The End of Retirement

Teresa Ghilarducci teaches economics at Notre Dame University and is a visiting professor at the New School for Social Research this spring. She was a member of the Advisory Board of the Pension Benefit Guarantee Corporation.

An esteemed colleague read three paragraphs of news clip on employer pensions before he realized it was from the satirical newspaper The Onion. The tip off was the interview with an eighty-seven-year-old machine shop worker struggling with widowhood, high stress, and early stage Alzheimer’s at General Electric. Early stage Alzheimer’s was the first clue, not the eighty-seven-years of age. Satire writers must have a holy grail of seconds before the earnest reader starts chuckling; my colleague’s delay might be a record. It takes three seconds to know “Cindy Sheehan loses second son in Katrina” is a lampoon. The reason it took so long to laugh at a news story that GE was adopting a new policy of “lifetime” jobs and a new forty-five-year vesting period for their pensions is that it is credible; the signs of the end of retirement are all around.

Scarcely a day passes without a new pension nightmare: Social Security privatization, pension terminations by corporations that are household names (such as Delphi and United Airlines), household saving reaching another new low, canceled retiree health benefits, and 401(k) accounts becoming “201(k)s” while replacing traditional pensions. Airlines were able to use bankruptcy court to abandon their defined benefit obligations. Most of the pensions were protected by the Pension Benefit Guaranty Corporation (PBGC), but future accrual was scrapped. The PBGC will not be able to handle many more obligations without more defined benefit plans coming in. That is why efforts to enhance the defined benefit system are so crucial—they grow or die (see sidebar).

These nightmares reflect an unpleasant reality. What we need to know is what has caused reality to change. The changes in the retirement future of Americans stem from the decline of union contracts, a dramatic shift in presidential and congressional attitudes about government responsibility for social insurance, and the substitution of defined contribution or 401(k)-type accounts for traditional defined benefit pensions.

The current federal policy of promoting an “ownership society” means shifting risks that were once spread out among many workers and employers to individuals, who are now expected to manage individual accounts: their 401(k) plans, their health savings accounts, and, if powerful forces get their way, Social Security individual accounts. None of this bodes well for working people. Twenty years of experience with 401(k) plans reveal that workers will never be able to accumulate enough assets in individual accounts and choose payout options that will provide a steady stream of income for life after retirement. This means that Americans will turn to the option that American adults have always relied on—contingent, low-paying jobs—and will lose one of the few remaining accomplishments of the American working class, retirement.

Whither Retirement?

The concept of retirement changed in the post-Second World War period, when U.S. workers, through their collective power, won retirement time. And retirement time increased dramatically as a result of two factors. First, there were large increases in longevity, primarily in the mid-1970s, because of three contemporaneous trends: reduced smoking among men, increased Social Security benefits, and increased access to health care through Medicare, each of which had its salubrious effects on health. At the same time, the same programs encouraged men to retire at younger ages. This meant that workers’ households spent more time in a period of life called retirement, which opened up to the working class a way of life heretofore reserved for the wealthy.

The expectation of time off at the end of one’s working life is a concept that has evolved, just as have other entitlements to time off, such as “the weekend” and the eight-hour workday. The entitlement developed through compromises among workers, organized labor, business firms, and the government. Throughout the post-Second World War period, workers, mainly those in unions, negotiated for holidays, vacations, and leaves, trading these for lower pay or increased productivity. That unionized workers wanted free time is evidenced by the fact that some of the most famous strikes in the 1950s were over pension benefits. Just last year, transit workers in New York City stranded millions of commuters in chilly December over a proposed cutback in benefits. Workers and their organizations came to regard pensions as one way to achieve middle-class status.

Despite corporate and political arguments to the contrary, pensions are as affordable as they were thirty years ago. The nation can still afford retirement, or more precisely, the ability of older people to choose not to work. Everything really depends on the distribution of power in the labor market. The loss of pensions and the ability to choose to work on one’s own terms coincided with the loss of union bargaining power, which began in the 1980s. We must be clear that workers losing the bargaining power to secure pensions is not the same thing as the economy losing the ability to pay for them. Those who argue that the elderly should work more imply that we cannot afford the same levels of retirement time because a growing number of retirees must be supported by fewer workers. But demography is not destiny: economic bargaining power is.

