions to a defined contribution costs will rise just as the baby-boom genera- plan are made to an individual account estab- tion approaches retirement. lished for each employee. The final benefit Table 6 (see page 56) reproduces a numeri- paid is based on the sum of these contribu- cal example of how the funding patterns for tions plus return on investment and some- individuals of various ages were disrupted by times forfeitures that have been allocated to the new rule. The plan in the example pays a 5 that account. Defined contribution plans are benefit of 1 percent of final-average salary for attractive to employers because they are less each year of service; it assumes a 5.5 percent complicated to administer than defined ben- annual rate of salary increase and a return on efit plans and because funding costs are pre- assets of 8 percent. The third column shows dictable, since contributions are based on a how the plan would be funded prior to predetermined formula. These plans are also OBRA87 using the projected unit credit attractive to some employees because they are method. Contributions rise slowly as a per- generally portable (i.e., they can be carried cent of compensation throughout the worker’s from employer to employer) and because it is career. The present value of contributions dur- easy to keep track of the current value of their ing the first half of the worker’s career is ex- benefit. Moreover, it is usually possible for actly equal to the present value of contribu- employees to receive their benefits in a lump tions during the second half. However, for the sum upon terminating employment with the 25-year-old worker whose pension is funded sponsoring employer. Because they are used exclusively under the OBRA87 rule, contribu- by small firms and individuals, the number of tions as a proportion of salary begin at a much defined contribution plans is rising rapidly. lower level and rise very steeply during the Successive legislation has generated a large worker’s forties and early fifties (see Figure number of types of qualified defined contri- 25, page 56). On a dollar-present-value basis, bution plans, sometimes making the choice quite complex. (See “Defined Contribution 75 percent of this employee’s funding is Plans,” pages 54-55 for a summary of plan pushed back to the second half of his or her types.) The most rapid growth has occurred in working life. For the 35- and 45-year-old 401(k) plans, which are often supplements to employees in this example who had already primary defined benefit retirement plans. received benefit credits when OBRA87 took effect, contributions had to be temporarily halted FUNDING LIMIT FOR DEFINED (for nine and six years, respectively) while BENEFIT PLANS accrued benefits caught up with the amount From a national saving perspective, the that had already been funded under the most serious attack on retirement saving was pre-OBRA87 rules.

51 Recent empirical studies have concluded 48 percent of the plans met or exceeded the that the full-funding limitation introduced in 150 percent limit and were consequently OBRA87 substantially reduced pension con- overfunded according to the new rule; by 1992, tributions.6 Data gleaned from surveys of busi- that percentage had dwindled to 37 percent as nesses sponsoring pension plans support the firms cut back on their contributions to avoid conclusion that the effects of OBRA87 were being overfunded.7 significant. Prior to the enactment of OBRA87, Firms can only ensure the viability of their the Wyatt Company concluded that 40 per- defined benefit plans if they are permitted cent of 664 defined benefit plans it examined flexibility to spread the costs of funding plans would have been affected by the new funding over time in a manner that realistically limit in 1987; that is, under pre-OBRA87 actu- reflects both expected plan liabilities and arial assumptions, these plans could have been the firms’ ability to make contributions. The expected to make contributions amounting to full-funding limit based on termination more than 150 percent of termination liability. liability denies such flexibility and places a A survey they conducted in 1987 revealed that disproportionate burden on firms in the later

ESTIMATING LIABILITIES AND FUNDING FOR DEFINED BENEFIT PLANS

A typical defined benefit plan credits an retirement. This salary figure is then multiplied employee with a unit of benefit for every year by the benefit percentage specified in the plan of participation in the plan, usually expressed formula (1.5 percent in the example) and then as a percentage of compensation, though some- by years of service to get the total annual retire- times as a flat dollar amount. In the former ment benefit. (The expected value of benefits for case, compensation is usually based on career- a group of participants in a plan would also be average, or final-average, earnings for a speci- reduced by the probabilities of preretirement fied number of years (often three to five) termination, disability, or mortality.) immediately prior to retirement. A simple 2. Calculating the value at age 65 of all post- example of the final-average case is an retirement benefit flows. This requires a mortal- employee who works for thirty years until ity assumption to estimate life span and an retirement at age 65, averages $50,000 in salary interest rate assumption to discount the value for his last five years of employment, and par- of benefit payments made over successive ticipates in a plan that credits him with 1.5 years. The number thus derived represents percent of final-average compensation for each the total pension benefit valued at age 65 (or year of service; he would be entitled to a pen- whatever the retirement age). sion of 45 percent of $50,000 ($22,500) for each year that he lives in retirement. In many cases, 3. Finding the present value of the number calcu- the pension benefit is in addition to Social lated in step 2. This calculation depends criti- Security benefits.a cally on the interest rate assumption. Because Because accrual and payment of benefits occur a long period of time elapses between benefit over a period of many years, it is necessary for accrual and retirement, a small change in the the plan sponsor, for purposes of funding, to interest rate causes an inordinately large move- first determine the present value of the pen- ment in the valuation of plan costs and conse- b sion. This involves several steps: quently in the funding status of the plan. 1. Estimating the final-average (or career- The plan sponsor next has to determine how to average) salary, which incorporates assign the cost of funding this liability over time assumptions about rate of salary growth within the constraints of minimum and maximum from commencement in the plan until annual funding requirements set out in the law.

52 years of workers’ careers. CED therefore rec- the basic pension policy objective of encour- ommends that the funding limit be restored aging retirement saving. Moreover, it turns to its pre-1987 level of 100 percent of pro- out that these rules, in combination with dis- jected plan liability. Projected plan liability crimination rules, have a substantial adverse in the case of flat-benefit plans should be effect on retirement saving by middle- and calculated to include anticipated increases lower- income workers. in the dollar benefit level over time. Dollar Limits on Contributions and Ben- efits. Pension plan benefits and contributions REDUCTIONS IN CONTRIBUTION are limited in a number of ways. There are AND BENEFIT LIMITS limits on deductible contributions for both Over the years, Congress has enacted a defined benefit and defined contribution plans, series of regulations intended to raise revenue backed up by an excise tax penalty. There is a by reducing retirement benefits for higher- limit on the benefit that can be received from a income workers, business owners, and key defined benefit plan, and there is an excise tax employees. These changes are in conflict with on aggregate distributions of more than

Allocation of costs to each year is determined funding pensions over time. This can be using one of six actuarial cost methods allowed by achieved by using an actuarial cost method ERISA. The allocation for any given year is called geared to making sufficiently high contribu- the normal cost of the plan for that year. The tions in the earlier part of a worker’s career so minimum-funding requirement stipulates that that contributions do not have to rise steeply in the normal cost of the plan must be funded, in later years. addition to amortization of unfunded liabilities, It should be noted that in 1987, the Financial experience gains or losses (which occur when actu- Accounting Standards Board mandated that arial assumptions such as the interest rate differ firms use the projected unit credit method for from actual plan experience), and gains or losses reporting pension liabilities in financial state- arising from changes in the actuarial assumptions ments. This is one of the less conservative of themselves. the six allowable funding methods because it Each of the actuarial cost methods is tailored for back-loads contributions (as a percent of com- different preferences. Some are designed to fund pensation) into the later years of an employee’s benefits at a fairly stable percentage of payroll, working life. Moreover, 1987 legislation that some front load contributions into the earlier lowered the maximum funding limit to 150 phases of a worker’s tenure, and others do the percent of termination liability effectively de- reverse. layed the funding of many workers’ pensions. Since the work-force will age as the baby Because plans are usually designed so that ben- boomers move toward retirement, any laws efits rise as a worker advances in age, it frequently that backload the funding of pensions will makes sense (particularly for mature firms on a make it particularly burdensome for firms to solid financial footing) to smooth out the cost of finance these plans.C a. Assuming that pension benefits are independent of Social Security benefits, the individual in the example would probably receive a total annual retirement benefit of about $33,000, or two-thirds of preretirement earnings. b. A 1 percent change in the interest rate assumption can alter the long-run cost of benefits for a typical plan by about 25 percent. Dan M. McGill and Donald S. Grubbs, Jr., Fundamentals of Private Pensions, 6th ed. (Philadelphia: Pension Research Council, Wharton School, University of Pennsylvania, 1989), p. 259. c. Many firms apparently also shifted to this method for funding their plans, away from more conservative methods, thus reducing contributions. William G. Gale, “Public Policies and Private Pension Contributions,” Journal of Money, Credit, and Banking, forthcoming.

53 $148,500 to any individual from one or more plan and the maximum annual benefit that plans. There is also a limit on considered com- could be paid out by a qualified defined ben- pensation (the compensation that can be taken efit plan. The Deficit Reduction Act of 1984 into account in a benefit or contribution for- (DEFRA) extended this freeze on contribu- mula). tions and benefits until 1988.8 The Tax Reform In 1982, TEFRA reduced by one-third and Act of 1986 (TRA) effectively extended the then froze until 1986 both the maximum freeze on contributions beyond 1988. It also annual contribution to a defined contribution reduced the contribution limit by individuals

DEFINED CONTRIBUTION PLANS MONEY PURCHASE PENSION PLANS (from after-tax dollars), although most such The important distinguishing characteristic of a plans also provide for full or partial employer money purchase pension plan is that contribu- matching. The employee is typically able to tions must be definitely determinable; that is, they choose from a range of rates of contribution, are made according to a predetermined for- and the employer match is normally a uni- mula that is independent of fluctuations in the form percentage of each employee’s contribu- profits of the sponsoring employer. The contri- tion. bution limit on behalf of each employee is the lesser of $30,000 or 25 percent of employee PROFIT-SHARING PLANS compensation. There are several different types Profit-sharing plans must meet many of the same of money purchase plans: regulatory requirements as money purchase pen- 1. Traditional money purchase plans. The sion plans. The principal difference is that contri- plan sponsor undertakes to make periodic butions are not definitely determinable and in contributions to employee accounts accord- fact need not be tied to profits at all. IRS regula- ing to a formula, usually a percentage of tions require only that contributions be “recur- employee compensation. These plans are ring and substantial.” The limit on tax-deductible frequently “contributory,” meaning that contributions is 15 percent of aggregate employee contributions are made by both employer compensation. There are three principal kinds of and employee. Except for plans established profit-sharing plans: before passage of ERISA, which were able 1. Conventional profit-sharing plans. All to convert to 401(k) plans, employee contri- required contributions are made by the butions are made from after-tax dollars and employer. are not tax deductible. 2. Thrift profit-sharing plans. Employees make 2. Target benefit plans. Contributions on contributions, and these are matched by the behalf of each employee are based on an employer, often using a formula based on actuarial estimate of the contributions corporate profits. required to achieve a specific target benefit upon retirement. 3. Cash or deferred arrangements [CODAs, or 401(k) plans]. Employees can choose to 3. Negotiated contribution plans. The for- receive distributions in the form of cash, or as mula for contributions usually arises tax-deferred contributions by the employer to from collective bargaining and covers the a trust. The earliest CODAs were established employees of more than one firm. Contribu- out of employer contributions that were in tions accumulate in a single fund, rather excess of regular compensation, but over time than individual accounts. the concept came to be applied to regular 4. Savings, or thrift, pension plans. These compensation, whereby employees enter into always involve employee contributions salary-reduction agreements with employers,

54 for one important category of defined contri- tions in profitable years to offset lower con- bution plans, 401(k) plans, from $30,000 to tributions in unprofitable years. TRA also $7,000 (before adjustment for inflation). reduced early retirement benefits from defined Profit-sharing plans were prohibited from benefit plans actuarially in line with the nor- applying the unused portion of their prior-year mal retirement-age benefit limit established contribution limit to exceed the contribution under TEFRA. limit in another year. This reduced the degree As a result of the changes to qualified con- to which they could make higher contribu- tribution and benefit limits, and after subse-

and the amount of the reduction is placed in an adjusted gross income in excess of a tax-deferred account by the employer on the $25,000 ($40,000 for a married couple), the employee’s behalf. The limit on these contri- tax deduction is scaled down, reaching zero butions was $9,240 in 1994. About half of at an income level $10,000 over the thresh- 401(k) plans provide for some amount of old. matching contribution by employers. The 2. Individual retirement annuities. This is an combined limit is the same as for a money annuity purchased from a life insurance purchase pension plan ($30,000). company; as with the IRA, the insurer can- not accept more than $2,000 annually in STOCK BONUS PLANS premiums. 1. Traditional plans. These are similar to profit- 3. Simplified employee pensions (SEPs). An sharing plans, except that the benefits can be employer establishes and finances an IRA made available in the form of stock in the for each eligible employee. The contribution employer’s company. limit is much higher than for an IRA: the 2. Employee stock ownership plans (ESOPs). lesser of $30,000 or 15 percent of employee These are stock bonus plans that are at least 50 compensation. Compared with conventional percent invested in the securities of the spon- pension plans, SEPs are easy and inexpen- soring employer. Most ESOPs are invested sive to establish and are therefore favored by exclusively in employer securities. A leveraged many employers. ESOP is one in which money is borrowed in order to purchase the stock. TAX-DEFERRED ANNUITIES UNDER SECTION 403(B) INDIVIDUAL RETIREMENT PLANS Section 403(b) plans are made available for non- The principal kinds of individual retirement profit organizations and educational institu- plans are individual retirement accounts, individual tions. They are similar to 401(k) plans in that retirement annuities, and simplified employee pen- employees and/or employers may contribute sions. The first two are funded by the worker pretax dollars to an investment pool. There and are distinguishable primarily by the type of are two limits on combined employer and funding instrument. The third is an employer- employee annual contributions: (1) the lesser financed IRA. of 25 percent of taxable compensation or 1. Individual retirement accounts (IRAs). A $30,000, or (2) a cumulative maximum exclusion worker can make tax-deferred contributions allowance limit of 20 percent of taxable com- to an IRA up to an annual limit of $2,000 pensation times years of service minus prior ($2,250 for a single-earner married couple). contributions. There is also a limit on tax- However, if an individual is already covered deferred employee contributions similar to under an employer-sponsored plan and has that of 401(k) plans ($9,500 in 1994).

55 quent indexing of these amounts in the years Figure 25 since, the following annual limits applied in 1994: Projected Lifetime Contributions • Benefits from a defined benefit plan: the as a Percent of Salary for 25-Year-Old lesser of $118,800 or 100 percent of average Participant compensation for the three consecutive Percent years of highest earnings. A plan that pro- 20 vides for benefits in excess of this amount After OBRA87 is disqualified. 15 • Contributions for a defined benefit plan

are subject to a funding limit of 150 percent 10 of termination liability. Excess funding is subject to a 10 percent excise tax. Pre-OBRA87 5 • Contributions to a defined contribution plan: the lesser of $30,000 or 25 percent of 0 the individual’s compensation. Excess con- 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 tributions are subject to a 10 percent excise Age tax.9 • Contributions by individuals to a 401(k): SOURCE: The Wyatt Company. $9,240.

Table 6 Effects of OBRA87 Full-Funding Limits on Contribution Rates Contribution Rates Under Projected Funding Limit of 150% of Accrued Benefit Unit Credit Age When Change Implemented Worker Contribution Age Pay Rate 25% 35% 45% 55%

25 $25,000 4.2% 0.9% 4.2% 4.2% 4.2%

30 32,674 4.7 1.4% 4.7% 4.7% 4.7%

35 42,704 5.3 2.6% 0.0% 5.3% 5.3%

40 55,812 5.9 4.4% 0.0% 5.9% 5.9%

45 72,944 6.7 7.4% 7.4% 0.0% 6.7%

50 95,335 7.5 12.3% 12.3% 0.0% 7.5%

55 124,599 8.4 16.2% 16.2% 16.2% 0.0%

60 162,846 9.5 11.5% 11.5% 11.5% 10.2%

64 201,737 10.4 11.1% 11.1% 11.1% 10.7% SOURCE: The Wyatt Company.

