The industrial revolution is more than two centuries old, but the first rich nations to provide for the retirement of their elderly did not do so until roughly a century ago. Until the late 1800s, when agriculture still dominated the newly industrializing economies, old people, especially in America, usually owned their farms or passed them on to their children, who looked after them. Those who worked the farms were also often taken in. Suffering among the elderly existed, of course, but life spans were relatively short.

As workers were forced off the farms and into factories and mines in the second half of the nineteenth century, however, more and more of them had no assets to fall back on, and the poverty of growing numbers of the elderly turned into a crisis that could no longer be ignored. Many people had to go on working until they were physically unable, and for most the possibility of a comfortable retirement did not exist.

In the late 1800s and early 1900s, as industrialization more fully transformed the old farm economies, continental Europe and Scandinavia began to establish government-run pension systems. But the US and Great Britain were slower to react. The US did not establish Social Security until 1935, fifteen to twenty-five years after similar programs were begun in many well-to-do European nations, and fifty years after Germany established its pioneering state system. Even Britain established a public pension program a generation before the US did.1

It was not until the post–World War II period, however, that many elderly Americans were able to maintain a standard of living in retirement that eventually came fairly close to the one they had as workers. At first, America’s Social Security benefits were small, but they were raised substantially over the next four decades, and financed by repeated increases in payroll taxes. At the same time, rapid US economic growth enabled corporations to expand their own pension programs aggressively. Some American companies, such as the Pennsylvania Railroad and American Express, had pioneered private pensions as early as the turn of the century, at first to encourage workers to retire. After World War II such company-sponsored pensions were designed to attract and keep workers in tight labor markets. The contributions to these plans were exempt from taxes. By the early 1960s the proportion of American workers with pension benefits had risen from 20 percent to more than 40 percent.

The corporate pension plans were far from adequate, however, often failing to invest sufficient funds to guarantee future benefits to all employees, imposing long job tenure requirements on workers before they qualified for their pensions, and using overly risky investment strategies. In 1974, the federal government passed new regulations requiring companies to correct many of these practices. The law also established the Pension Guaranty Benefit Corporation, a government insurance agency that would pay workers’ pensions if their companies failed.

By the 1970s, the combination of public and private pensions thus created an extraordinary new situation for many American workers: the prospect of a fairly comfortable old age. The average age at which men retired fell from nearly seventy-five in 1910 to less than sixty-five today. Still, benefits were linked with how long one worked and how much one earned. The tax exemption for pension contributions withheld from workers’ salaries cost the federal government tens of billions of dollars each year and contributions mostly benefited better-off workers who paid the highest tax rates. But the combined public and private system was, nevertheless, one of the remarkable achievements of twentieth-century American society. In the 1950s, one in three of the elderly lived in poverty; today, one in ten or so do.

Now, according to the British historian and political scientist Robin Blackburn, these gains are just as assuredly being reversed in the United States as well as in other rich nations, and lower-income workers hurt more than others. Blackburn believes that many elderly will fall back into outright poverty, be forced to take jobs in their late sixties and early seventies that are too demanding, if they can find jobs at all, and will be unable to afford needed health care as life spans increase and costs rise. Yet, as Blackburn writes in his impressive book Age Shock, there is little political interest in the problem: climate change, epidemics, nuclear proliferation—and in the US the rising costs of health care and the stagnation of wages—all seem more urgent and dramatic than the problem of poverty in old age.

To Blackburn, how rich nations treat their elderly is a measure of their decency and, perhaps more important, their ability to devise new ways of living. Modern societies now have the capacity to make old age a productive “third age,” he argues. But, as restraints on government spending become national priorities, and an ideology of privatization and free-market capitalism spreads, the hard-won state systems and the private pensions that supplement them have come under threat.