The loss of pensions has not generated a national crisis or a political backlash because of an ambiguity about whether this is such a bad thing. Cultural and economic leaders keep telling us that the concept of retirement has changed. They say that older people want to work longer and do not want pensions as much as they did before. This thinking seems to fit in nicely with the United States’ distinctive consumer and complementary work culture. Americans work more hours per year and spend relatively fewer years in school, and American mothers and the elderly work more too. This assumed pro-work culture helps make it socially acceptable to declare that the elderly who can work, should. Thus, it should not be surprising—though it is ironic—that the capacity to work is an “asset” in an American’s “retirement” portfolio, and income from paid work is becoming an important income source for retired Americans. Since the 1978 passage of the Age Discrimination in Employment Act, the United States became, and still is, the only OECD nation that bars age discrimination in employment. Employers may not demote employees, change work assignments, alter pay structures, layoff or fire anyone because of their age. Mandatory retirement is illegal in the United States (except for commercial pilots, which is one reason they bargained for high pensions).1

Using history as our guide, the best guess about whether workers will have pensions depends on leverage, negotiation, and workers’ sense of their entitlement to retirement. Clearly, no matter what happens, some workers will want to work longer, at least part time. But it is not clear that employers will provide the jobs older workers want. U.S. Labor Department economist Richard Johnson found that older workers report that less physical strength is required on their jobs, but their jobs are requiring more intense concentration and keen eyesight and involve more stress. So, reversing this deterioration of retirement income will require policies that encourage more widespread coverage of pension plans.

The Erosion of the U.S. Retirement Security System

For the first time in U.S. history, every source of retirement income is under siege: Social Security, personal savings, and occupational pensions.2

Social Security: The cornerstone of national retirement security is Social Security which currently pays benefits to more than 47 million people, including retired workers, disabled workers, the spouses and children of retired and disabled workers, and the survivors of deceased workers. It is the largest single source of income to retirees. Although Social Security helps keep most retirees out of poverty, it is not sufficient to provide a middle-class worker with their accustomed standard of living. The system replaces about 41 percent of income for workers who earn average wages throughout their careers; financial planners recommend replacing 70 percent of pre-retirement income. Alicia Munnell at Boston College found that the already low 41 percent replacement rate will fall in two decades to 30.5 percent as normal retirement age creeps up to age 67, which reduces income for those collecting at earlier ages, as do income taxes, and Medicare premium increases.

Personal Saving: For the first time since the Great Depression, at the end of 2005, Americans spent more than they received: American workers are now, on average, consuming all of their income and more. Just twenty years ago workers (aged 25–65) were saving 10 percent of their income. Brookings Institute economists Barry Boswell and Lisa Bell found that the United States ranks last among any industrialized country in personal saving rates, with other developed nations having rates as high as 12 percent. They argue that the fall in the personal saving rate is actually a great mystery to economists. We know that middle-aged workers save more and there are more middle-aged workers today than they were twenty years ago; educated people save more and workers are more educated than they have ever been; and most personal saving comes from households at the top one-third of the income distribution and this group has had the most income gains over the last decade.

Though the reason for the lack of personal saving is an academic mystery, the fact that nearly all saving by households is now in the form of contractual, institutionalized, “pension savings” is well-known. American spendthrift spending habits (household saving started to decline in the 1980s) are legendary, but up until the 1990s, contractual or institutionalized savings, again mainly pensions, helped staunch the red ink. Today a deep decline in contractual savings—especially by employer-financed defined benefit pension plans—explains a large part of the savings rate decline, because employers had to allocate funds in advance to fulfill the promises made for workers’ pensions, and if the firm provided a defined benefit pension, almost everyone at the firm was included.

The decline in defined benefit plans has corresponded with a rise of defined contribution plans. However, twenty years of experience with such plans show clearly that workers have been either unwilling or unable to fill the savings gap. An extensive survey by Vanguard found that only one-third of workers were on track to achieve a 70 percent retirement replacement rate. Although 90 percent of workers say they have thought about their retirement needs, only 41 percent of workers had specific plans about how to obtain the assets they need. Further, 40 percent of households at risk of not having enough income in retirement would need to double their savings rates or defer retirement until age 70 to maintain living standards. There appears to be a large disconnect between thinking and doing. Education is of no help. Only 40 percent of those who receive counseling or attend a financial education class change their saving behavior.