56 • Distributions from defined contribution limit that applies to other defined contribu- plans are subject to the overall limit on tion plans, such that the combined employee benefits from one or several plans: a 15 and employer contributions do not exceed percent excise tax on aggregate distribu- that limit. This will give workers in smaller tions received by an individual from one firms an opportunity to expand their retire- or more plans in excess of $148,500.10 ment savings and will provide a greater The data indicate that although the vast incentive for managers of small firms to majority (86 percent in 1987) of 401(k) plans offer these plans to their workers. are primary plans (i.e., the firm’s sole or prin- The contributions of all workers are also cipal pension plan), aggregate participation in limited by discontinuities in employment and secondary 401(k) plans is substantially higher earnings. As we discussed in Chapter 3, pen- (49 percent of participants versus 26 percent sion plan participation is substantially lower in primary plans).11 This suggests that small among workers who are younger and have firms are much more reliant on 401(k) plans as lower earnings (see Figures 20 and 21, page their principal retirement saving plan. 30). Many of these workers are not offered a The regulations described above tend to pension plan by their employers. Others do have two effects: (1) They reduce pension con- not have sufficient means to begin saving for tributions and saving. (2) Because a substan- their retirement, or they do not give saving for tial impact is felt by managers of businesses, retirement a high priority because they are far who generally make the decision whether or from the end of their working lives. They per- not firms operate a pension plan, the regula- manently lose tax benefits available for pen- tions frequently lead to the abandonment of sion saving during their younger years, and some plans and discourage the formation of compounding of pension contributions and new ones, particularly for small firms. CED interest cannot occur over a long-enough believes that there are legitimate reasons to period of years to build sufficient retirement limit the pension tax preference received by income. CED believes that it would be more any one individual. But the limitations just equitable and a stimulus to saving if the described, together with the limits on consid- retirement-saving tax preference were based ered compensation, are duplicative, discour- on accumulated lifetime income. This would age investment in high-yielding assets, and enable parents, who often must leave the are too damaging to the basic pension policy labor force temporarily to care for children, objective of encouraging retirement saving. and other individuals who undercontribute Therefore, CED urges Congress to eliminate to their pension plans early in their careers the excise tax on distributions above the to exceed the contribution limit in later years. stated limit. Limits on benefits from quali- In practice, this would allow individuals to fied defined benefit plans, which discour- make larger tax-deferred contributions when age retirement saving and encourage invest- they are financially better equipped to do so, ment in lower-yielding assets, should be usually later in their careers.12 increased. Contributions for qualified Reductions in Considered Compensation defined contribution plans should also be Limits. In addition to the dollar limit on con- raised to more reasonable levels (and indexed tributions, pension contributions for both for inflation) in order to encourage small- defined benefit and defined contribution plans business managers to provide pension plans have been reduced by limits on considered for all employees. compensation, which is the maximum amount CED also believes that individual contri- of income that may be taken into account when butions to 401(k) plans that are primary pen- calculating the eligible contributions for sion plans should be subject to the same a defined contribution plan or the benefits

57 under a defined benefit plan. Although the individuals with low incomes but whose pro- intent may have been to limit contributions jected earnings exceed $150,000. The easiest made for highly compensated individuals, way to see this is by examining Figure 26, in practice these limits are expected to have which illustrates how the limit on considered a substantial effect on contributions for compensation could affect the contributions lower-income workers. The Tax Reform Act for employee earnings of less than $20,000. of 1986 reduced the maximum from $235,000 Lowering the cap on considered compensa- to $200,000, and the Omnibus Budget Recon- tion also affects middle- and lower-income ciliation Act of 1993 (OBRA93) reduced it workers participating in defined contribution further, to $150,000.13 The most obvious effect plans, such as 401(k) plans, because it increases of the rule is to reduce the pension contribu- the likelihood of a plan failing to comply with tions and benefits for highly compensated nondiscrimination rules peculiar to these employees.14 However, the lower limit on con- plans. Plan administrators must divide par- sidered compensation can also severely limit ticipating employees into two groups: highly contributions to defined benefit plans for compensated and all others. A plan is consid-

Figure 26 Considered Compensation Limits May Affect Non-Highly Compensated Employees The chart shown illustrates how the $150,000 the chances that employers will be unable or limit on considered compensation may affect unwilling to continue to fund their pension contributions of employees with less than plans. $150,000 of current income but whose pro- jected earnings exceed $150,000. The solid line represents the lifetime projected compensation Effect of Considered Compensation path for the same individual as in Table 6. The Limit at Each Age for Participant in dashed line represents the OBRA93 earnings Defined Benefit Plan path for purposes of calculating his pension Compensation (dollars) benefit. This line is derived by taking $150,000, the maximum allowable, and discounting back to year 1 at 5.5 percent a year (the same interest 200,000 rate as the salary growth assumption). This

gives a compensation figure of $18,589. Thus, 150,000 instead of a benefit based on a percentage of $25,000, this individual’s benefit is funded as a Pre-OBRA93 percentage of $18,589. The vertical distance 100,000 between the two lines at every point represents the amount of reduction in considered com- OBRA93 pensation for each age. An important qualifier 50,000 is that the $150,000 is indexed for inflation, which will allow employers to increase contri- 0 butions over time and leave the ultimate ben- 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 efit payment to the employee unaffected. Age However, as with the OBRA87 full-funding rule, delay of funding in the early years in- NOTE: Assumes individual enters service at age 25 earning $25,000 a year and expects annual salary growth of 5.5%. creases the burden in later years and increases

58 ered nondiscriminatory only if elective con- CED also believes that the nondiscrimi- tributions for the highly paid group do not nation tests under Sections 401(k) and 401(m) exceed those for the lower-paid group by an are unnecessarily complicated and reduce amount determined by two alternative for- pension saving among middle-income earn- mulas.15 For high-wage workers already con- ers as well as among more highly compen- tributing the maximum allowable to their sated employees. These tests could be elimi- 401(k) plans ($9,240 in 1994), reducing the con- nated and replaced by a vastly more simple sidered compensation limit automatically and administratively inexpensive safeguard increases the ratio of contributions to com- against discrimination. For example, employ- pensation and thus increases the likelihood ers who make contributions to 401(k) that plans will fail the nondiscrimination test. accounts could be required to contribute the Rebalancing contributions among high-wage same percentage of payroll for all partici- employees in order to correct this problem is pants up to the maximum contribution limit.16 expected to result in reduced contributions for middle-income employees unfortunate enough to be just above the cut between the REGULATIONS TO PREVENT highly compensated and nonhighly compen- DISCRIMINATION IN PENSION sated groups. PLANS The primary purpose of the pension limi- tations in OBRA93 was to raise federal rev- The limit on considered compensation and enue. Advocates of considered compensation limits on maximum benefits and contributions limits believed that such limits would prevent directly reduce pension saving. There are also discrimination in favor of key employees. But nondiscrimination rules that are extremely clearly this objective will be undermined as costly to administer and have more subtle some plans are abandoned, as some plan for- effects on pension saving. Generally, nondis- mulas are amended to negate the effects of the crimination rules provide for some rebalancing provision, and as lost benefits are recouped of pension benefits and contributions between by other forms of payment. CED urges Con- high- and low-salary individuals. These rules gress to discontinue the recent practice of are often expressed as percentages by which reducing considered compensation as a the contributions and benefits of high-salary means of raising federal revenue. We workers cannot exceed those of low-salary believe this practice is harmful because it workers. Falling into this category are the dis- can affect pension saving by reducing con- crimination tests for 401(k) plans, Social Secu- tributions for lower-income employees. It rity integration rules, coverage rules, and also reduces the willingness and ability of top-heavy rules. These rules are extremely sponsors to create and maintain qualified complicated, often redundant, and impose plans. enormous costs on pension administration. Assuming the elimination of the excise CED believes that these rules should be tax on distributions and increased limits on streamlined and simplified in a manner that benefits as recommended by CED, we reduces costs. Ideally, all discrimination tests believe that some limits on considered com- would be replaced by a vastly more simple pensation should be retained. However, the and less costly safeguard against discrimina- considered compensation limit should be tion. For example, all employees who meet raised to a more reasonable level, at least to nondiscriminatory age and service eligibil- its pre-OBRA93 level (and indexed); it should ity requirements should be covered and apply only to those whose incomes actually receive the same ratio of contributions to equal or exceed that limit, not to those whose wages up to the prescribed limit. (In the case projected incomes equal or exceed it. of defined benefit plans, the same benefit

59 formula should be applied.) If such a radical mately the same percentage of compensation reform could not be adopted in its entirety, for all employees. simplification should proceed along the lines The Tax Reform Act of 1986 introduced described in this section. new rules that restricted the degree to which employers could integrate plans with Social TOP-HEAVY RULES Security. Essentially, it stipulated percentages Top-heavy rules, a product of TEFRA, are a by which benefits and contributions with re- special set of particularly stringent nondis- spect to compensation above the Social Secu- crimination standards that must be met by rity taxable wage base could not exceed those plans in which 60 percent or more of accumu- with respect to compensation below the Social lated benefits accrue to key employees (cer- Security taxable base (see the Appendix for tain officers and highly paid employees). They details). Because of these limits, many firms include minimum benefits for nonkey employ- have had to restrict contributions to highly ees in defined benefit plans, minimum contri- compensated employees.17 CED believes that butions for employees in defined contribution restrictions on the integration of Social plans, a complete prohibition against Social Security and private pensions are not needed Security integration, and a faster vesting sched- in cases where the employer contribution for ule (see the Appendix). The rules are costly to all employees equals or exceeds an appropri- administer because they involve annual test- ate minimum threshold. ing of all plans. They also reduce pension ben- efits for more highly compensated employ- COVERAGE AND VESTING RULES ees. Consequently, many top-heavy plans The Tax Reform Act of 1986 also intro- (primarily smaller ones) were abandoned af- duced highly complicated nondiscrimination ter passage of the regulations. tests relating to coverage and vesting in The Tax Reform Act of 1986 subsequently defined benefit and defined contribution strengthened nondiscrimination rules for all plans.18 These rules are in addition to the strin- pension plans by enacting more stringent cov- gent participation and vesting rules for indi- erage and participation rules and requiring vidual employees.19 Provided they are not too more rapid vesting of benefits. In light of these complex, rules that broaden coverage and requirements, the special rules for top-heavy plans serve little useful purpose. CED believes accelerate vesting are generally desirable be- that the other nondiscrimination rules ap- cause they encourage the growth of pension plying to all pension plans are adequate and saving. At present, however, these rules are that it is desirable therefore to completely excessively complex. eliminate top-heavy rules. There are two primary effects of the cover- age rules: (1) They make pensions complex SOCIAL SECURITY INTEGRATION and costly to administer because of the nature Because contributions and benefits under of the calculations involved. (2) They tend to OASDI are greater for lower-salary workers, redistribute benefits from high- to low-salary it has been a common practice in private pen- employees rather than simply increasing ben- sion plans to compensate for this distribu- efits for the low-salaried (and therefore total tional effect by weighting pension contribu- pension saving). tions and benefits toward the higher-salaried CED believes the current coverage rules employees. The desired outcome was that are not cost-effective and should be simpli- either the combined contributions or the com- fied into a single test. For example, the rule bined benefits accruing from Social Security could state that all employees who meet non- and the private pension would equal approxi- discriminatory eligibility requirements must

60 be covered. Moreover, employers who pro- PORTABILITY OF PENSIONS vide a uniform ratio of contributions to wages Previously enacted reforms involving vest- for all eligible employees should be given ing in retirement plans and proposals for fur- safe harbor from all further discrimination ther improvements in pension coverage will tests. help preserve pension funds for retirement. CED also recommends that minimum A needed improvement in pensions relating vesting requirements be revised for recogni- to preservation of funds for retirement involves tion of the greater mobility of today’s labor the portability of defined benefit plans and force. For example, vesting could be conferred the rollover of pension distributions when on one of the following two schedules: (1) a workers change jobs. CED recommends that three-year cliff (down from five) or (2) regulators investigate options for improving five-year graded (down from seven). the portability of vested benefits from defined benefit plans. Individuals who change employers should be strongly encour- PRESERVING PENSION FUNDS FOR aged to roll over preretirement lump-sum RETIREMENT distributions into alternative retirement sav- ing instruments such as individual retire- PRERETIREMENT LUMP-SUM ment accounts (IRAs) and defined contribu- WITHDRAWALS AND BORROWING tion plans maintained by their new FROM PENSIONS employers. As we noted in Chapter 2, many observers are greatly concerned that lump-sum preretirement withdrawals are often not be- PENSION BENEFIT GUARANTY ing rolled over when participants change jobs. CORPORATION There is also concern about borrowing from According to the PBGC’s 1993 annual pension funds. It is feared that both these prac- report, its liabilities for payment of benefits to tices place the economic security of future participants of terminated plans exceeded its retirees in jeopardy. It is also argued that they assets by $2.9 billion. The six-year period that are an abuse of tax preferences provided to ended in 1993 saw an alarming trend: encourage pension saving. Other observers Although the number of plan terminations fear that prohibition of these practices will fell sharply compared with those in the pre- discourage participation and pension saving, ceding six years, the net claims of these plans especially when applied to supplementary sav- on the PBGC (plan liability minus the sum of ing plans that typically involve voluntary em- plan assets and recoveries from employers) ployee contributions. However, CED believes rose from $1.7 billion (between 1982 and 1987) that preretirement withdrawals and borrow- to $2.4 billion (between 1988 and 1993). ing from pensions pose an even greater threat Unfortunately, there is a strong likelihood to pension saving and conflict with the basic that the $2.9 billion is just the tip of the ice- objectives of pension policy. berg. The PBGC has developed three alterna- CED recommends that in-service pre- tive forecasts of future losses. The most opti- retirement withdrawals and borrowing of mistic of these forecasts projects a deficit of employer contributions to pension plans be $1.9 billion (in 1993 dollars) by the end of prohibited. Preretirement withdrawals and 2003, but it assumes a continuation of the same borrowing of voluntary employee contribu- average annual net losses the agency has tions to pension plans should not be prohib- incurred over its entire lifetime. On the other ited, but existing penalties should be hand, if companies in a precarious financial retained. state, whose plans are classified by the PBGC

61 as “reasonably possible losses,” actually ter- tion, were supposed to fund the PBGC’s minate their plans, the PBGC expects a deficit liabilities. The 1994 legislation phases out the of $13.8 billion by 2003.20 In this scenario, the cap on the variable part of the premium over possibility of a taxpayer bailout looms omi- three years. This reform still does not permit nously. premiums to be set in a way that satisfies Although the true financial condition of insurance market principles. the PBGC is reflected in a comparison of its In a private insurance market, premiums assets and the present value of its liabilities, would be set for each plan according to two the relevant comparison for federal budgeting criteria: is between current-year income and expenses; 1. The statistical probability of plan failure, in this case, the PBGC now has a surplus. which in turn is dependent on the firm’s Thus, the PBGC actually contributes to reduc- financial condition and future prospects. ing the budget deficit, which in turn reduces the incentive for Congress to focus on the 2. The amount of the insurer’s exposure in underlying long-term problem. However, a the event of plan termination. This is rep- change in the way that the PBGC’s activities resented by the unfunded actuarial liabil- are accounted for in the federal budget ity minus employer liability payments.22 from an operating-income basis to a change- Although phasing out the variable pre- in-net-liability basis would help to get the mium cap is a move toward risk-based premi- attention of Congress because it would then ums, the current PBGC premium structure is be adding to the deficit. still deficient in two respects: There are three underlying causes of the PBGC’s dire financial straits: (1) Premium lev- 1. It does not reflect the likelihood of plan els are not set to cover future claims failure. A number of the companies whose adequately. (2) Premiums are structured in a pension plans are most underfunded are way that creates an adverse-selection prob- nevertheless in sound financial condition lem in the insurance pool. (3) Companies that and expose the PBGC to relatively little underfund their plans have insufficient incen- risk. tive to increase funding, because of weak 2. Although the variable part of the premium PBGC compliance authority. Moreover, mini- is now more closely related to underfund- mum- funding requirements under current law ing, it does not distinguish between plans allow companies to amortize certain kinds of on the basis of asset risk.23 unfunded liabilities, including those that arise because of an increase in promised benefits, Because of these shortcomings in the pre- over a long period of time.21 mium structure, some underfunded plans are, in effect, being subsidized by healthy ones. THE PREMIUM STRUCTURE Consequently, the companies that sponsor Until passage of the Retirement Protection weaker plans have an incentive to stay in the Act of 1994 (incorporated in the GATT- system, and those that sponsor stronger ones enabling legislation; see the Appendix), plans have an incentive to terminate them. More- paid $19 a year per participant, plus $9 for over, there is a clear danger that raising pre- each $1,000 of unfunded vested benefits. There miums to a level that would cover the PBGC was an overall cap of $72 per plan participant. for its expected losses would further exacer- The sum of these premiums and the income bate the adverse selection problem by causing earned from investing them, plus recoveries strong firms to terminate their plans and with- from employers in the event of plan termina- draw from the insurance pool.24

62 MORAL HAZARD Moreover, the PBGC now has claim to 100 The PBGC does not have sufficient author- percent of unfunded liabilities from plan spon- ity to enforce responsible behavior by pension sors. Although these changes have amelio- plan sponsors whose actions expose the agency rated the problem of moral hazard, they have to loss. Thus, firms in a precarious financial by no means eliminated it. Bankrupt compa- position have an incentive to garner additional nies frequently do not have sizable assets for resources by reducing pension contributions. the PBGC to seize; and in any case, the PBGC They are also likely to embark on riskier claim is not necessarily given priority above investment strategies. Moreover, such firms those of other creditors. The data indicate that have an incentive to offer pension benefits liability payments from employers to the PBGC that are insured by the PBGC and whose fund- have actually dropped slightly as a percent- ing is deferrable in lieu of wage compensa- age of uncovered benefit liabilities since the tion. Workers accept such arrangements new rules were passed. because the PBGC protects them from loss. It CED recommends the following reforms: is for these reasons that unfunded plan liabili- • The PBGC insurance premium should be ties tend to rise sharply in the period just prior restructured so that it more closely to plan termination (usually when the firm is resembles what would be offered in a bankrupt), thus increasing the PBGC’s expo- private market. Such a redesign should sure. In the words of the PBGC’s own annual include a stronger linkage between the report: premium level and the actual risk posed The insurance program’s weakness cre- by a pension plan to the PBGC. For ex- ates incentives for financially distressed companies to take actions that further in- ample, the premium calculation could take crease the program’s exposure to loss. Such into account the financial strength of the actions include: plan sponsor, the marketability of plan • using pension increases as a form of com- assets, and the proportion of assets tied pensation, the costs of which can be up in the firm’s own equities. deferred. Workers are more willing to • The PBGC should be given more power agree to these promises because they are backed by federal pension insurance. . . . to influence the behavior of sponsors of underfunded plans, including the author- • forgoing required pension contri- ity to prevent plans operating below a butions while in bankruptcy with judges’ approval. . . . particular funded ratio from granting ben- 26 • allowing a pension plan to run out of efit increases to employees. money without violating the minimum • The amortization schedules for unfunded funding standards. . . . liabilities should be simplified by reduc- Lenders rarely put pressure on troubled ing the number of categories, and the companies to fund their plans, believing in amortization periods should be shortened the optimistic funding assumptions and to accelerate the funding of liabilities. expecting the PBGC’s pension claims will have • The PBGC’s benefit guaranty structure no priority. On the contrary, creditors are more should be revised to correlate more closely likely to pressure distressed companies to ter- with the minimum funding requirements minate plans rather than fund them.25 attributable to specific benefits, such as It is no longer possible, as it was prior to shutdown benefits. 1986, to terminate an underfunded pension • The status of the PBGC’s claims in bank- plan and pass the liability on to the PBGC ruptcy should be reviewed and enhanced without demonstrating financial distress. as appropriate.