Blackburn is not talking about the solvency of Social Security. It is, he concedes, a serious concern, but like many analysts he believes that the expected shortfall between tax revenues and Social Security payments can be closed without great difficulty. The Social Security Administration, based on a seventy-five-year forecast, figures that as the proportion of elderly in the population increases, payroll taxes will not be adequate to meet the annual payouts by about 2041. It is true that all the rich Western nations, as well as Japan, are aging rapidly. In the US, the process is occurring less rapidly than in other countries, but as large numbers of people reach retirement age, the proportion of workers to retired people is shrinking. In the US, there are now somewhat more than three workers for every retired person, but by 2030, there will be only two.

If current estimates are right, and they may well be pessimistic, to make the system solvent will require raising social security taxes two percentage points, or reducing benefits by an equivalent amount. The current payroll tax for Social Security is 12.4 percent of wages, and another 3 percent is required for Medicare, shared equally by the employer and employee. Raising taxes will not be easy, but it is certainly manageable. Even if taxes were not raised, Social Security will still be able to pay 75 percent of its benefits after 2041. That’s hardly the end of the system.

Less well known, however, is that even if Social Security is made solvent, its benefits as a proportion of a worker’s average income over a lifetime will recede significantly in the next two decades. First, because of changes made in the 1980s, the retirement age is being raised step by step to sixty-seven, reducing the payout over the years. Second, Medicare payments for hospitalization are taken out of Social Security checks and these will continue to rise rapidly. Third, more of the income from Social Security will be subject to income tax because companies will not be required to adjust existing exemptions upward to reflect inflation. As a result, Alicia Munnell and Steven Sass, of the Center for Retirement Research at Boston College, predict in their forthcoming book, Working Longer: The Solution to the Retirement Income Challenge, that Social Security benefits will, on average, replace only 30 percent of pre-retirement income for a retiree in the middle of the income distribution by 2030, compared to 40 percent today.2 This “replacement state” may in fact be less than 30 percent because the reductions do not include the premiums retirees will have to pay for the new Social Security drug prescription plan.

At the same time, private retirement plans offered by employers are likely to replace a smaller share of income as well—for some a far smaller share. One problem is that many large companies are increasingly defaulting on their payments, including Bethlehem Steel, Kemper Insurance, US Airways, and Polaroid. The Pension Benefit Guaranty Corporation compensates for only part of these unpaid liabilities, and even so its obligations are soaring.

In his recent book, The Great Risk Shift, Yale political scientist Jacob Hacker describes what happened to the family of Victor Saracini, the pilot of the United Airlines jet plane that was the second to crash into the World Trade Center on September 11, 2001. UAL declared bankruptcy the following year and turned over its pension liabilities to the Pension Benefit Guaranty Corporation. But the government agency imposes a maximum on the amount it will guarantee for any single worker, leaving Vincent’s widow, Ellen Saracini, with only half the retirement income the company promised.

The second and more widespread concern is that the traditional pensions, known as defined benefit plans, are being abandoned rapidly in favor of personal savings plans, known as defined contribution plans. These newer employer-sponsored plans—mostly 401(k)s—are not backed by the Pension Benefit Guaranty Corporation and can be far more risky for workers. Today, some three in five workers have such a plan—compared to one in five with traditional pension plans. In the early 1980s, it was the reverse. In the 401(k) plans, workers contribute from every paycheck; the amount they contribute and any income earned on the savings are exempt from taxes until withdrawn on retirement. Employers typically also supplement employee contributions. Among the attractions of the personal savings plans is that they are completely portable; a worker has them for life, regardless of whether the employer changes. A worker who leaves a job with a traditional pension, by contrast, will often lose some or all of the benefits because they are tied to years of work at that company.

But unlike traditional pensions, the personal savings plans rely on workers themselves to decide whether and how much to contribute, after which companies may also also match or otherwise contribute to the worker’s plan. And workers on average are not contributing nearly enough to sustain retirement income that is comparable to that of the old-style pensions. Economists estimate that a worker with a median salary of roughly $45,000 today who is enrolled in a 401(k) for his or her working life and contributes 6 percent of each paycheck to the account would accumulate some $300,000 discounted for inflation by age sixty-five. Yet as of 2004, Munnell and Sass observe, the typical worker aged fifty to sixty-four had actually accumulated only $50,000 or so through a 401(k) or similar kind of plan—enough to replace only 10 or 20 percent of pre-retirement income.