There is little evidence that average- and lower-income workers have the capability to anticipate, manage, and save enough money to retire. Generally, workers who report having reasonable expectations and a retirement plan to meet those expectations have higher incomes in the first place. This relationship between income and retirement preparation is partially rooted in public policy. A major impetus for retirement planning stems from the tax code’s exemption of income that is diverted into retirement accounts such as 401(k) plans, individual retirement accounts (IRAs), and similar savings vehicles. Since they have more income and because income taxes are progressive, it is the higher-income quintiles that have the greatest financial incentive to take advantage of professional counseling to maximize their tax-preferred retirement savings.

The bottom line is that automatic institutionalized pension saving, particularly through defined benefit plans, are the critical ways workers save for retirement.

Employer Pensions: After Social Security, employer-based pensions are the second largest source of retirement income, providing almost 20 percent of the income received by elderly households. These contractual forms of savings are a key way middle-income workers retain “middle-class status” in retirement. The amount of taxes not collected because pensions get a tax break equaled a full fourth of total annual Social Security contributions—$114 billion in 2004. Yet, despite taxpayers forgoing a huge amount of revenue in the form of federal tax breaks to pensions, coverage has stagnated for the last decade, with participation of only about 60 percent of the workforce.

Chief among the causes for pension stagnation is the decline in union density and the corresponding reduction in collective bargaining agreements. Compared to the compensation of nonunion workers, union compensation consists of less cash and more insurance and employee benefits. Unions inform and educate members about the importance of insurance and pensions, and more senior workers tend to have more say in setting union priorities than the marginal, younger worker just hired, who likely prefer cash relative to insurance. Unions form communities, and communities often lengthen time horizons and expand the concerns of their members beyond individual consumption and competition. Instead of striving for positional goods—those that set a person apart from another—individuals in communities tend to put more value on public goods and longer-term security. Whatever the reasons, union workers have almost twice the coverage rates for the lower-income workers and over 10 percent more for higher-paid workers.

Along with the overall stagnation of pension coverage is the development of the increasing dominance of defined contribution plans over defined benefit plans. The number of single-employer defined benefit plans has declined significantly, from about 95,000 in 1980 to fewer than 35,000 in 2002, while the number of active defined benefit plan participants—employed workers covered by a defined benefit plan—has declined as a percent of all national private wage and salary workers from 27.3 in 1980 to about 15 percent in 2002.

Explaining the Shift from Defined Benefit to Defined Contribution Plans

Explanations for the trend toward defined contribution plans and away from defined benefit plans range from those that emphasize the shift as an inevitable evolutionary development in a global economy and those that view the result as a consequence of policies that created a hostile environment for defined benefit plans to grow. The first view implies that defined benefit plans are dinosaurs that did not adapt to the environment. The other view is that defined benefit plans are pandas, a worthwhile species endangered by short-sighted policies and decisions. The dinosaur interpretation argues that employees have come to prefer the 401(k)-type, defined contribution plans. Yet, despite much popular wisdom, there is little evidence that defined benefit pensions are stagnating because workers do not want them. Rather, defined contribution plans are, to many workers, a second-best option given the uncertainty and the lack of commitment to defined benefit plans displayed by employers.

The panda interpretation is that defined benefit coverage rates are stagnating because many companies are adopting 401(k)-style plans to take advantage of temporary changes in accounting standards and the economy in order to reduce pension costs, not to respond to supposed worker preferences for defined contribution plans, or to changes in labor processes and technology. A sponsor’s pension costs may also be reduced by 401(k) plans because a substantial number of workers choose not to participate in a defined contribution plan, even when it means foregoing the sponsor’s contribution match. Extrapolating from a detailed study of large companies, economist Bridgett Madrian and her colleagues calculate that between 2002 and 2004, if all eligible workers participated in their employers’ 401(k) plans, employers would have had to contribute 26 percent more—for an annual total of $3.18 billion.