63 RISING REGULATORY COSTS defined benefit plans (covering 15 and 10,000 DISCOURAGE PENSION SAVING persons, respectively) increased at an average The number of defined benefit plans annual rate of more than 10 percent between peaked at just over 175,000 in 1983 and subse- 1982 and 1991, with costs for intermediate plan quently declined by 24 percent in the 1984– sizes increasing at a slightly lower rate. Small 1989 period. Of the net 43,000 defined benefit plans were particularly burdened by one-time plans lost during that period, about 90 per- expenditures arising from frequent changes cent were plans servicing fewer than 100 work- in regulations (Figure 27). For the smallest ers. The Employee Benefit Research Institute plans, administrative costs per covered employee reports that 60 percent of the net terminations increased from $162 to $455 (in 1990 dollars), between 1985 and 1989 had fewer than 10 compared with an increase from $19 to $54 for participants.27 There is no doubt that rising large plans.29 Equally telling is that by 1991, administrative costs, which are particularly small-plan administrative costs were equal to burdensome for small employers, contributed about one-third of the benefits accrued in that significantly to the decimation of defined bene- year. In other words, for every $3 of retire- fit plans in the 1980s. ment benefits accruing to participants, an A study by Hay/Huggins Company for extra $1 was being spent on administrative the PBGC28 analyzed the impact of eleven costs. For a large plan with 10,000 employees, major regulatory changes, enacted between administrative costs were estimated at just 4 1982 and 1989, on the cost of administering percent of benefits. defined benefit and defined contribution plans According to the Hay/Huggins study, of varying sizes. It found that real administra- defined contribution plans fared slightly bet- tive costs for both the smallest and the largest ter under the constantly changing regulatory

Figure 27 One-Time Administrative Costs per Worker in Defined Benefit Plans, by Plan Size, 1983 to 1990 (in 1990 dollars) Dollars 400

15 employees

75 employees 300 500 employees

200

100

0 1983 1984 1985 1986 1987 1988 1989 1990 Year

SOURCE: Hay/Huggins Company, Inc.

64 regime; the increase in real administrative Reconciliation Act of 1987 (OBRA87). (See the costs for these plans between 1982 and 1989 Appendix for details.) was in the 6 to 7 percent range, depending on Although each regulatory change can be plan size. More important, ongoing costs for examined in isolation for the costs it imposed the most severely affected defined contribu- on pension plans, it is important to recognize tion plans were still only half as large as those that the sheer regularity of the changes has for small defined benefit plans. The results of made it difficult (if not impossible) for smaller direct interviews with fifty pension service firms to stay in compliance with the law and providers, conducted for the Small Business to have any assurance of future stability. It is Administration in 1989, indicated similar also important to note that there have been a trends. Annual administrative costs for a small number of regulatory changes since 1989 (the defined benefit plan were 50 percent higher last year studied), which continue to impose a than for a defined contribution plan of the heavy administrative cost burden on pension same size and 270 percent higher on a plans.31 per-participant basis than for a larger defined IMPACT OF COST INCREASES benefit plan.30 ON PLAN TERMINATIONS AND FORMATIONS SOURCES OF COST INCREASES The costs of maintaining defined benefit According to the Hay/Huggins study, con- plans have become too high for many em- sulting fees for actuarial and legal services ployers, a fact that has contributed to their made up the bulk of the cost increases for decision to terminate plans. In a survey con- small plans, whereas consulting fees and ducted in 1992, 50 percent of the respondents PBGC premiums were of equal importance in stated that they terminated their defined ben- large plans. The consulting fees were paid to efit plans because federal regulations were actuaries and lawyers who were needed “too costly or burdensome.”32 The data indi- repeatedly by plan sponsors to explain regu- cate that a large number of employers who latory changes, redraw plan documents, test terminated defined benefit plans did not plans for discrimination, and calculate maxi- replace them with successor pension plans of mum and minimum contribution levels. These any kind, and most of those who did so opted activities were accompanied by increases in for defined contribution plans.33 in-house administrative costs for such pur- These trends indicate that it is critical that poses as collecting and preparing data and policy makers simplify and stabilize regula- communicating plan changes to participants. tions pertaining to defined contribution The most costly legislative changes were iden- plans in order to reduce compliance costs. tified as: Tax Equity and Fiscal Responsibility The retirement security of the 40 million par- Act of 1982 (TEFRA), Retirement Equity Act ticipants of defined benefit plans may depend of 1984 (REA), Tax Reform Act of 1986 (TRA), to a significant degree on policy makers’ Single Employer Pension Plan Amendments actions that influence plan administrative Act of 1986 (SEPPAA), and Omnibus Budget costs.34

65 APPENDIX A Brief History of Government Regulation of Private Pensions

Legislation governing private pension maladministration of plans more easily plans was only sporadic until passage of the detectable. Employee Retirement Income Security Act 5. Self-Employed Individual Retirement Act of (ERISA) in 1974. Since then, however, pension 1962 (Keogh Act). Expanded tax-favored law has been subject to incessant revision by retirement plans to include unincorporated Congress. small business and the self-employed. What follows is only a thumbnail sketch of the evolution of private pension regulation. However, it is sufficient to demonstrate how Employee Retirement Income Security Act obsessively Congress has tinkered with the of 1974 (ERISA) law and how complicated that law has 1. Concerns addressed by legislation: become. a. Plans were exclusionary, with limits on Before ERISA participation and vesting of benefits. 1. Revenue Act of 1921. Deferred tax (until ben- b. Some plans were inadequately funded efit disbursement) on contributions to, and to meet their obligations. income from, stock bonus or profit-sharing c. There were insufficient incentives for plans established for employees. The Rev- some employers to offer pension plans. enue Act of 1926 extended the concession to d. Some plans were not administered other pension plans. according to acceptable fiduciary standards. 2. Revenue Act of 1938. Denied tax-exempt sta- e. Pension benefits for key employees were tus to any pension plan that revoked pen- sometimes viewed as excessive in relation sion promises to its participants. This leg- to perceived retirement needs. islation responded to the problem of plans being terminated so that the funds could f. Pensions were not always protected dur- be used for purposes other than meeting ing takeovers or other kinds of company pension obligations. restructuring. 3. Revenue Act of 1942. Responded to concerns 2. Major provisions: about discrimination against rank-and-file a. Required more information to be provided employees by establishing participation by employers to plan participants: an eas- tests. ily intelligible plan description, subsequent 4. Welfare and Pension Plans Disclosure Act of plan modifications, an annual financial 1958. Required pension plan documents report, and a statement of the participant’s and annual reports to be submitted to the accrued benefits upon request. This Secretary of Labor and to plan participants. strengthened the provisions of the 1958 dis- The objective was to make fraud and other closure act, which was formally repealed.

66 b. Strengthened participation rules. Required j. Established the Pension Benefit Guaranty that employees 25-years-old and over with Corporation. Purchase of insurance was one year of service could not be excluded made mandatory for most plans receiving from a plan.1 tax concessions. The PBGC guaranteed pay- c. Established vesting rules. Employers ment of vested benefits to a certain level. could choose one of three alternative for- The premium was set at $1 per plan par- mulas: ticipant. i. Full vesting after ten years with no vesting until then. Revenue Act of 1978 ii. Graded vesting, achieving 100 per- 1. Contained incentives to encourage smaller cent after fifteen years of service.2 firms to set up pension plans for their employees. iii. The “rule of 45”: at least 50 percent vesting when the employee’s age and 2. Major provisions: years of service add to 45, increasing by a. Established cash or deferred arrangements 10 percent each succeeding year until (CODAs) by adding Section 401(k) to the full vesting is attained.3 Internal Revenue Code.6 d. Required that a joint and survivor annuity b. Established simplified employee pensions be provided to an employee retiring at the (SEPs), in which the employer sets up and normal age unless the employee specifi- finances IRAs for eligible employees. How- cally waived that right. ever, the maximum contribution limit, 7 e. Set minimum funding rules, by requiring $7,500, was higher than for IRAs. that the normal cost of a pension plan be c. Created tax-credit ESOPs, or TRASOPs, a funded currently.4 Past service costs were form of employee stock ownership plan to be amortized over thirty or forty years. whereby an employer could receive a tax f. Set fiduciary standards. Plan assets had to credit equal to contributions. be invested prudently and for the sole ben- efit of plan participants. Relevant financial Multiemployer Pension Plan Amendments and participation data must be provided Act of 1980 (MEPPAA) periodically to the government. 1. This legislation was designed to address g. Allowed a person not covered by a pen- underfunding of multiemployer pension sion plan to establish an individual retire- plans. ment account. IRA contributions up to the 2. Major provisions: lesser of $1,500 or 15 percent of earned income would be tax deductible. a. Reduced the incentive for individual employers to withdraw from multi- h. Limited contributions to profit-sharing and employer plans by obliging them to con- money purchase plans5 to the lesser of 25 tinue funding the liability of workers they percent of annual compensation or $25,000. had hired in the past. i. Limited annual pension benefits that could b. Increased PBGC premiums for multi- be paid to highly compensated employees employer plans. to the lesser of $75,000 or 100 percent of average compensation for the three years c. Required faster funding of unfunded of highest career earnings. liabilities.

67 Economic Recovery Tax Act of 1981 (ERTA) d. Introduced top-heavy rules. A top-heavy plan was defined as one in which 60 per- 1. Contained further incentives to increase cent of accumulated benefits had accrued pension saving: to key employees (officers and highly com- a. Replaced TRASOPs with payroll-based pensated employees). Top-heavy plans had ESOPs (PAYSOPS), whereby an employer to comply with special standards for vest- received a tax credit equal to a percentage ing, contributions and benefits, and Social of payroll. Security integration.11 b. Commercial lenders facilitating leveraged Deficit Reduction Act of 1984 (DEFRA) ESOPs8 were allowed to deduct a portion of their interest income on these loans. This 1. Major provision: effectively lowered the cost of setting up a. Delayed indexing of contribution and ben- an ESOP. efit limits until 1988. c. Permitted IRAs for all workers, and raised the contribution limit from $1,500 to $2,000. Retirement Equity Act of 1984 (REA) d. Increased limit on contributions to SEPs 1. Major provisions: from $7,500 to $15,000 per participant. a. Strengthened participation rules by low- ering the minimum age a firm can require for enrollment in a plan from 25 to 21; the Tax Equity and Fiscal Responsibility Act law also lowered the minimum age for vest- of 1982 (TEFRA) ing service from 22 to 18. 1. This legislation signaled a shift in pension b. An employee could now have a break in policy away from concerns about pension service of up to five consecutive years or security and the adequacy of retirement the period of eligibility or vesting service income toward measures to reduce tax rev- accumulated prior to the break without los- enue loss. ing that eligibility or vesting service. Ma- ternity and paternity leaves were to be 2. Major provisions: treated as though the employee was still at a. Reduced contribution limits for defined con- work through the period of absence. tribution plans to the lesser of 25 percent of c. Greater survivor protection: compensation or $30,000.9 i. Preretirement death benefit was b. Reduced maximum annual pensions from extended to all vested employees. defined benefit plans from $136,425, to the lesser of 100 percent of average cash com- ii. Written spousal consent was required pensation in the three years of highest earn- to exclude death benefits in order to ings or $90,000. obtain a more generous pension. iii. On some domestic relations orders, c. For defined contribution plans integrated private pensions could be divided upon with Social Security, contributions based on divorce. that part of income above the Social Secu- rity taxable wage base (the so-called inte- gration level or breakpoint) could not exceed Single Employer Pension Plan Amendments contributions in respect of income below Act of 1986 (SEPPAA) (the base contribution percentage) by more 1. Congress enacted SEPPAA largely because than 5.4 percent.10 of the moral hazard implicit in the pension

68 insurance system, whereby under existing f. Limited the amount of compensation that law, it was possible for a firm to terminate could be considered in contribution or ben- a pension plan and shift the unfunded efit calculations to $200,000, the same as liabilities onto the PBGC. for top-heavy plans. 2. Major provisions: g. Temporarily capped contributions to de- fined contribution plans at $30,000. a. Raised PBGC premium. h. Reduced the limit on employee contri- b. Limited the circumstances under which butions to 401(k) plans from $30,000 to a voluntary plan termination could occur: $7,000. i. Standard termination: A voluntary ter- i. Changed rules for integration of plans mination in which liabilities were cov- with Social Security: ered by assets. i. For a defined contribution plan, con- ii. Distressed termination: Permitted at tributions in respect of compensation the discretion of the PBGC. Required above the integration level could not ex- the plan administrator to show that the ceed the base contribution percentage firm was financially unable to continue by more than the lesser of the base con- 12 the plan. tribution percentage or 5.7 percent. ii. For a defined benefit excess plan,13 the Tax Reform Act of 1986 (TRA) excess benefit percentage could not ex- 1. Major provisions: ceed the base percentage by more than a. the lesser of the base benefit percentage The minimum vesting requirement was 14 now defined by the following two options: or 0.75 percent. iii. For a defined benefit offset plan,15 the i. Five-year cliff. maximum offset could not exceed the ii. Graded, under which participants are lesser of 50 percent of the benefit ac- 20 percent vested after three years, with crued without regard to the offset or an additional 20 percent each subsequent 0.75 percent of final-average compensa- year until full vesting is attained after tion multiplied by years of service. seven years. j. New coverage rules were introduced. b. Where plan vesting was 100 percent upon One of three tests now had to be satisfied: enrollment, an employer could now require i. Percentage test: The plan must cover only two years of service before enrolling at least 70 percent of all non–highly an employee. compensated employees.16 c. Instituted a 10 percent tax penalty on ii. Ratio test: The percentage of non– distributions made prior to age 59 1/2. highly compensated employees covered d. Established a 10 percent excise tax on under a plan must be at least 70 percent excess pension assets that reverted to the of the percentage of highly compensated employer upon termination of a pension employees covered. plan. iii. Average benefits percentage test: e. For participants in employer-sponsored Both of the following must be satisfied pension plans whose adjusted gross income to pass this test: was greater than $25,000 ($40,000 for a mar- (a) The plan must cover a non– ried couple filing jointly), pretax IRA con- discriminatory classification of em- tributions were phased out. ployees.17

69 (b) The ratio of employer-provided and losses was reduced from fifteen to five benefits or contributions to the years. participant’s compensation for non– d. Introduced a new minimum contribution highly compensated employees must standard for underfunded defined benefit be at least 70 percent that of highly plans with more than 100 participants. The compensated employees. minimum contribution may henceforth k. A new nondiscrimination rule was intro- include a “deficit reduction contribution” duced for 401(k) plans. The average ratio that would effectively speed up the fund- of contributions to compensation of the ing of underfunded plans.18 highly compensated group cannot exceed: e. The full-funding limitation for defined ben- i. 125 percent of the ratio of the efit plans was capped at 150 percent of rank-and-file group if the ratio for the termination liabilities. As a consequence of latter group is 8 percent or more. rising asset values, many firms were unable to make further deductible contri- ii. 200 percent of the ratio of the butions. rank-and-file group if the ratio for the latter is 2 percent or less. Technical and Miscellaneous Revenue Act iii. 2 percent in all other cases. of 1988 (TAMRA) l. Profit-sharing plans could no longer ap- 1. Major provision: ply the unused portion of their prior-year contribution limit (15 percent of the cash a. Increased the excise tax on employer compensation of plan participants) to ex- reversion of assets from 10 percent to 15 ceed their contribution limit in another year. percent.

