Chairman Smith, Senator Kohl, and members of the Committee, thank you for inviting me to discuss the future of Social Security. I am here representing The Concord Coalition, a nonpartisan organization dedicated to strengthening the nation's long-term economic prospects through sound and sustainable fiscal policy.
Concord's co-chairs are former senators, Warren B. Rudman (R-NH) and Bob Kerrey (D-NE). They, along with Concord's President former Commerce Secretary Peter G. Peterson and our nationwide membership, have consistently urged Washington policymakers to produce a credible plan for dealing with Social Security's long-term challenges in a fiscally responsible and generationally equitable manner.
My testimony today will address the three questions posed in your invitation:
- How do the issues facing Social Security's future fit into the larger retirement security challenge?
- What are the consequences of delaying action?
- What kinds of broad prescriptions for change would be the most effective?
For over 65 years Social Security has provided a vital floor of protection. Its broad range of retirement, disability, and survivors' benefits for millions of Americans makes it an important issue for people of all ages. But changing demographics render the current pay-as-you-go system fiscally unsustainable and generationally inequitable over the long-term. Reversing this trend will require facing up to some hard choices and making far-sighted decisions.
Social Security's future must be assessed within the broader context of retirement security for a population that is living longer, retiring earlier and spending more on health care than was assumed when our senior entitlement programs were created.
Social Security faces real difficulties, primarily its growing costs and an ever-widening gap between dedicated revenues and benefit promises beginning in 2018, but its projected shortfall is only part of a much bigger problem. Put simply, we have promised ourselves an array of future retirement and health care benefits that is unaffordable.
Our nation, along with the rest of the developed world, is about to undergo an unprecedented demographic transformation for whose vast cost it has no idea how to pay. The coming age wave is not a temporary challenge that will recede once the baby boom generation passes away. The boomers' retirement is ushering in a permanent transformation in the age structure of America's population--and a permanent rise in the cost of programs such as Social Security, Medicare and Medicaid.
It is true that no immediate crisis is confronting Social Security. Nor is an immediate crisis facing Medicare and Medicaid - the other two large entitlement programs for the aged. Yet, a broad bipartisan consensus exits that these three programs are on an unsustainable course. No one can say exactly when a crisis will hit, but by the time it does we will have likely burdened the economy with a debilitating amount of debt; leaving painful benefit cuts and steep tax increases as the only solutions. Waiting for this gut-wrenching outcome, knowing full well that is coming, would be an act of fiscal and generational irresponsibility on a grand scale.
The Social Security trustees project that program expenditures will grow from roughly 4.3 percent of the nation's gross domestic product (GDP) today to 6.3 percent in 2030, rising modestly thereafter to 6.6 percent in 2075.
Perhaps if viewed in isolation, this cost might be bearable. What makes the problem worthy of immediate attention, however, is that it reflects the impact of a rapidly aging society. Older people rely heavily on government entitlement programs and their numbers are soon expected to grow as the post World War II baby boomers enter their advanced years. There are 37 million people in the population age 65 and older today. By 2025, it is estimated that there will be 62 million. By 2045, they will rise to nearly 80 million, or more than double their current number. In contrast, by 2025 the number of people working in the economy is estimated to rise by only 13 percent, and by 2045, only 20 percent.
This dynamic has consequences that go far beyond Social Security. It is equally troublesome for Medicare and Medicaid. And for them, the looming demographics are only part of the issue. Health care prices continue to outpace economic growth and this phenomenon will only compound the growing costs attributable to the rising number of aged. The Congressional Budget Office projects that the combined cost of Medicare and Medicaid could increase from 4.8 percent of GDP today to 11.5 percent in 2050.
Without a policy response, the overall cost of government as a share of the economy could reach levels not seen since World War II. Today, governmental expenditures absorbs almost 20 percent of GDP. Under what CBO sees as the most plausible range, they could rise to as low as 23 percent of GDP in 2050 or as high as 53 percent. While it may be unrealistic to assume that half the nation's economic output such a large share of the economy could be consumed by government programs, even if the cost of government it rose to only 30 percent of GDP, the share of the economy needed would be 50 percent greater than it is today.
Federal tax receipts have hovered in the range of 18 percent of GDP over the past half century. Today they stand at 16.8 percent. The federal budget deficit, now standing at 3 percent of GDP, hovers around $400 billion. If senior entitlements are allowed to grow on autopilot pushing, total federal spending to 30 percent of the economy, and Americans' intolerance for taxes above 20 percent of GDP holds true, the resulting deficits will rapidly escalate to dangerous levels. A deficit of 10 percent of GDP in today's terms is the equivalent of $1.2 trillion a year. That amount is equal to roughly half of today's total government expenditures. The prospects of being able to carry that amount of new debt year after year without stifling the economy are doubtful.
Whether through increased taxes or constrained spending (or some combination thereof), action by lawmakers will likely be necessary to restore balance between future governmental receipts and expenditures. Economic growth alone will not be enough to close the gap. Moreover, the sooner action is taken, the more gradual the remedies could be. The political system can adjust to unexpected good news. More problematic are the potentially harsh adjustments of deferring action on bad news projections that prove correct.
In addition to addressing the growth of senior entitlements as a share of the economy, a critical strategy for preparing for the demographic transformation is to increase savings to build a bigger economy.
Given demographic trends, the economy in the future will be called upon to transfer a rising share of real resources from workers to retirees. These resources will be much easier to find in a healthy growing economy than in a stagnant one. The best way to achieve economic growth and increase real income in the future is to increase savings today. Savings provide the capital to finance investments, which will enhance productivity and increase the amount of goods and services each worker can produce. Without new savings reform is a zero–sum game.
The final report of the President's Commission to Strengthen Social Security aptly linked the idea of prefunding more of our future benefit promises and the need for higher savings:
Advance funding raises national savings, increasing the nation's capital stock and productive capacity and reducing Social Security's financial burden on future generations….To ensure that Social Security's financing burdens are equitably shared, it is imperative that a portion of these revenues be devoted to advance funding. The resulting increases in national saving will raise the country's capital stock, and therefore boost our productivity and output. In essence, increased national savings increases the size of the economic pie that is available for everyone, old and young alike, to consume in the future.
