Tuesday, 16 July 2013

Policy Challenges Posed by the Aging of America

library Leonard E. Burman, Rudolph G. Penner, C. Eugene Steuerle, Eric Toder, Marilyn Moon, Lawrence H. Thompson, Michael Weisner, Adam Carasso

A discussion briefing prepared for the Urban Institute Board of Trustees meeting, May 20, 1998. Contributing authors include Urban Institute researchers Len Burman, Rudolph Penner, Gene Steuerle, Eric Toder, Marilyn Moon, Larry Thompson, Michael Weisner, and Adam Carasso.

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

Note: This report is also available in the PDF format, which many users find more convenient when printing.

Introduction

The nation is about to experience a great demographic shock. Between 2010 and 2030, the over 65 population will rise over 70 percent while under current law the population paying payroll taxes will rise less than 4 percent (CBO, 1997). The implied shortage of workers and the rapid increase in the number of retirees has profound implications for the private sector and the amount of economic growth that can be expected in the future. It also has important implications for public policies affecting the economic and physical well-being of the elderly population.

Not only are retirement and health benefits promised to a rapidly growing proportion of the population, but the real benefits per person also rise significantly. In the public sector alone, there are large unfunded commitments for Social Security, Medicare, and Medicaid benefits covering long-term care, as well as for other programs such as Supplemental Security Income for the elderly. In part because of these public benefits, individuals privately save only modest amounts for their retirement. Private employers, in turn, face the prospect of rising costs for defined benefit pension plans and largely unfunded retiree health benefits, as well as requirements to pay expensive health insurance costs of older workers in lieu of Medicare if they offer any health care to younger workers. Moreover, increases in life expectancy threaten the retirement security of even those who have saved on their own by forcing them to stretch their retirement savings over a longer period of time. Meanwhile, the private sector provides strong incentives for older workers to retire early—in response partly to incentives in government programs, partly to their own pay compensation structure—and individual workers have responded by taking advantage of those early retirement opportunities.

The public policy debate has, thus far, been focused on Social Security, the huge public pension system that has survived up to this point by taxing a large labor force to pay for increasingly generous retirement pensions. Now, the baby boomers who paid for the expansion of Social Security are getting older themselves, and when they retire, they both deplete the ranks of workers and add substantially to the list of beneficiaries dependent on transfers from others. Although Social Security has been accumulating a surplus to pay for future liabilities, that fund covers only a tiny fraction, perhaps 10 percent, of these future liabilities. If current policy were to continue, the Social Security trust fund would be exhausted around the year 2032 and current tax rates would then be only sufficient to support 75 percent of current commitments (OASDI Trustees, 1998).

The situation for Medicare—public health insurance for the elderly—is even more challenging. The program must serve the same rapidly growing population as is served by Social Security, but the cost of serving each recipient is also increasing faster than the growth in wages and income per capita. The same dynamic—rapidly escalating health costs for growing numbers of elderly people whose long-term care is paid for by Medicaid—threatens to cause that part of the budget, financed entirely out of general revenues, to be a growing economic burden. And since individuals bear a substantial share of the costs of care, they are spending an ever larger share of their own incomes for health care. Right now some states are responding by simply denying permits to build nursing homes, for fear that once built the state will have to use them. And these problems are arising even while a baby bust population—those born in the depression and World War II—is filling the ranks of the elderly.

The aging of America is partly a triumph of affluence and major advances in medical technology that have increased life expectancies and the quality of life for the aged. Indeed, one might say medical advances are actually making an aging America younger. Social Security, Medicare, and the expansion of private pensions have made it possible for unprecedented numbers of people to enjoy long retirements free from poverty and in better health than could have been imagined when Medicare was invented. President Clinton suggested in his 1998 State of the Union address that a person born in 1998 might live to see the beginning of the 22nd century—that is, live to 102. Without changes to current law, these persons will be able to retire at age 62 and receive 40 years of benefits.

But how will society pay for its good fortune? Social Security and Medicare currently account for about 7.4 percent of gross domestic product (GDP). Assuming current policies continue, the percentage will increase to about 13 percent of GDP by the year 2030 if full benefits are financed despite the depletion of the trust fund, and the share of GDP devoted to the elderly will continue to grow after that (OASDI Trustees, 1998). Thus our children will face unprecedented tax burdens if all promised benefits are funded.

The private sector will also face extraordinary stresses. The same economic and demographic forces that threaten the public sector will also cause private sector pension and health costs to grow. We have mentioned the requirement to pay health benefits for older workers, an incentive that discourages maintaining or hiring these workers. Many private pension plans built around "defined benefits" also discourage hanging onto older workers. And some believe that seniority pay scales tend to make compensation costs of older workers exceed their productivity, thus again encouraging earlier retirement. In many ways, the nation has spent very large shares of its incremental wealth over the past half century building up a system where individuals now expect to retire younger, live longer in retirement, and have increasingly higher levels of retirement and health benefits.

All these forces come to a head in the very near future, as labor becomes much more scarce. Indeed, the current debate over a federal budget surplus hides the fact that this is merely the calm in the eye of the storm—a calm caused by the retirement of a baby bust population today right before the baby boomers both swell the ranks of the elderly and simultaneously deplete the ranks of the working population. Their demands will play out not only in the public sector but also in the private sector, as when they draw down their pensions and potentially reduce national saving.