Omnibus Budget Reconciliation Act of 1986 Omnibus Budget Reconciliation Act of 1989 1. Major provisions: (OBRA89) a. Benefit accruals could no longer be fro- 1. Major provision: zen beyond normal retirement age. a. Defined benefit plan valuations would b. Employees hired after age 60 could no now be required annually instead of trien- longer be excluded from participation. nially.

Omnibus Budget Reconciliation Act of 1987 Omnibus Budget Reconciliation Act of 1990 (OBRA87) (OBRA90) 1. Major provisions: 1. Major provision: a. Increased the PBGC premium from $8.50 a. Enabled employers to transfer excess pen- to $16 per participant, plus an additional sion assets tax-free to an account for the premium of $6 per $1,000 of unfunded current health benefit expenses of retirees. liability (although the premium was capped Otherwise, the excise tax on asset rever- at $50 per participant). sions increased to 20 percent or 50 percent unless the employer transferred a portion b. Introduced quarterly contribution require- of the assets to a replacement plan or ments. increased benefits under the terminating c. The period for amortizing experience gains plan.

70 Omnibus Budget Reconciliation Act of 1993 ii. Provided for accelerated funding of (OBRA93) unpredictable contingent event benefits.19 1. Major provision: iii. Lowered the maximum interest rate a. Reduced considered compensation from that could be used for calculating plan $235,840 to $150,000. liability to 105 percent of the weighted average of thirty-year Treasury securi- ties, for the four most recent years prior Retirement Protection Act of 1994 (RPA, to the plan year.20 incorporated in Uruguay Round Agreements Act) iv. Mandated use of the 1983 Group An- nuity Mortality Table for determination 1. The Retirement Protection Act accelerated of current plan liability.21 funding of the liabilities of underfunded pension plans and contained other mea- c. Prohibited certain underfunded plans sures that strengthened the position of the from changing actuarial assumptions for PBGC. The net effect of RPA on govern- determining current liability without first ment revenues is positive; consequently, gaining approval of the Secretary of the it was incorporated as part of the financing Treasury. package for the General Agreement on d. Prohibited an employer in bankruptcy Tariffs and Trade (GATT). from amending an underfunded plan to 2. Major provisions: increase benefits unless the increase a. Delayed indexation of dollar limits on became effective after the planned date of contributions and benefits by stipulating that the firm’s reorganization. inflation adjustments be implemented in e. Authorized the PBGC to obtain certain round dollar amounts, rather than annu- information from sponsors of underfunded ally in step with the precise rate of infla- plans that would assist the PBGC in deter- tion. Indexation will now occur in the fol- mining plan assets and liabilities. lowing increments: f. Gave the PBGC authority to sue plan i. Benefit limit for a defined benefit plan: sponsors to enforce minimum-funding $5,000. requirements where the amount of the ii. Contribution limit for a defined con- deficient contributions exceeds $1 million. tribution plan: $5,000. g. Phased out, over three years, the cap on iii. Limit on elective deferrals under a the variable part of the PBGC premium. 401(k) plan: $500. h. Required employers who pay the vari- b. Strengthened minimum funding standards able premium to notify plan participants for underfunded defined benefit plans: of the plan’s funded status and the extent i. Increased the required contribution for of the PBGC’s guaranty in the event of unfunded new liability. plan termination.

71 NOTES

CHAPTER 1 9. Defined benefit plans promise a specific benefit; de- fined contribution plans promise a specific contribution 1. The baby-boom generation refers to the bulge in the age to a fund. The benefits provided by defined benefit plans distribution caused by a rise in birthrates between 1946 are normally determined by a formula relating to com- and 1964 and the decline thereafter. The earliest baby pensation and length of service. Benefits from a defined boomers will reach 65 in 2011, but they will affect the contribution plan depend on the annuity value of the total retirement-age population much earlier because many accumulation. workers retire before the age of 65. 10. Pension Benefit Guaranty Corporation, news release, 2. B. Douglas Bernheim, “Is the Baby Boom Generation December 5, 1994. Preparing Adequately for Retirement?” Technical Re- port, mimeographed, 1992, and a “Summary Report” 11. The PBGC insures individual benefits promised by with the same title published by Merrill Lynch, 1993. 66,000 private defined benefit plans up to $30,886 (ad- justed annually for inflation). Although it charges plan 3. The ratio is expected to remain at that level or one sponsors a premium for the insurance, the premiums do slightly lower through to 2070. not cover expected liability, which has risen rapidly. 4. Information supplied by the Office of Management Reforms enacted as part of GATT-enabling legislation and Budget. should help correct this situation over time. (See Chapter 5, page 48.) However, if underfunding of pension plans 5. Hospital insurance includes only Part A of Medicare. continues to grow rapidly as some anticipate, an eventual Under current policies, the HI program is expected to bailout by taxpayers cannot be ruled out. realize a $7 billion cash-flow deficit in calendar 1994, increasing to $1.2 trillion in 2030. See U.S. Department of 12. Office of Management and Budget. Health and Human Services, Social Security Administra- 13. Merrill Lynch, Retirement Savings in America (Princeton, tion, Annual Report of the Board of Trustees of the Federal N.J.: Merrill Lynch, Pierce, Fenner & Smith, Inc., 1993). Old-Age and Survivors Insurance and Disability Insurance Trust Funds, 1994 (Washington, D.C.: U.S. Government 14. Prior to OBRA93, the considered compensation limit Printing Office, 1994), and the U.S. Department of Health was $235,840. and Human Services, Health Care Financing Adminis- tration, Annual Report of the Board of Trustees of the Federal Hospital Insurance Trust Fund, 1994 (Washington, D.C.: CHAPTER 2 U.S. Government Printing Office, 1994). 1. The aging of the population is, in fact, a worldwide 6. Although the Social Security projection includes the in- phenomenon; and old-age security systems, including crease in the normal retirement age enacted earlier, cur- both community- or family-based systems and formal rently legislated benefits are assumed to be unchanged. retirement programs, appear to be breaking down in 7. Currently, about three-quarters of SMI funding comes many countries. International Bank for Reconstruction from general tax revenues and one-quarter from enroll- and Development, Averting the Old Age Crisis: Policies to ees. SMI is expected to grow to be as large as HI. C. Protect the Old and Promote Growth (New York: Oxford Eugene Steuerle and Jon M. Bakija, Retooling Social Secu- University Press, 1994). rity for the 21st Century: Right and Wrong Approaches to 2. Employee Benefit Research Institute, Baby Boomers in Reform (Washington, D.C.: The Urban Institute Press, Retirement: What Are Their Prospects?, Special Report and 1994), pp. 54-55. Issue Brief, no. 151 (Washington, D.C.: EBRI, July 1994), 8. The full commission did not accept the Kerrey-Danforth Table 12, page 19. proposals, of which the four most important in relation to 3. About 2.6 million federal workers and 11.3 million state Social Security were: (1) further increasing the Social and local government workers participate in these plans. Security retirement age, (2) reducing growth of initial EBRI, Baby Boomers in Retirement: What Are Their Pros- benefits relative to earnings, (3) reducing benefits for pects?, Table 12, page 19. upper-income workers, and (4) replacing part of the Social Security program with a mandatory saving pro- 4. Although foreign capital inflows may provide consid- gram. The leadership of the 104th Congress has also erable support for investment, they cannot be counted on indicated that no changes will be made to Social Security to make up for the decline in domestic saving. Economic for the time being. studies find that saving and investment rates are highly

72 correlated both across countries (which suggests that 11. These policy changes are described in Chapter 5. barriers to international capital flows are still important) Other factors that may have contributed to the decline in and over time within individual countries. Martin S. business contributions include the rising costs of health Feldstein and Charles Horioka. “Domestic Saving and care programs, the growing importance of small firms International Capital Flows” Economic Journal (June 1990). that often have no (or less generous) pension plans, and It has been estimated that it would take a national saving the rising value of assets in pension portfolios. rate of 5 to 5.5 percent simply to maintain productivity 12. This increase in the normal retirement age, enacted in growth at about 1 percent annually. See Charles L. the early 1980s, was intended to compensate for the rise Schultze, “Of Wolves, Termites, and Pussycats, or Why in life expectancy from that time until 2025 but not the We Should Worry about the Deficit,” The Brookings growth in life spans before or after. C. Eugene Steuerle Review (Summer 1989): 33. and Jon M. Bakija, Retooling Social Security for the 21st 5. On a current-account basis, the U.S. net foreign invest- Century: Right and Wrong Approaches to Reform (Washing- ment position fell from a surplus of $234 billion in 1984 ton, D.C.: The Urban Institute Press, 1994), pp. 195-198. to a deficit of $591 billion in 1992, the latest year for which 13. The current Social Security offset rules are as follows: data are available. Economic Report of the President (Wash- (1) for those under 65, benefits are reduced $1 for every $2 ington, D.C.: U.S. Government Printing Office, February of annual earnings in excess of $8,040; (2) for those 65 to 1994), p. 385. 69 years old, benefits are reduced $1 for every $3 of 6. Some have speculated that the saving rate will rise earnings in excess of $11,160; (3) for those over 69, no sharply as the baby boomers age. But the decline in the benefit reduction. personal saving rate does not appear to have occurred 14. OBRA93 increased the percentage of Social Security because of demographic change; therefore, the aging of benefits subject to tax for higher-income taxpayers. Social the baby boomers may not increase the overall saving Security benefits for individuals with incomes below rate significantly. Martin Neil Baily, Gary Burtless, and $25,000 ($32,000 for joint returns) are not taxable. For Robert E. Litan, Growth with Equity: Economic Policymaking individuals with combined incomes between $25,000 for the Next Century (Washington, D.C.: The Brookings and $34,000 ($33,000 and $44,000 for those filing jointly), Institution, 1994), p. 147. 50 percent of benefits are taxed. The act made 85 percent 7. See Restoring Prosperity: Budget Choices for Economic of benefits subject to tax for those with higher incomes. Growth, p. 6. 15. Merrill Lynch, Retirement Savings in America. 8. Congressional Budget Office, An Economic and Budget 16. A recent survey of 1,000 adult Americans sponsored Outlook Update (Washington, D.C.: U.S. Government by Merrill Lynch found that 82 percent were unable to Printing Office, August 1994), p. 8. The standardized pass a basic financial literacy test; 55 percent were not employment budget measure, which adjusts for the able to correctly answer which of stocks, bonds, savings business cycle, suggests that nearly half of the deficit accounts, and CDs had generated the highest historical decline reflects a change in fiscal policy. returns; 29 percent could not make a simple compound 9. CBO, An Economic and Budget Outlook Update, p. 29. interest calculation; and 87 percent could not accurately estimate the maximum annual Social Security benefit. 10. A recent survey conducted by Merrill Lynch indi- Using Merrill Lynch data, Professor B. Douglas Bernheim cated that among those who are not self-employed, only found a causal connection between substandard finan- 46 percent of those in the 45- to 64-year age bracket cial literacy and inadequate saving. See B. Douglas reported that they were saving specifically for retire- Bernheim, “The Determinants and Consequences of ment; among the baby-boom generation, only 25 percent Financial Literacy,” (manuscript, October 1994). indicated that they were currently saving for retirement. See Merrill Lynch, Retirement Savings in America (Princeton, N.J.: Merrill Lynch, Pierce, Fenner & Smith, CHAPTER 3 Inc., 1993). Based on a telephone survey of workers who are not self-employed, p. 3. Surveys conducted in 1993 1. These projections included 5.0 million recipients of and 1994 by the Gallup Organization for the Employee disability in 1993; this number is expected to rise to 10.5 Benefit Research Institute found that two-thirds of re- million by 2030. The projections assume that the fertility spondents who had not retired had begun to save for rate will decline from 2.07 in 1990 to 1.9 in 2020 and hold their retirement, although the amounts actually being steady thereafter. (The fertility rate is the average number saved appear to be quite inadequate. See Employee of children that would be born to a woman in her lifetime Benefit Research Institute, Public Attitudes on Retirement if she were to experience the birthrate prevailing in a Income, 1994, Report no. G-55 (Washington, D.C.: EBRI, particular year.) Life expectancy at birth is assumed to 1994); and EBRI, Public Attitudes on Retirement Age and rise from about 75 years in 1990 to about 79 years in 2030. Planning, 1993, Report no. G-46 (Washington, D.C.: For details, see U.S. Department of Health and Human EBRI, 1993). Services, Social Security Administration, Annual Report

73 of the Board of Trustees, Federal Old-Age and Survivors GATT-enabling legislation. The maximum interest rate Insurance and Disability Insurance Trust Funds, 1994 (Wash- that could be used to calculate pension liabilities was ington, D.C.: U.S. Government Printing Office, 1994), lowered to 105 percent of the preceding four-year p. 59. weighted Treasury bond rate (from 110 percent). This will cause some pension plans that were previously fully 2. See Social Security Administration, Annual Report of the funded to become underfunded and to have to increase Board of Trustees of the Federal OASDI Trust Funds, 1994, contributions. p. 119. 11. The term pension coverage refers to the percentage of 3. In addition to economic and demographic factors, the workers eligible to participate in an employer- or magnitude of the necessary payroll tax increase would union-sponsored pension plan. Some eligible workers depend on such factors as when that tax increase would choose not to participate. Although the absolute num- be implemented and the size and rate of the drawdown bers of covered and participating workers differ, cover- of the surplus. The Social Security Administration calcu- age and participation rates exhibit similar trends over lated that the OASDI rate would have to be raised by time. 2.13 percentage points beginning in 1994 in order to bring the program into actuarial balance over the seventy- 12. Employee Benefit Research Institute, Pension Cover- five-year projection period (ending in 2068). (The pessi- age and Participation Growth: A New Look at Primary mistic projection suggests a much larger tax increase.) and Supplemental Plans, Issue Brief no. 144 (Washington, However, it is unlikely that rates will be increased for this D.C.: EBRI, December 1993), p. 4. Available data from purpose until much later, perhaps until the surplus be- one source, the 1988 Current Population Survey, indicated gins to run down. Thus, the realized increase is likely to that coverage declined slightly between 1979 and 1988, be substantially higher than 2.13 percentage points. an event that raised some concern. Data for more recent Social Security Administration, Annual Report of the Board periods suggest that coverage has since increased or at of Trustees of the Federal OASDI Trust Funds, 1994, p. 22. least leveled off. Moreover, IRS data on participation rates did not show the decline. 4. It is sometimes argued that the ratio of covered work- ers to beneficiaries overstates the burden on workers. 13. The obligation assumed by an employer in establish- Robert Ball and Henry Aaron, “Social Security: It Is ing a pension plan often takes one of two forms: (1) a Affordable,” Washington Post, February 15, 1994, p. A17. defined benefit plan, whereby the employer provides specified benefits that are established in advance by 5. Two recent estimates place the cost of supporting the formula (usually affected by wage and years of service) elderly at 2.5 times and 1.76 times the cost of supporting and financed by variable employer contributions to a a child dependent. See Dennis A. Ahlburg and James W. pension fund, and (2) a defined contribution plan, whereby Vaupel, “Immigration and the Dependency Burden” the annual employer contributions are specified (usually (manuscript, 1993), p. 6. a percent of the wage) but the benefits are variable, 6. For example, the median age of immigrants to the depending on the performance of the fund. For further United States is currently 28 years, but a 28-year-old description, see Dan M. McGill and Donald S. Grubbs, arriving in 1990 will be 88 by 2050 and probably retired. Jr., Fundamentals of Private Pensions, 6th ed. (Philadelphia: Therefore, in order to maintain the worker-retiree ratio at Pension Research Council, Wharton School, University its 1990 level in the year 2050, additional immigrants of of Pennsylvania, 1989), p. 105. working age would be needed to fill the retiree’s place. 14. David Kennell et al., Retirement Coverage in Small and 7. Thomas Espenshade, “Can Immigration Slow U.S. Large Firms, Final Report, (Paper prepared for the Office Population Aging?” Journal of Policy Analysis and Man– of Advocacy, U.S. Small Business Administration, agement (1994); Ahlburg and Vaupel, “Immigration and Washington, D.C.: June 1992), p. ES.8. the Dependency Burden.” 15. Based on tax status determination requests to the IRS 8. In 1988, the median government employee pension from companies instituting new plans, it appears not was $8,049, but many federal retirees and some state and only that the total number of new plans has declined but local retirees were not eligible for Social Security ben- also that most of the new plans introduced have been of efits. Joseph S. Piacentini and Jill D. Foley, EBRI Databook the defined contribution type. In 1982 alone, for example, on Employee Benefits, 2d ed. (Washington, D.C.: Employee 85,000 plan sponsors applied for favorable determina- Benefit Research Institute, 1992), p. 103. tion status, of whom 28,000 sponsored defined benefit plans; but in 1990, 1991, and 1992 combined, only 40,000 9. Increased saving out of current income increases in- sponsors applied, of whom just 3,000 were of the defined vestment resources, whereas a rise in the market value of benefit type. However, this measure of recent trends is existing assets does not. not ideal because application for favorable determina- tion status is not compulsory. 10. Indeed, a similar situation was engineered by one of the pension reform measures contained in the 1994 16. Those who favor defined benefit plans argue that

