Social Security: The Case for Pay-As-You-Go

By Dan Cornwell

The Topic and Scope of this Paper

For 60 years, Social Security has served the nation well. With modest adjustments, it can continue to do so indefinitely. History will not deal kindly with political leaders who preside over the dismantling of this excellent system in favor of one which is less dependable, far less efficient, risky for individuals, and regressive in its impact.

This paper deals with the pension part of Social Security, that is, the Old-Age and Survivors Insurance and Disability Insurance (OASDI) programs. These programs provide income to a worker in case of disability, to a worker and spouse at old age, and to children and elderly spouses in case of death of the worker.

The Author

The author is Professor Emeritus at the University of Wisconsin-Madison. Since retiring from the Department of Chemistry, he has devoted some of his time to the study of several national policy issues.

Madison, Wisconsin, April, 1997
Revised, October, 1997

The Madison Institute
P.O. Box 5304
Madison, WI 53705

Table of Contents

Preface to Political Leaders i
Summary iii

Introduction 1

A Vision 1
Two Problems 1
Pay-as-you-go vs. Advance Funding 2
The Central Issue 2
Pay-as-you-go: The System of Choice 4
A Lesson from Recent History 5
Earn More by Investing the Current Trust Fund in Stocks? 5
Inequities in the Transition to an Advance-funded System. 6
Advance Funding: Faith in Financial Paper 6
National Savings and Investment 8
Two-Tier Systems 10

Making Pay-as-you-go Work 11

What to Do 11
What Not to Do 12
Would the Suggested Changes Suffice? 14
Reversing the Culture of Early Retirement 14
Maintaining the Ratio of Workers to Retirees 15

Privatization 17

Steps Toward Privatization 17
Fundamental Issues 17
Comparison of Benefits 19
The National Pension System as a Tax-Haven for the Rich 21

Social Security Advisory Council Proposals 21

Common Features of All Three Plans 21
The Maintain Benefits (MB) Plan 22
The Two-Tier Plans 25
Further Comments 28

Conclusions 30

Acknowledgment 30

References 31

Preface to Political Leaders

For a national retirement system, pay-as-you-go funding has a number of inherent advantages. It is simple, efficient, and robust. It is "honest" in that tying current benefits to current wages recognizes the economic fact that, however the financing is arranged, each generation of retirees is supported by the nation's concurrent production of goods and services. For a national basic pension program, pay-as-you-go funding is superior to advance funding on all counts.

Why then is no one making the case for it in the current debate on Social Security? The discussion is focused on proposals which abandon pay-as-you-go and move toward privatization. There appear to be two distinct reasons:

Keeping the pay-as-you-go system requires acknowledging the need to correct Social Security by increasing the age of retirement and then indexing it to life expectancy so as to maintain a fair and economically sustainable ratio of workers to retirees. Most people see this as sensible and necessary for a "safety net" program, but relative to current law it can be portrayed as a cut because it halts what is really a continuing increase in benefits.

There are fortunes to be made if the nation can be persuaded to adopt an advance funded, privatized system. Those who stand to reap these profits have supported an unrelenting campaign to destroy public confidence in the current system while touting a privatized alternative. No matter that the campaign has been marked by exaggerated claims and misleading comparisons. It has been effective.

Why has there been so little response? Apparently, because no one will make any money out of keeping the current simple, efficient system.

The alternative to staying with pay-as-you-go and facing the demographic challenge directly is to adopt an advance funding strategy-counting on stock and bond returns to fund retirement benefits. Such an approach is unnecessarily costly and poses great risk for the system and the people who would depend on it.

Advance funding is intrinsic to any privatized system. A public plan which adopts advance funding takes a dangerous step toward privatization.

Hybrid and fully privatized plans, in which workers have separate accounts which they own and control, would exchange part or all of the present secure social insurance program for a tax-sheltered investment opportunity. The investment opportunity may be advantageous for upper income families, but for low and middle income families, especially young ones, it is no substitute for the assured benefits of Social Security.

Who would benefit by adoption of a privatized system? A system which mandates the flow of 10% of the nation's payroll annually into the stock and bond markets would be a bonanza for financial businesses and advertising firms. This is money which would otherwise go directly out to beneficiaries, or be left in the hands of the workers who had earned it. The ceaseless churning of these funds over a 30-year average holding period would make for enormous profits for the finance industry. Wealthy families who already have large holdings in equities would stand to gain from the stock price inflation likely to occur during the build-up period. Low and middle income families, lacking the best advice, often driven by fear or false expectations and without the resources to ride out market fluctuations, would be put at great risk. Even "playing it safe" risks serious losses due to inflation.

Policy makers should consider whether such outcomes fit the goals of a national pension system. It is hard to believe that the public would endorse such a program if the implications were honestly portrayed.

Seeking consensus in the middle will not do. Moving part way toward privatization takes the nation to a hybrid system which, already compromised, would be harder to defend than the existing one. It would have the same enemies and a weaker rationale.

For 60 years, Social Security has served the nation well. With modest adjustments, it can continue to do so indefinitely. History will not deal kindly with political leaders who preside over the dismantling of this excellent system in favor of one which is less dependable, far less efficient, risky for individuals, and regressive in its impact.

[ Preface to Political Leaders | Introduction | Making Pay-as-you-go Work | Privatization | Social Security Advisory Council Proposals | References ]

Summary

Introduction

Social Security faces real problems, arising mainly from demographics. These long-term problems can be dealt with most efficiently by making adjustments in the current system, but without changing its basic structure as a completely pay-as-you-go social insurance program.

The purpose of this paper is to show the logic and feasibility of this approach.

Pay-as-you-go vs. Advance Funding

A pay-as-you-go national retirement system has several inherent advantages over advance-funded systems. It has built-in protection against inflation, because pay-in and pay-out are both in the same current dollars. Backed by the commitment of the nation, it can respond to challenges of national scope as no market-based system can. With suitable adjustments, it can respond effectively and fairly to demographic trends and surges. In tying benefits to current wage levels, it stays close to reality, because current wages track the nation's current production of goods and services, the real determinant of the nation's living standard.

Advance-funded systems, lacking all these advantages, rely on financial paper of uncertain future value. Advance funding on a national scale is naive. It would use the power of government to move funds into financial markets at a rate far higher than that which market incentives could attract. Stock and bond prices would tend to rise artificially during the build-up phase, but false expectations thereby created could probably not be sustained. A demographic surge could produce a boom-bust sequence. There is no need to take this risk.

A further disadvantage is that going over to an advance-funded system involves a huge transition cost, amounting to trillions of dollars. There is no need to incur this cost.

The familiar argument that advance funding is desirable because it would provide needed investment in business cannot withstand scrutiny. To begin with, there is scant evidence for a scarcity of funds for investment in financial markets. Indeed, a stronger case can be made for increasing public investment in infrastructure, education, basic research, and health care.

Beyond this, money tied up for an average of 30 years or more in advance funding might be put to better use if left in the hands of individual families, as it would be under a pay-as-you-go system. Spending at the family level for education, health care, and housing can enhance national productivity while also contributing to the future well-being of families.