Social Security's current pay-as-you-go financing works against higher savings. In the first place, the program's widening cash deficits threaten to trigger a huge new run-up in the publicly-held debt starting in the 2020s. Moreover, many economists believe that Social Security's pay-as-you-go structure discourages household savings, and hence capital formation, because it promises households future benefit income while creating no real economic resources to generate that income. As a result, households put less into other (fully funded) forms of savings.
The oldest segment of the 75-million strong baby boom generation, now turning 59, will begin drawing on their Social Security benefits in three years … in six years they will be eligible for Medicare. And while Medicare is projected to grow faster than Social Security, this faster growth only makes achieving savings in Social Security more urgent.
The choices we make now will determine what kind of America our children and grandchildren inherit 20 and 30 years from now. With the first of the 76 million baby boomers on the verge of retirement, the window of opportunity to counteract the generationally inequitable consequences of inaction is rapidly slamming shut.
II. What are the consequences of delaying action?
We have a crisis today only because of the threat of political gridlock. Inaction now increases the prospects of severe changes later. Every year that alterations are put off greatly raises the risk of large tax increases or sudden benefit reductions in the future. Reforming Social Security today would not free society from that future stress, but it would be a good start.As noted, in just three years the baby boomers begin to receive their first Social Security checks. From that moment on, the number of workers whose wages are taxed, relative to the number of beneficiaries who receive the proceeds of the tax, will sharply decline. Here are the facts:
In 1960 there were 5.1 workers for each Social Security beneficiary. Today the ratio is 3.3 workers for each beneficiary. As the huge baby boom generation retires the ratio will fall to 2 workers for each beneficiary.
This dynamic has a profound effect on the system's fiscal sustainability. Social Security will generate ample surpluses, in the range of $100 billion, for the next few years. But in 2009, the year after the first baby boomers qualify for benefits, the annual cash surplus will begin to shrink, and by 2018 Social Security's cash flow will turn negative.
From 2018 through 2041 Social Security will need to draw upon interest income and eventually liquidation of its trust fund assets--special issue Treasury bonds--to pay benefits.
Redeeming Social Security's trust fund assets will have an impact on the rest of the budget and the economy because these “assets” are liabilities to the Treasury. To come up with the money for Social Security, Treasury will have to cut other spending, raise taxes, use any surpluses that may exist, or borrow from the public.
At first the gap will be relatively small - $16 billion in 2018 - but it will grow very quickly as those who were born in the peak of the baby boom begin to retire in large numbers during the 2020s.
The annual shortfall grows to $250 billion by 2030 and in 2041, the last full year of trust fund “solvency,” Social Security faces a cash deficit of $370 billion. All told, between 2018 and 2041 paying off the trust fund bonds will require a cash infusion of $5.4 trillion. 
In 2042, when all the trust fund bonds have been liquidated, Social Security's spending authority will be limited to its cash income. This will be sufficient to pay just 73 percent of promised benefits.
The alternative to cutting benefits by 27 percent to meet available income would be a payroll tax increase of 36 percent.
Over the trustees' 75-year horizon Social Security's cash deficit of $26 trillion in today's dollars far outweighs the cash surplus of less than $1 trillion through 2017.
As a percentage of the economy, Social Security will grow by more than 50 percent from 4.3 percent today to 6.5 percent in 2042, according to the 2004 trustees report.
More importantly, this growth in Social Security's cost will take place in the context of rising costs for other entitlements. The combined cost of Social Security, Medicare, and Medicaid will more than double from 8.4 percent of the economy today to over 18 percent by 2050. By comparison, all of government this year equals 19.8 percent of GDP, and revenues equal 16.8 percent.
This trend leads to one of three outcomes: large tax hikes, resurgent and unsustainable deficits, or the withering away of the rest of government - allowing spending on the poor, on infrastructure, and on defense to steadily decline decade after decade.
No one believes that the federal government's sole function should be to transfer income to retirees at the expense of all other government functions. But that is the inevitable consequence of adhering to two widely held - and entirely contradictory - goals: limiting the size of government and leaving senior benefits on autopilot.
Suppose that one of your colleagues introduced legislation called The Social Security Do Nothing Act. Under this bill, promised retirement benefits would be cut by roughly 15 percent for today's 30-year olds, by 30 percent for today's 20-year olds, and by 35 percent for today's newborns. Alternatively, payroll taxes would suddenly go up by 36 percent in 2042.
How many of you would rush to endorse this bill? None, I suspect. And yet, these are the choices under the Do Nothing Plan.
What is remarkable is not that reform plans engender such heated debate, but that the Do Nothing Plan engenders so little outrage. Worse yet is the fact that no one will have to endure the scrutiny and ridicule of specifically advocating the Do Nothing Plan in order for its consequences to take effect. The Do Nothing Plan has already been enacted. It is current law.
To put it in more personal terms, consider the table below which looks at where four different generations will be at various times in their lives relative to Social Security's current outlook. What may sound like a distant and abstract problem becomes more immediate and relevant when we consider that today's 30-year old will qualify for full retirement benefits in 2042 - the year of projected trust fund insolvency - and that the system will begin to run growing annual cash deficits even before today's newborns enter the workforce.
Ages of Persons in Four Generations at Significant Dates for the Social Security Program
90 years old
103 years old
60 years old
73 years old
97 years old
105 years old
30 years old
43 years old
67 years old
75 years old
105 years old
13 years old
37 years old
45 years old
75 years old
- In 2018, Social Security's dedicated revenues will no longer cover all of its expenses. At this point Social Security will become a net drain on the budget as it begins to draw upon its claims on general revenues. The pay-as-you-go tax rate will be 13.24, up from 10.87 today. Including Medicare Part A, the payroll tax cost rate will be 17.14, up from 14.02 today.
- In 2042, all of the assets in the Social Security trust fund will be exhausted, leaving the program able to pay only 73 percent of promised benefits. The pay-as-you-go tax rate will be 17.79 percent of taxable payroll. Including Medicare Part A, the tax rate will be 25.23.
- By 2050, the Congressional Budget Office estimates that the cost of Social Security, Medicare and Medicaid combined will consume nearly 18 percent of GDP, almost all total federal revenues assuming that taxes remain in the range of about 18 percent of GDP as they have over the past 40 years. The pay-as-you-go tax rate will be 17.90 percent of taxable payroll. Including Medicare Part A, the tax rate will be 26.31.