Policy-makers have long recognized that reforming the financing and operation of entitlements—especially Social Security, Medicare, and Medicaid—is essential, but entitlement reform is dreaded as the "third rail" of American politics. (See Figure 1.) Unfortunately, unlike the meteor, this problem won't go away by itself. Policy makers are now showing signs that they recognize this, and some are starting to compete with each other to put their own Social Security plans on the table.

Now is the best time to start to deal with programs for the elderly both from a policy perspective and a political one. The policy imperative is obvious. The longer we defer addressing entitlements, the more daunting the task becomes, and the greater the decline in public and private funds available for the non-elderly, including education and other investment in the young. But the political stars are also aligned particularly well, though only briefly. For the next few years, barring a recession or irresponsible spending increases and tax cuts, the budget is likely to produce significant surpluses. As Figure 2 shows, however, this happy state of affairs will be transitory unless the growth of entitlements can be slowed. For this reason, the budget provides both the imperative and the means for tackling entitlements now. The solutions cannot be expected to be permanent. Policies will be continually adapted to changing conditions, but major changes are required quickly.

Defining the Problem

The problem is peculiar—largely the result of good news. Americans are living longer than ever before, enjoying longer retirements, receiving ever higher annual Social Security benefits, as well as rapidly rising levels of medical care to ameliorate or even avoid many of the once inevitable ills of old age. In 1940, the average 65-year-old male could expect to live 12 more years, the average female, 13 years. (See Figure 3.) By comparison, life expectancies for men reaching age 65 today have increased by almost 4 years to 16 years in retirement; for women, an increase of 6 years to 19 years. Projecting current trends in mortality, life expectancies for both men and women will increase by another three years by 2075, but the current revolution in medical technology could lengthen actual life spans even faster. In many ways, those entering their "golden years" in the next century will have picked the best time in history to grow old.

One issue is the way in which this happy confluence of developments interacts with public and private sector policies designed to address completely different realities. Most attention has focused on Social Security, the public retirement system enacted as part of the New Deal. Although Social Security was obviously aimed at alleviating poverty among the elderly, it is easy to forget now that another purpose of its depression-era sponsors was to induce older workers to leave the workforce to make what they believed naively to be a fixed supply of scarce jobs available to younger workers. With respects to both objectives, Social Security has been a remarkable success. The poverty rate among the aged has declined rapidly, and the retirement age for males has declined steadily since its inception. In 1950, the typical male retired at age 69. (See Figure 4.) By 1994, the average male retiree was under 64 years of age. Combined with longer lives, this means that the typical person retires almost one decade earlier than in 1950 and now expects to live close to one-third of his or her adult life in retirement.

Employers in the private sector have had their own reasons to reinforce incentives created under Social Security. Defined benefit pension plans, which still cover most Americans who have pensions, typically include strong incentives to retire after a certain age (usually 65 or younger). Pension benefits rise with compensation level and years of service up to a certain age, but not beyond. Employers often encourage early retirement for several reasons, including the high cost of medical insurance for older workers and the possible decline in the productivity of workers as they age.

Demographic effects compound the imbalance between retirees and workers. The huge post-war baby boom provided a surge of workers just as the number of Social Security recipients was burgeoning. (See Figure 5.) The increase in workers muffled the effects of ever more generous Social Security benefits, but only temporarily. As the baby boomers retire, they will be followed by fewer workers per beneficiary. This, by itself, would threaten the viability of a pay-as-you-go system. But in combination with earlier retirement and longer life expectancy, it means that workers 30 years from now will be supporting many more retirees than current workers do. Whereas there are more than three workers for every retiree at present, by the year 2030, that ratio will have fallen to two-to-one, and it continues to fall after that, barring an increase in fertility. (See Figure 6.)

The bottom line? The Old Age and Survivors and Disability Insurance program is out of balance, and the private pension system will not provide adequate benefits for a majority of workers. Over the long run, Social Security receipts will fall far short of expenditures, and the problem will get worse over time because people are living longer, retiring earlier, and real Social Security benefits have increased. (See Figure 7.)

Medicare and Medicaid face even more severe problems for the same reasons, but also because medical costs have historically increased far faster than wages or income in the economy. There is also a debate as to whether longer life spans mean better health or simply more time for people to contract costly diseases and suffer the failure of joints and organs that can be repaired or replaced, but only at great cost. Another issue is that modern health care may allow people with severe illness to survive longer but exhaust their savings in the process and place them in nursing homes where they are wards of the state under Medicaid. Note that current designs of public and private insurance may add to costs by encouraging cost-increasing technology, rather than cost-saving technology.

In summary, the public and private costs imposed by an aging and longer-living population will absorb a higher and higher share of our economic resources for the foreseeable future, especially if current practices are sustained. The private sector will see pension outlays rise to the point where there could be net outflows from private pension funds (Schieber and Shoven, 1994). An aging population will also face increased health costs, but it is important to note that the health costs for younger workers are also likely to rise faster than wages and income under existing incentives.

The distributional consequences of private sector developments are not clearly understood. In particular, the replacement of defined contribution for defined benefit plans may have a significant negative effect on the lower middle class. On the other hand, as noted above, defined benefit plans may especially encourage employers to offer earlier retirement because of their high compensation costs in later years.