Making Pay-as-you-go Work

The principal change needed in order to address the demographic challenge is to assure a reasonable ratio of retirees to workers. Specifically, this means increasing the normal retirement age and reversing incentives which now encourage early retirement. The fundamental issue here is fairness to younger workers. Whatever the financing system, retirees are in fact supported by the goods and services produced by workers.

As life spans lengthen, we will not prosper by encouraging ever-earlier retirement and limiting job opportunities for seniors who want to work. With a growing backlog of unmet needs, we can and should reverse these trends.

Several other changes would improve equity while also helping to achieve fiscal balance. (1) Taxation of benefits on the same basis as taxation of other government and private pensions would provide a modest increase in progressivity while improving tax equity. (2) Extending mandatory coverage to newly-hired state and local government employees would move the system closer to the goal of universal coverage while improving balance in the near term. (3) Protecting the payroll tax base against further erosion through expansion of untaxed fringe benefits is a matter of fairness to workers whose compensation is mainly in cash.

COLAs should not be reduced as a cost-saving measure. While it is appropriate to adopt improved measures of the cost of living, COLAs should continue to include price increases attributed to improvements in the quality of products even if these are excluded from the CPI.

With these changes, the present pay-as-you-go system can continue to work well throughout and beyond the 75-year period of current projections. Dealing directly with problems which otherwise could cause the system to drift out of balance over time helps to assure that the system can remain strong far into the future.

Privatization

Privatization schemes bring all the disadvantages and costs of advance funding. In addition, they eliminate (or in the case of hybrid plans, severely weaken) the insurance provided by Social Security. They would replace it by a tax-sheltered investment system. Such systems predict only average benefits, so there is much uncertainty and risk to the individual.

Benefit predictions by advocates of privatization are highly exaggerated. They ignore the effect of demographic surges on stock prices. They overlook enormous administrative costs. They often neglect huge transition costs or pass them off to the general taxpayer. Comparisons of benefits with those of today's Social Security are misleading at best because they ignore the value and cost-effectiveness of insurance, which no investment account can replace. Often these comparisons fail even to mention the disability, survivor, and spousal benefits of Social Security.

Advisory Council Proposals

The three proposals offered by the Advisory Council on Social Security would all abandon pay-as-you-go financing. While only one has a large privatization component, the other two would take steps toward privatization. All three would build up substantial investments in the stock and bond markets, thus incurring considerable risk. All three would impose increases in the payroll tax of the order of 1.5%.

All these risks and added costs to workers are unnecessary.

Conclusions

Cost estimates based on data provided in the Advisory Council's report show that the pay-as-you-go plan proposed in this paper easily meets the test of long-term balance. There is no need for a payroll tax increase. There is no need for investment in private financial markets. What is needed is a willingness to face honestly the necessity of dividing America's work fairly across generations in future decades.

[ Preface to Political Leaders | Introduction | Making Pay-as-you-go Work | Privatization | Social Security Advisory Council Proposals | References ]

Introduction

A Vision

The challenge for a mandatory national retirement system is to assure an income for the elderly which is adequate for a decent standard of living, and to do this, despite the aging of the population, without unduly burdening younger workers.

To meet this challenge efficiently, the system must be simple, universal, and focused. Simplicity promotes efficiency. Universality allows the system to be simple and also helps to ensure that the program achieves its goals. Keeping the system focused on the single objective means that scarce resources are not diverted to other purposes.

To command broad public support, the system should also be fair within and across generations. To be effective, it should be robust under challenges of national scope.

Two Problems

Social Security faces two problems. The first is demographic-the aging of the population and the population surge known as the baby boom. Projections by the Social Security Board of Trustees are that 50 to 75 years from now, if there is no change in current law, the income stream will be sufficient to fund only about 70% of benefits.

More serious is the political problem. Those who stand to reap enormous profits by adoption of alternative schemes have waged an unrelenting campaign to destroy public confidence in the system. This campaign has fed on and added to widespread public confusion about the operation of the current system.

In truth, the current system is efficient and fundamentally sound. With changes in some of its rules, but not in its basic structure, it can perform better than any alternative which has been proposed. My purpose is to make this case.

Pay-as-you-go vs. Advance Funding

The Central Issue

There is one central issue: Pay-as-you-go vs. advance funding.

The two contrasting systems are illustrated simplistically in Fig. 1. In the pay-as-you-go system (approximated by Social Security now) the funds collected are paid out to beneficiaries the same year. The purpose of the trust fund is to ensure a smooth cash flow. It is rather like a checking account. Historically, the trust fund reserve has been of the order of the amount needed to cover one year's payments to beneficiaries.

Currently there is some deviation from pay-as-you-go because of yearly surpluses. The trust fund value is scheduled to rise to a maximum around 2013 and then to decline rapidly. However, at its peak, the accumulation is projected to be only about 3.1 times one year's payments.

In an advance-funded system, the trust fund is large enough to cover the present value of all benefits which have been earned but not yet paid. In practice, this means that payments into the system are invested for an average of about 30 years before the benefits are paid out. The size of the trust fund is therefore roughly 30 times that for the pay-as-you-go system. (Many factors determine the relative sizes, but a ratio of this general magnitude is to be expected.)

a)

b)

Fig. 1. Relative magnitudes of trust funds for a) pay-as-you-go system and b) advance-funded system. The pay-as-you-go trust fund, which plays the role of a checking account, is required only to assure a steady flow of funds to beneficiaries and typically is about equal to one year's pay-out. The advance-funded trust fund has accumulations equal to the present value of all benefits earned but not yet paid. At steady-state, it is roughly 30 times as large as the trust fund for the pay-as-you-go system.

Pay-as-you-go: The System of Choice

For a national retirement system, pay-as-you-go has several inherent advantages:

1. Since pay-in and pay-out are in the same current dollars, benefits can be so structured as to make inflation all but irrelevant. The purchasing power of retiree benefits is automatically tied to current wages. Current wages are as good an index as we have of the current cost of living at the current national standard of living.

In the existing Social Security system, indexing of wages in the calculation of the initial benefit protects against inflation over the period of worker payments, and COLAs protect against inflation over the pay-out period. In a pay-as-you-go system, COLAs as well as wage indexing are sustained by nominal wage growth.

No market-based system can provide this automatic, accurate protection against increases in cost of living.

2. The ultimate guarantee of a national pay-as-you-go system is the commitment of the nation to assure an adequate basic pension for its elderly. If challenges of national scope arise, such as war, depression, pestilence or global warming, the nation can respond, making the best use of its resources. No private, market-based system can do this.

3. A pay-as-you-go system can deal easily with population aging by indexing retirement age to life expectancy in such a way as to maintain an appropriate ratio of workers to retirees. Indeed, this is the only generationally fair and economically sustainable response to demographic trends.

Population surges (such as the baby boom phenomenon) will stress any retirement system. The inherent advantage of a pay-as-you-go system is that the effects on it of such surges are predictable and therefore can be dealt with fairly and efficiently. In contrast, predictions of effects of demographic surges on any market-based system are full of uncertainties, putting those who would depend on the system at great risk.

4. The purpose of a retirement system is to produce a flow of goods and services from current production to current retirees. A pay-as-you-go system recognizes this reality, in that income to the trust fund is directly related to current wages, which in turn are directly related to the nation's current production. There is no reliance on the value of financial paper.