- 2080 is the last year of the trustees' projection. By then the program will be able to pay just 68 percent of that year's promised benefits. The pay-as-you-go tax rate will be 19.39 percent of taxable payroll. Including Medicare Part A, the tax rate will be 32.78.
The table above underscores an important point: Social Security reform is a much more critical issue for today's young than today's elderly. The current system is more than adequate to meet its obligations to those who are already retired. However, the system can't afford all of the benefits it promises to today's workers. Those with the greatest stake in this debate are therefore the so-called Gen X'ers and younger, and it is this segment of the population most overlooked in the Social Security reform debate.
Public opinion surveys have indicated declining confidence in Social Security over the past 25 years. Many younger workers are beginning to discount Social Security entirely in their retirement planning. This decline in public confidence is itself a major problem for a system that depends critically on everyone's approval and trust. Social Security is a generational compact in which each generation's welfare depends directly upon the willingness of the next generation to participate. If the next generation grows disaffected, the survival of the system is thrown into question.
It is worth recalling that President Bush is not the first president in recent years to put Social Security on the political agenda. In 1998, President Clinton made Social Security reform one of his top domestic priorities. Here is how President Clinton summarized the problem at a forum hosted by The Concord Coalition and AARP in July 1998:
Today, the system is sound, but we all know a demographic crisis is looming. There are 76 million of us baby boomers now looking ahead to retirement age and longer life expectancies. By 2030, there will be twice as many elderly as there are today, with only two people working for every one person drawing Social Security. After 2032, contributions from payroll taxes to the Social Security trust fund will be only enough to cover about 75 cents on the dollar of current benefits.
We know the problem. We know that if we act now it will be easier and less painful than if we wait until later. I don't think any of you want to see America in a situation where we have to cut benefits 25 percent, or raise inherently regressive payroll taxes 25 percent, to deal with the challenge of the future and our obligations to our seniors.
I can tell you, I've spent a lot of time talking to the people I grew up with; most of them are middle-class people with very modest incomes and they are appalled at the thought that their retirement might lower the standard of living of their children, or undermine their children's ability to raise their grandchildren. So let's do something now in a prudent, disciplined way that will avoid our having to make much more dramatic and distasteful decisions down the road.
President Clinton ended with an admonition that is as relevant today as it was in 1998:
We dare not let this disintegrate into a partisan rhetorical battle. Senior citizens are going to be Republicans and Democrats and independents. They're going to come from all walks of life, from all income backgrounds, from every region of this country, and therefore, so will their children and their grandchildren. This is an American challenge and we have to meet it together.
Any genuine reform has a fiscal and political price, so it's tempting to pretend that the status quo can continue indefinitely. It can't. Not acting is itself a choice, and one that will have grim consequences for today's midlife adults and even grimmer ones for their children.III. What kinds of broad prescriptions for change would be the most effective?
There are just two ways to close Social Security's financing gap without burdening tomorrow's workers and taxpayers:
· Reduce Social Security's long-term cost, and:
· Make the remaining cost more bearable by increasing national savings and hence the size of the economy.
A workable plan should do both. Ideally, it should also pay for itself from day one and find savings within the Social Security system through some combination of reduced benefits and new contributions.
The bottom line is that the system requires change and this cannot happen without sacrifice in one form or another. The choice we face is not between guaranteed future benefits under the current system and a risky path of reform. It is between reform options that, in different ways, attempt to ensure the fiscal sustainability of fair and adequate benefits over the long-term.
Despite widespread recognition that hard choices are unavoidable, this difficult work is forced to compete for attention with an assortment of arguments for inaction and reform ideas that purport to fix the problem without asking anyone to give up anything.
Here are four of the most frequently used arguments:
Argument #1: Social Security can pay full benefits until the year 2042.
This argument is true as far as it goes, but it does not tell the full story. The trustees now project that Social Security will be “solvent” until the year 2042 -- meaning that its trust funds will possess sufficient assets, and hence budget authority, to cover benefits until that date. However, trust fund solvency says nothing about fiscal sustainability.
The problem is that the trust funds are primarily an accounting device. Social Security's assets consist of Treasury IOUs that can only be redeemed if Congress raises taxes, cuts other spending, uses surpluses, or borrows from the public. Thus, their existence, alone, doesn't ease the burden of paying future benefits. It is true that when trust fund surpluses are used to reduce the publicly-held debt it does result in higher savings. But experience has shown that trust fund surpluses are just as likely to be spent as saved. It therefore cannot be assumed that a trust fund surplus will result in higher savings.
Trust fund accounting minimizes the magnitude of the problem because it implies that there really are resources being held in reserve… real assets that can be drawn down in the future to pay benefits. However, real assets are not created by giving the trust fund an IOU and promising to sell the IOU to the public when the money is needed to pay benefits. The IOUs will no doubt be honored, but that's not the point. The real issue is how the government and society will afford them. The debate over Social Security reform should concentrate more on economic and budgetary consequences than on governmental bookkeeping. Fiscally, it is not the trust fund balance, but the program's operating balance that matters -- that is, the annual difference between its outlays and earmarked tax revenues. Social Security's current operating surplus is due to begin falling in 2009 and turn into an operating deficit in 2018. This deficit is projected to widen indefinitely into the future.
Argument #2: A mere 1.89 percent of payroll increase would cure the problem.
A related argument is that a tax hike of merely 1.89 percent of payroll is all that is needed to restore Social Security to long-term solvency - technically an increase of less than one percent each for employers and employees. This claim is based on the program's actuarial balance, which averages projected trust-fund surpluses and trust-fund deficits over the next seventy-five years. In 2004, Social Security's actuarial balance was a shortfall of 1.89 percent of payroll. In theory, this is the amount that Congress would have to raise payroll taxes or cut Social Security benefits, starting immediately, in order to keep the trust funds “solvent” for 75 years.
Proponents of this idea neglect to mention a couple of important caveats. For one thing, “mere” is a relative term. A tax hike of 1.89 percent of payroll is equivalent to a $1.2 trillion tax increase over the next 10 years. For another, the solution is not permanent. If the combined 12.4 percent tax rate were raised by 1.89 percent (the amount of the average 75-year deficit), it would only keep the program's cash flow positive through 2023, or for five additional years. If the payroll tax were hiked to cover the subsequent shortfalls, by 2025, it would have to be 14.8 percent; by 2040, it would have to be 16.9 percent; and by 2075, 18.2 percent. Those rates reflect increases that are a long way from what the 75-year “averaging” method implies. A 14.8 percent tax rate translates into a 20 percent increase in taxes; a 16.9 percent rate translates into a 36 percent increase, and an 18.2 percent rate translates into a 47 percent increase.