The growth in defined contribution plans has, so far, not raised overall pension coverage. In 2005, employers reported that 43 percent of workers participate in defined contribution plans and 21 percent participation in defined benefit plans. If these shares are added, 64 percent of the workforce would be in retirement plans, but, in fact, only 50 percent are. If defined contribution coverage and participation growth expanded access to pension plans rather than replaced or supplemented an already existing defined benefit plan, then growing defined contribution rates should have pulled up total pension coverage and participation rates. This has clearly not happened so far.

Participation rates for all workers in all pension plans rose only 4.2 percent and defined contribution coverage went up 19.4 percent in the six years up to 2005. Union workers experienced one of the largest increases in pension coverage rates—from 79 percent in 1999 to 85 percent in 2005—and this group had already started from a high base. Remarkably, this group also had one of the largest increases in defined benefit coverage, from 70 percent to 72 percent, and a large increase in defined contribution coverage, from 39 to 43 percent. Because unions almost always bargain defined contribution plans to complement defined benefit plans, we can be fairly certain that in this case, the increase in the defined benefit rate likely boosted pension access.

There is a strong correlation between general pension and defined benefit participation rates. The correlation between the overall expansion of pension participation and defined benefit participation rates is a strongly positive 79 percent. But the correlation between all pension and defined contribution growth is negative 10 percent, suggesting that 401(k) plans do not boost pension coverage. The recently announced terminations of traditional defined benefit plans and their replacement with 401(k) defined contribution plans by IBM, Hewlett Packard, Motorola, and other major corporations support the finding.

Although the number of defined benefit plans, and total participation in them, continues to decline, they remain an important component of pension income and of total retirement income. There is also some evidence that some new, smaller professional firms are forming defined benefit plans, although this at best may only slow the decline. Further, despite their declining number and the growing number of plan freezes among the remainder, a majority of Fortune 500 companies—especially in the pharmaceuticals and manufacturing sectors—continue to maintain defined benefit plans, and defined benefit plans continue to control over 1 trillion dollars in assets. In the public sector, employers at all levels of government, the armed forces, public schools and universities, and hospitals continue to offer traditional defined benefit plans with over $2 trillion in assets. At least in the short run, many employers, public and private, will continue to sponsor defined benefit plans.

The Implications for Workers

Most workers who have 401(k) plans have not saved nearly enough for retirement, nor could they ever do so, even if a 401(k) was their only plan. The median annual salary deferral into a 401(k)–type plan was just $1,896 in 2004—an adequate retirement savings rate would be twice that. (Financial planners recommend saving 12 percent of salary in a retirement account; $5,400 is required per year for workers earning the median salary of $45,000.) Further, the share of workers in 401(k) plans who say they cannot afford to save has risen from 15 percent in 1998 to 19 percent in 2003.

The several decades of experience with 401(k) type plans—since 1978—do not instill hope that savings behavior will improve. The average size of defined contribution account balances remains quite low. As of 2004, the mean value for 401(k) plans was $53,600 for near retirees (those aged 55–64) with a far lower median value of $23,000. These accounts may not seem small, but the annuities they would generate are. A $50,000 lump sum would buy a 65-year-old person less than $50 per month if indexed for inflation. Low incomes, low savings, the potential to borrow from the accounts, and the ability to consume the account balance upon switching employers, despite the tax penalties, are all likely suspects in the current weak balance accumulation performance of defined contribution plans.

Why are people not accumulating sufficient balances? Besides low incomes, at least some individuals are hampered by events and human traits that interfere with increased retirement asset accumulation. Unexpected emergencies, the desire for increased consumption, or simple shortsightedness can lead to decisions that favor placing more immediate priorities and desires ahead of retirement saving. For example, even if people save a lot in their 401(k) plan, they may tend to spend down or borrow against their 401(k)s to pay for costs associated with job changes—over half of people who change jobs spend their 401(k) plan balances—health emergencies, home purchases, and financing a child’s education. Seventy-two percent of workers have 401(k) plans that allow 401(k) loans (which reduce accumulations), and 10 percent of participants borrowed from their 401(k)s. The median outstanding balance of a loan from a 401(k) plan is $2,000. Such temptation is not only common, it is popular among plan participants. The Government Accountability Office found that the ability to borrow from one’s 401(k) plan was a key incentive in workers choosing to participate in the plan.