Pay-as-you-go plans have gotten a bad name in the private sector, but there is an essential difference between the federal government and a family business or a corporation. Family businesses and corporations come and go, they expand and contract, they change hands. Their ability and commitment to make good on obligations extending many decades into the future cannot be assured. On the other hand, the nation is expected to endure. A national commitment which is so clearly in the national interest can be depended upon to maintain a pay-as-you-go system and allow the nation to enjoy its inherent efficiency, made possible by the direct link between current wages and benefits.

A Lesson from Recent History

With any retirement system, changed circumstances may require some adjustment in the parameters which determine income and benefits. To put the present situation in context, recall that during and after World War II, inflation seriously eroded the purchasing power of Social Security benefits. Methods adopted in the 1970s, wage indexing and COLAs, have virtually eliminated the effects of inflation.

Now demographic changes require adoption of measures which will maintain an adequate ratio of workers to retirees. If we deal with the demographic problem directly, any other problems will be tractable. If we do not do this, the effects will be far more severe than they need to be.

Earn More by Investing the Current Trust Fund in Stocks?

For a pay-as-you-go system, it doesn't matter very much how the fund is invested, as long as it is liquid and safe. Like interest on your checking account, the extra income won't change your life style.

To the extent that it matters, the trust fund does get market rates for a practically risk-free investment. For example, for the years 1993-96, the interest earned was around 8%. Treasury bonds are continually bought and sold on the market. The rates are competitive for a virtually risk-free investment.

Higher returns for the trust fund will make a real difference only if the surplus in the fund is increased at least several-fold so that the system becomes advance-funded to a significant extent. It will be argued below that such an approach is unwise.

Inequities in the Transition to an Advance-funded System.

There are significant inequities inherent in going from a pay-as-you-go to an advance-funded system. The problem is to build up the fund accumulation (Fig. 1b) while concurrently maintaining a flow of funds to sustain the phase-out of the pay-as-you-go system. The added burden roughly equals the cost of filling up the trust fund. This amounts to trillions of dollars.

Who would bear this burden? Those who advocate going to a funded system and who acknowledge the transition problem typically suggest doing it by some combination of reducing benefits to those in the current system and shifting the rest of the burden to the general taxpayer, usually through government borrowing.

Transition costs may be creatively disguised or shifted out of sight, but they are always there. Someone has to "fill up the box".

Advance Funding: Faith in Financial Paper

In an advance-funded system, the value of the funds is supposed to equal the present value of all benefits earned but not yet paid. The funds are invested in market securities. Benefits may or may not be tied to the performance of the investments, but if they are not, there is at least the expectation that investment returns will be sufficient to pay the benefits.

There are many ways of implementing such a system. There may be a single investment fund overseen by a government agency which is also responsible for paying benefits. Alternatively, individuals may have separate accounts which they own and manage. In the latter case, the account at retirement may be used to purchase an annuity or for other purposes.

The present section deals only with characteristics common to all advance-funded systems. The implications of various features which are included in some plans but not others are discussed in later sections.

Fundamentals. The essential weakness of all advance-funded systems is that they rely on financial paper to fund retirement benefits in a distant future. There is no logic in tying retirement income to the compounded values of stocks and bonds, which are strongly influenced by speculation, fluctuations in interest rates, and price inflation. Unlike wages, stock and bond returns have no clear relationship to Gross Domestic Product (GDP) or to national income. There is no fundamental reason for expecting that the value of such financial paper will track the cost of living.

Advocates of advance funding, citing historical records, assure us that returns will be more than enough to meet our needs. There is a certain unreality about this. If real stock returns over long periods are 6-7% and real growth in GDP is 2-3%, does this mean that by pouring enough money into the stock market we can earn so much that everyone can retire early in comfort? Obviously not. A relatively small segment of the population can feed on stock price inflation, as the paper will buy more goods and services. But as stock income becomes a more significant share of total national income, the feast must come to an end. The national standard of living can rise no faster than does the production of goods and services. It is not reasonable to expect that the baby boom retirement can be financed by bloated stock market returns.

To put this another way, there can be no substance to the promise that benefits tied to stock returns can rise faster than GDP, when the beneficiaries make up a large segment of the nation's population.

Effect of Demographic Surges on Financial Markets. If most of the nation's basic retirement system were to be advance-funded and invested in stocks and bonds, portfolio values would be affected by demographic surges. The initial effect would be positive, as the fund build-up occurred. In the next phase, the effect would be adverse as the tide reversed when a massive systematic withdrawal began. We have no precedent for these circumstances.

To see the problem more concretely, consider the scale of the investments being contemplated. Each year, some $350 billion now comes into the OASDI trust funds. Presently, private savings are on the order of $200 billion/year and undistributed corporate profits are roughly $100 billion/year. Thus, diverting even a substantial fraction of the OASDI payroll taxes into private financial markets could dramatically affect stock prices and interest rates. As a consequence, the value of the fund would be affected significantly by demographic surges which determine the rates of payment into the fund and of withdrawals.

Note: Bear in mind that with a balanced pay-as-you-go alternative, these funds would not be used for net purchase of Treasury bonds. They would either be left in the hands of workers or be paid out quickly to beneficiaries.

How well would this system cope with the baby boom phenomenon? The answer is obvious. Baby boomers pay in; stock prices rise. Baby boomers retire and sell to finance their retirement; stock prices fall, or rise less rapidly.

In these circumstances, there would be pressures for the government to take action to prop up the market. If it did act, then inflation might follow. In either case, retirees would find that their financial paper would not have the purchasing power they had counted on.

National Savings and Investment

Advocates of an advance-funded system claim that Americans are not saving enough and that diverting payroll taxes to investment in private-sector financial markets would create jobs and enhance productivity.

These assertions do not stand up under scrutiny. Consider the following points.

Not Investing Enough? Claims that we are not investing enough in the private sector are often based on faulty data. OECD reports (see reference list), widely used for comparisons among industrialized nations, do not include investments by federal, state, and local governments of the U.S. but do include such investments by most other nations. Thus, for the U.S., spending on highways, airports, port facilities, school buildings, and government-supported research laboratories are not counted as investment. The resulting picture distorts reality. Recent NIPA tables (see reference list) do report government investment data for the U.S., and they show that the U.S. lags behind Japan and Germany most strikingly in public-sector investment.

investment in Financial Markets. Where is the evidence for a shortage of funds for investment in financial markets? The rise in stock prices over recent decades suggests no lack of funds for private-sector investing. Diverting additional billions each week into the stock market will push stock prices up faster, and rising stock prices will attract more investing. There is no clear connection with either job creation or productivity enhancement.

Job Creation. Availability of funds cannot alone bring about business expansion. As often as not, the very purpose of investment by business is to reduce labor costs. Jobs may be lost or moved to low-wage countries.

Business expansion with net job creation in the U.S. will occur only as there is anticipation of increased consumer demand. Demand will not be increased if payroll tax receipts are diverted from beneficiaries, who will spend them, to the stock market.

Investments by Individuals and Government. The tacit assumption being made by those who would mandate increased investment in private-sector markets appears to be that spending by government and by individuals is pure consumption and contributes little or nothing to national productivity.