Moreover, this “solution” assumes that the horizon for trust-fund solvency will forever remain fixed at seventy-five years from today. In other words, it assumes that while we would require the trust funds to be in balance over a full seventy-five years, our children will be satisfied with forty years and our grandchildren will be satisfied with an empty cupboard.
And there's a more fundamental problem. Any trust-fund surplus is immediately lent to Treasury, leaving Congress free to spend the money it is supposedly saving. For the 1.89 percent solution to ease Social Security's burden on the economy, legislators would have to allow the program's extra interest-earning assets to accumulate unspent for more than 30 years -- a proposition that seems highly unlikely and in any event cannot be guaranteed.
Argument #3: The Trustees are too pessimistic about the future.
Another frequently heard argument is that the Social Security Trustees are too pessimistic--that the projections are unduly gloomy about future economic growth and that with more realistic assumptions the Social Security problem disappears.
It is true that the Trustees project that the economy will grow more slowly in the future than it has in the past. But this is a matter of prudence, not pessimism. Economic growth (GDP) depends, in part, on workforce growth, and this will fall to near zero when the boomers start retiring.
- Since 1973, the U.S. workforce has grown by 1.7 percent per year.
- Over the next seventy-five years, it is projected to grow by just 0.3 percent per year.
- Given the demographics, it is unlikely that GDP growth will not slow.
A more legitimate question is whether the trustees are too pessimistic about the growth in productivity, or output per worker hour. In the future, the trustees may have to raise their assumption. Since 1995, productivity has unexpectedly surged. Some believe that this heralds the arrival of a “new economy” in which information technologies and globalization will lead to permanently higher rates of productivity growth. But there are reasons to be skeptical:
- The new-economy thesis remains just that: a thesis. No one yet knows whether the surge in productivity that began in the mid-1990s will persist. The trustees' current long-term assumption for productivity growth--1.6 percent per year--is in line with the record of the past twenty-five years.
- Even if the enthusiasts are right about the new economy, higher growth is not a long-term fix for Social Security. When productivity goes up, average wages go up, and this adds to long-term tax revenues. But when average wages go up, average benefit awards also go up, and this adds to long-term outlays.
- Practically, the only way to get big savings from higher productivity growth is to sever the link between average wages and new benefit awards. Without such a fundamental change, higher productivity growth alone cannot possibly save Social Security.
There is one aspect in which the Trustees are indeed pessimistic--but here greater optimism would obviously add to Social Security's costs. The Trustees project that mortality rates will decline more slowly in the future than they have in the past--and that longevity will therefore grow more slowly.
- According to the Trustees, life expectancy at age 65 will grow at just half the pace over the next seventy-five years as it has over the past seventy-five.
- Some biotech optimists are now predicting that a life expectancy of 100 or more is attainable within a generation.
- If anything approaching that came to pass, the entire structure of old-age entitlements would be rendered instantly and massively unaffordable.
- But one doesn't have to agree with these visionaries to conclude that the Trustees are too conservative. Accepting their projections means believing that Americans will have to wait until the mid-2030s to achieve the life expectancy that the Japanese already have today.
Argument #4: Investment returns provide a “pain free” solution.
Moving toward a more funded Social Security system could indeed have enormous benefits: not just higher returns to retirees, but greater national savings and productive investment, and hence greater wage growth for workers in the years before retirement. It would also be the surest method of locking up any new contributions because it would prevent the government from spending the money on other programs. But it cannot be supposed that directly funding more of Social Security's benefits is a way to avoid the hard choices. It is the hard choice:
- The challenge is that, until the transition is complete, workers will have to pay more, retirees will have to receive less, or both. Reform plans that do not face up to this transition cost will not result in new net savings or a larger economy. Any gains for future beneficiaries will necessarily come at the expense of future taxpayers.
- It is neither realistic, nor economically sound to count on the historic spread between the investment returns on stocks and bonds to fund a reform plan without cost reductions or higher contributions.
- The fundamental issue is not whether the system should be public or private, but the extent to which it should be unfunded or funded. Unfunded personally owned accounts would neither add to national savings nor reduce the burden of today's system on future generations, even if they earn a higher rate of return than the current pay-as-you-go system. A new system of unfunded accounts, like trust fund solvency, avoids the real challenge, which is to ensure that adequate resources are set aside to meet the cost of future benefits.
The Social Security challenge is first and foremost a cost challenge. Any responsible reform plan must start with measures that reduce the projected growth in benefits and makes the system fiscally sustainable over the next 75 years and beyond.
But reducing Social Security's cost is not the only challenge. There are also the issues of benefit adequacy and individual equity. Reform must ensure that future retirees have adequate benefits. It must also ensure that workers do not pay an ever-rising payroll tax burden in return for ever-diminishing paybacks on contributions.
That is why, along with measures to reduce its long-term cost, greater funding is an essential part of Social Security reform. To make a difference, however, the funding must be genuine. It isn't enough to simply credit more Treasury bonds to the trust funds or to redirect existing payroll contributions into marketable securities, with or without personal accounts.
Without new savings, without real funding, a plan cannot increase the productivity of tomorrow's workers, and thus becomes a zero-sum game of pushing liabilities from one pocket to another or from one generation to another.
The Concord Coalition does not support raising the payroll tax rate. For one thing, it is regressive tax that falls most heavily on middle and low-income workers who might wonder why they must pay more to subsidize the high-income old. A payroll tax increase would also deepen the generational inequities within the system. Young workers might ask why they must pay more than today's midlife boomers for the same (or worse) benefits.
Some advocate getting the wealthy to contribute more by raising, or eliminating, the payroll tax cap on wages, now at $90,000. A modest increase in the wage base would bring in a modest amount of new revenue, but wouldn't do much to reduce the system's long-term cash deficit. Eliminating the cap would have a bigger impact, but would substantially alter Social Security's traditional focus on both fairness to individuals and protection of the needy. It would destroy the whole presumption of a contributory system--that what people get back be at least somewhat proportional to what they pay in.