74 retirement income will be higher because fewer lump-sum 22. A number of alternative concepts are used to estimate distributions are made from defined benefit plans and pension liabilities that can significantly affect the percep- because the professional investors who manage the plans tion of the adequacy of funding. Because of the need for are likely to get a higher return on investment. On the federal revenues, Congressional tax-writing committees other hand, the defined contribution plans are generally tend to choose less stringent standards of funding to fully portable, a feature that may improve retirement determine the eligibility of contributions. For example, benefits for those who have several jobs during their tax provisions of OBRA87, which had the effect of sharply working years. See Richard V. Burkhauser and Dallas L. reducing business contributions to pension funds, re- Salisbury, eds., Pensions in a Changing Economy, (Wash- quired that the calculation of maximum eligible funding ington, D.C., and Syracuse, N.Y.: Employee Benefit Re- be based on 150 percent of the termination-of-plan concept search Institute and National Academy on Aging, 1993); of liabilities. This concept assumes the pension is termi- and Andrew Samwick and Jonathan Skinner, nated at the valuation date and therefore ignores in- “How Will Defined Contribution Pension Plans Affect creases in obligations that will arise because of future Retirement Incomes?” (Paper prepared for conference on wage increases and years of service. Broader measures of “Public Policy Toward Pensions,”sponsored by the Asso- funding adequacy, such as the concept required by the ciation of Private Pension and Welfare Plans and the Financial Accounting Standards Board (FASB Statement Center for Economic Policy Research, Stanford Univer- 87), do include projected increases in earnings and, con- sity, Stanford, Calif.: October 7-8, 1993). sequently, generally show higher liabilities. 17. Available data on lump-sum distributions are dis- 23. See Chapter 2 for recommendations for changes in the cussed in Burkhauser and Salisbury, Pensions in a Chang- Social Security system. ing Economy, pp. 27 and 43-45. 24. Those single men and women who have an average 18. Employee Benefit Research Institute, “Retirement wage and turn 65 years old in 1995 and 2010, respec- Program Lump-Sum Distributions: Hundreds of Billions tively, will be the first cohorts with average wages to in Hidden Pension Income” Issue Brief no. 146 receive a negative return. Returns remain positive in (Washington, D.C.: EBRI, February 1994), p. 8. those years for all but the highest-income married cou- ples. For high-wage single individuals who turn 65 in 19. Private pension funds will be drawn down as the 2010, the present value of taxes paid will exceed the baby-boom generation retires. Indeed, one projection annuity value of benefits by $135,000 for men and $84,000 indicates that the pension system will cease being a net for women. See C. Eugene Steuerle and Jon M. Bakija, source of savings by 2024. Given the magnitude of the Retooling Social Security for the 21st Century (Washington, shift, there is concern that this will have a depressing ef- D.C.: Urban Institute Press, 1994), p. 107. fect on asset prices. Sylvester Schieber and John Shoven, “The Consequences of Population Aging on Private Pen- 25. The payroll tax rate for employees and employers in sion Fund Saving and Asset Markets,” Working Paper 1994, including Medicare, was 7.65 percent, consisting no. 4665 (Cambridge, Mass.: National Bureau of Eco- of 5.60 for retirement (OASI), 0.60 for disability (DI), nomic Research, March 1994). and 1.45 for health (HI). Self-employed persons pay a higher rate. 20. The PBGC’s single-employer insurance fund contin- 26. Social Security Administration, Annual Report of the ues to be in deficit on a net liability basis; and at the end of 1993, PBGC liabilities ($11.3 billion) exceeded its assets Board of Trustees, OASDI Trust Funds, 1994, pp. 5 and 177. ($8.4 billion) by $2.9 billion. Although ERISA states that 27. The full commission did not accept the Kerrey- the United States is not liable for any obligation incurred Danforth proposals, of which the four most important in by the PBGC [Title IV Section 4002(g)(2)], Congress has relation to Social Security were: (1) further increasing the behaved as if the PBGC’s liabilities are the liabilities of the Social Security retirement age, (2) reducing growth of federal government because it wants to avoid a default on initial benefits relative to earnings, (3) reducing benefits federally insured pensions. Consequently, many have for upper-income workers, and (4) replacing part of the concluded, as the Congressional Budget Office has, that Social Security program with a mandatory saving pro- “the federal government is almost certainly responsible gram. The leadership of the 104th Congress has also for those liabilities of the PBGC that exceed PBGC’s indicated that no changes will be made to Social Security assets.” CBO, Controlling Losses of the Pension Benefit Guar- for the time being. anty Corporation (Washington, D.C.: U.S. Government Printing Office, January 1993), p. 3. 28. In August 1994, the CBO projected that the total federal deficit would be $202 billion in fiscal 1994. The 21. Pension Benefit Guaranty Corporation, Annual Report OASDI surplus was expected to total $58 billion, indicat- 1992 (Washington, D.C.: U.S. Government Printing Of- ing a non–Social Security debt of $260 billion. U.S. Con- fice, 1992), pp. 4 and 10; and PBGC, Annual Report 1993 gress, Congressional Budget Office, The Economic and (Washington, D.C.: U.S. Government Printing Office, Budget Outlook: An Update (Washington, D.C.: U.S. Gov- 1993), pp. 7, 20-21. ernment Printing Office, August 1994), p. 29. Because of

75 the fungibility of government income, strictly speaking, justed for inflation. Many private sector pensions are it cannot be stated that the Social Security surplus is not adjusted to compensate for inflation, but the adjustment used to finance government investment spending is nearly always discretionary and frequently intermit- instead of consumption spending. Some would like to tent. William J. Wiatrowski, “On the Disparity between see the annual surplus invested in private assets in an Private and Public Pensions,” Monthly Labor Review attempt to ensure that pension savings are increased, (April 1994): 3-9. though this would not necessarily increase national sav- 33. It is also argued that automatic cost-of-living adjust- ing. Others argue that the non–Social Security budget should be in balance so that the Social Security surpluses ments for Social Security and other pension benefits raise social costs by reducing the voting constituency that can be used to retire federal debt and, thereby, increase the nation’s saving. See Rudolph G. Penner, Social Secu- supports anti-inflationary policies. rity and National Saving (New York: Committee for Eco- 34. Elderly income has also grown relative to that of the nomic Development, 1989). nonelderly as a result of economic conditions. A large share of their income has been shielded from inflation 29. Office of Management and Budget, “Budget Baseline, and unemployment; consequently, the elderly may be Historical Data and Alternatives for the Future,” Budget relatively well off in bad times. For example, high unem- of the United States Government for Fiscal Year 1994 (Wash- ployment, high inflation, and high interest rates in the ington, D.C.: U.S. Government Printing Office, January 1979–1984 period helped to reduce the real income of the 1993), pp. 187-196. nonelderly by 0.3 percent while raising the income of the 30. There have been widespread news reports about elderly by 3.4 percent. abuses involving excessively generous pensions. For 35. According to the example given in “Retirement Ben- example, local government pensions that base benefits efits and Saving Requirements” (see page 39), if the real on earnings during the final years of work occasionally return were 6 percent annually, the individual’s required have been manipulated by permitting workers in those rate of saving to attain the same retirement income target final years to work long hours of overtime and to include would be only 7 percent. This illustration also presents overtime pay in earnings counted for retirement. Some estimates of required saving to achieve alternative federal employees also qualify for very generous pen- replacement ratios. sions, often from legislative service or from combined civilian and military service. 36. Many managers of defined contribution pension plans fear that participants are too risk averse to invest in 31. Pension coverage is higher in the public sector, and portfolios with higher potential returns. Educating par- benefits are sometimes more generous than in the private ticipants about the required rate of saving and asset sector. But it is not always the case that public-sector returns needed to attain a desired level of retirement pensions are a “better deal” for workers than income is increasingly emphasized. For example, private-sector pensions. Most public-sector workers are TIAA-CREF, a very large defined contribution plan, required to contribute to their defined benefit plan, encourages its members to include equities in their port- whereas most private-sector workers make contribu- folios. Based on its historical rates of return on a fifty-fifty tions to Social Security only. Private-sector workers are split between a traditional annuity and a common-stock more likely to have a supplementary defined contribu- fund, TIAA-CREF reports that a contribution of 10 per- tion plan, and many public-sector workers are not cov- cent of salary for thirty years yields a replacement rate of ered by Social Security. (New workers in the public 50.7 percent of final pay. See John J. McCormack, State- sector are now required to be covered by Social Security.) ment presented to U.S. Department of Labor, ERISA A recent study of retirement income provided by defined Advisory Council, Washington, D.C., September 23, 1993. benefit plans and Social Security found that the initial retirement income replacement ratio for typical newly 37. Employee Benefit Research Institute, EBRI Data Book retired public-sector workers not covered by Social Secu- on Employee Benefits, 2d ed. (Washington: D.C.: EBRI, rity was similar to or slightly better than the income 1992), pp. 28 and 36. replacement ratio for private-sector workers but that the 38. The study does concede that future retirement looks contribution to pensions made by public-sector workers very bleak for some groups, particularly the poorly edu- typically exceeded contributions made by private-sector cated and single women with few marketable skills. workers. Those public-sector workers who were covered Congressional Budget Office, Baby Boomers in Retirement: by Social Security on average had very high ratios of An Early Perspective (Washington, D.C.: U.S. Government replacement income, but their contributions were simi- Printing Office, September 1993). larly high (Table 3). 39. Alan J. Auerbach and Laurence J. Kotlikoff, “The 32. About half of state and local government employees United States Fiscal and Savings Crisis and Their Impli- covered by defined benefit pensions and all full-time cations for the Baby Boom Generation” (Paper presented federal employees are in programs that provide auto- at a policy forum sponsored by the Employee Benefit matic cost-of-living increases. Social Security is also ad- Research Institute, Washington, D.C., May 4, 1994).

76 40. Hilary Williamson Hoynes and Daniel McFadden, There is a certain irony about the existence of saving “The Impact of Demographics on Housing and and investment incentives in the context of an income Non-Housing Wealth in the United States” (National tax, inasmuch as the principal reason for believing Bureau of Economic Research, Inc., Working Paper no. that special incentives for these activities are needed 4666, Cambridge, Mass., March 1994). is the tax itself. The rules that we think of under this heading are largely aimed not at mitigating some 41. These and other arguments have led some scholars inherent defect in the economy but at offsetting a to conclude that the CBO calculations actually present a problem created by the tax system in the first place: very pessimistic picture about the economic status of namely, the disincentive effect of income taxes on baby boomers in retirement. B. Douglas Bernheim, “The saving and investing. Adequacy of Saving for Retirement: Are the Baby Boom- ers on Track?” (Paper presented at EBRI Policy Forum, Describing tax-qualified savings as an incentive (under Washington, D.C., May 4, 1994). the comprehensive accrual-income theory) may also con- tribute to excessive and complicated regulation of what 42. See B. Douglas Bernheim, “Is the Baby Boom Genera- is deemed to be a special exception. Under the consump- tion Preparing Adequately for Retirement?” Technical tion theory, the exclusion of savings from the tax base is Report, mimeographed 1992, and a “Summary Report” treated as normal, not a special exception, and therefore with the same title published by Merrill Lynch, 1993. The is less likely to be subject to overregulation. methodology is also discussed by Bernheim and John Karl Scholz in “Private Saving and Public Policy,” Tax 2. The federal government’s classification of the deferral Policy and the Economy, vol. 7, ed. James M. Poterba of taxes on pension savings as a tax expenditure is consis- (Cambridge, Mass.: National Bureau of Economic Re- tent with the income tax view, not the consumed-income search and MIT Press, 1993), pp. 76-87. view. (See “Tax Expenditures and Pension Tax Prefer- ences,” page 46.) 43. Karen Martin Gibler and Joseph Rabianski, “Elderly Interest in Home Equity Conversion,” Housing Policy 3. Target savers are likely to be low-income individuals Debate 4, no. 4 (1993): 565-588. who save the minimum needed for a specific purpose, such as a down payment on an automobile or an appli- ance, whereas those who increase savings in response to CHAPTER 4 a higher return are likely to be higher-income earners who have discretion to save for more than minimum 1. Over the years, CED has sponsored critical research on future consumption. reforming the income tax, including perhaps the most basic source on the consumed-income tax. In the 1980s, 4. See, for example, Stephen F. Venti and David A. Wise, CED commissioned a study by David Bradford, pub- “IRAs and Saving,” in The Effects of Taxation on Capital lished as Untangling the Income Tax, a Committee for Accumulation, ed. Martin Feldstein (Chicago: University Economic Development publication, (Cambridge, Mass: of Chicago Press, 1987), pp. 7-40; and James Poterba, Harvard University Press, 1986). Bradford described the Steven F. Venti, and David A. Wise, “401(k) Plans and U.S. income tax system as a hybrid that has sometimes Tax-Deferred Saving,” National Bureau of Economic followed the comprehensive accrual-income theory, which Research, Working Paper no. 4181 (Cambridge, Mass.: includes in the income tax base all consumption and NBER, October 1992). saving, and sometimes followed the consumed-income 5. Poterba, Venti and Wise, “401(k) Plans and theory, which does not include savings in the tax base. Tax-Deferred Saving.” Bradford believes that our hybrid approach produces a muddle that is worse than would occur if we were to 6. This is because the positive impact on savings result- follow either theory consistently. He recommended ing from the increased return will dominate over the moving from the accrual-income theory toward the con- shifting of savings as funds available for shifting decline. sumption theory. See Eric M. Engen, William G. Gale, and John Karl Scholz, “Do Saving Incentives Work?” in Brookings Papers on Bradford makes clear that neither theory is the true Economic Activity, vol 1., ed. William C. Brainard and theory. Rather, he believes that the contending theories George L. Perry (Washington, D.C.: Brookings Institu- present a choice. The language of the debate on particular tion, 1994), pp. 85-152. questions (e.g., taxation of pensions) is influenced might- ily if one accepts one or the other theory as true. Exclud- 7. To promote household saving, Bernheim and Scholz ing retirement savings from the tax base is described as proposed premium saving accounts (PSAs) as an alterna- the grant of a subsidy or incentive from the perspective tive to IRAs. A PSA would require each taxpayer to save of the comprehensive accrual-income theory, whereas in total some fixed amount before becoming eligible to the inclusion of those savings in the tax base is described make contributions to a tax-preferred account. The as a penalty or disincentive from the perspective of the amount of eligible contributions would vary with ad- consumed-income theory. On page 208, Bradford states: justed gross income. B. Douglas Bernheim and John Karl

77 Scholz, “Private Saving and Public Policy,” in Tax Policy 6. See William G. Gale, “Public Policies and Private and the Economy, ed. James Poterba (Cambridge, Mass.: Pension Contributions,” Journal of Money, Credit, and NBER, 1993), p. 98. Banking, forthcoming; and Scott B. Smart and Joel Waldfogel, “Tax Policy, Saving, and Pension Funding” 8. B. Douglas Bernheim and John Karl Scholz, “Private (manuscript, Indiana University, October 1992). Saving and Public Policy,” National Bureau of Economic Research, Working Paper no. 4215 (Cambridge, Mass.: 7. Sylvester J. Schieber and John B. Shoven, The Con- NBER, November 1992), p. 24. sequences of Population Aging on Private Pension Fund Saving and Asset Markets, CEPR Technical Paper no. 363 9. Alicia Munnell, “It’s Time to Tax Employee Benefits,” (Stanford, Calif.: Center for Economic Policy Research, New England Economic Review (July-August 1989): 49-63. Stanford University, September 1993), p. 12. A 1989 sur- 10. Some researchers have also found that policies that vey by Mercer-Meidinger-Hansen found comparable reduce pension tax benefits are likely to increase inequal- results. See William M. Mercer-Meidinger-Hansen, The ity in the distribution of wealth. See Robert L. Clark and Effect of OBRA on Pension Plans: A Survey (New York: Elisa Wolper, “Pension Tax Expenditures: Magnitude William M. Mercer-Meidinger-Hansen, 1989). Distribution and Economic Effect,” (Paper prepared for 8. Contributions to defined contribution plans were conference on “Public Policy Toward Pensions,” spon- capped at $30,000 until the indexed benefit limit for sored by the Association for Private Pension and Welfare defined benefit plans reaches $120,000. Contributions Plans and the Center for Economic Policy Research, have also been reduced by limits on considered compen- Stanford University, October 1993). sation. 11. The advantages and disadvantages of compulsory 9. If an employee participates in a defined benefit and a saving plans are discussed in William G. Gale and Robert defined contribution plan, a special overall limitation on E. Litan, “Saving Our Way Out of the Deficit Dilemma,” contributions and benefits is applied. This limit uses a The Brookings Review (Fall 1993): 7-11. complicated formula based on the limits for individual 12. This has occurred yet again in the 1994 legislation plans. Thus, if the individual-plan limits are changed, the enabling GATT. For example, one measure effectively multiple-plan limit also changes. delays indexation of pension benefit and contribution 10. This limit applies to annuities; if a lump sum is limits purely as a revenue-raising measure. (See Retire- involved, the limit is $742,000. ment Protection Act of 1994, Title VII, Section 732). 11. Emily S. Andrews, “The Growth and Distribution of 13. For the most recent CED statement on the budget 401(k) Plans,” in U.S. Department of Labor, Pension and deficit, see Restoring Prosperity: Budget Choices for Eco- Welfare Benefits Administration, Trends in Pensions 1992 nomic Growth (1992), p. 2. (Washington, D.C.: U.S. Government Printing Office, 14. See Restoring Prosperity: Budget Choices for Economic 1992), p. 164. Growth, pp. 6-7. 12. Section 403(b) plans (see “Defined Contribution Plans, pp. 54-55) already include a catch-up provision that allows additional employer and employee deferrals for CHAPTER 5 individuals who have been with an employer fifteen years. However, these plans are available only to non- 1. Pension plans for public-sector employees are not profit organizations and educational institutions. covered under the federal laws governing private plans. 13. There is also a family aggregation rule, which stipu- 2. Employee Benefit Research Institute, Pension Funding lates a combined $150,000 limit for a married couple if and Taxation: Implications for Tomorrow (Washington, DC: they both work for the same employer and one spouse is EBRI, 1994), p. 44. in the top group of highly compensated employees. 3. Many propose that the United States abandon the 14. For example, in a defined benefit plan, an individual current practice of taxing income and adopt a earning $175,000 and participating in a plan that specifies consumption-based tax system, which would exempt all a pension benefit of 40 percent of salary would be cred- saving from taxation. In such a system the limits on tax ited with only 40 percent of $150,000 under the new rule, preferences would not be necessary. (See Chapter 4.) a $10,000 reduction in retirement benefit. Of course, 4. American Council of Life Insurance, The Pension Legacy: because the contributions are limited, benefits are re- The Case for Preserving the Private Pension System (Wash- duced accordingly. ington, D.C.: 1993), p. 6. 15. They are referred to as the actual deferral percentage 5. A forfeiture is a situation whereby an individual ceases tests: (1) The average percentage of compensation de- employment before being vested with the pension ben- ferred (i.e., contributed to the plan) for the highly com- efits. pensated group cannot exceed 125 percent of the average