This narrow view of investment must be challenged. Investments by individuals in education, health care, adequate housing, and reliable transportation all contribute to productivity. Government investments in public transportation, education, health care and basic research are crucial to future productivity. The prevalent notion that we do future generations a favor by sharply curtailing such expenditures is absurd.

Investment vs. Consumption. Even the idea that it is always good to increase investment at the expense of consumption is simply wrong. Our future prosperity depends on a wise allocation of today's resources between consumption and investment. The family which eats its seed corn is looking for trouble, but so is the family which withholds nourishment from its children in search of future rewards.

Economic Inefficiency. An advance-funded system would use the power of government to divert to financial markets a far higher level of investment than market incentives could attract. Such intervention would create a huge market distortion and carries a heavy burden of justification.

Regressive Impact. Building up and then sustaining a huge trust fund requires diverting payroll taxes to financial markets, where they would be held and churned for some 30 years before being paid out to beneficiaries. The impact would clearly be regressive. Low and middle income families are likely to be harmed by a mandated shift of their limited resources from spending for needed consumption or for their own education, health care, or housing, and they are likely to be affected directly by reductions in government investment in such infrastructure as mass transportation, public parks, and public libraries. Wealthy families can readjust their investment portfolios to maximize after-tax returns or meet other objectives. Their total savings may be changed very little.

Overview. It would be unjust and unwise to mandate that a regressive tax be used to channel funds extravagantly into private-sector markets. It would be especially bad policy to tie our hands by committing to such a diversion for the indefinite future.

Two-Tier Systems

Various two-tier or hybrid systems have been proposed. The first tier is a government-run pay-as-you-go pension system intended to guarantee less than adequate basic support. The second tier is an advance-funded system intended on average to bring the total benefit up to an adequate level. The second tier is funded by mandatory payments into an individual account which the worker invests, perhaps with some restrictions. At retirement, the worker uses the account for purchase of an annuity or for other purposes.

In short, the first tier would be a social insurance plan while the second would be a tax-sheltered investment opportunity with considerable risk. Thus, two-tier systems draw scarce resources away from serving the main purpose of a national pension system, which is to assure an adequate level of basic support.

Few question the need for a mandatory system sufficient to meet basic needs of beneficiaries. A simple, focused system does this best. Two-tier systems are less efficient because they divert resources to serve other purposes.

[Preface to Political Leaders | Introduction | Making Pay-as-you-go Work | Privatization | Social Security Advisory Council Proposals | References ]

Making Pay-as-you-go Work

What to Do

The only effective approach will be to address the demographic problem directly. This means adjusting the age of retirement so as to maintain an adequate and fair ratio of workers to retirees. The other measures recommended here can also be regarded as matters of equity.

Taken together, these proposals are more than sufficient to restore the Social Security system to a condition of long-range fiscal balance. The contribution of each proposal is shown as a percentage of the total adjustment required for the achievement of balance. These percentages are only rough estimates, but they provide an idea of relative magnitudes.

Note: Estimates are based on data provided in the report of the 1994-96 Advisory Council on Social Security (Vol. I, Appendix III), the staff report of the Bipartisan Commission on Entitlement and Tax Reform, and the 1997 report of the Board of Trustees of the OASDI trust funds.

· Gradually increase the normal retirement age (NRA) and then index it to life expectancy at the NRA. (45%)

For example, beginning in 2002, increase the NRA at the rate of 2 months every year until it reaches 69 in 2030; then index it to life expectancy at the NRA. (This would mean a life expectancy at retirement of about 16 years or more for those retiring in the period 2020-2070, compared with 17.3 for those retiring in 1998.) Increase the age of eligibility for early retirement benefits (EEA) at half the rate of the NRA increase.

· Replace present incentives for early retirement by incentives and opportunities for continuing work, including part-time. (40%)

This includes extending the number of years for which wages are averaged in determining the initial benefit, and eliminating penalties for earnings after retirement income has begun.

These proposals are intended to go beyond just changing Social Security rules (although some changes are needed). Our future prosperity and productivity depend on reversing the current trend toward early retirement. This is discussed further in later sections.

· Make benefits subject to income taxation on the same basis as other government and private pension income. (10%)

This means including in taxable income all amounts received in excess of previously-taxed contributions by the employee. This is a matter of simple tax equity. It also increases the progressivity of the net benefit much more fairly than does means testing. New revenues from this source would go into the OASDI trust funds.

Note: Current revenues from taxation of benefits go partly to the OASDI trust funds and partly to the Medicare HI trust fund. The cost estimate above relates only to additional revenues.

· Avoid further erosion of the payroll tax base through expansion of fringe benefits not subject to the tax. (20%)

A significant factor contributing to the long-range funding shortfall is the continuing expansion of forms of compensation exempt from the payroll tax. Exclusion of fringe benefits from the payroll tax base requires rates to be higher than necessary and is unfair to workers whose compensation is all in cash.

· Extend Social Security coverage to newly-hired state and local government employees. (5%)

The success of Social Security is due in part to the fact that it is nearly universal. Coverage is already mandatory for most private-sector and federal government employees.

Spousal Benefits. It is generally recognized that these are over-generous in some cases and inadequate in others. Consideration should be given to adopting an "earnings sharing" plan with earnings credits awarded for rearing children (Steuerle and Bakija). This is so clearly an equity issue that it may be useful to construct a budget-neutral solution so that it can be considered separately and presented to the public as a plan to improve the fairness of the program.

What Not to Do

· Do not subject benefits to means testing.

Means testing would turn Social Security into a welfare program and erode support for it.

There is also an equity issue. The form usually proposed, in which benefits are sharply reduced as total taxable income rises, is unfair because it imposes much higher marginal tax rates on beneficiaries than are applied to everyone else. Social Security income should be taxed on the same basis as other income, as suggested above.

· Do not raise the payroll tax rate.

Maintaining a reasonably constant payroll tax rate is a matter of fairness across generations. A constant rate means approximately constant sharing of goods and services between current workers and current beneficiaries.

The payroll tax rate is now about 1% higher than it should be for a pay-as-you-go system, although this situation is projected to continue only until about 2020. While the tax could be reduced by that amount for the period 2000-2020, it may be preferable to use the surplus to cover transition costs associated with introduction of reforms such as earnings sharing.

· Do not reduce COLAs by political action as a cost-cutting measure.

COLAs should reflect the true increases in cost of living as accurately as possible. Providing less than this defeats the very purpose of a basic pension. Because the effect is cumulative, it is especially hard on those who live many years after retirement.

COLAs and the Adjustment of the CPI. The Advisory Commission to Study the Consumer Price Index (CPI) estimated in its report (1996) that the CPI as previously calculated has overstated increases in the cost of living by about 1.1% per year. However, 0.6% of this bias is attributed to quality improvement (see p. 68 in the report). The cost of improvements in quality of products is properly considered to be part of the increase in cost of living. Therefore, the correct figure for adjusting Social Security COLAs relative to the previous CPI is 0.5% rather than 1.1%.

The Bureau of Labor Statistics is already making a technical correction which will have the effect of lowering the CPI by about 0.5%. This may be justified, but no "quality" corrections should be made.