Workers also must manage their own investments in 401(k) plans, usually through arrangements made by the companies with financial institutions that administer the company plan and offer a variety of investment options with lesser or greater degrees of risk. Many workers have little idea how to manage these sophisticated investments and will lose—or have already lost—a large part of their holdings. Nothing is quite so telling regarding contemporary greed as the 2005 documentary The Smartest Guys in the Room, which showed Jeffrey Skilling, the CEO of Enron, urging his employees to invest their 401(k) savings in Enron stock—rather than in a diversified range of investments—while he was selling his own shares.3

The Enron stock fell from roughly $90 a share to a quarter of that price after evidence of fraud and flagrant earnings manipulation was divulged and led to bankruptcy. Hacker cites the case of one worker who lost almost all of his retirement savings of $300,000, none of which was backed by federal insurance. Thousands of Enron workers similarly lost most of what they had saved by placing their savings in Enron shares. The Pension Protection Act of 2006 has since made investing in company shares more difficult, but critics worry that the safeguards can still be circumvented.

Still more troubling, millions of workers have no private retirement plans available to them at all. Even during the golden age of retirement gains in the 1960s and 1970s, more than half of US workers—particularly those employed by small businesses or engaged in contract or part-time employment—were not covered by an employer-sponsored retirement plan. The proportion has not improved since the 1970s, and today, businesses are reducing coverage. Thus, seventy-five million people, many of them lower-income workers, depend now, or will depend, mostly on Social Security for their retirement income—and Social Security replacement rates are falling. Even if they are home owners, many such low-income workers have often been constrained in recent years to borrow heavily against the value of their homes and have relatively low equity.

Jacob Hacker attributes the revolutionary replacement of traditional pensions with 401(k)s to the success over the past two decades of what he calls the “Personal Responsibility Crusade,” a broad movement to shift economic risk from government to workers themselves. Advocates of this new approach encourage government to support social benefits through tax deductions and credits, not outright subsidies. The advantage, they argue, is that workers have freedom of choice, and the economy operates more efficiently.

The crusaders for personal responsibility described by Hacker emphasize the potential for high returns from personal investment in 401(k)s, not the risks. As one conservative policy expert and early advocate of privatized retirement savings confided to Hacker, we try to “wean people gradually off of social-insurance risk management into private risk management without making them fearful about it. You have to do it in steps….” George Bush calls it the “ownership society.” Thus, he has advocated replacing Social Security with personal savings accounts—“privatization.” The creation of personal savings accounts for health care and higher education are also conservative priorities. “It is…a world where,” Blackburn writes, “the different stages of the life course require the purchase of an appropriate financial product.”

For Social Security at least, the American public is not going along. As Greg Anrig recounts in his spirited book, The Conservatives Have No Clothes, Bush, after defeating John Kerry in 2004, proclaimed boldly that he would reform the Social Security system, which he insisted was facing “bankruptcy.” His campaign made little headway, however, and even the Republican-dominated Congress let it die. To Anrig the failure of such efforts to privatize savings, at least for education, health care, and Social Security, is that they are based on ideology rather than on a pragmatic understanding of personal finance and the economics of social programs. Reducing government involvement is the main goal, not increasing the well-being of workers.

Still supporters of privatization have a powerful ally in the financial industry, which has profited enormously from flows of funds into stocks and from the management fees and brokerage commissions collected on retirement accounts since the rise of the 401(k) twenty-five years ago. As Alicia Munnell and others, such as William Wolman and Anne Colamosca, authors of The Great 401(k) Hoax, foresaw years ago, the 401(k) revolution has done far more good to Wall Street than to the financial security of retired workers.