Programs for the elderly should not be considered in isolation. As the nation absorbs itself more and more in these efforts, it has less available for other needs. The non-retirement and health part of the federal budget has been in decline since the mid-1970s whereas mandatory spending programs, such as Social Security and Medicare, which currently consume about 7.4 percent of GDP, will increase to about 13 percent of GDP by 2030 if currently scheduled benefits are fully funded and if economic growth and demographic developments follow the intermediate assumptions used by the Trustees of the OASDI Trust Funds. (See Figure 8.) In short, current policies are unsustainable.

Easy Solutions Do Not Exist

In a sense, the solution to the coming entitlement problem is straightforward: slow the growth in benefits and/or increase the budget and private resources available in the next century. But straightforward does not mean easy. Not meeting outstanding promises will spark formidable political opposition. Under today's law, younger workers now in the labor force will receive a very low rate of return on the payroll taxes that they paid and that were paid on their behalf by employers. Of course, this is true of any transfer program, but for decades programs for the elderly managed to provide higher rates of returns to retirees simply by raising tax rates consistently for half a century on younger workers. But for the long-term future, any tax increase or benefit cut to bring the system back into balance can only whittle the rate of return down further. Moreover, people who have made retirement plans based on the current benefit structure will have their plans disrupted by changes in the law. The pain can be spread over the generations in different ways using different techniques, but there is no avoiding the fact that adjusting our many promises downward will be politically difficult.

The political barriers that make reform difficult are buttressed by institutional barriers. For example, one option is to encourage people to retire later. But public and private trends are at cross purposes with this objective because they both have combined to drive retirement ages down over time (Burkhauser and Quinn, 1997). That decline seems to have abated, but reversing the decades-long trend will be extremely difficult.

Another possibility is to increase the rate of economic growth and so increase the supply of resources available to retiring baby boomers. Typically, this approach would increase the supply of saving to enhance investment and productivity growth.

But private saving has been sliding downward for the past two decades, and economists know little about why it is happening or how to reverse it. A good case can be made for either increasing public or private saving, but government has limited ability to stop individuals from offsetting those moves through reductions in their other saving or increases in their borrowing. Exacerbating this problem is a decline in employer contributions to private pensions as a percentage of GDP, as the rising stock market has reduced the need for new inflows. At the same time rising health insurance costs have put downward pressure on real wages. (See Figure 9.)

Benefit increases and tax increases have, for the most part, risen hand-in-hand over the past four decades. (See Figure 10.) For about 75 percent of workers, payroll taxes (including what employers contribute) now exceed income taxes. Certainly, these tax rates cannot be raised continually without creating hardships and economic distortions. Other approaches to increasing the trust fund surplus will be explored below, but few can be expected to increase national saving.

Labor force growth rates may be enhanced by immigration, but it must be remembered that immigrants also earn a claim to benefits in the long run. Also, immigration rates pale in significance relative to fertility rates. Even if the immigrant population is continually increased at twice its current rate, it has little effect on the long-run financial health of the system. Finally, it is unrealistic to count on foreign investments bailing out the United States. Virtually every developed country faces demographic problems at least as serious as, and often more serious than, ours. (See Figure 11.) Thus, the only solution is getting our own house in order. It is a daunting task, and one that must be initiated with only severely inadequate information at hand. The research proposed by the Urban Institute should begin to close that information gap.

The Implications of Current Policy

The previous section argued that current policies are not sustainable. Almost all experts agree with this proposition. But how much policies must be changed to ensure a stable future remains open to debate.

Most assessments of the coming entitlement squeeze rely on the "intermediate" economic and demographic assumptions of the trustees of the Social Security system. The estimates provided in the introduction to this proposal were based on those assumptions, as are those of nearly all the many groups working on the problem. But just how accurate will those assumptions prove to be?

At one extreme, it is not difficult to imagine an economic disaster. Growth is sure to slow as massive waves of retirements slow the growth of the labor force. Such a slowdown will be exacerbated by potential reductions of saving and investment as the baby boomers retire. Net withdrawals will deflate private pension funds (Schieber and Shoven, 1994), and without reforms, Medicare and Social Security will continue to grow apace and add to public deficits. Any increased absorption of private saving by government will further reduce investment and productivity growth. In theory, the process can feed on itself: slower growth and rising obligations for interest on the public debt beget higher deficits that further reduce growth, raise interest rates, and so on, until the whole system collapses. This disastrous scenario was outlined by the Congressional Budget Office (CBO) in March 1997 and the General Accounting Office (GAO) in 1998. Such outcomes assume that neither tax laws nor benefits are changed, and therefore, a growing portion of benefits continue to be debt financed after the trust funds are exhausted.

But such scenarios may greatly underestimate the private sectors ability to adjust to shocks. After the oil shocks of the 1970s, some prognosticators were forecasting $200 per barrel prices of oil by now. Instead, the world is swimming in oil. It is possible that the private market will respond equally well to demographic changes.

However, the analysis must begin in a cloud of ignorance. The effects of demographic change on the private sector have barely been examined. Most important is understanding the implications for economic growth. If there is a significant chance that the CBO disaster scenario will come true, reforms should be swift, sweeping, and dramatic. But if the private sector is much more resilient than that analysis implies, or if the problem has been exaggerated, because it assumed far larger budget deficits in the future than now seem likely, then there is more time for incremental reform.