That people can spend their 401(k)s when they leave employment makes 401(k)s, ironically, a problem in a mobile society. The seemingly attractive portability feature of 401(k) plans can become counterproductive for retirement saving because half of workers participating in 401(k) plans cash them out when they change jobs rather than roll them over for retirement. In such instances, 401(k) plans may actually serve as severance plans that help alleviate the costs associated with being out of work and changing jobs. There are other leakages from the 401(k) retirement system that also diminish needed accumulations. Many workers use 401(k) plans to pay for their children’s education, household expenses, and housing needs. Although such leakage is often more a consequence of broader weaknesses in the social safety net—low unemployment insurance benefits, the lack of health insurance, etc.—it nevertheless erodes the basic goal of pension plans, which is to help provide an adequate retirement income.

This temptation risk does not just apply to workers’ decisions to spend their 401(k) plan assets before retirement. Policymakers also yield to temptation to let people spend down their 401(k) accumulations. For example, Congress waived tax penalties for those victims of Hurricane Katrina who withdrew assets from their 401(k) accounts. No one can deny these victims’ financial desperation, but the temptation to let a devastated forty-five-year-worker have easier access to his or her 401(k) plan in 2005 means lower and possibly much lower income for that worker in 2025.

Similarly situated victims who had defined benefit accumulations will have more at retirement than 401(k) holders. 401(k) participants pay higher retail money management fees, while defined benefit plan sponsors pay wholesale fees. The level of fees crucially affects overall financial returns, a 1 percentage point fee can lower returns by 20 percent. Typical fees charged to defined contribution accounts can reduce account values by 21–30 percent depending on the size of the account.

Defined benefit plans also typically pay annuities, and workers cannot access pension income before retirement. Defined contribution participants usually are paid in a lump sum. But annuities provide more piece of mind to retirees: a Boston College and Rand study found that retirees having both a defined benefit plan and a 401(k)-type plan report an 8 percent boost in satisfaction. However, just having a 401(k)-type defined contribution plan alone does not improve an elderly persons’ self-reported satisfaction with life.

The reasonable interpretation is that the elderly would rather have an equivalent level of income coming from a defined benefit plan or a combination of a defined benefit and a defined contribution plan, rather than just from a defined contribution plan. This is because retirees, like many of us, would trade some income for security.

The employer, rather than the employee, bears the risk of investment loss and employers can bear the risk better than individuals—they typically have a longer time horizon, have more financial resources, can tap expert advice more easily, and can bargain for cheaper investment fees than individual workers can. Defined benefit plans are especially valuable to middle-aged workers, who are poised to experience the largest rates of accrual under traditional defined benefit plans. On the other hand, workers like defined contribution plans because they are easier to understand and are portable. To the extent that workers use their accounts as a form of general savings, it is also nice to have savings to fall back on emergencies. The problem is that many emergencies occur before the golden age of retirement.

Pension Reform

The right kinds of policies can encourage plans with the favorable characteristics of both defined benefit and defined contribution plans. For example, cash balance plans have increased in number over the last twenty years. These plans resemble 401(k) accounts, and their value is expressed similarly, but the employer invests the funds and bears all the performance risk. Under cash balance plans, the sponsor guarantees a return on the employer’s contributions and is obligated to pay the balance when the individual retirees and benefits are insured by the Pension Benefit Guaranty Corporation. Although cash balance plans express their benefit as in 401(k)s and in fact are touted as offering participants a lump-sum option, these plans are also required to offer an annuity option. The notional balances of the cash balance plans cannot be spent or borrowed against by the employee.3

Cash balance plans, properly designed, could provide adequate benefits for all workers, especially if they ban lump sum payments. Some unions, like the Communication Workers of America, have negotiated cash balance plans for some of their members.

The other major type of hybrid is over seventy years old—the defined benefit multi-employer pension plan. Multi-employer plans cover approximately 20 percent of defined benefit participants and exist in industries where workers are often skilled and mobile, for instance in the mining, needle trades, trucking, and construction sectors. TIAA-CREF, the largest pension plan in the nation, which covers research, university, and college professionals, is a type of multi-employer plan incorporating many defined benefit and defined contribution characteristics.