78 deferral percentage for the low-paid group. (2) The aver- Protection Act of 1994 speeds up amortization of differ- age deferral percentage for the highly compensated group ent kinds of unfunded liabilities but also further compli- cannot be more than two times that of the other group, cates the amortization schedules. and the difference between the percentages cannot be 22. McGill and Grubbs, Fundamentals of Private Pensions, more than 2 percent. Matching employer contributions are subject to similar tests under Section 401(m) of the p. 611. Internal Revenue Code. 23. Carolyn L. Weaver, “Government Guarantees of 16. In the case of saving plans with matching Private Pension Benefits: Current Problems and Likely employer-employee contributions, nondiscrimination Future Prospects” (Paper prepared for conference on rules would require that the employer contribution be a “Public Policy Towards Pensions,” sponsored by the uniform percent of the contribution made by individual Association of Private Pension and Welfare Plans and employees. Retention of a ceiling on contributions to the Center for Economic Policy Research, Stanford Uni- secondary 401(k) plans would prevent highly compen- versity, September 1993), pp. 13-14. sated employees from gaining an inordinately large ben- 24. Weaver, “Government Guarantees of Private Pen- efit from this arrangement sion Benefits: Current Problems and Likely Future Pros- 17. John Trutko and John Gibson et al. Cost and Impact of pects,” pp. 14-15. Federal Regulation on Small Versus Large Business Retire- 25. Pension Benefit Guaranty Corporation, Annual Re- ment Plans, Final Report (submitted to the Office of the port 1992 (Washington, D.C.: U.S. Government Printing Chief Counsel for Advocacy, Small Business Administra- Office, 1992), p. 11. tion, June 1990), p. 66. 26. The Retirement Protection Act of 1994 gives the 18. A pension plan had to satisfy one of these tests in order PBGC authority to sue plan sponsors to enforce mini- to be in compliance: mum-funding standards where contributions are defi- 1. Percentage test: The plan must cover at least 70 percent cient by more than $1 million. of all non–highly compensated employees. 27. Celia Silverman and Paul Yakoboski, “Public and 2. Ratio test: The percentage of non–highly compensated Private Pensions Today: An Overview of the System,” in employees covered under a plan must be at least 70 Pension Funding and Taxation: Implications for Tomorrow, percent of the percentage of highly compensated ed. Dallas L. Salisbury and Nora Super Jones (Washing- employees covered. ton, D.C.: EBRI, 1994), p. 19. 3. Average benefits percentage test: Both of the follow 28. Hay/Huggins Company, Inc., Pension Plan Expense ing must be satisfied to pass this test: Study for the Pension Benefit Guaranty Corporation (Wash- a. The plan must cover a nondiscriminatory classifi- ington, D.C.: Hay/Huggins, September 1990). cation of employees. 29. Small plans are at a significant cost disadvantage in b. The ratio of employer-provided benefits or contri- trying to comply with constantly changing regulations butions to the participant’s compensation for non– because the variable costs of plan administration are not highly compensated employees must be at least 70 closely linked to the number of participants. Pension percent that of highly compensated employees. plan service providers typically charge a fixed fee plus a per-participant fee; however, the latter may be reduced 19. The participation rule requires that no individual who if the number of plan participants exceeds a threshold. has reached the age of 21 and has one year of service can See Trutko and Gibson et al., Cost and Impact of Federal be excluded from a plan. Employees must, at a minimum, Regulation on Small Versus Large Business Retirement Plans, be vested according to one of two schedules: (1) five-year p. 74. cliff or (2) 20 percent after three years and an additional 20 percent for each of the next four years. 30. The two plan sizes considered were 25 and 200 participants. Trutko and Gibson et al., p. 80. 20. Independent projections by the Office of Manage- ment and Budget put the PBGC’s exposure at about $24 31. These changes include OBRA93, which further re- billion over the next thirty to forty years. duced considered compensation, and requires compli- cated calculations for compliance. See pages 57-59 of this 21. Changes in a plan’s unfunded liability arising from an document. increase in plan benefits can be amortized over thirty years. Losses or gains resulting from changes in actuarial 32. Another 7 percent cited limitations on benefits to assumptions can be amortized over ten years; experience owners resulting from regulations governing maximum losses or gains can be amortized over five years. There are benefits, considered compensation, and nondiscrimina- a number of other amortization schedules that apply to tion. Other major reasons were plant shutdowns or un- other categories of gains and losses. The Retirement certainty of employer income (22 percent) and employer/

79 employee preferences (17 percent). David Kennell et al. 6. The most common type of 401(k) plan entails an em- Retirement Plan Coverage in Small and Large Firms, Final ployee salary reduction, the amount of which is placed in Report (submitted to Office of Advocacy, U.S. Small Busi- a tax-deferred account. About half of these plans involve ness Administration, June 1992), Table III-19. See also some amount of matching employer contribution. Trutko and Gibson et al., pp. 60-61. 7. The contribution limit for SEPs is now the lesser of 33. Although many workers were left without coverage $30,000 or 15 percent of compensation. because of the failure of employers to form successor plans, the rapid decline in pension coverage under de- 8. Under a leveraged ESOP, the plan’s trustee takes out a fined benefit plans cannot be attributed exclusively to loan and uses it to purchase employer stock, which itself regulation. In addition to administrative costs, research- is pledged as collateral for the loan. As repayments of ers believe that labor market shifts have been a crucial principal are made on the loan, stock can be released as factor: A decreasing number of workers are in industries collateral and allocated to employee accounts. dominated by large, unionized firms that traditionally 9. Down from $45,475. sponsor defined benefit plans. Daniel J. Beller and Helen H. Lawrence, “Trends in Private Pension Plan Coverage” 10. Previously 7 percent. in U.S. Department of Labor, Pension and Welfare Ben- efits Administration, Trends in Pensions 1992 (Washing- 11. The major rules are as follows: ton, D.C.: U.S. Government Printing Office, 1992), p. 70. a. Vesting: Either three-year cliff or graded to achieve 34. The impact of regulatory reform on the formation of full vesting after six years. defined benefit plans by very small firms may be less b. Minimum benefits: For nonkey employees, the mini- critical. Many smaller firms still do not offer their employ- mum annual benefit from a defined benefit plan was ees any pension plan. Evidently, economic factors such as set at the lesser of 20 percent of average compensation uncertainty about income have much to do with this. or 2 percent of average compensation multiplied by Smaller firms also tend to have a greater percentage of the number of years of service. employees who are young, part-time, and/or not likely to c. Minimum contributions: For nonkey employees, the stay for a long period in their job. See Kennell et al., minimum contribution to a defined contribution plan Retirement Plan Coverage in Small and Large Firms, Final was set at the lesser of 3 percent of cash compensation Report, p. ES.8. or the highest contribution rate for a key employee. d. Considered compensation: Only the first $200,000 of compensation could be taken into account when cal- APPENDIX culating benefits. 1. Where the plan provides for full immediate vesting, the e. Social Security integration: Neither Social Security employer can require three years of service. Also, em- benefits nor taxes could be counted against minimum ployees who begin service before age 25 can have three pension benefits. years of that service credited toward vesting when they attain 25. f. Aggregate benefit and contribution limits on multiple plans: Normally, if an employer operates a plan of 2. The vesting grades were now as follows: at least 25 each type (i.e., one defined contribution and one de- percent after five years, increasing by 5 percent in each of fined benefit), the total benefits and contributions the following five years and 10 percent annually in the cannot exceed 140 percent of the limit governing one third five-year tranche. plan if the percentage of compensation limit applies or 3. However, an employee with ten years of service must 125 percent of the limit if the dollar limit applies. For be 50 percent vested even if the sum of age and service top-heavy plans that would otherwise come under the does not exceed 45. 125 percent rule, benefits and contributions from multiple plans were reduced to 100 percent unless 4. The normal cost of a pension plan in a particular year minimum benefits and contributions for nonkey em- is the cost of benefits accrued in that year. ployees are increased 1 percentage point and no more than 90 percent of benefits and account balances is 5. A money purchase plan is a defined contribution plan designated to key employees. whereby periodic contributions are made according to a formula, usually a percent of salary. Some plans may 12. However, the plan was made liable to the PBGC for require contributions by the employee as well as the the sum of total unfunded benefits up to 30 percent of net employer. A profit-sharing plan is another type of defined worth and the excess (if any) of 75 percent of all unfunded contribution plan, with the distinctive characteristic that benefits minus 30 percent of net worth. contributions are likely to be variable because they are tied to a firm’s profits. 13. An integrated defined benefit excess plan is one that

80 provides a higher percentage of benefits for earnings 17. This is a subjective test administered by the IRS. above the integration level than for earnings below the integration level. Under prior law, it was possible to 18. The deficit reduction contribution consisted of two design a plan so that there was no benefit below the parts: (1) the amount necessary to amortize unfunded old liability in equal annual installments over eighteen years integration level. and (2) a percentage of unfunded liability relating to the 14. Here the integration level is defined as covered compen- current plan year. sation. Covered compensation is the average Social Secu- 19. Unpredictable contingent event benefits are benefits rity taxable earnings base for the thirty-five years imme- promised in the event of the closure of a facility or some diately preceding the participant’s normal retirement other occurrence that cannot be reliably anticipated. age. 20. Reduced from 10 percent above the same average. 15. Defined as a plan in which the benefit attributable to employer contributions is reduced by a specified amount. 21. Amortization of increases in current liability as a result of the new rules regarding the allowable interest 16. TRA also introduced a definition of highly compen- rates and mortality table had to be amortized in equal sated employee that differed from that in the top-heavy annual installments over a twelve-year period beginning rules contained in TEFRA. in 1995.

81 MEMORANDUM OF COMMENT, RESERVATION, OR DISSENT

report properly puts the retirement savings Page 14, JAMES Q. RIORDAN issue in the context of the debate now going The report is excellent and timely. The U.S. forward on whether savings of any kind retirement system is indeed underfunded, should be included in the income tax base. overpromised, and overregulated. As a result Indeed it briefly states the case for moving to we have less savings for retirement than we a tax system that does not include private need. This is, however, a special case of the saving in the income tax base. Unfortunately, more general problem that government policy it does not carry the discussion to its logical generally discourages savings and encourages conclusion and instead makes the assumption consumption. CED has long warned about the that the government will persist in accepting lack of savings and the consequences of gov- as revealed wisdom the comprehensive ac- ernment dissaving (deficits). The deficit has, at crual-income theory that treats every instance times, shown a year-to-year decline in the last of deferred taxation on savings as a prefer- decade; but in the best of years it has been too ence. The assumption has led the report to large, and the outlook is bleak. accept continuing limits and regulatory com- If net saving is to increase, it will need to plications that will limit the increase in sav- come from increased private saving — includ- ings. It inevitably leads to the use of language ing increased retirement savings. that distorts the dialogue. I wish that we had The report makes many excellent incremen- grasped the nettle and recommended taxing tal suggestions to improve the tax and regula- consumption and not savings. tory scheme governing retirement savings. I Nevertheless, this is a fine report, and I wish, however, that we had recommended a hope that all of the improvements recom- more fundamental cure for our problem. The mended will be adopted.

82 OBJECTIVES OF THE COMMITTEE FOR ECONOMIC DEVELOPMENT For more than 50 years, the Committee for independent, nonprofit, nonpartisan, and Economic Development has been a respected nonpolitical. influence on the formation of business and Through this business-academic partner- public policy. CED is devoted to these two ship, CED endeavors to develop policy state- objectives: ments and other research materials that To develop, through objective research and commend themselves as guides to public and informed discussion, findings and recommenda- business policy; that can be used as texts in tions for private and public policy that will contrib- college economics and political science courses ute to preserving and strengthening our free soci- and in management training courses; that will ety, achieving steady economic growth at high be considered and discussed by newspaper employment and reasonably stable prices, increas- and magazine editors, columnists, and com- ing productivity and living standards, providing mentators; and that are distributed abroad to greater and more equal opportunity for every citi- promote better understanding of the Ameri- zen, and improving the quality of life for all. can economic system. To bring about increasing understanding by CED believes that by enabling business present and future leaders in business, govern- leaders to demonstrate constructively their con- ment, and education, and among concerned citi- cern for the general welfare, it is helping busi- zens, of the importance of these objectives and the ness to earn and maintain the national and ways in which they can be achieved. community respect essential to the successful CED’s work is supported by private functioning of the free enterprise capitalist voluntary contributions from business and system. industry, foundations, and individuals. It is

83 CED BOARD OF TRUSTEES

Chairman

JOHN L. CLENDENIN, Chairman and DENNIS C. BOTTORFF, President and Chief Executive Officer Chief Executive Officer BellSouth Corporation First American Corporation & First American National Bank C. L. BOWERMAN, Executive Vice President Vice Chairmen and Chief Information Officer PHILIP J. CARROLL, President and Chief Phillips Petroleum Company Executive Officer MIKE R. BOWLIN, President and Shell Oil Company Chief Executive Officer ROBERT CIZIK, Chairman and Chief ARCO Executive Officer DICK W. BOYCE, Chief Executive Officer Cooper Industries Inc. FTD A.W. CLAUSEN, Retired Chairman and ERNEST L. BOYER, President Chief Executive Officer Carnegie Foundation for the Advancement BankAmerica Corporation of Teaching ALFRED C. DECRANE, JR., Chairman of the RICHARD J. BOYLE, Vice Chairman Board and Chief Executive Officer Chase Manhattan Bank, N.A. Texaco Inc. JOHN BRADEMAS, President Emeritus MATINA S. HORNER, Executive Vice President New York University TIAA-CREF EDWARD A. BRENNAN, Chairman and JAMES J. RENIER Chief Executive Officer Renier & Associates Sears, Roebuck & Co. ELI BROAD, Chairman, President and Treasurer Chief Executive Officer JOHN B. CAVE, Principal SunAmerica Inc. Avenir Group, Inc. STEPHEN L. BROWN, Chairman and Chief Executive Officer John Hancock Mutual Life Insurance Company REX D. ADAMS, Vice President - Administration Mobil Corporation JOHN H. BRYAN, Chairman and Chief Executive Officer PAUL A. ALLAIRE, Chairman and Sara Lee Corporation Chief Executive Officer Xerox Corporation J. GARY BURKHEAD, President Fidelity Management and Research Company IAN ARNOF, President and Chief Executive Officer THEODORE A. BURTIS, Retired Chairman First Commerce Corporation of the Board Sun Company, Inc. EDWIN L. ARTZT, Chairman of the Board and Chief Executive *OWEN B. BUTLER, Retired Chairman of the Board The Procter & Gamble Company The Procter & Gamble Company WILLIAM F. BAKER, President and JEAN B. BUTTNER, Chairman and Chief Executive Officer Chief Executive Officer WNET/Channel 13 Value Line Inc. RICHARD BARTH, Chairman, President and *FLETCHER L. BYROM, Chairman Chief Executive Officer Adience, Inc. Ciba-Geigy Corporation DONALD R. CALDWELL, Executive Vice BERNARD B. BEAL, Chief Executive Officer President M. R. Beal & Co. Safeguard Scientifics, Inc. HANS W. BECHERER, Chairman and FRANK C. CARLUCCI, Chairman Chief Executive Officer The Carlyle Group Deere & Company PHILIP J. CARROLL, President and HENRY P. BECTON, JR., President and Chief Executive Officer General Manager Shell Oil Company WGBH Educational Foundation EDWARD M. CARSON, Chairman ALAN BELZER, Retired President and First Interstate Bancorp Chief Operating Officer ROBERT B. CATELL, President and AlliedSignal Inc. Chief Executive Officer PETER A. BENOLIEL, Chairman of the Board Brooklyn Union Gas Company Quaker Chemical Corporation JOHN B. CAVE, Principal MICHEL L. BESSON, President and Avenir Group, Inc. Chief Executive Officer JOHN S. CHALSTY, President and Saint-Gobain Corporation Chief Executive Officer ROY J. BOSTOCK, Chairman and Donaldson, Lufkin & Jenrette, Inc. Chief Executive Officer RAYMOND G. CHAMBERS, Chairman D’Arcy, Masius, Benton & Bowles, Inc. of the Board Amelior Foundation