The demographic assumptions used in most projections are also dogged by uncertainty. In particular, baby boomers could live longer and be healthier in old age than is anticipated. If so, pension costs might rise along with total health costs but annual health costs per person do not necessarily need to rise at the same rate as today. Again, much of this depends upon incentives in the public and private health insurance markets and whether they continue to induce so much cost-increasing technology.

Still more uncertainty revolves around projecting the rate of economic growth. Projecting growth requires projecting labor force size, its skill level, the rate of investment, and the rate of technological change. At least three main issues arise in such projections:

1. Labor Force - When the baby boomers retire, labor markets will lose valuable human capital and the pool of potential replacements is growing very slowly. The question is the extent to which the private labor market will cushion the blow by taking steps to increase labor force participation rates. For example, will employers reduce incentives for early retirement? Will they make it easier for workers to scale back their hours while remaining at their career jobs? Will rising wages delay retirement and induce younger groups to work harder for more hours? Will older workers remain healthier, and will good health induce them to work longer?

2 . Investment - The rate of investment is limited by the quantity of private saving, the amount of saving absorbed by public deficits, and the amount that the United States can borrow from abroad.

Private saving - As baby boomers withdraw private pension funds, private saving will be depressed and rates of return will be pushed upward. That push might reduce the fall in saving, but economists are not in agreement on this issue. A totally separate question is whether baby boomers are now saving enough for their retirement (Bernheim, 1996; CBO, 1993). If not, can we soon expect a surge in saving as they try to catch up?

Public deficits - Deficits in the long run will partly be determined by our deficit policy over the next few years. If the economy avoids recession, current policy implies moderate surpluses for awhile. If the surpluses last into the early 21st century, much of the national debt could be retired. The reduction in the interest bill relative to GDP could conceivably offset a significant portion, although not a majority, of projected future costs for the elderly. Thanks to compounding, relatively small changes in budget policy now or soon can have huge long-run effects on the deficit and, in turn, even the economic growth rate. Spelling out the consequences of various scenarios using simple growth models, as we plan to do, will help policy makers weigh the options.

The difficulty of forecasting the relative increase in health costs and their impact on Medicare and Medicaid spending heighten long-run uncertainties over public deficits. Also, as noted, current budget projections are based on actuarial estimates that may understate baby boomers' expected life spans.

International borrowing - In recent years, capital inflows from abroad have enabled the United States to maintain a decent rate of capital investment despite a low saving rate. As a rough rule of thumb, an abrupt fall in U.S. saving of one dollar increases the capital inflow by about fifty cents. But will this flow be enough in the 21st century? Probably not—unless we pay a much higher interest rate on such borrowing, since almost all developed nations will be in the same demographic boat. Most likely, the shortage of saving will be worldwide and some nations, such as Japan (which supplied a significant share of world saving in the past), will suffer a much more severe shock, and one that comes sooner, than in the United States.

3. Technological Change - The rate of technological change is an extremely important determinant of economic growth. In early models of economic growth, technological change was considered independent of the growth rate of the economy or of the capital stock (Solow, 1956). Later, modelers assumed that investment was needed in order to take advantage of technological advances (Solow, 1957). More recently, economic growth, human capital formation, and technological change are seen to feed on each other as growth creates a market for technological change and provides more resources for research and development and human capital formation (Romer, 1996; Romer, 1990). The decline in labor force participation could, therefore, reduce that supply of inventiveness and creativity that was the real source for technological advance.

The accuracy of these different views is unlikely to be settled soon. So various theoretical approaches must be used to investigate various policy changes. If the pace of technical change is independent of the growth rate and level of physical investment, policies that raise saving or investment will work slowly and poorly. If the rate of growth and investment does influence the rate of technical change, such policies will work faster and better.

A particularly important question, relatively untouched by economic research, is the extent to which the nature of capital investment will evolve to reflect the increasing scarcity of labor. Typical growth models assume that labor and capital are to a large degree complementary—that is, additional machines become less and less productive unless accompanied by additional workers to operate them. The success of strategies to save to mitigate the effects of future labor shortages will require that capital become more of a substitute for labor. Thus, the standard growth model must be extended to account for this kind of endogenous technical change. More importantly, we need to look for evidence on the extent to which this kind of adaptation is possible.

Federal Policy

The invention of Social Security during the New Deal forever changed the nature of the federal government and its budget. For the first time, the federal government accepted major responsibility for the well-being of the elderly. That commitment was expanded greatly as Social Security eligibility was broadened in 1950; Medicare provided health benefits starting in 1965; and the pension system was automatically indexed in 1972. By 1997, Social Security (including disability and survivors insurance) and Medicare constituted 55 percent of total domestic spending apart from interest payments.

In a way, the federal government wrote a social contract in which each generation agreed to help support the retired elderly in return for receiving similar support from subsequent generations. This New Deal substituted for traditional transfers from children to their parents—a system that had been supplemented in the late 19th and early 20th centuries by almost universal Civil War pensions, the beginnings of a corporate pension system, and transfers to the elderly through union pensions, fraternal organizations, state and local governments, and charities. By the 1930s, the traditional system was under enormous pressure. Civil War pensions had largely disappeared. Better health meant that many more persons were making it to old age, and the Great Depression reduced the ability of children and others to support the aged financially.