Price increases which reflect improvements in quality of a product are not inflation, but they are true increases in the cost of living.

Would the Suggested Changes Suffice?

Yes. Since the proposed adjustments add up to 120% of the total required, there is room to spare.

Furthermore, there is no reason to expect that the system would drift out of balance even beyond the 75-year period for which projections are made. Two major factors which could cause long-term drift are aging of the population and erosion of the payroll tax base. Both are addressed directly in these proposals.

Projections by the Advisory Council of the effect of various proposals for advancing the retirement age include estimates of cost savings but assume no increase in revenues from payroll taxes. If the incentives and expanded opportunities proposed above and elaborated in later sections are effective, there will be a significant increase in participation of seniors in the workforce, relative to what is now projected. Seniors, earning more, will pay more taxes and may also earn higher Social Security benefits. There should be a net increase in income to the system as well as a benefit to the economy.

Reversing the Culture of Early Retirement

There is much evidence that current Social Security rules encourage early retirement and discourage work after retirement (Steuerle and Bakija; 1994-96 Advisory Council report, Vol. II, pp. 12-27). Policies of many employers contribute to the trend toward early retirement and also limit opportunities for work by people over 50. Even opportunities for part-time work are limited. The culture of encouraging early retirement and excluding seniors from the workforce must be reversed.

The above proposals for delaying retirement will not achieve their objectives unless there are jobs available for those who want to work. Both government and private sector employers should work in this direction. Many retirees in the 21st century will have computer skills and experience in service industries, so that work in later years should be an increasingly attractive and viable option.

Many older workers would welcome opportunities for working at a reduced pace. To maintain adequate income in such circumstances, Social Security benefits might be phased in so as to provide a fraction of the normal full benefit to supplement the part-time wage of a senior moving toward retirement. A similar phase-in of Medicare coverage would help to make employment of seniors more attractive to employers by reducing their cost for health insurance (Steuerle and Bakija).

America will not achieve prosperity by retiring larger and larger fractions of its potential workforce. Support must be provided for those who need to retire for reasons of health, but opportunities must be assured for those who want to work longer.

This conclusion is not altered by contemplation of future productivity gains. The benefits of such gains should be spread across the entire population, in part through shorter work weeks or longer vacations. They should not go mainly to the support of an increasing proportion of retirees.

Maintaining the Ratio of Workers to Retirees

Maintaining an adequate ratio of workers to retirees is essential if the system is to be both sustainable and fair. This is true for advance-funded systems as well as for pay-as-you-go systems because, no matter how the financing is arranged, retirees are supported by the goods and services produced by those who are working.

One way of judging sustainability and fairness is to compare the expected number of years in retirement for someone at the normal retirement age (NRA) with the potential number of years of work (taken to equal the NRA minus 20). These data are given in the following table for three cases. The first case shows what the situation would be if the NRA had been left at 65. In this case, all the increase in longevity goes into the retirement period. The last column, which gives the ratio of the numbers in the preceding two columns, shows the expected number of years in retirement for each year in the workforce.

The second case illustrates the situation under current law, which provides for a gradual increase in NRA from 65 to 67, with no further change. After 2027 (year of retirement), all increases in longevity go toward increasing the retirement period.

The third case shows the effect of adopting the proposal suggested above. After reaching age 69 in 2030, the NRA would be indexed to a constant ratio. This would mean that for every four years added to longevity, three would be in the potential work period and one in the retirement period. Thus, longevity gains would be "shared" between work and retirement.

Table. Potential number of years working and expected number in retirement for several cases.

Case 1: Retirement at 65

Case 2: Current Law

Case 3: Proposed plan*

*Starting in 2002, increase NRA by 2 months every birth year until it reaches 69 in 2030, then index to constant ratio (last column).

[Preface to Political Leaders | Introduction | Making Pay-as-you-go Work | Privatization | Social Security Advisory Council Proposals | References ]

Privatization

Steps Toward Privatization

A system with a single trust fund invested by a government agency in private financial markets has many political liabilities. If a proposal along these lines were to be adopted, there would be ceaseless pressure to move on toward a privatized system.

In the first place, many people have strong misgivings about empowering a government agency to invest so heavily in the private sector. The possibility of bungling and the opportunities for mischief are too great.

Secondly, there would be the constant cry that returns could be improved if individuals could own and manage their accounts. In a situation where higher returns can be expected, on the average, by acceptance of increased risk, prudence is difficult to defend. Thus, political pressures would build to give the individual both ownership of an account and some degree of control over its investment.

The underlying reality is that the financial businesses which would profit from privatization would gain little from an advance-funded public system. The opportunities for profit would come from the multiplicity of choices and constant churning of a privatized system. These interests have the resources to wage an insistent campaign to move public opinion toward privatization.

Fundamental Issues

In the present context, a privatized system is one in which the worker makes mandatory payments into an individual account which he or she owns, with considerable freedom as to how the funds are invested. At retirement, the account may be used in any way.

Privatized systems have all the shortcomings of any advance-funded system as described above, along with others associated specifically with ownership and individual control of the accounts. The major points will be summarized here.

1. Sensitivity to Market Fluctuations. While any advance-funded system will be affected by gradual movements in stock prices caused by the business cycle or demographic changes, individually invested accounts are at the mercy of day-to-day fluctuations as well. Fluctuations matter for anyone dependent on the account for retirement income because income would be frozen at a level determined by the value of the individual's account on the day on which it is converted to an annuity.

The stock market declined 70% over the period 1929-34 and it dropped 30% in only a few days in the 1987 crash. The knowledge that the market always seems to recover-perhaps within a few weeks, perhaps over a period of years-would be small comfort to the worker who is about to annuitize an account when the market drops. Wait, or act quickly?

Such circumstances make it impossible for the plan to serve its primary purpose-to assure a dependable level of income for the retiree or a beneficiary.

2. Ownership. Ownership detracts from the central purpose of a national pension system, which is to assure an adequate income after retirement and to provide insurance support in case of disability or death. Scarce resources should not be diverted to other personal purposes such as building an estate or engaging in tax-sheltered speculation.

3. Choice. Choice of investment vehicle is clearly a benefit for wealthy people who can afford to accept risk and who have the advantage of the best financial advice. It is a disadvantage for those low and middle income families who have neither of these advantages and who are therefore more likely to misread market signals and to be taken in by false and misleading advertising.

If there is variety in the annuity plans offered, most consumers will find it difficult to compare plans which have different advantages and limitations. Will the widow who has just lost a spouse remember to ask about inflation protection? How secure are the guarantees? If the basic purpose is to assure adequate lifetime income, why encourage diversion of resources to other purposes?

While it is politically attractive to offer each individual the opportunity of choosing his or her own level of risk, there will be winners and losers. Whose responsibility will it be to provide for the losers and their dependents?

There is also the matter of efficiency. Choice complicates administration and brings in substantial costs for advertising, market research, client consultation, handling of transactions, and profits. Net benefit payments would suffer.

A mandatory national investment program with choice for investors is a bonanza for financial institutions, insurance companies, and advertising firms. Money invested for an average of 30 or more years would be ceaselessly churned. The churning is driven by speculation and fear, with a time horizon seldom exceeding six months.