The declining retirement security faced by growing numbers of Americans is being exacerbated by increasing longevity and quickly rising health care costs. Teresa Ghilarducci, an economist at the New School, writes in her passionate forthcoming book, When I’m Sixty-four, that somehow Americans may have to replace 100 percent of middle- and low-wage worker incomes in coming decades to avoid slipping into poverty or going without adequate health care. In 1988, 66 percent of employers offered health-care benefits for retirees, but today only 33 percent do.

Can 401(k)-style retirement plans be redesigned so that they can replace an adequate level of pre-retirement income? Regular employee contributions that are invested tax-free in equities from an early stage in a worker’s career can accumulate to handsome sums. A few years ago, the Employee Benefit Research Institute (EBRI) in Washington, D.C., under the direction of Jack VanDerhei, a Temple University professor, completed a complex analysis of the likely outcomes—simulations based on samples of actual behavior. Assuming average stock market returns between 1926 and 2001, the study found that a middle-income worker who invested in a typical mix of equities and bonds consistently over his or her working life would be able to retire at age sixty-five with an income that amounts to a “replacement rate” of about 60 percent of pre-retirement income, assuming that the accumulated sum were then invested in an annuity. Social Security benefits, at their current level, would replace roughly another 25 percent.4

Even if equities equaled only the worst annual returns experienced over a fifty-year period—between 1929 and 1978—replacement rates from accumulation in the 401(k) would still replace roughly 50 percent of pre-retirement income. Added to Social Security, retirement earnings, according to this model, would amount on average to 70 percent or more of pre-retirement income.

But VanDerhei points out that many workers do not remain in or contribute to 401(k)s their entire working lives. In particular, when people switch jobs or temporarily leave the workforce, they often need the money and do not reinvest the lump sum they take with them, despite being hit with income taxes on their early withdrawal, and an additional penalty of 10 percent if they are younger than fifty-five. Not to mention the high fees they want paid to mutual funds and Wall Street brokers, which can greatly reduce a retirement amount over the years. As Hacker writes,

Workers who are laid off are 47 percent less likely to roll over their pension distributions. Workers who relocate to obtain a new job are 50 percent less likely. And workers who leave work to care for a family member are 77 percent less likely.

Munnell and her coauthors find that, based on 2004 data, 43 percent of retired people fall short of adequate retirement income—estimated to be roughly 70 percent of pre-retirement earnings—by at least ten percentage points. For those born between 1965 and 1975, the number of people with inadequate retirement savings rises to 49 percent.5 So far, Americans have, as noted, accumulated far too little, and Hacker observes that “roughly three-quarters of account holders have less than the widely cited average.”

Such results have been challenged by a group of economists from the University of Wisconsin and the Urban Institute. In a much-publicized 2006 study they reported that fewer than 20 percent of workers were not meeting their optimal retirement goals.6 But VanDerhei and Munnell say the study relied on assumptions that are unrealistic and data that are outdated. Munnell points out that the economists based their findings on a 1992 sample of older workers. By updating and broadening the sample to all workers, she finds a far higher number of workers who are saving too little for retirement. Too much has changed since 1992, including reduced Social Security replacement rates and the shift away from traditional pensions, she says, to make the study’s findings relevant today.

Until Barack Obama criticized Hillary Clinton in October for failing to propose a plan to make Social Security solvent, retirement security was hardly mentioned in the current presidential race. But Obama missed the important point. For the Democrats, raising the specter of a Social Security crisis plays into the hands of those seeking to privatize Social Security. To shore up the system’s finances, Obama was at least willing to face the issue and proposed raising the limit on incomes that are subject to payroll taxes. Right now, payroll taxes are paid only on the first $97,500 of earnings. Clinton would establish a commission to decide how to proceed. Despite the failure of privatization under Bush, and its abandonment by the Republican Congress, John McCain and the other remaining Republican presidential candidates still favor channeling part of Social Security taxes into private accounts, and most have ruled out any payroll tax increases.

Aside from assuring that Social Security benefits continue to be met, there are two basic aims that any US government urgently needs to address. First, as many as possible of the 75 million American workers who do not belong to an employer-sponsored retirement program should be covered under an additional pension plan of some kind. Second, steps should be taken to make sure that those who do have employer-sponsored plans, now mostly 401(k)s or similar programs, are making adequate contributions and managing their funds reasonably—and not withdrawing them prematurely—especially lower-income Americans.