Social Security was invented in response to these growing pressures. In many ways, Social Security and Medicare are among the most successful social programs ever created by government. Government-provided pensions and health care were much more generous than the traditional systems that they replaced and have become ever more generous over time. They have converted the elderly population from one of the poorest segments of society to one now having a poverty rate lower than that of the rest of the population. (See Figure 12.)

Because the burden imposed by the system is projected to grow rapidly in the 21st century, the debate over Social Security and Medicare has begun in earnest. There is no shortage of policy proposals: various groups have advocated differing solutions with varying degrees of ideological fervor. But there is a severe shortage of knowledge on the effects of these proposals on current and future beneficiaries and on the economy. This knowledge gap allows some groups to make exaggerated and unsupported claims for their proposals and to sweep major problems under the rug. The knowledge gap also greatly increases the probability that even well-considered and well-intended proposals will have severe unintended consequences.

The effects of commonly discussed policy options can differ significantly for different generations and for different income groups within each generation. Few if any proposals are examined on the simple basis of whether they serve well those most in need. Impacts on the private sector can also vary dramatically. Some proposals for reducing the negative growth impact of the baby boomers' retirement emphasize incentives to get workers to retire later while others aim to increase national saving, increase the progressivity of Social Security and Medicare through means testing, or deliver Medicare more efficiently.

Later retirement - Large government savings and a smaller negative impact on future economic growth could be achieved by inducing people to work longer. However, as described earlier, such proposals buck a national trend that has been reinforced by corporate downsizing and the private sector's desire to rid itself of the high cost of providing health insurance to older workers. Even so, it is ironic that people have been retiring earlier and earlier as the expected life spans and the health of the older population have been improving. Part of this trend is simply explained by the fact that the population can more easily afford to enjoy more leisure time as it has become richer, and we have decided for some time as a nation to spend our new wealth there rather than on other needs such as those facing many of our children.

Under current law, the normal retirement age under Social Security is scheduled to gradually increase to age 67 by the year 2027. A number of proposals would raise the normal retirement age faster and to a greater extent than under current law. Although most of these proposals keep the earliest retirement age at 62, an increase in this threshold could also be contemplated. Increasing the normal retirement age, but not the age of actual retirement, basically results in a larger actuarial reduction in benefits and, therefore, lowers the portion of recent earnings that is replaced at any specific age of retirement, say at age 65. Also, raising the normal retirement and/or early retirement ages might prompt some employers to reduce or drop early retirement incentives. More pressure could be put on the private sector by raising the age at which private tax qualified plans can start paying benefits. In any case, a cut in benefits is likely to induce some increase in work effort by delaying retirement (Lumsdaine, 1996; Lumsdaine, Stock, and Wise, 1994), but how much is in dispute and how individuals or the private sector might react are unknowns requiring further study.

For Medicare, the eligibility age remains at 65 under current law even though the normal retirement age for Social Security will increase to 67. One proposal is to set the eligibility age for Medicare at the normal retirement age for Social Security. But Medicare operates very differently than Social Security. For example, many of those losing eligibility for Medicare might find it impossible to purchase private insurance without major reforms in how the private sector operates. In response, one variant would be combining this change with a "buy-in" to the system as early as age 62. (President Clinton recently proposed allowing a buy-in for even younger individuals if they met certain conditions.) But then if people continually live longer Medicare would continue to gradually expand to cover larger and larger shares of the population.

Savings from any increase in the eligibility age would be moderate over time because younger retirees consume much less health care than older ones do. Raising the age for Medicare eligibility is likely to have a powerful effect on increasing the age of retirement, although allowing people under age 65 to buy into Medicare is likely to have the opposite effect. It could also force the private sector to adjust compensation patterns to avoid cost increases because employers would have to provide private health insurance to a large portion of those who stay in the workforce longer. Alternatively, they might accelerate the decline in retiree health insurance. Finally, increasing the age of retirement or the eligibility age for Medicare would probably also increase pressure on the disability system.

Many proposals would eliminate the current Social Security retirement earnings test for those aged 65 to 69. The test reduces benefits by one dollar for every three dollars earned above $14,500 in 1998, but will rise to $30,000 by 2002 under legislation enacted in 1996 (Annual Statistical Supplement, 1997). A large body of research indicates that the test influences those over 65 to retire earlier even though working longer increases future benefits (Friedberg, 1997). The reward for longer work has been increased over time and will continue to be adjusted over time toward an actuarially fair amount if one does not take into account that almost all new taxes contributed receive no return at all (Blinder, Gordon, and Wise, 1980; Steuerle and Bakija, 1994). But this incentive is poorly understood and is not having much of an effect on work decisions. Intensely studied already, the retirement test is not now a high research priority. But the issue should be revisited as part of the panoply of institutional means by which our society has encouraged earlier retirement.

Other possible techniques for delaying retirement would lower the cost to employers of employing older workers or reduce the tax bill of those workers themselves. Such policies include eliminating the Social Security tax after age 65, making Medicare the primary payer of health costs for those eligible, or creating wage subsidies (e.g., an Earned Income Tax Credit) for those over 65.

Increase national saving - Many individuals and groups have advocated proposals to increase national saving by moving toward full funding of our Social Security obligations.