Through collective bargaining, the union representing mechanics and other workers at US Airways agreed to a new defined benefit plan after the old one was terminated in bankruptcy in the summer of 2005. When that single-employer defined benefit plan failed, the International Association of Machinists’ multi-employer plans stepped in as a replacement, offering better benefits than the airline’s proposed defined contribution plans for the same contribution. The plan is comparatively more stable for the workers, because it is a defined benefit plan that includes the other airlines. These workers could lose their jobs at US Airways or any other airline, but as long as they obtained employment at another carrier participating in the plan, they could maintain active membership in their pension plan.

In the late 1990s, nurses in a New Jersey hospital finally achieved their longstanding demand to join the multi-employer pension plan of the hospital’s operating engineers. Why a multi-employer plan? Why the operating engineers? The hospital had changed ownership so many times that each single employer plan ended when another firm bought the hospital. The employees didn’t move—it was the employers who were mobile. Joining the multi-employer plan—the only remaining plan at the site—allowed the nurses build up credits in one defined benefit plan and provided for a more secure retirement.

Jobs and the Older American

Under what conditions can society provide jobs that older people want, not jobs that older people have to take? Currently it seems employers are offering jobs to older people that employers have previously reserved for other marginal workers. Instead of sixty being the new thirty, it would be the new seventeen as older people fill the area with the predicted largest growth in new jobs—retail clerks. This seems to be the direction in which we are going.

Since 1949, men and women over age sixty-five have said “ciao” to the labor market in recessions, and men withdrew from the labor force at the average yearly rate of 2.4 percent and women by 1.5 percent. Yet, in the most recent recession, men over age sixty-five still said “ciao” by 3.9 percent; but women said “hello” by increasing work effort by 5.3 percent. The labor force participation rate for slightly younger men and women, aged fifty-five to sixty-four, was higher over the most recent business cycle. However, if we remember our economics lessons, labor force participation and working are not the same thing. AARP analyst Sara Rix characterized 2002, the year the upturn began, as a mixed year for older workers. Despite the rapid increase in labor force participation, many of the elderly found unfavorable working conditions. If the elderly were laid off, they had half as much chance of being reemployed as younger people. The average duration of unemployment is higher for older workers and rose in 2002; the average search for job seekers over age fifty-five was sixteen weeks, up from 12.7 in 2001. Significantly, older men’s job security has gotten much worse. Their median years of tenure—the number of years a person has been employed by their current employer—has fallen dramatically, by almost 50 percent from 15.3 years for men aged 55–64 to just 10.2 years. (The decline is much smaller for women, from 9.8 years in 1983 to 9.6 years in 2002.)

Just because some of us will expect to live longer than our parents does not mean that people should work longer. The logic is based falsely on the assumptions that workers do not value free time, that older people can do the new jobs, and that improved longevity is not related to retiring sooner. The first is obviously false; as the nation grew richer work hours fell and vacation time soared. Second, there is little evidence that the ability of older people to work longer has improved. Since 1981, the share of older workers reporting limitations in the ability to work stayed steady at about 15–18 percent. Jobs demanding heavy lifting, stooping, and kneeling, and overall physical effort are declining, especially for men. But, older workers report an over 17 percent increase in their jobs involving a lot of stress and intense concentration. Older women report an over 17 percent increase in their jobs requiring good eyesight. Only magical thinking can conclude that the computer has made jobs easier for older workers.

That older people found jobs harder to get, more difficult to perform, and unemployment duration much longer suggests that older people are forced onto the labor market because of eroding pensions. The unemployment rate for older women went up faster than any other’s in the recession, and older men’s and women’s unemployment rate recovered less than younger peoples in the expansion.

Since older people are working, we need policies that speak to the special needs of the elderly workers. The elderly will likely be attracted to the fastest-growing jobs, which, ironically, serve the elderly and are dominated by women. The second two largest areas of job growth are registered nurses and home health care and personal health care aids. These jobs are better for both the client and worker when collective bargaining determines wages, hours, and working conditions, no matter how old the workers are. But older people in these jobs need the protections that the Americans with Disabilities Act provides—the right for reasonable accommodations in scheduling and equipment. In the end, the only way to ensure that older workers have jobs on their terms depends on them having secure retirement income.