It is generally accepted that the basic retirement system must be mandatory-that individuals should not be given the choice of participating or not. For assuring basic lifetime income after retirement and some support for dependents, simplicity is a virtue. "One size fits all" is exactly the right approach. Thoughtful leaders should make the case for and support limiting choice.

4. Insurance vs. Investment. Privatization turns a social insurance program into a tax-sheltered investment opportunity. Social Security now seeks to assure basic disability, survivor, and retirement benefits for every worker and necessarily is redistributive. Its claim to fairness rests on the proposition that no one can be sure that personal savings will meet these needs. Privatization may appeal to some who expect never to need those guarantees, but no one is immune to misfortune bringing a devastating loss of savings and income.

Comparison of Benefits

The chief argument advanced for replacing Social Security by a privatized system is that individual ownership and control of accounts offer the opportunity of earning far higher retirement benefits than would be provided by Social Security. In fact, the claims are highly exaggerated and the comparisons are often misleading.

1. Projections of total compounded returns over periods of 60 years or more are extremely sensitive to assumptions about annual rates of return and inflation. Advocates of privatization emphasize returns for all-stock portfolios, which are inappropriate for assuring the basic "safety-net" retirement income.

To illustrate the point, increasing the assumed real return rate from 4% to 6% roughly doubles the projected monthly benefit. This would be the difference between a rate for an all-stock portfolio and the rate for a conservative mixture, say 40% stocks and 60% high-quality bonds. Thus, assuming an all-stock portfolio inflates the projected benefit by a factor of around two.

2. Advocates typically assume annuity returns for the pay-out period which do not allow for inflation. For Social Security, that would be equivalent to eliminating COLAs. If the inflation rate is 4%, the purchasing power of an annuity after 25 years is only 37% of its initial value. For an elderly retiree dependent on such an income, the effect would be devastating. If the annuity were re-structured to provide lifetime income of constant purchasing power, the benefit level would be reduced by about 25% below the initial amount given for benefits without such protection.

3. Advocates of privatization seldom mention the spousal benefits, survivor insurance, or disability insurance benefits of Social Security. The latter two insurance benefits are especially valuable to a young low or middle income family. For such families, ownership of the account can not begin to provide the equivalent of this insurance protection. Purchasing equivalent insurance as part of a private plan would reduce the pension benefit by at least a third. For a one-earner couple, regardless of income level, the average spousal and survivor benefits are worth more than the worker's benefit (Steuerle and Bakija, Table A.3).

4. Social Security benefits allow for administrative costs, which are less than 1% of pay-out amounts. Estimates of benefits made by privatization advocates seldom mention administrative costs. An estimate of such costs might be based on charges made by mutual funds, typically in the range 1-3% of the accumulated account value each year If one assumes a rate of 2%, the effect is to reduce the annual return by that amount relative to annual returns estimated from stock indexes such as the Dow or S&P 500. As a rough guide, a 2% reduction over a 46-year pay-in period reduces the projected retirement benefit by about 40%.

Even this estimate does not allow for the cost of keeping track of and reporting large numbers of relatively small federally-mandated deposits.

5. Comparisons of returns are often flawed by lack of consideration of transition costs. When they are mentioned at all, the usual recommendation is a combination of benefit cuts and shifting of huge costs to future general taxpayers.

Conclusion. If an appropriately conservative portfolio is assumed and allowance is made for administrative costs, COLAs , the value of disability insurance, survivor insurance, and spousal benefits, and transition costs, it becomes clear that the benefits for most workers in an alternative privatized system would be well below those from Social Security. Only high income or independently wealthy workers could expect to come out much better.

The National Pension System as a Tax-Haven for the Rich

There is just one sense in which comparisons made by advocates of privatization are correct. Wealthy individuals could expect to profit from a privatized system by using it for tax-sheltered investment in speculative portfolios. With expert financial advice and enough resources to be able to ride out market fluctuations, they could, more often than not, achieve higher benefits than would be available from Social Security. Under many privatization proposals, they could pass unused accumulations tax-free to heirs.

Surely such outcomes do not fit the goals of a national pension system.

[Preface to Political Leaders | Introduction | Making Pay-as-you-go Work | Privatization | Social Security Advisory Council Proposals | References ]

Social Security Advisory Council Proposals

The 1994-96 Advisory Council Report offers three proposals, each supported by several members. Let's look at these proposals from the viewpoints developed previously in this paper.

Common Features of All Three Plans

All Council members concluded that means testing is a bad idea. All support the principle that COLAs should be adjusted according to technical criteria to match the cost of living as accurately as possible, but should not be reduced through a political process as a cost-cutting measure. All three plans would extend OASDI coverage to state and local government employees hired after 1997. All these recommendations are in agreement with proposals supported in this paper.

All plans assume a COLA adjustment of 0.21% downward relative to the CPI as previously calculated. This figure was the most recent Bureau of Labor Statistics (BLS) estimate available when the calculations were started and was believed to be a conservative estimate of future corrections. The report of the Advisory Commission on the CPI had not yet come out.

All three of the Council's plans involve advance funding to some degree, with investment of a part of the assets in stocks. These and other features which seem to go in the wrong direction will be reviewed critically in this section.

I will start with the plan which, despite some serious flaws, comes closest to meeting the basic criteria set forth here.

The Maintain Benefits (MB) Plan

Under this plan, supported by six Council members, the current structure and level of benefits would be maintained. It would remain a defined-benefit program.

The elements of this plan are as follows:

1. Adjust COLAs according to revised BLS calculations of the CPI. For making projections, the Council assumed that the adjustment would reduce COLAs by 0.21% after 1997.

2. Cover under OASDI all state and local government employees hired after 1997.

3. In calculation of average wage for determination of benefits, increase from 35 to 38 the number of years included. Phase in from 1997 to 1999.

4a. Subject benefits to federal income taxation in the same manner prescribed for private and government defined-benefit plans. Phase in between 1998 and 2007.

4b. Redirect revenues from taxation of benefits from the Medicare HI trust fund to OASDI. Phase in from 2010 to 2019.

5. Invest a portion of the OASDI trust funds in stocks beginning in 2000, reaching 40% of assets in stocks for 2015 and thereafter.

6. Increase the payroll tax rate in 2045 by 0.8% each for employees and employers.

These proposals together are projected to eliminate the deficit in OASDI funding over a 75-year period and also meet a long-range criterion designed to ensure that the system will not rapidly drift out of balance in subsequent years. Despite all this, the plan does not provide a satisfactory solution.

Failure to Address the Demographic Problem. The MB plan offers a financial fix without addressing the underlying cause of most of the problem. By 2070, retirees would spend well over a quarter of their adult years in retirement, supported by the workers of that day. As life spans continued to grow, the burden on workers would continue to grow.

The MB plan seeks only to find the money to compel an ever-increasing flow of goods and services from future workers to an ever-increasing proportion of retirees. This approach is neither fair nor economically sustainable.