Fuller funding of Social Security can be approached in a variety of ways. First, the payroll tax can be decreased and individuals can either be encouraged to save the tax cut voluntarily for retirement (Moynihan, 1998) or the savings can be mandated (Tanner, 1996). Such proposals are almost always combined with significant cuts in prospective benefits, especially for younger workers, and the hope is that the proceeds from the savings accounts will replace some or all lost benefits.

The development of the private accounts is often referred to as "privatizing" Social Security. But that is a misnomer for several reasons: the savings are mandated, and the government dictates how the savings can be invested, when they can be withdrawn, and whether they must be annuitized when withdrawn. In fact, individual accounts can be managed by government (just as todays thrift accounts owned by civil servants), or by private investment funds. Under most proposals, private accounts are combined with a guaranteed minimum benefit, either through a traditional but slimmed-down Social Security system or through a government guaranteed floor under income from the investments (as in Chile). (For a summary of Chilean pension reform, see Diamond and Valdés-Prieto, 1994.)

Although a vital issue for research, it is difficult to know how much voluntary or mandatory individual saving accounts would actually increase saving. It is naive to assume that every dollar of mandated savings is a dollar that would not be saved otherwise. More affluent people can simply relabel existing saving to satisfy any mandate. Less affluent people may choose to borrow more. But there are limits on how much people can borrow at reasonable interest rates, and some individuals may appreciate the discipline imposed by mandated accounts. For these reasons, mandated accounts may have some positive effect on saving.

As noted above, almost all plans that call for individual savings accounts also reduce Social Security benefits. Many recipients will decide to save more in order to replace some or all of the lost benefits, and some would work longer, thus reducing the amount of saving they would need for retirement.

Individual accounts face three formidable hurdles. First is the huge transition problem that confronts most reforms of today's retirement systems. Individuals working when individual accounts are introduced will be expected to finance two retirements. They must continue to pay payroll taxes to the retired population and they will be expected to make deposits in individual accounts to help finance their own retirement.

Second, the current Social Security system is highly redistributive. Huge transfers occur from rich to poor, from men to women (because women live longer), and from two-earner families to one-earner families (because the second earner, despite increased payroll taxes, typically receives little or no increase in benefits over and above the spousal and survivor benefits already available). In contrast, individual accounts are not redistributive. The eventual pension depends solely on the individuals work history and success as an investor. To retain any of the current redistributions in the present system or dampen the political opposition from those who lose in a reform, other mechanisms will be necessary. In most proposals, individual accounts are complemented by a residual Social Security system or a new pay-as-you-go system that provides minimum benefits. The former retains the same redistributions present in the current system, but gives them less weight. The latter helps the poor some, but does little for other losers. In designing new approaches, distributional analyses are clearly needed so that some groups are not allowed to fall through the cracks while others reap a bonanza from reform.

Third, independent accounts can be much more expensive to administer than Social Security. To some degree, this is because they would be expected to provide more services—regular reports on wealth accumulation and brokerage services for trading among different types of assets. But even in the absence of such services, private individual accounts are likely to be relatively expensive to administer because they do not enjoy the economies of scale enjoyed by a more uniform, but perhaps less flexible, system administered by a single agency, the Social Security Administration.

Despite these problems, individual accounts of some type retain considerable political appeal. Social Securitys rate of return on the payroll taxes paid by each successive cohort of retirees will be falling rapidly under current law. To the extent that the current pay-as-you-go system is balanced by cutting benefits or raising taxes, the rate of return falls still further. Thus, many have concluded that individuals should be encouraged, or perhaps forced, to enjoy the higher rate of return available on private assets in providing for their own retirement, even though they would face an increase in investment risk.

Other proposals would attempt to increase national saving by increasing the surplus in the Social Security trust funds. However, Congress has the goal of balancing the "unified budget," which includes not only the Social Security trust fund but also nearly all other government activities. Consequently, any increase in the Social Security surplus is likely to be offset by an increase in the deficit of the rest of the government, so there would be no net increase in national saving. Changes in the Social Security surplus would affect national saving if the non-Social Security budget had to be balanced, but since the Social Security surplus is now running about $100 billion and rising in the short run, such a change would be traumatic and politically difficult.

Numerous types of benefit cuts have been suggested aside from raising the normal retirement age. Among them are cost-of-living adjustment (COLA) cuts for the currently retired on the grounds that the current consumer price index (CPI) overcompensates for inflation. Changes in the indexing and the structure of the formula that determines initial retirement benefits (mainly reductions in the rate of growth of annual benefits from one generation to the next) have also been suggested to cut benefits. The approaches vary in terms of how different age cohorts bear the pain associated with lower benefits.

One faction of the recent Advisory Council on Social Security suggested improving the financial status of the trust fund partly by allowing it to invest in equities and not just Treasury securities. But at current relative rates of return, swapping bonds for equities, taken by itself, would be a zero-sum game since the gain in trust funds income would be exactly offset by a loss in private income. The private sector cannot be expected to play in this game until the rate of return on bonds rises relative to that on equities. This could significantly raise the interest bill on the public debt and partially offset any gain to the trust fund. (Each basis point in added interest cost raises the interest bill on the public debt about $100 million in the first year, $200 million in the second, and more than $300 million per year in the long run.)