*Life Trustee ROBERT CIZIK, Chairman and FOREST J. FARMER Chief Executive Officer Rochester Hills, Michigan Cooper Industries Inc. KATHLEEN FELDSTEIN, President RICHARD A. CLARKE, Chairman of the Board Economics Studies, Inc. Pacific Gas and Electric Company RONALD E. FERGUSON, Chairman, President and A. W. CLAUSEN, Retired Chairman and Chief Executive Officer Chief Executive Officer General RE Corporation BankAmerica Corporation WILLIAM C. FERGUSON, Chairman and JOHN L. CLENDENIN, Chairman and Chief Executive Officer Chief Executive Officer NYNEX Corporation BellSouth Corporation RICHARD B. FISHER, Chairman NANCY S. COLE, President Morgan Stanley Group, Inc. Educational Testing Services *EDMUND B. FITZGERALD, Managing Director *EMILIO G. COLLADO Woodmont Associates Palm Beach Gardens, Florida *WILLIAM H. FRANKLIN, Retired Chairman KATHLEEN COOPER, Chief Economist of the Board Exxon Corporation Caterpillar Inc. GARY L. COUNTRYMAN, Chairman and HARRY L. FREEMAN, President Chief Executive Officer The Freeman Company Liberty Mutual Insurance Company ELLEN V. FUTTER, President ROBERT L. CRANDALL, Chairman, President American Museum of Natural History and Chief Executive Officer JOHN W. GALIARDO, Vice Chairman and General Counsel AMR/American Airlines Becton Dickinson and Company RONALD R. DAVENPORT, Chairman ALBERT R. GAMPER, JR., President and of the Board Chief Executive Officer Sheridan Broadcasting Corp. The CIT Group, Inc. ROBERT J. DAYTON, Chief Executive Officer RICHARD L. GELB, Chairman of the Board Okabena Company Bristol-Myers Squibb Company ALFRED C. DECRANE, JR., Chairman of the JOHN A. GEORGES, Chairman and Board and Chief Executive Officer Chief Executive Officer Texaco Inc. International Paper Company LINNET F. DEILY, Chairman, President and RAYMOND V. GILMARTIN Chief Executive Officer President and Chief Executive Officer First Interstate Bank of Texas Merck & Company, Inc. JERRY E. DEMPSEY, Chairman and BOYD E. GIVAN, Senior Vice President and Chief Executive Officer Chief Financial Officer PPG Industries, Inc. The Boeing Company JOHN DIEBOLD, Chairman CAROL R. GOLDBERG, President John Diebold Incorporated The AVCAR Group, Ltd. WILLIAM H. DONALDSON, Chairman and ELLEN R. GORDON, President and Chief Executive Officer Chief Operating Officer New York Stock Exchange Tootsie Roll Industries, Inc. JOSEPH L. DOWNEY, Director JOSEPH T. GORMAN, Chairman and The Dow Chemical Company Chief Executive Officer FRANK P. DOYLE, Executive Vice President TRW Inc. GE DENNIS J. GORMLEY, Chairman and E. LINN DRAPER, JR., Chairman, President Chief Executive Officer and Chief Executive Officer Federal-Mogul Corporation American Electric Power Company EARL G. GRAVES, SR., Publisher and GEORGE C. EADS, Vice President, Chief Executive Officer Worldwide Economic & Market Analysis Black Enterprise Magazine General Motors Corporation HANNA H. GRAY, President Emeritus and Harry Pratt W.D. EBERLE, Chairman Judson Distinguished Service Professor of History Manchester Associates, Ltd. The University of Chicago WILLIAM S. EDGERLY, Chairman WILLIAM H. GRAY, III, President and Foundation for Partnerships Chief Executive Officer WALTER Y. ELISHA, Chairman and United Negro College Fund Chief Executive Officer ROSEMARIE B. GRECO, President and Springs Industries, Inc. Chief Executive Officer JAMES D. ERICSON, President and CoreStates Bank Chief Executive Officer GERALD GREENWALD Northwestern Mutual Life Insurance Company Greenwald & Freeman Associates WILLIAM T. ESREY, Chairman and BARBARA B. GROGAN, President Chief Executive Officer Western Industrial Contractors Sprint CLIFFORD J. GRUM, Chairman of the Board JANE EVANS, Vice President and General and Chief Executive Officer Manager, Home and Personal Services Division Temple-Inland Inc. U.S. West Communications, Inc. JOHN H. GUTFREUND New York, New York

*Life Trustee JOHN R. HALL, Chairman and HARRY P. KAMEN, Chairman and Chief Executive Officer Chief Executive Officer Ashland Oil Inc. Metropolitan Life Insurance Company JUDITH H. HAMILTON, President and EDWARD A. KANGAS, Chairman and Chief Executive Officer Chief Executive Officer Dataquest Deloitte Touche Tohmatsu International RICHARD W. HANSELMAN, Retired Chairman HELENE L. KAPLAN, Esq., Of Counsel Genesco Inc. Skadden Arps Slate Meagher & Flom JOHN T. HARTLEY, Chairman and JOSEPH E. KASPUTYS, Chairman, President Chief Executive Officer and Chief Executive Officer Harris Corporation Primark Corporation LANCE H. HERNDON, Managing Consultant JULIUS KATZ, President Access, Inc. Hills & Company EDWIN J. HESS, Senior Vice President EAMON M. KELLY, President Exxon Corporation Tulane University RODERICK M. HILLS, Partner THOMAS J. KLUTZNICK, President Mudge Rose Guthrie Alexander & Ferdon Thomas J. Klutznick Company DONALD P. HILTY CHARLES F. KNIGHT, Chairman and Economic Strategy Institute Chief Executive Officer HAYNE HIPP, President and Emerson Electric Co. Chief Executive Officer ALLEN J. KROWE, Vice Chairman and The Liberty Corporation Chief Financial Officer DELWIN D. HOCK, Chairman and Chief Texaco Inc. Executive Officer RICHARD J. KRUIZENGA, Senior Fellow Public Service Company of Colorado ISEM HARRY G. HOHN, Chairman and CHARLES R. LEE, Chairman and Chief Executive Officer Chief Executive Officer New York Life Insurance Company GTE Corporation LEON C. HOLT, JR., Retired Vice Chairman FRANKLIN A. LINDSAY, Retired Chairman Air Products and Chemicals, Inc. Itek Corporation MATINA S. HORNER, Executive Vice President EDWARD E. LUCENTE, Executive In Residence, GSIA TIAA-CREF Carnegie Mellon University AMOS B. HOSTETTER, Chairman and EDWIN LUPBERGER, Chairman and Chief Executive Officer Chief Executive Officer Continental Cablevision, Inc. Entergy Corporation JAMES R. HOUGHTON, Chairman and BRUCE K. MACLAURY, President Chief Executive Officer The Brookings Institution Corning Incorporated COLETTE MAHONEY, RSHM, Chair BILL HOWELL, President Educational Consulting Associates, Ltd. Howell Petroleum Products, Inc. MICHAEL P. MALLARDI, President, Broadcast WILLIAM R. HOWELL, Chairman of the Board Group and Senior Vice President J.C. Penney Company, Inc. Capital Cities/ABC, Inc. CORDELL W. HULL, Executive Vice President DERYCK C. MAUGHAN, Chairman and and Director Chief Executive Officer Bechtel Group, Inc. Salomon Brothers Inc. ROBERT J. HURST, General Partner WILLIAM F. MAY, Chairman and Goldman, Sachs & Co. Chief Executive Officer SOL HURWITZ, President Statue of Liberty - Ellis Island Foundation, Inc. Committee for Economic Development JEWELL JACKSON MCCABE, President ALICE STONE ILCHMAN, President Jewell Jackson McCabe Associates, Inc. Sarah Lawrence College R. MICHAEL MCCULLOUGH, Senior Chairman GEORGE B. JAMES, Senior Vice President and Booz Allen & Hamilton Inc. Chief Financial Officer ALONZO L. MCDONALD, Chairman and Levi Strauss & Co. Chief Executive Officer ERIC G. JOHNSON, President and Chief Avenir Group, Inc. Executive Officer JAMES L. MCDONALD, Co-Chairman Tri-Star Associates, Inc. Price Waterhouse JAMES A. JOHNSON, Chairman and JOHN F. MCGILLICUDDY, Retired Chairman of Chief Executive Officer the Board and Chief Executive Officer Fannie Mae Chemical Banking Corporation ROBBIN S. JOHNSON, Corporate Vice President, EUGENE R. MCGRATH, Chairman, President and Public Affairs Chief Executive Officer Cargill, Incorporated Consolidated Edison Company of New York, Inc. THOMAS W. JONES, President and HENRY A. MCKINNELL, Executive Vice President Chief Operating Officer and Chief Financial Officer TIAA-CREF Pfizer Inc. PRES KABACOFF, President and DAVID E. MCKINNEY Co-Chairman Westport, Connecticut Historic Restoration, Inc.

*Life Trustee R. CLAYTON MCWHORTER, Chairman, WILLIAM R. RHODES, Vice Chairman President and Chief Executive Officer Citicorp/Citibank, N.A. Healthtrust Inc. WILLIAM C. RICHARDSON, President JEAN C. MONTY, President and The Johns Hopkins University Chief Executive Officer JAMES Q. RIORDAN Northern Telecom Limited Stuart, Florida J. RICHARD MUNRO, Chairman, JOHN D. ROACH, Chairman, President and Executive Committee Chief Executive Officer Time Warner Inc. Fibreboard Corporation GARY L. NEALE, Chairman, President VIRGIL ROBERTS, President and Chief Executive Officer Dick Griffey Productions/Solar Records Nipsco Industries DAVID ROCKEFELLER, JR., Chairman KENT C. NELSON, Chairman and Rockefeller Financial Services, Inc. Chief Executive Officer United Parcel Service of America, Inc. JUDITH S. RODIN, President University of Pennsylvania MARILYN CARLSON NELSON, Vice Chairman Carlson Holdings, Inc. IAN M. ROLLAND, Chairman and Chief Executive Officer JOSEPH NEUBAUER, Chairman and Lincoln National Corporation Chief Executive Officer ARAMARK Corp. DANIEL ROSE, President Rose Associates, Inc. BARBARA W. NEWELL, Regents Professor Florida State University CHARLES O. ROSSOTTI, Chairman American Management Systems PATRICK F. NOONAN, Chairman and Chief Executive Officer LANDON H. ROWLAND, President and The Conservation Fund Chief Executive Officer Kansas City Southern Industries, Inc. RICHARD C. NOTEBAERT, Chairman and Chief Executive Officer NEIL L. RUDENSTINE, President Ameritech Corporation Harvard University JAMES J. O’CONNOR, Chairman and GEORGE E. RUPP, President Chief Executive Officer Columbia University Commonwealth Edison Company GEORGE F. RUSSELL, JR., Chairman DEAN R. O’HARE, Chairman and Frank Russell Company Chief Executive Officer VINCENT A. SARNI, Retired Chairman Chubb Corporation and Chief Executive Officer JOHN D. ONG, Chairman of the Board, President PPG Industries, Inc. and Chief Executive Officer JOHN C. SAWHILL, President and The BFGoodrich Company Chief Executive Officer ANTHONY J.F. O’REILLY, Chairman, President The Nature Conservancy and Chief Executive Officer HENRY B. SCHACHT, Chairman of the H.J. Heinz Company Executive Committee JAMES F. ORR III, Chairman and Cummins Engine Company, Inc. Chief Executive Officer THOMAS SCHICK, Executive Vice President, UNUM Corporation Corporate Affairs and Communications ROBERT J. O'TOOLE, Chairman and American Express Company Chief Executive Officer JONATHAN M. SCHOFIELD, Chairman and A. O. Smith Corporation Chief Executive Officer WILLIAM R. PEARCE, President and Airbus Industrie of North America, Inc. Chief Executive Officer DONALD J. SCHUENKE, Chairman IDS Mutual Fund Group Northern Telecom Limited JERRY K. PEARLMAN, Chairman and ROBERT G. SCHWARTZ Chief Executive Officer New York, New York Zenith Electronics Corporation J. L. SCOTT VICTOR A. PELSON, Executive Vice President, Salt Lake City, Utah Chairman-Global Operations Team ERVIN R. SHAMES AT&T Corp. Wilton, Connecticut PETER G. PETERSON, Chairman WALTER V. SHIPLEY, Chairman and The Blackstone Group Chief Executive Officer DEAN P. PHYPERS Chemical Banking Corporation New Canaan, Connecticut C. R. SHOEMATE, Chairman, President and S. LAWRENCE PRENDERGAST, Vice President Chief Executive Officer and Treasurer CPC International Inc. AT&T Corp. WALTER H. SHORENSTEIN, Chairman of the Board WESLEY D. RATCLIFF, President and The Shorenstein Company Chief Executive Officer *GEORGE P. SHULTZ, Distinguished Fellow Advanced Technological Solutions, Inc. The Hoover Institution ALLAN L. RAYFIELD ROCCO C. SICILIANO Lowell, Massachusetts Beverly Hills, California JAMES J. RENIER L. PENDLETON SIEGEL, President and Renier & Associates Chief Operating Officer Potlach Corporation

*Life Trustee ANDREW C. SIGLER, Chairman and JAMES A. THOMSON, President and Chief Executive Officer Chief Executive Officer Champion International Corporation RAND IRBY C. SIMPKINS, JR., Publisher and CHANG-LIN TIEN, Chancellor Chief Executive Officer University of California, Berkeley Nashville Banner RANDALL L. TOBIAS, Chairman of the Board and FREDERICK W. SMITH, Chairman, President Chief Executive Officer and Chief Executive Officer Eli Lilly and Company Federal Express Corporation ALEXANDER J. TROTMAN, Chairman, President RAYMOND W. SMITH, Chairman of the Board and Chief Executive Officer and Chief Executive Officer Ford Motor Company Bell Atlantic Corporation RICHARD A. VOELL, President and SHERWOOD H. SMITH, JR., Chairman of the Chief Executive Officer Board and Chief Executive Officer The Rockefeller Group Carolina Power & Light Company HAROLD A. WAGNER., Chairman, President and WILLIAM D. SMITHBURG, Chairman and Chief Executive Officer Chief Executive Officer Air Products & Chemicals, Inc. The Quaker Oats Company ADMIRAL JAMES D. WATKINS, USN (Ret.), TIMOTHY P. SMUCKER, Chairman President The J.M. Smucker Company Joint Oceanographic Institutions, Inc. ALAN G. SPOON, President ARNOLD R. WEBER, Chancellor The Washington Post Company Northwestern University ELMER B. STAATS, Former Comptroller LAWRENCE A. WEINBACH, Managing Partner– General of the United States Chief Executive JOHN R. STAFFORD, Chairman, President Arthur Andersen & Co, SC and Chief Executive Officer HARVEY A. WEINBERG, Consultant American Home Products Corporation HSSI, Inc. STEPHEN STAMAS, Chairman ROBERT E. WEISSMAN, President and New York Philharmonic Chief Executive Officer JOHN L. STEFFENS, Executive Vice President Dun & Bradstreet Corporation Merrill Lynch & Co., Inc. JOHN F. WELCH, JR., Chairman and Chief Executive Officer RICHARD J. STEGEMEIER, Chairman of the Board Unocal Corporation GE W. THOMAS STEPHENS, Chairman, VIRGINIA V. WELDON, M.D., Senior Vice President, President and Chief Executive Officer Public Policy Manville Corporation Monsanto Company PAUL G. STERN JOSH S. WESTON, Chairman and Potomac, Maryland Chief Executive Officer Automatic Data Processing, Inc. PAULA STERN, Senior Fellow Progressive Policy Institute CLIFTON R. WHARTON, JR., Former Chairman TIAA-CREF DONALD M. STEWART, President The College Board DOLORES D. WHARTON, Chairman and Chief Executive Officer WILLIAM P. STIRITZ, Chairman of the Board The Fund for Corporate Initiatives, Inc. Ralston Purina Company EDWARD E. WHITACRE, JR., Chairman and ROGER W. STONE, Chairman, President and Chief Executive Officer Chief Executive Officer SBC Communications, Inc. Stone Container Corporation GORDON L. WILLIAMS, Corporate Vice President MATTHEW J. STOVER, President and and General Manager for the Commercial Chief Executive Officer Aircraft Division NYNEX Information Resources Company Northrop Grumman Corporation CARROLL W. SUGGS, Chairman of the Board and HAROLD M. WILLIAMS, President Chief Executive Officer The J. Paul Getty Trust Petroleum Helicopters, Inc. J. KELLEY WILLIAMS, Chairman and JAMES N. SULLIVAN, Vice Chairman of the Board Chief Executive Officer Chevron Corporation First Mississippi Corporation RICHARD F. SYRON, Chairman and MARGARET S. WILSON, Chairman of the Board Chief Executive Officer Scarbroughs American Stock Exchange WILLIAM S. WOODSIDE, Vice Chairman ALISON TAUNTON-RIGBY LSG Sky Chefs Lincoln, Massachusetts MARTIN B. ZIMMERMAN, Executive Director, ANTHONY P. TERRACCIANO, Chairman, President Governmental Relations and Corporate Economics and Chief Executive Officer Ford Motor Company First Fidelity Bancorporation CHARLES J. ZWICK RICHARD L. THOMAS, Chairman and Coral Gables, Florida Chief Executive Officer First Chicago Corporation