Weakness of the Advance Funding Strategy. Rather than face up to these facts, the MB plan would try to prop up the retirement system by expanding the trust fund through investment in equities. The vulnerability of advance funding strategies to demographic effects has already been emphasized. In the present instance, the flow of OASDI funds into the stock market would change from a net investment growing to nearly $100 billion per year in 2014 to a net withdrawal of over $100 billion per year by 2020 (in current dollars) as the demographic tide reversed.

These changes are not insignificant in comparison with the changes in net flow of new private funds into the market. If a nervous market showed signs of a systematic decline, or even of a slowing rate of growth, what would be the reaction of private investors? Would foreign and U.S. investors seek better opportunities for growth in other markets? Would there be pressure to invest some of the trust fund assets in foreign markets in order to "diversify" and get better returns?

The inherent weakness of the MB plan is evident from the need to add a payroll tax rate increase. According to estimates which do not take into account the market reactions just described, the first five of its provisions (1 through 5 above) would bring the system into balance over a 75-year period, but the system at that point would be going out of balance again. The payroll tax increase beginning in 2045 only staves off the day of reckoning.

Fairness. This issue has already been alluded to. As people live longer, it will not do to have a growing proportion of able retirees supported by younger workers. Even if gains in productivity were to make such an outcome possible, it would not be fair. A major share of any productivity gains should go toward improving the quality of life of younger families, in part through a shorter work week or longer vacations. The way to do this is to keep a reasonable ratio of workers to retirees.

The Pay-as-you-go Alternative. If the stock-purchase approach is replaced by an alternative which addresses the demographic trend adequately, the payroll tax increase will not be needed. The appropriate alternative is to accelerate the increase in normal retirement age and to stop encouraging early retirement and penalizing work after retirement.

Projections contained in the Council's report (Vol. I, Appendix III) make clear that such changes could easily achieve fiscal balance without the stock-purchase option. Further, by addressing the long range problem at its roots, this alternative obviates the need for the oppressive payroll tax increase.

Opposition to Raising the Normal Retirement Age. While eight of the 13 Council members support some increase in the NRA, five of the six who supported the MB proposal are opposed to raising the retirement age.

There are legitimate concerns. Some workers, for reasons of health or because of the nature of their occupations, may find it difficult or impossible to work beyond age 65, at least at the jobs they had held earlier. Finding jobs is not easy for people in their 60s. Increasing the NRA has a regressive impact because of the correlation between lower income and shorter lifespan.

Despite the compelling nature of these concerns, it would be increasingly costly and inefficient to deal with health problems and unemployment by retiring everyone at an early age. These problems must be addressed and solved outside the retirement system.

The key elements of any solution must be to expand efforts to assure good health care and job opportunities for all. Joint efforts of government, business, and non-profit institutions may be required. The Supplemental Security Income (SSI) program should provide a "safety-net" for seniors who for health reasons cannot work.

It will be essential to broaden opportunities for part-time work for seniors, perhaps supplemented by phasing in partial retirement benefits (Steuerle and Bakija). Joint public-private efforts must be made to assure availability of jobs. With increasing needs for health care, day care, and other services for children and the elderly, it will be unacceptable to have large numbers of people able and willing to work while at the same time real needs are not being met.

One often hears the argument that older people should retire and leave the jobs to younger workers. What this really comes down to is "Let the younger workers produce the goods and services, while seniors who could work and would like to earn more should just retire and live on their pensions." In fact, what would help everybody would be for those seniors who can work to do so, adding to the national product, paying payroll taxes, consuming more, and thereby helping to create jobs for others.

The Two-Tier Plans

The other two plans proposed by the Council share important characteristics and will be discussed together.

Both are two-tier plans, with one tier being a defined-benefit plan structured like Social Security but with substantially reduced benefits, the second tier being an advance-funded defined-contribution plan with workers having ownership and some control of separate accounts.

Individual Accounts (IA) Plan. In this plan, supported by two Council members, the first tier provides OASDI benefits reduced to about 75% of the levels defined in present law. While this tier is basically a pay-as-you-go system, there is some advance funding through further accumulation of an OASDI trust fund surplus invested in Treasury bonds.

For the second tier, workers would be required to make an additional 1.6% payment (not matched by the employer) into one of several government-regulated but privately-managed investment funds. Upon retirement, the worker's accumulated balance would be used for purchase of a lifetime annuity. The benefit from the second tier is intended on average to make up for the cuts in the OASDI part, but this result would be contingent on good performance of the stock and bond markets relative to inflation. The outcome for each individual would also depend on the investment strategy selected and on the condition of the stock market on the date of conversion to an annuity.

Although the IA plan in its second tier has many of the characteristics of privatization, the rather severe constraint on investment choices makes it less risky than the PSA plan (see below) and public management of the funds makes it more efficient.

Nevertheless, it has all the disadvantages associated with diversion of resources to an advance-funded investment tier. For the well-to-do, the effect of mandating an investment tier would be mostly to allow shifting of investments from stocks and corporate bonds to tax-sheltered IA accounts. For low and middle income workers, the impact would often be a diversion of scarce resources from spending for education, health care, nutrition, or housing, to the government-mandated investment in financial markets.

Personal Security Accounts (PSA) Plan. This plan is supported by five Council members. The first tier is an emaciated OASDI plan, with benefits reduced to about 45% of current levels (at or below poverty lines). The funding for this tier would become strictly pay-as-you-go.

The second tier would be a completely privatized system. The worker would direct 5% of payroll (part of the employee's current 6.4% tax) into a "personal security account" and would control its investment. "Security" is a euphemism, as the only secure benefit is that from the first tier. After reaching the retirement age, the owner could use the account in any way.

The retirement system would borrow some $2 trillion or more from the U.S. Treasury to finance the acknowledged part of the transition cost. The bonds would be paid off by an added 1.52% payroll tax (split equally between employee and employer) for 70 years.

The PSA plan suffers from all the defects of a two-tier plan with a major privatization component which have been described in earlier sections. Although its supporters claim that the second tier would more than make up for the severe cut in OASDI benefits, workers and their families would be at the mercy of long-term market performance, fluctuations, and inflation. They could lose out badly by being either too adventuresome or too timid in choice of investments.

Further concerns about the PSA plan:

1. Advocates of the PSA plan assert that low income workers would be well treated, but they acknowledge none of the regressive aspects of such plans which have been emphasized in this paper.

2. Allowing for costs of handling accounts, advertising, and profits would probably reduce the projected average benefits at least by half.

3. PSA plan advocates never mention deception, fraud, or default. Most of the financial wizards who gave us the Savings and Loan scandal are still on the street. Any night on AM radio one can hear something like "Invest $5,000 now and get back $20,000 in six weeks. (There is some risk.)". Regulation of such a complex system as PSA would be difficult and costly. Insurable accounts would be so conservative as to invite devastation by inflation. Even modest regulatory fees paid by the industry and transaction fees paid by account holders would-because of their compounding effect over decades-drastically erode benefits.

Other objections to the PSA plan are thoroughly aired in the Council's report by members opposed to it.

Insurance vs. Investment. Advocates of both of the two-tier plans persist in judging Social Security as an investment. It is not an investment plan, it is social insurance. It insures the worker and his or her family against loss of income due to disability, retirement, and death of the worker.