Moreover, rearranging the ownership of assets between the public and private sectors does nothing to address the fundamental problem that there will be too few resources available to finance the obligations of the government to the elderly. In this sense, policy is only successful if it increases the amount of capital available in the future. Public investment in equities may have little effect other than to change who owns bonds and equities.

Medical savings accounts (MSAs), backed up by catastrophic health insurance, have been proposed to increase saving and pay for health insurance more efficiently. Funds deposited in tax-favored accounts could be used to buy health insurance and to pay deductibles and co-payments. Any surplus in the accounts could eventually be used for other purposes. Presumably, health insurance companies would compete to provide efficient health insurance to private buyers.

One difficulty is that if MSAs exist along with the traditional Medicare system then healthier individuals are likely to opt for MSAs while the unhealthy stay with Medicare, thus greatly increasing its per capita costs. A similar problem arises in some proposals to allow individuals to buy into health maintenance organizations (HMOs). This "adverse selection" problem is more severe between the healthy and unhealthy elderly and less a problem between old and young (Burman, 1997). Further, to achieve saving, deductibles would have to be set very high, placing low-income individuals at a substantial disadvantage.

Would MSAs increase national saving? That depends on whether they would increase private saving by more than their favored tax status increases the public sector deficit—a great controversy. It also depends on whether these accounts would be fully used for health expenses. The answer largely reflects whether the assets in such accounts result from new saving or from shifts by investors of current savings from fully taxable to tax-favored accounts. On the one hand, any tax concessions granted such accounts do not alter the incentive to save of someone who would have saved more than the contribution limits in any case. On the other hand, individuals may not be well-informed or rational in making saving decisions (Thaler, 1992). To the extent that financial institutions advertising such accounts publicize the benefits of saving and the magic of compound interest, saving might be enhanced.

Research on the effect of tax-favored savings accounts on total saving is voluminous. But no consensus has evolved (Engen, Gale, and Scholz, 1996; vs. Poterba, Venti, and Wise, 1996).

Increase the progressivity of Social Security and Medicare - The Concord Coalition has suggested means testing Social Security and Medicare. But any highly progressive scheme can severely penalize those who save for their own retirement. When superimposed upon the current tax system, additional tax payments combined with benefit reduction as income increases could impose the equivalent of a marginal tax rate exceeding 100 percent! Where the implicit marginal tax rate is high, individuals may increase consumption before retirement, buy excluded assets such as housing, or transfer assets to their children. This has partly been the experience in Australia, which allowed lump sum payments in a Social Security type system, only to find many persons spending the lump sum and then falling back on a welfare-like part of the system.

Social Security benefits can be means-tested by reducing benefits by a certain portion of each additional dollar of income. Another approach that avoids many of the difficulties of annual means testing is to alter the benefit structure to reduce the replacement rate for higher earners. The approach bases reductions upon past lifetime earnings, rather than current annual income. (Other types of income besides earnings wouldn't count under this scheme.) Medicare similarly can be means-tested by increasing the part B premium for physicians insurance as income rises or by increasing more affluent recipients' deductibles and co-payments. However, many details of these proposals have yet to be examined, such as whether or not it is possible to change monthly premiums on the basis of annual income.

Deliver health care to the elderly more efficiently - Increasing the efficiency of the health care system and hence lowering the cost of care has been a long-sought goal of many reforms, but one that remains elusive. Today's Medicare program offers a wide range of choice of service providers and few constraints on the use of care. Greater reliance on managed care and on competition among providers may hold promise, but troubling questions remain. Analysts do not fully agree on the extent to which managed care, which is now common among the working population, has held down costs while maintaining high quality care. And, as implied in the above discussion of MSAs, markets work imperfectly in health care, often rewarding those insurers who cover healthier customers rather than those who provide care most efficiently. Moreover, more efficiency almost inevitably means fewer services, even if those services by another standard are not cost-effective.

Many proposals to enhance efficiency have the elderly pay a higher proportion of the cost of their own health care. This is supposed to dissuade them from buying treatments of dubious value. It is clearly important to examine the interaction between such proposals and proposals for cutting Social Security benefits. If the elderly must pay a greater share of their health costs for the sake of efficiency, perhaps the minimum benefit under Social Security should be higher and the SSI program more generous.

An alternative might be distributing health care vouchers to the elderly to help them buy their own health insurance or to help them pay any additional deductibles or coinsurance. Medicare's counterpart to "privatization" would be to give beneficiaries a fixed dollar amount and let them negotiate their own insurance arrangements. Presumably this would put pressure on beneficiaries to become more aware of the costs of their care and to seek insurance plans that are able to more efficiently offer care. Those who wish for less restrictive plans would be required to pay these costs out of pocket. This emphasis on the private market would require beneficiaries to become very sophisticated consumers. It would create an individual insurance market, however, which is often characterized by high administrative costs to cover marketing, sales, and processing of enrollment that would forego some of Medicare's efficiencies.

Conclusions

The nation faces a great challenge as it moves into the 21st century. The elderly population will live longer on average and the numbers of elderly will rise. It is impossible to keep the promises made to them without large tax increases on the young, and even then, reducing the share of the budget spent on all other societal needs. Numerous proposals have been made that would scale back benefits, and some would use individual savings accounts or increased government and Social Security surpluses to try to increase national saving and, eventually, to replace lost benefits. Proposals should be judged according to their ability to meet various goals or principles. Among the most important are fairness, economic efficiency, and economic growth.