In any case, the savings would diminish over time. While higher FICA contributions would initially swell the Social Security trust funds, these contributions would eventually require that higher benefits be paid out. And even though the benefits would represent a low or even negative return on contributions, they would still be large in absolute terms. The ultimate effect of eliminating the cap would be to increase the cost of Social Security and the total fiscal burden of government. It wouldn't generate enough new revenue to balance the system and in the absence of some mechanism to save the money its primary short-term effect would be to simply bring in higher taxes to fund current government operations.
Cost reduction options
Several benefit reduction strategies are possible, including raising the so-called normal retirement age and shifting from wage-indexing to price-indexing for calculating new benefits. The most reasonable strategies stress gradualism and fair warning.
The Concord Coalition believes that the necessary savings could be achieved using some variation of the following options:
1. Raise the “normal retirement age” for full benefit eligibility
One of the most logical options to consider is raising the age for full benefit eligibility. It makes good sense for two reasons:
- Longevity is increasing steadily, and longer life spans mean longer, and more costly, benefit spans.
- In coming decades, the pool of working-age Americans will virtually stop growing, depriving our nation of this engine of economic growth. Raising the full benefit-eligibility age could help augment the labor force by encouraging older people to remain at work for a few more years.
It's conventional wisdom that our population soon will be growing older because the huge baby boom generation is poised to begin retiring. But that's only part of the picture. Even if there were no baby boom, the rising longevity and fall in birth rates mean an older America and would spell serious trouble for Social Security (and Medicare as well). Increasing life spans have already increased benefit spans.
- In 1940, when the first benefits were paid, 65-year-old men could expect to live almost another 12 years and women another 13.4 years.
- Today, men retiring at 65 can expect, on average, 16 years of benefits and women can expect 19 years.
- By the time today's high-schoolers begin retiring in the 2050s, 65-year old men are expected to live another 19 years and women more than 21 years.
- Or to turn it around, for people today to spend the same number of years collecting benefits as the typical 65-year-old when the program began, they would have to wait until 72, and by 2050 they would have to wait until 75.
But the problem posed by an aging population is not just that benefit spans will lengthen. We also expect to be coping with a labor shortage. Instead of increasing our supply of working age people by 2 percent each year as in recent decades, or even the current 1.3 percent rate today, between 2010 and 2050, workforce growth will slow to a crawl: just 0.3 percent per year. There will be just barely enough new workers each year to replace those who are leaving.
Growing our economy could help finance benefits for a mushrooming retiree population. But, boiled down to essentials, economic growth depends on two factors: increasing the number of workers, and increasing how productive each worker is. Since no one has a sure-fire recipe for boosting worker productivity enough to make up for the slowdown in workforce growth, anything we can do to encourage people to work a few more years and encourage employers to accommodate older workers will help our economy.
2. Index for Longevity
Any reform plan should also index initial benefits to changes in elder life expectancy. Without this provision, Social Security will once again drift out of balance; with it, the system's long-term cost will be stabilized relative to worker payroll.
Social Security retirement benefits are paid in the form of a defined benefit annuity. An annuity purchased with a defined contribution personal account balance would naturally take into account expectations about future longevity. The more years the annuity provider expects to have to pay benefits, the smaller the annual benefit a given account balance would buy. The current Social Security system makes no such adjustment. The benefit annuity it promises is set by a formula that yields the same result no matter how fast and far life expectancy rises. Cutting benefits by a fixed percentage may balance the system for a while. But unless reform also adjusts benefits for ongoing gains in life expectancy, the system will drift out of balance again.
The impact of rising longevity on Social Security's long-term cost is large. Over the next 75 years, the Trustees project that life expectancy at 65 will rise from 17.5 to 21.6 years, or by 23 percent. Over the long run, this 23 percent rise in life expectancy will translate into a roughly equivalent percentage rise in total benefits. The Trustees' projection, moreover, assumes that longevity will increase more slowly in the future than it has in the past. If the historical trend continues, the impact on Social Security costs will be even greater.
There are two ways to index Social Security to longevity. The minimum eligibility age for benefits could itself be indexed--that is, the early retirement age could be raised in tandem with average life expectancy. Or else--and this is the approach the President's Commission took--annual benefits could be reduced so as to offset the greater number of years that will be spent collecting those benefits. This is the equivalent of indexing the so-called normal retirement age, the age at which full or unreduced benefits are payable.
3. Treat Social Security Benefits Like Private Pensions for Tax Purposes
Making 85 percent of all benefits taxable is fair, and should be on the table as a means of increasing Social Security's revenues. The 85 percent taxability rule that now applies to beneficiaries with incomes over high thresholds could apply to all beneficiaries. The 15 percent exemption reflects an estimate of the dollar value of most beneficiaries' prior FICA contributions that have already been subject to personal taxation. It would thus bring the tax treatment of Social Security in line with the tax treatment of private pension benefits.
Since this provision would affect only those households with enough income to pay income taxes, it would maintain the progressivity of the program. It's worth noting that because current law does not index the thresholds at which benefit taxation applies, a rising share of total OASDI benefits are now becoming taxable--and eventually 85 percent of all benefits will be taxable. Full benefit taxation is therefore already due to be instituted in the future (and future revenues from it are already included in current projections). What this option would do is to move to full benefit taxation right away.
The new revenue from this provision is not large but it is available immediately and thus generates critical near-term budget savings, which may be needed for the transition costs of any reform plan.
4. Affluence test
An affluence test for upper income beneficiaries could be designed as an alternative to full benefit taxation and generate roughly the same aggregate savings in every future year, which makes the two provisions substitutable. The appeal of full benefit taxation is its simple equity: It would merely subject Social Security beneficiaries to the same tax code as everyone else. The possible drawback is that it reaches deep down into the middle class. The appeal of the affluence test is its greater progressivity. The possible drawback is that it may be regarded as arbitrary.
5. Change the Cost of Living Adjustment (COLAs)
Cost-of-Living Adjustments (COLAs) are used in Social Security, the federal income tax code, and other programs to ensure that specified dollar amounts are adjusted every year for inflation, as measured by the Consumer Price Index (CPI).
There has been substantial debate about how accurately the CPI measures the true cost of living. There are many sources of bias, some very technical. For example, the particular market basket of goods used can rapidly become out of date. The basket based on surveys taken between 1982 and 1984 was still in use through 1997. This means, for one, that new products can be ignored completely.