Insurance is inherently a defined-benefit system. Anyone who doubts this attribute should contemplate the purchase of a defined-contribution fire insurance policy for the family residence. If the house burns down during the first year, you get one year's premium plus interest. Over time, as additional premiums are paid and investment returns accumulate, the account value rises and the insurance amount might even reach 100%. As an investment, it might look great, but it would not meet the needs of most people for fire insurance.

It is also curious that the advocates of the IA and PSA plans seem unable to appreciate the difference between having insurance and having the money which might have been used to purchase insurance. The difference would be clear enough to anyone whose house had just burned down.

Taxation of Benefits. A particularly objectionable aspect of both the IA and PSA plans is the discriminatory way in which they tax benefits. Benefits from the OASDI tier would be taxed under "income tax principles" while benefits from the second tier (the individual or personal accounts) would be taxed according to "consumption tax principles". Translated, this means that, for OASDI benefits, any amounts received above the amount originally contributed by the worker would (appropriately) be subject to income taxation. On the other hand, none of the second tier benefits would be taxed.

The consumption tax principle alluded to here is that "income is taxed once and only once". Translated, this means that wage income is taxed, while returns on capital investments are never taxed. This may be recognized as the theme of various "flat tax" proposals.

Diversion of Resources. The two-tier systems, as already emphasized, divert resources from the unique role which a national insurance system can fulfill, that of providing a defined-benefit plan to assure certain levels of support. There is no justification for such a diversion, especially when the money comes from a regressive tax and the second tier is likely to benefit primarily upper-income families.

Further Comments

Advance Funding. A good measure of the extent of advance funding is the ratio of total assets at any one time to the cost of benefits for one year. For a pay-as-you-go system, the ratio is expected to be in the range 1.0 to 1.5. The ratios for the Council's three plans after they have been fully phased in can be estimated from data given in the report. They are

MB 4.5

IA 15

PSA 30.

By way of contrast, under current law, the ratio would rise to about 2.4 and then decline again as the baby boom retirement phase began. If proposals urged in this paper were adopted, the ratio would remain in the normal pay-as-you-go range thereafter.

Impact of Advance Funding on the Market. This concern as it relates to the MB plan was discussed above. In the case of the IA and PSA plans, the level of market investment in the individual or personal accounts grows to be much greater than that for the MB plan. The concern is for systematic effects due to demographic changes. The phasing of the build-up of the IA and PSA accounts appears to be timed so as to avoid the flow reversal of the MB plan, but the sustained heavy rate of investment in the stock market would contribute to inflation of stock prices during the build-up phase. The PSA plan also impacts financial markets through first building up and then paying off a huge government debt. This could have a major secondary effect on the stock and bond markets.

Payroll Tax Increases. All three proposals of the Council include payroll tax increases of about 1.5%. For the MB plan, the increase is the price of not addressing the demographic challenge directly.

For the IA plan, the tax increase is the cost of setting up the individual accounts. (The supporters of the plan do not call this mandated investment a tax.)

For the PSA plan, the tax increase, extending over 70 years, would pay the acknowledged part of the cost of transition to advance funding.

Long-Term Drift Out of Balance. The Council report expresses concern over the tendency for the system to go out of balance as each new 75-year projection is shifted further into the future. The report refers to this as the effect of passage of time. At least three contributing causes are clear:

1. Increasing longevity coupled with a trend toward earlier retirement.

2. The increasing use of fringe benefits as a form of compensation which escapes the payroll tax.

3. Poor real wage growth for low and middle income workers.

It is disappointing that only 8 of the 13 members were willing to endorse a step to address the first item, which is the major cause of the problem. Only one Council member suggested doing anything about the second. The Council may have considered the third item to be out of its purview, but it found that the drop in real wage growth since 1983, relative to what had been assumed at that time, accounts for about 23% of the long range deficit now projected (Vol. I, p. 164).

Poor real wage growth for low and middle income workers has not been an accident. It is the consequence of consistent monetary and labor policies designed to assure a generous supply of low-cost labor. It is to be hoped that recently adopted welfare policies will not be allowed to further depress wages.

Conclusions

The public has been presented with a number of proposals on Social Security which would scrap the present system or would set it on a course toward privatization. This is madness. It is clear from the data presented here that there is no need to give up the inherent advantages of the simple, efficient pay-as-you-go system.

Furthermore, the major change needed, which is to maintain a reasonable ratio of workers to retirees, will have to come about under any plan if the system is to be generationally fair and economically sustainable.

The existing system is efficient, robust, and well-focused on the objective of providing workers' basic income protection covering disability and retirement, with survivor and spousal benefits. We should make the adjustments needed to preserve it.

Acknowledgment

I thank Midge Miller for suggesting the preparation of this paper and for stimulating discussions on the subject. Of course, I remain fully responsible for the content and opinions expressed herein.

[Preface to Political Leaders | Introduction | Making Pay-as-you-go Work | Privatization | Social Security Advisory Council Proposals | References ]

References

1994-96 Advisory Council on Social Security, Volume I: Findings and Recommendations; Volume II: Reports of the Technical Panels. (Washington, DC, January, 1997). Edward Gramlich, Chairman.

Advisory Commission to Study the Consumer Price Index, "Toward a More Accurate Measure of the Cost of Living". Final report to the Senate Finance committee (December, 1996). Michael J. Boskin, Chairman.

Bipartisan Commission on Entitlement and Tax Reform, "Staff Report on Entitlement Reform Options" (Washington DC, December, 1994). This two-part report suggests many options for reform. Useful in providing rough estimates of cost savings for various options.

Bipartisan Commission on Entitlement and Tax Reform, "Final Report to the President" (Washington DC, January 1995). Senator J. Robert Kerrey, Chairman, Senator John C. Danforth, Vice-Chairman.

Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds, "1997 Annual Report" (U.S. Government Printing Office, Washington DC, 1997).

Karl Borden, "Dismantling the Pyramid: The Why and How of Privatizing Social Security", Paper SSP No. 1 (Cato Institute, Washington DC, 1995).

Bureau of Economic Analysis (BEA), Department of Commerce, "National Income and Product Accounts (NIPAs)", published in Survey of Current Business.

Dan Cornwell, "Pay-As-You-Go: The Way to Go With Social Security" (The Madison Institute, Madison, WI, October, 1997). This paper displays the fiscal balance of the current proposal over the time period 2000-2075.

Congressional Budget Office, "Implications of Revising Social Security's Investment Policies", CBO Paper (Washington, DC, September, 1994)

Organisation for Economic Cooperation and Development (OECD), Department of Statistics, "Quarterly National Accounts" (Paris).

William G. Shipman, "Retiring with Dignity: Social Security vs. Private Markets", Paper SSP No. 2 (Cato Institute, Washington DC, 1995).

C. E. Steuerle and J. M. Bakija, "Retooling Social Security for the 21st Century", (Urban Institute Press, Washington DC, 1994). This excellent book describes the background, operation, and problems facing Social Security. It also discusses many options for reform and suggests principles for judging their merit. The authors present extensive tables of estimated lifetime taxes and benefits for various cohorts.

[Preface to Political Leaders | Introduction | Making Pay-as-you-go Work | Privatization | Social Security Advisory Council Proposals | References ]