More research is needed on the fine points of proposals to fend off a retirement crisis and put Social Security and Medicare on a secure footing. Such research should feed into a national dialogue on Americans' shared retirement future. # # #

Bernheim, B. Douglas. 1993-96. "Is the Baby Boom Generation Saving Adequately for Retirement? Summary Report.quot; New York: Merrill Lynch, Pierce, Fenner & Smith Inc.

Blinder, Aland S., Roger H. Gordon and Donald E. Wise. 1980. "Reconsidering the Work Disincentive Effects of Social Security." National Tax Journal Vol. 33, pp. 431-442.

Burkhauser, Richard V., and Joseph F. Quinn. 1997. "Pro-Work Policy Proposals for Older Americans in the 21st Century." Policy Brief No. 9/1997. Syracuse, NY: Syracuse University Maxwell School of Citizenship and Public Affairs, Center for Policy Research.

Burman, Leonard E. 1997. "Medical Savings Accounts and Adverse Selection." Washington, DC: The Urban Institute Press.

Congressional Budget Office (CBO). 1997. "Long-Term Budgetary Pressures and Policy Options." Washington, DC: U.S. Government Printing Office (March).

__________. 1993. "A CBO Study: Baby Boomers in Retirement: An Early Perspective." Washington, DC: U.S. Government Printing Office.

Diamond, Peter, and Salvador Valdés-Prieto. 1994. "Social Security Reforms." In The Chilean Economy: Policy Lessons and Challenges, edited by Barry P. Bosworth, Rudiger Dornbusch, and Raúl Labán. Washington, DC: The Brookings Institution, pp. 257-328.

Engen, Eric M., William G. Gale, and John Karl Scholz. 1996. "The Illusory Effects of Saving Incentives on Saving." Journal of Economic Perspectives, Vol. 10, No. 4, pp. 113-138.

Friedberg, Leora. 1997. "The Labor Supply Effects of the Social Security Earnings Test." Discussion Paper 97-01, January. San Diego, CA: University of California, San Diego.

General Accounting Office (GAO). 1998. Budget Issues: Long-Term Fiscal Outlook. Testimony before the Committee on the Budget, U.S. Senate, of Paul L. Posner, Director, Budget Issues, Accounting and Information Management Division. GAO/T-AIMD/OCE-98-83.

Lumsdaine, Robin L. 1996. "Factors Affecting Labor Supply Decisions and Retirement Income." In Assessing Knowledge of Retirement Behavior, edited by Eric A. Hanushek and Nancy L. Maritato. Washington, DC: National Academy Press, pp. 61-122.

Lumsdaine, Robin L., James H. Stock, and David A. Wise. 1994. "Retirement Incentives: The Interaction Between Employer-Provided Pensions, Social Security, and Retiree Health Benefits." NBER Working Paper No. 4613. Cambridge, MA: National Bureau of Economic Research, Inc.

Moynihan, Daniel P. 1998. "Social Security Saved!" Address to the Institute of Politics: Spring Exercise on Social Security Reform. John F. Kennedy School of Government, Harvard University. Senate web site: http://www.senate.gov/~moynihan/socsec.htm

Poterba, James M., Steven F. Venti, and David A. Wise. 1996. "How Retirement Saving Programs Increase Saving." Journal of Economic Perspectives, Vol. 10, No. 4, pp. 91-112.

Romer, Paul M. 1996. "Why, Indeed, in America? Theory, History, and the Origins of Modern Economic Growth." New Growth Theory and Economic History: AEA Papers and Proceedings,Vol. 86, No. 2, pp. 202-206.

__________. 1990. "Endogenous Technological Change." Journal of Political Economy, Vol. 98, No. 5, Pt. 2, pp. S71-S99.

Schieber, Sylvester J., and John B. Shoven. 1994. "The Consequences of Population Aging on Private Pension Fund Saving and Asset Markets." NBER Working Paper No. 4665. Cambridge, MA: National Bureau of Economic Research, Inc.

Solow, Robert. 1962. "Technical Progress, Capital Formation, and Economic Growth." The American Economic Review (May), Vol. 52, No. 2, pp. 76-86.

__________. 1956. "A Contribution to the Theory of Economic Growth," Quarterly Journal of Economics (February), Vol. 70, pp. 65-94.

Steuerle, C. Eugene, and Jon M. Bakija. 1994. Retooling Social Security for the 21st Century: Right and Wrong Approaches to Reform. Washington, DC: The Urban Institute Press.

Stock, James H., and David A. Wise. 1990. "Pension, the Option Value of Work, and Retirement." Econometrica (September), Vol. 58, No. 5, pp. 1151-1180.

Tanner, Michael. 1996. "It's Time to Privatize Social Security." Challenge (November-December), pp. 19-20.

Thaler, Richard H. 1992. "How to Get Real People to Save." In Personal Saving, Consumption, and Tax Policy, edited by Marvin H. Kosters. Washington, DC: The AEI Press, pp. 143-150.

U.S. Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds. 1998. Annual Report. Washington, DC: U.S. Government Printing Office.

__________. 1997. Annual Report. Washington, DC: U.S. Government Printing Office.


View the original article here