*Life Trustee CED HONORARY TRUSTEES

RAY C. ADAM LYLE EVERINGHAM, Retired Chairman New York, New York The Kroger Co. *O. KELLEY ANDERSON THOMAS J. EYERMAN, President Boston, Massachusetts Delphi Associates Limited ROBERT O. ANDERSON, Chairman JOHN T. FEY Hondo Oil & Gas Company Jamestown, Rhode Island ROY L. ASH JOHN M. FOX Los Angeles, California Sapphire, North Carolina SANFORD S. ATWOOD, President Emeritus DON C. FRISBEE, Chairman Emeritus Emory University PacifiCorp ROBERT H. B. BALDWIN, Retired Chairman W. H. KROME GEORGE, Retired Chairman Morgan Stanley Group Inc. Aluminum Company of America JOSEPH W. BARR WALTER B. GERKEN, Chairman, Hume, Virginia Executive Committee GEORGE F. BENNETT, Chairman Emeritus Pacific Mutual Life Insurance Company State Street Investment Trust PAUL S. GEROT HAROLD H. BENNETT Delray Beach, Florida Salt Lake City, Utah LINCOLN GORDON, Guest Scholar JACK F. BENNETT The Brookings Institution Greenwich, Connecticut KATHARINE GRAHAM, Chairman of HOWARD W. BLAUVELT the Executive Committee Charlottesville, Virginia The Washington Post Company FRED J. BORCH JOHN D. GRAY, Chairman Emeritus New Canaan, Connecticut Hartmarx Corporation MARVIN BOWER, Director WALTER A. HAAS, JR., Honorary Chairman McKinsey & Company, Inc. of the Board Levi Strauss & Co. ALAN S. BOYD Washington, D.C. ROBERT A. HANSON, Retired Chairman Deere & Company ANDREW F. BRIMMER, President Brimmer & Company, Inc. ROBERT S. HATFIELD, Retired Chairman The Continental Group, Inc. HARRY G. BUBB, Chairman Emeritus Pacific Mutual Life Insurance ARTHUR HAUSPURG, Member, Board of Trustees JOHN L. BURNS Consolidated Edison Company of New York, Inc. Greenwich, Connecticut PHILIP M. HAWLEY, Retired Chairman THOMAS D. CABOT, Honorary of the Board Chairman of the Board Carter Hawley Hale Stores, Inc. Cabot Corporation WILLIAM A. HEWITT PHILIP CALDWELL, Senior Managing Director Chadds Ford, Pennsylvania Lehman Brothers Inc. OVETA CULP HOBBY, Chairman EDWARD W. CARTER, Chairman Emeritus H&C Communications, Inc. Carter Hawley Hale Stores, Inc. ROBERT C. HOLLAND, Senior Economic EVERETT N. CASE Consultant Van Hornesville, New York Committee for Economic Development HUGH M. CHAPMAN, Chairman GEORGE F. JAMES NationsBank South South Bristol, Maine E. H. CLARK, JR., Chairman and Chief HENRY R. JOHNSTON Executive Officer Ponte Vedra Beach, Florida The Friendship Group GILBERT E. JONES, Retired Vice Chairman GEORGE S. CRAFT IBM Corporation Atlanta, Georgia CHARLES KELLER, JR. DOUGLAS D. DANFORTH, Retired Chairman Keller Family Foundation Westinghouse Electric Corporation GEORGE M. KELLER, Chairman of the JOHN H. DANIELS, Retired Chairman Board, Retired and Chief Executive Officer Chevron Corporation Archer-Daniels Midland Co. DAVID M. KENNEDY RALPH P. DAVIDSON Salt Lake City, Utah Washington, D.C. JAMES R. KENNEDY ARCHIE K. DAVIS, Chairman of the Essex Falls, New Jersey Board (Retired) Wachovia Bank and Trust Company, N.A. TOM KILLEFER, Chairman Emeritus United States Trust Company of New York DOUGLAS DILLON New York, New York CHARLES M. KITTRELL Bartlesville, Oklahoma ROBERT R. DOCKSON, Chairman Emeritus CalFed, Inc. PHILIP M. KLUTZNICK, Senior Partner Klutznick Investments

*Life Trustee HARRY W. KNIGHT ROBERT M. PRICE, Retired Chairman and New York, New York Chief Executive Officer ROY G. LUCKS Control Data Corporation San Francisco, California R. STEWART RAUCH, Former Chairman ROBERT W. LUNDEEN, Retired Chairman The Philadelphia Savings Fund Society The Dow Chemical Company AXEL G. ROSIN, Retired Chairman RAY W. MACDONALD, Honorary Chairman Book-of-the-Month Club, Inc. of the Board WILLIAM M. ROTH Burroughs Corporation San Francisco, California IAN MACGREGOR, Retired Chairman GEORGE RUSSELL AMAX Inc. Bloomfield, Michigan RICHARD B. MADDEN, Retired Chairman JOHN SAGAN, President and Chief Executive Officer John Sagan Associates Potlatch Corporation RALPH S. SAUL, Former Chairman of the Board FRANK L. MAGEE CIGNA Companies Stahlstown, Pennsylvania GEORGE A. SCHAEFER, Retired Chairman STANLEY MARCUS, Consultant of the Board Stanley Marcus Consultancy Caterpillar, Inc. AUGUSTINE R. MARUSI MARK SHEPHERD, JR., Retired Chairman Red Bank, New Jersey Texas Instruments, Inc. OSCAR G. MAYER, Retired Chairman RICHARD R. SHINN, Retired Chairman Oscar Mayer & Co. and Chief Executive Officer GEORGE C. MCGHEE, Former U.S. Metropolitan Life Insurance Co. Ambassador and Under Secretary of State NEIL D. SKINNER Washington, D.C. Indianapolis, Indiana JAMES W. MCKEE, JR., Retired Chairman ELLIS D. SLATER CPC International, Inc. Landrum, South Carolina CHAMPNEY A. MCNAIR, Retired Vice Chairman DAVIDSON SOMMERS Trust Company of Georgia Washington, D.C. J. W. MCSWINEY, Retired Chairman of the Board ELVIS J. STAHR, JR. The Mead Corporation Chickering & Gregory, P.C. CHAUNCEY J. MEDBERRY, III, Retired Chairman FRANK STANTON, President Emeritus BankAmerica Corporation and Bank of America CBS, Inc. N.T. and S.A. EDGAR B. STERN, JR., Chairman of the Board ROBERT E. MERCER, Retired Chairman Royal Street Corporation The Goodyear Tire & Rubber Co. J. PAUL STICHT, Retired Chairman RUBEN F. METTLER, Retired Chairman and RJR Nabisco, Inc. Chief Executive Officer ALEXANDER L. STOTT TRW Inc. Fairfield, Connecticut LEE L. MORGAN, Former Chairman of the Board WAYNE E. THOMPSON, Past Chairman Caterpillar, Inc. Merritt Peralta Medical Center ROBERT R. NATHAN, Chairman CHARLES C. TILLINGHAST, JR. Nathan Associates, Inc. Providence, Rhode Island ALFRED C. NEAL HOWARD S. TURNER, Retired Chairman Harrison, New York Turner Construction Company J. WILSON NEWMAN, Retired Chairman L. S. TURNER, JR. Dun & Bradstreet Corporation Dallas, Texas LEIF H. OLSEN, President THOMAS A. VANDERSLICE, Chairman, President Leif H. Olsen Investments, Inc. and Chief Executive Officer NORMA PACE, Senior Advisor M/A-Com, Inc. The WEFA Group ROBERT C. WEAVER CHARLES W. PARRY, Retired Chairman New York, New York Aluminum Company of America JAMES E. WEBB JOHN H. PERKINS, Former President Washington, D.C. Continental Illinois National Bank and Trust Company SIDNEY J. WEINBERG, JR., Limited Partner HOWARD C. PETERSEN The Goldman Sachs Group, L.P. Radnor, Pennsylvania GEORGE WEISSMAN, Retired Chairman C. WREDE PETERSMEYER Philip Morris Companies Inc. Vero Beach, Florida ARTHUR M. WOOD RUDOLPH A. PETERSON, President and Chicago, Illinois Chief Executive Officer (Emeritus) RICHARD D. WOOD, Director BankAmerica Corporation Eli Lilly and Company EDMUND T. PRATT, JR., Chairman Emeritus Pfizer Inc. CED RESEARCH ADVISORY BOARD

Chairman JOHN P. WHITE FRANK LEVY JUNE E. O‘NEILL Director, Center for Business and Daniel Rose Professor of Urban Director, Center for the Study of Government Economics Business and Government John F. Kennedy School of Government Department of Urban Studies and Baruch College, City University Harvard University Planning of New York Massachusetts Institute of Technology PETER PASSELL REBECCA MAYNARD The New York Times Trustee Professor of Education DOUGLAS J. BESHAROV Policy CHRISTINA D. ROMER Resident Scholar University of Pennsylvania Professor of Economics American Enterprise Institute for University of California, Berkeley Public Policy Research JANET L. NORWOOD Senior Fellow BERNARD SAFFRAN SUSAN M. COLLINS Urban Institute Franklin & Betty Barr Professor Senior Fellow, Economic Studies of Economics Program Swarthmore College The Brookings Institution

CED PROFESSIONAL AND ADMINISTRATIVE STAFF

SOL HURWITZ President

VAN DOORN OOMS SANDRA KESSLER HAMBURG Senior Economic Consultant Senior Vice President and Vice President and Director of ROBERT C. HOLLAND Director of Research Education Studies WILLIAM J. BEEMAN TIMOTHY J. MUENCH Advisor on International Vice President and Director Vice President and Director of Economic Policy of Economic Studies Finance and Administration ISAIAH FRANK William L. Clayton Professor ANTHONY P. CARNEVALE EVA POPPER of International Economics Vice President and Vice President, Secretary of the The Johns Hopkins University Director of Human Resource Board of Trustees and Director Studies of Development CLAUDIA P. FEUREY NATHANIEL M. SEMPLE Vice President for Vice President, Secretary of Communications and the Research and Policy Corporate Affairs Committee, and Director of Business-Government Relations

SHARON O'CONNELL MARK FRANCIS THOMAS R. FLAHERTY Director of Special Projects Director of Information Comptroller and Director of Operations

Research Development Administration MICHAEL K. BAKER JULIA R. HICKS DOROTHY M. STEFANSKI Economist Assistant Director Deputy Comptroller JESSE W. ELLIS MICHAEL BORNHEIMER KAREN CASTRO Policy Analyst Staff Associate Assistant Comptroller LORRAINE MACKEY ANA SOMOHANO ARLENE M. MURPHY Research Administrator Campaign Coordinator Administrative Assistant to the President MARIA L. LUIS WILFORD V. MALCOLM Publications Coordinator Campaign Production Administrator SHIRLEY R. SHERMAN Office Manager, Washington Conferences VALERIE MENDELSOHN Manager STATEMENTS ON NATIONAL POLICY ISSUED BY THE COMMITTEE FOR ECONOMIC DEVELOPMENT

SELECTED PUBLICATIONS:

Putting Learning First: Governing and Managing the Schools for High Achievement (1994) Prescription for Progress: The Uruguay Round in the New Global Economy (1994) *From Promise to Progress: Towards a New Stage in U.S.-Japan Economic Relations (1994) U.S. Trade Policy Beyond The Uruguay Round (1994) In Our Best Interest: NAFTA and the New American Economy (1993) What Price Clean Air? A Market Approach to Energy and Environmental Policy (1993) Why Child Care Matters? Preparing Young Children For A More Productive America (1993) Restoring Prosperity: Budget Choices for Economic Growth (1992) The United States in the New Global Economy: A Rallier of Nations (1992) The Economy and National Defense: Adjusting to Cutbacks in the Post-Cold War Era (1991) Politics, Tax Cuts and the Peace Dividend (1991) The Unfinished Agenda: A New Vision for Child Development and Education (1991) Foreign Investment in the United States: What Does It Signal? (1990) An America That Works: The Life-Cycle Approach to a Competitive Work Force (1990) Breaking New Ground in U.S. Trade Policy (1990) Battling America's Budget Deficits (1989) *Strengthening U.S.-Japan Economic Relations (1989) Who Should Be Liable? A Guide to Policy for Dealing with Risk (1989) Investing in America's Future: Challenges and Opportunities for Public Sector Economic Policies (1988) Children in Need: Investment Strategies for the Educationally Disadvantaged (1987) Finance and Third World Economic Growth (1987) Toll of the Twin Deficits (1987) Reforming Health Care: A Market Prescription (1987) Work and Change: Labor Market Adjustment Policies in a Competitive World (1987) Leadership for Dynamic State Economies (1986) Investing in Our Children: Business and the Public Schools (1985) Fighting Federal Deficits: The Time for Hard Choices (1985) Strategy for U.S. Industrial Competitiveness (1984) Strengthening the Federal Budget Process: A Requirement for Effective Fiscal Control (1983) Productivity Policy: Key to the Nation's Economic Future (1983) Energy Prices and Public Policy (1982) Public-Private Partnership: An Opportunity for Urban Communities (1982) Reforming Retirement Policies (1981) Transnational Corporations and Developing Countries: New Policies for a Changing World Economy (1981) Fighting Inflation and Rebuilding a Sound Economy (1980) Stimulating Technological Progress (1980) Helping Insure Our Energy Future: A Program for Developing Synthetic Fuel Redefining Government's Role in the Market System (1979) Improving Management of the Public Work Force: The Challenge to State and Local Government (1978) Jobs for the Hard-to-Employ: New Directions for a Public-Private Partnership (1978) An Approach to Federal Urban Policy (1977) Key Elements of a National Energy Strategy (1977) Nuclear Energy and National Security (1976) Fighting Inflation and Promoting Growth (1976) Improving Productivity in State and Local Government (1976) *International Economic Consequences of High-Priced Energy (1975) Broadcasting and Cable Television: Policies for Diversity and Change (1975) Achieving Energy Independence (1974) A New U.S. Farm Policy for Changing World Food Needs (1974) Congressional Decision Making for National Security (1974) *Toward a New International Economic System: A Joint Japanese-American View (1974) More Effective Programs for a Cleaner Environment (1974) The Management and Financing of Colleges (1973) Financing the Nation's Housing Needs (1973) Building a National Health-Care System (1973) High Employment Without Inflation: A Positive Program for Economic Stabilization (1972) Reducing Crime and Assuring Justice (1972) Military Manpower and National Security (1972) The United States and the European Community: Policies for a Changing World Economy (1971) Social Responsibilities of Business Corporations (1971)

*Statements issued in association with CED counterpart organizations in foreign countries. CED COUNTERPART ORGANIZATIONS

Close relations exist between the Committee for Economic Development and independent, nonpolitical research organizations in other countries. Such counter- part groups are composed of business executives and scholars and have objec- tives similar to those of CED, which they pursue by similarly objective methods. CED cooperates with these organizations on research and study projects of common interest to the various countries concerned. This program has resulted in a number of joint policy statements involving such international matters as energy, East-West trade, assistance to developing countries, and the reduction of nontariff barriers to trade.

CE Circulo de Empresarios Madrid, Spain

CEDA Committee for Economic Development of Australia Sydney, Australia

EVA Centre for Finnish Business and Policy Studies Helsinki, Finland

FAE Forum de Administradores de Empresas Lisbon, Portugal

IDW Institut der Deutschen Wirtschaft Cologne, Germany

IE Institut de l’Entreprise Brussels, Belgium

IE Institut de l’Entreprise Paris, France

Keizai Doyukai Tokyo, Japan

SMO Stichting Maatschappij en Onderneming The Netherlands

SNS Studieförbundet Naringsliv och Samhälle Stockholm, Sweden