Although the government has made some improvements to the CPI in recent years some experts, including Federal Reserve Board Chairman Alan Greenspan, believe that CPI still overstates inflation and thus over compensates beneficiaries. Additional adjustments to the CPI may thus be in order. However, a great deal of caution must be used in deciding whether to make ad hoc COLA reductions. While over-indexing Social Security squanders budget resources, setting COLA's beneath a fair measure of inflation is not the right way to balance the system since it would unfairly penalize the oldest and poorest beneficiaries.
6. Change the formula for determining initial benefits
a. Bend points
The determination of a retiree's initial Social Security benefit check is based on the calculation of Averaged Indexed Monthly Earnings (AIME). The amount of money earned by an individual each year of work is multiplied by the increase in average wages that has occurred up to the year of eligibility for Social Security, and then the average of the highest 35 years (fewer for those receiving disability benefits) of indexed wages is taken and divided by 12 to get the AIME.
Once the AIME is calculated, the Primary Insurance Amount (PIA) is determined by applying the “primary insurance amount formula.” This progressive formula is designed to replace a share of annual pre-retirement income based on three “bend points.” (90 percent, 32 percent, and 15 percent.) For example, in 2005 the replacement rates are 90 percent of the first $627 of average monthly earnings, 32 percent for earnings up to $3,779, and 15 percent of higher earnings up to the taxable maximum.
One way to reduce Social Security's long-term cost would be to lower the bend points across the board. Or if preferred, reduce the replacement rate within each bend point bracket on a progressive basis that would protect low-income workers. This later approach would work particularly well with a system of personal accounts, which in the absence of some other mechanism such as savings matches paid out of general revenues, would make the overall system less progressive than it is now.
Another option would be to index initial benefits to the growth in prices (CPI) rather than to the growth in wages. Under current law, initial benefit awards are indexed to wages--that is, the wage history on which benefits are based is updated at the time of retirement to reflect the rise in the economy's overall wage level over the course of the beneficiary's working career.
In effect, wage-indexing ensures that the living standard of retirees keeps pace with society's overall living standard. Re-indexing initial benefit awards to prices merely ensures that the absolute purchasing power of retirees keeps up with inflation. Note that this reform effects only initial benefit awards; current benefits are already price indexed.
The reform has two advantages: its simplicity and its large savings. If real wages are growing 1 percent per year faster than inflation, price indexing will result in a roughly 35 percent cut in initial benefits relative to current law for the first cohort to spend a complete career under the new regime. Under this assumption, the savings would be roughly sufficient to close Social Security's long-term cash deficit.
Under current law, it is virtually impossible to close Social Security's deficit through an acceleration in productivity growth. Higher productivity would result in higher wages and this would boost payroll tax revenue. But higher wages would also result in higher benefits, and this would largely cancel out the gain. With price-indexing, however, benefits would shrink indefinitely relative to taxable payroll and GDP--and the faster wages grow, the more benefits would shrink as a share of the economy.
This dynamic, of course, means that the living standards of retirees will diverge from those of the working population. To the extent that we view Social Security as a pure floor of projection, this does not pose a public policy problem. To the extent that we view it as an income replacement program, it does.
For this reason, price-indexing makes most sense as part of an overall reform that also incorporates funded benefits like personal accounts. The price indexed pay-as-you-go benefit would ensure that the purchasing power of benefits would remain the same for each new generation of retirees. The funded benefits would help ensure that the relative living standard of retirees is not eroded. The rate of return to a funded system, after all, is the rate of return to capital--and historically, this has been faster than the rate of growth in wages.
Options for prefunding future benefits
Funding Social Security cannot substitute for measures that raise new contributions or reduce pay-as-you-go benefits. But, in conjunction with cost-saving reform, funding can help create a Social Security system that is not only more sustainable, but that offers a fairer deal and a sounder floor of protection.
The case for funding is simple and compelling. At the macro level, a funded system means higher savings and hence higher productivity and higher national income. At the micro level, it means higher returns and hence higher benefits at any given contribution rate.
Unfortunately, the way that the Social Security trust funds work undermines the whole purpose of funding. Any trust-fund surplus is immediately lent to the Treasury, leaving Congress free to spend the money that it pretends to save. As noted earlier, Social Security's assets consist of interest-earning Treasury IOUs whose sole function is to keep track of budget authority. They constitute a claim on future tax revenues, not economic savings that can be drawn down to finance future benefits.
In recent years, much attention has been given to various methods of prefunding future benefits. The main options are:
· A budgetary “lockbox” for the Social Security surplus · An independent board to mange trust fund investments
· Personally owned accounts
While ideological factors often cloud the debate over these options, the real issue is which is most likely to result in genuine savings. What legal, political and fiscal incentives best ensure that resources are actually reallocated from the present to the future?
1. Budgetary trust fund “lockbox”
In the late 1990s, political leaders promised that henceforth they would translate Social Security's current surpluses into genuine savings by balancing the budget excluding the trust funds. The goal was fiscally responsible, but achieving it rested on a chancy proposition--namely, that policy makers would have the fiscal discipline to “lockbox” large unified budget surpluses year in and year out.
With a booming economy generating windfalls for Treasury, keeping the lockbox promise was initially painless. Even President Bush's first tax cut observed it. But when the economy slowed and September 11 created new spending needs, the promise was quickly forgotten--and forgotten in a big way. This year, the CBO estimates that the budget will run a deficit of at least $540 billion excluding the trust funds. This is the amount that Congress would have to raise taxes or cut spending in order to save the Social Security surplus.
Regardless of intent, and despite any bookkeeping devices such as a lockbox, the government can only save the Social Security surplus if it continues, year after year, to take in more money than it needs to pay all of its other bills without dipping into the Social Security trust funds. This has only happened twice from 1983 to the present, and is not projected to happen again for the foreseeable future.
Success of the lockbox concept is therefore critically dependent on the willingness of future political leaders to maintain a level of fiscal discipline that is not currently discernable.
2. Investment by an independent board
To get around the porous nature of trust fund lockboxes, some have proposed to set up a Social Security reserve fund administered by an independent trustee and invested in marketable securities. This mechanism would probably provide a more reliable method than the budgetary lockbox for promoting savings but here too, there are important questions. What would prevent the federal government from borrowing against its own Social Security investments? When all is said and done, government would still own the reserve, and whatever government owns it can contrive to spend. Moreover, the public would have no particular incentive to ensure that the savings are genuine because Social Security's defined benefit promise is not contingent on the system being funded.