Obama and Clinton have proposed that the government match contributions to 401(k)s or IRAs of at least $500 a year and at most $1,000 a year. Many believe this is a step, if a modest one, in the right direction. Obama would also require all small businesses above a minimum size to put workers automatically in a payroll deduction plan, even when the company does not offer a retirement program of its own, an idea long proposed by economists at the Brookings Institution. The company need not contribute and the money must still be managed by the individual workers, though employers would provide access to investment programs.

Employers have also made constructive reforms, adopting programs that automatically enroll workers in their defined contribution plans. Economists have shown that this is an effective way to promote participation. The new Pension Reform Act of 2006 also enables companies to adopt automatic escalation clauses that require workers to raise the amount of their contributions as their salaries rise.

But many experts believe such steps are inadequate. Among those most critical of the current retirement system in the US, Alicia Munnell and Teresa Ghilarducci offer very different solutions. Munnell believes that most of the Social Security deficit as well as some of the private retirement issues can be resolved if more of the elderly continue to work a few years more, especially those who now take their Social Security benefits early at sixty-two or sixty-three. More years of employment will increase their benefits and their 401(k) savings as well as reduce their number of years in retirement that must be financed. Munnell calculates that working a few more years reduces the savings burden more significantly than most people realize. As for the low-income workers who have been left out of the private system, Munnell believes that the federal government should provide some kind of support for them in the form of direct subsidies or tax credits.

Ghilarducci, in contrast, is strongly opposed to raising the age at which retirees may begin to receive Social Security benefits. It is difficult to find decent jobs at that age, she argues, and working longer is a cause of ill health. To make Social Security whole, she would raise the limit on payroll taxes. Ghilarducci wants to reduce seriously the appeal of 401(k)s by ending the federal tax exemption on new contributions. She would use the resulting tax revenues to provide a $600 tax credit for all workers, thus making it easier even for low-income workers to establish a retirement savings account. For workers to earn this tax credit, she would make more savings mandatory. All workers and their employers would be required to contribute 2.5 percent of wages, or a total of 5 percent, to a personal savings plan, called a Guaranteed Retirement Account. In her plan, the funds would not be managed by Wall Street but sent to Washington, where the federal government would guarantee a minimum return of 3 percent a year. Washington would have the money professionally managed. If there were a shortfall, the federal government would still guarantee the 3 percent annual return.

It is difficult to imagine that Ghilarducci’s plan to end the tax deductibility of new contributions to 401(k)s will find political support anytime soon. Moreover, the mandatory 5 percent she recommends instead is far from trivial. If more mandatory savings are the goal, then why not simply raise payroll taxes substantially—more for the better-off worker than the lower-income worker—and with the higher proceeds, expand benefits for middle- and lower-income workers more than for others?

This could make Social Security benefits significantly more progressive, but it may be a worthwhile philosophical change when we consider the harsh prospects for lower-income workers—and the relatively small part played by Social Security in the retirement savings of affluent workers who have other forms of savings. An advantage of her Guaranteed Retirement Account over Social Security, Ghilarducci argues, is that it will be pre-funded and belong to a person for life, so that no one, Republican or Democrat, can reduce benefits in coming years. Even she subscribes to fears about the future of social programs.

Since health-care costs will also rise significantly in coming years, the conception behind Ghilarducci’s dramatic proposals makes sense. If Americans cannot, without heavy sacrifice, save enough themselves to ensure adequate retirement, perhaps government, backed with subsidies, should, as she suggests, make them save. What has become clear is that an “ownership society,” financed by federal tax giveaways, has been of great benefit to the well-off in higher tax brackets, who also often have access to sophisticated financial advice. But for most Americans, it has greatly increased their exposure to financial risk, while at the same time lowering the level of income and security that they can reasonably expect in retirement.

February 21, 2008

This Issue

March 20, 2008