3. Personally owned accounts
A third method of prefunding is to establish a system in which some portion of workers' contributions are saved and invested in personally owned accounts. The advantage of this method is that it would provide a lockbox no politician could pick.
The current system provides a statutory right to benefits that Congress can cut at some future date. Personally owned accounts would offer workers ownership of constitutionally protected property which, under some circumstances, could be passed on to their heirs - something the current system does not allow. The funds would be put beyond the reach of government. Congress could not double-count personal account assets in the budget. And if it tried to shut down the flow of funds into personal accounts, voters would have a huge incentive to object.
Personal account reforms come in two basic types: “carve outs” and “add ons.” In a carve out, a portion of the current payroll tax would be diverted to personal accounts. For the carve out to result in genuine funding, the diversion must be paid for by reductions in pay-as-you-go benefits beyond those that would need to be made in any case simply to eliminate Social Security's projected cash deficits. In an add on, the accounts would be funded partly or wholly from additional worker contributions. The contributions would be personally owned savings, and so would not constitute a tax--or at least would not function like one.
A pure carve out necessarily entails cuts in current-law Social Security benefits. Because personal account contributions would earn a higher return than contributions to the existing system, a carve-out plan might be able to pay retirees higher total benefits than today's purely pay-as-you-go system can afford. However, it cannot guarantee that retirees will receive everything that the existing system promises. In practice, most personal account carve outs rely on borrowing to substitute for the lost FICA revenue and mitigate benefit cuts. This entails the “free lunch” problem discussed below.
The “add on” approach offers a way to ensure the adequacy of future benefits without recourse to budgetary shell games. In fact, with a 2 percent of payroll add on it may be possible to ensure that every cohort of workers will receive benefits at least as large as what current law now promises but cannot afford. Is it worth paying a bit more to achieve these superior results? In the end, after all the shell games are played out, this is the central choice that the American public must confront.
To be clear, current law must eventually result in either a steep cut in benefits or a steep hike in taxes. If the choice is to avoid any hike in the Social Security contribution rate, a personal accounts carve out might generate larger benefits than today's pay-as-you-go system can afford. If the choice is to avoid any reduction in promised benefits, an add on might allow for this at a lower ultimate contribution rate. It is impossible to have it both ways: no cuts in total benefits and no new contributions.
The transition cost
Transitioning out of the current pay-as-you-go system into a partially funded system, with or without personally owned accounts, inevitably requires some group of workers to pay for the pre-funding of the new system while at the same time maintaining funding for those still receiving benefits under the old system. There is no avoiding this cost. Workers will thus have to save more, retirees will have to receive less, or both.
Investment in higher return assets might provide a way to mitigate the extent of benefit cuts or tax increases that might otherwise be required. However, no conceivable rate of return on investments, standing alone, would be enough to fund currently projected benefits at today's contribution rate. Indeed, the President's Commission to Strengthen Social Security confirmed this proposition. The Commission's Model One, does nothing more than dedicate 2 percent of the current payroll tax to personal accounts. As the Commission's final report states, under this approach, “Workers, retirees and taxpayers continue to face uncertainty because a large financing gap remains requiring future benefit changes or substantial new revenues.”
No Free Lunch on the Menu
Genuine funding requires genuine resource trade-offs. To save more, we must temporarily consume less--at least until the productivity benefits of higher savings kick in. Unfortunately, many personal accounts advocates pretend that there's a free lunch on the menu. Just divert current payroll contributions to personal accounts, they say, and the problem will be solved.
These advocates know that the Treasury will have to borrow to make up for the missing revenue, thus offsetting the new private savings. According to their logic, however, the mere fact that contributions are invested in private capital markets will ipso facto make everybody a winner. Apparently, they believe that each worker's personal account can indefinitely earn greater returns (at no greater risk) on the new equity assets than government would lose on the new debt liabilities.
The truth is that any plan that tries to cash in on the spread between stocks and bonds is a dicey and perhaps even dangerous proposition. Such financial arbitrage cannot work in the long run. Over time, the yield on bonds would rise and the yield on stocks would fall, narrowing and possibly even erasing the favorable spread on which the plan depends. Either that, or we have to suppose that markets are irrational, and that the general public will willingly disadvantage itself by buying bonds and selling stocks (with no change in the yield spread) so that personal account owners can enrich themselves by doing the opposite.
Plans that issue debt directly to Social Security participants in the form of “recognition bonds” raise an additional concern. By translating implicit benefit liabilities (which have no constitutional protection) into formal Treasury debt (which does), they would in effect render Social Security unreformable. Giving workers property rights to a pay-as-you-go entitlement is folly. The economy might collapse or the nation go to war. But short of default on the national debt, Congress could never reduce taxpayers' liability for Social Security.
Other reform ideas try to conceal the trade-offs by raising taxes or cutting spending outside the Social Security system. Some propose enacting a national sales tax or a value-added tax to help fund personal accounts. Others say we should pay for the transition by cutting discretionary spending.
The problem here is that there is no direct link between sacrifice and reward. The savings measures may never be enacted--particularly if, as is usually the case with discretionary spending cuts, they are just vague injunctions to reduce “government waste.” And even if they are enacted, the measures may not result in overall budget savings. In the case of the national sales tax or the value–added tax, the public may view these new forms of taxation as a substitute for existing taxes and demand an offsetting tax cut. If so, a larger deficit would neutralize the private savings boost.
To restate the bottom line: Without new savings, any gain for the Social Security system must come at the expense of the rest of the budget, the economy, and future generations. Issuing debt to finance the transition to a funded Social Security system undermines a fundamental purpose of reform because it would not boost, and may even lower, net national savings.
To be sure, reform plans that rely on debt financing usually promise that the debt will be paid back. But in most plans the borrowing is so large and the payback is so distant that it doubtful the payback will ever occur.
Adding personal accounts without using the current payroll tax is not a cost free solution either. It would require higher payroll contributions or a substantial and permanent infusion of general revenues. In the absence of budget surpluses, diverting general revenues to “fund” an add on plan would have the same deficit effects as a carve out.
The problems with debt-financing of a personal accounts were recently addressed by The Concord Coalition in a statement published in the New York Times. Concord Board of Directors members, Warren Rudman, Bob Kerrey, Pete Peterson, Chuck Bowsher, Donald Marron, Sam Nunn, Bob Rubin and Paul Volcker, joined in saying:
Ensuring a more sustainable system will require change, meaning that someone is going to have to give up something - either in the form of higher contributions, lower benefits or a combination of both. No Social Security reform will succeed unless this fact is acknowledged up front.
One reform idea that has received much attention lately is establishing personally owned accounts and “funding” them with borrowed money. Most of the undersigned believe that personal accounts have potential advantages if they are properly funded and adopted as part of a comprehensive reform plan. They are not a free lunch. Simply funding personal accounts with further borrowing, and not with new contributions or contemporaneous benefit cuts, raises many concerns:
· It would not add to national savings. A fundamental goal of reform should be to improve national savings. As America ages, the economy will have to transfer a rising share of resources from workers to retirees. This will be easier in a prosperous growing economy. The best way to ensure this is to raise national savings, and ultimately productivity growth. Social Security reform that relies on deficit financing will not boost net national savings, and may even result in lower savings if households respond to the new personal accounts by saving less in other areas. Without additional savings, any gain for the Social Security system must come at the expense of the rest of the budget, the economy, and future generations.
· It would worsen the already precarious fiscal outlook. The 10-year transition cost of roughly $2 trillion would come on top of the $5 trillion deficit that appears likely if current fiscal policies are continued. Yet the greater fiscal danger with most such plans is that they require additional borrowing for decades to come. In the most widely discussed plan produced by the 2001 President's Commission to Strengthen Social Security, the magnitude of the borrowing equals or exceeds the cost of the new Medicare drug benefit well into the 2020s. Meanwhile, the increased deficits and debt exceed the promised savings until the 2050s. Official projections already indicate that current fiscal policies are unsustainable long before then and the new deficits would only make the problem worse. Savings programmed for the 2050s won't be enough to prevent us from going over the cliff well before that time.
· It would send a dangerous signal to the markets that we are not taking our fiscal problems seriously. With our large budget deficit and low domestic savings rate we are borrowing record amounts from abroad. This year's increase in foreign debt is likely to approach $700 billion. If we “pay for” Social Security reform by running up the debt further, rather than making hard choices, it would signal to increasingly wary financial markets that Washington has no intention of doing what is necessary to get its fiscal house in order. This would increase the risks of a so-called “hard landing” such as a spike in interest rates, rising inflation and a plunging dollar. Promises that all the new debt will be paid back starting in about 50 years are unlikely to satisfy the concerns of those who are watching to see what Washington does now to improve its fiscal position. If markets looked out 50 years, current interest rates would be through the roof.
Personal accounts need not mean “privatization”
Critics of personal accounts often charge that they would shift unacceptable risks to individuals. But in fact, a personal accounts system is consistent with any degree of government regulation. It need not and indeed should not amount to “privatizing” Social Security.
A system of personally owned accounts need not allow people to recklessly undersave during their working years. Participation can be made mandatory and restrictions can be placed on the use of account balances. Nor need it put low-income (or simply unlucky) workers at greater risk of poverty and hardship in old age. The government can require workers to shift from equities into fixed-income assets as they grow older, thus protecting them from sudden market declines--even a crash on par with 1929. The government can also match savings contributions for low-earners and provide a guaranteed floor of old-age income protection, thus preserving or even enhancing the progressivity of the current system.
Many personal account advocates, including the President, believe that the accounts should be voluntary. That would be a mistake. Society has an interest in ensuring that people do not under-save during their working lives and become free riders on the means-tested safety net in old age. Choice is not important in a compulsory social insurance program whose primary function is to protect people against poor choices.IV. Conclusion
The rationale for reforming Social Security now has nothing to do with today's retirees or those who are about to retire. For them, there is no crisis. What's at stake is the retirement security of future generations -- those who have many working years ahead, or who have yet to enter the workforce. For them, doing nothing is the worst option. The issue is what makes sense for the world of 2040, not what made sense in the world of 1940.
The longer reform is delayed, the worse the problems inherent in the current system will become and the more difficult they will be to remedy. Delay risks losing the opportunity to act while the baby boom generation is still in its peak earning years, and the trust fund is running an ample cash surplus. Squandering this opportunity would be an act of generational irresponsibility.
As the debate gets started, it should be emphasized that despite the vitriolic rhetoric often surrounding Social Security reform, a widespread consensus exists that any viable plan will probably include some combination of benefit cuts, increased contributions, higher returns and general revenues. Each involves trade-offs and each comes with a fiscal and political price, regardless of whether it aims to prop up the existing pay-as-you-go system or aims at transitioning to a partially prefunded system.
Because the current system is substantially under financed, the proper comparison for any reform plan is between the benefits payable under a reformed system and the benefits payable under the Do Nothing plan. Some have argued that reform plans would result in deep benefit cuts when compared to the current system in a hypothetically solvent condition. This is neither fair, nor realistic. No realistic reform plan looks good when compared to the false hypothetical of a perfectly solvent system. It is fundamentally unfair to judge any reform plan against a standard that assumes the current system can deliver everything it promises. It can't. Today's Social Security system promises far more in future benefits than it can possibly deliver.
Moreover, in assessing the adequacy of benefits under a reformed system that includes personal accounts it must be kept in mind that a person's retirement income would come from both sources--a basic level of benefits from the defined benefit portion and the additional benefit financed from the lifetime accumulation of the personally owned account. In comparing benefit levels the entire benefit of a reformed system must be included.
We should stop playing political shell games with this issue. If we do not have the political will to solve the Social Security problem now, there is no hope of doing so when the baby boomers start collecting benefits -- not just for Social Security but for Medicare and Medicaid as well. The problems facing our health care programs are much more daunting and difficult than Social Security. These three programs together are expected to double as a share of the economy within the lifetime of today's younger workers putting unthinkable pressure on tax rates, the economy and the budget.
Now is the time to begin preparing for the aging of America by designing a retirement system that is both more secure for the old and less burdensome for the young. Demographic circumstances will never again be so favorable for Social Security reform. With a small (Depression) generation in retirement and a large (Baby Boom) generation still in the workforce, America is enjoying the last years of a “Demographic Indian Summer.” However, this window of opportunity is closing fast.
 This number and all others herein are expressed as 2004 constant (i.e., inflation adjusted) dollars. They are based on the so-called “Intermediate” or central forecast of the 2004 Social Security Trustees' report.