The U.S. system of retirement security is in transition. Despite their limitations in providing retirement without significant reform, defined contribution plans are here to stay. However, there is barely a regulatory framework in place to protect the public interest. Sensible protections that already govern the investment behavior of defined benefit plan fiduciaries could be a good place to start. The percent of a portfolio that is allowed to be in the employee’s company’s stock could be limited; lump sum payments could be banned; and worker representation on pension boards could be mandated.

Ultimately, it matters less what vehicles we use to get to a secure national retirement system than that we ensure a secure retirement for all American workers and their families. Unions, progressive activists, advocates for the elderly (principally the AARP), and the Democratic Party were impressively united against Social Security privatization. The puzzle is why the same solid coalition is not being built around diminishing pension expectations and the chipping away of good pensions. The answer may be that many workers are settling for individual accounts, and most workers do not have access to either 401(k) or defined benefit plans. The only way most workers get pensions is to unionize and organize. In order to have a balanced work life—with adequate pay, time off, security, and safety—it seems to always come down to that same ineluctable reality.

Notes

1. ↩ Americans view these restrictions as protections—an acceptable part of what constitutes labor rights. Interestingly, though, at the time of its passage, the labor movement was ambivalent about the legislation.One of the sources of the ambiguity was that being legally protected to be able to work makes retirement “voluntary ” and weakens a claim to pensions. Therefore, in the United States, unlike any other developed nation, retirement income is based on four pillars: the state pension, the ccupational pension, personal assets, and the capacity and performance of paid labor.
2. ↩ The declining availability of retiree health insurance and the rise of housing costs are critical issues in American retirement security but are outside the scope of this essay.
3. ↩ See Cooper v. IBM Pers. Pension Plan, F. Supp 2d.1010 (S.D. Ill. 2003).Cash balance plans are controversial.Over the last ten years, many employers converted DB plans to cash balance plans to save money, and many older workers lost expected benefits.One recent district court case has ruled that such plans are age discriminatory and the Internal Revenue Service has placed an effective moratorium on the applications for approval of cash balance plan designs.

A Note on Sources

Savings rates: Alicia H. Munnell, Francesca Golub-Sass, & Andrew Varani, “How Much Are Workers Saving?” Center for Retirement Research; Barry Bosworth & Lisa Bell, “The Decline in Saving: What Can We Learn from Survey Data?” Center for Retirement Research. Articles published by the center can be found online at http://www.bc.edu/centers/crr.

Annuities and elderly satisfaction: Keith A. Bender & Natalia A. Jivan, “What Makes Retirees Happy?” Center for Retirement Research; Constantijn W. A. Panis, “Annuities and Retiree Satisfaction,” Pension Research Council Working Paper (2003-19),
http://rider.wharton.upenn.edu/~prc/PRC/WP/wp2003-19b.pdf.

Retirement wealth: Elizabeth Bell, Adam Carasso, & C. Eugene Steurele, “Strengthening Private Sources of Retirement Savings for Low-Income Families,” Urban Institute, http://www.urban.org; Alicia H. Munnell, “The Declining Role of Social Security,” Center for Retirement Research; “The Enron Collapse,” Congressional Research Service, http://fpc.state.gov/documents/organization/8038.pdf; Social Security Administration, “Income of the Population 55 and Older, 2002,” (March 2005), http://www.ssa.gov/policy/docs/statcomps/income_pop55/2002; Vanguard Group, “Expectations for Retirement: A Survey of Retirement Investors,” 2004, https://institutional4.vanguard.com/; Vanguard Group, “Lifetime Income Program,” (October 15, 2005), http://flagship.vanguard.com/VGApp/hnw/content/AccountServ/Retirement/
ATSAnnuitiesOVContent.jspvanguard.com; Christian Weller & Edward N. Wolff, Retirement Income (Washington, D.C.: Economic Policy Institute, 2005).

Individual retirement accounts, 401(k)s: Alicia H. Munnell & Annika Sundén, Coming Up Short (Washington, D.C.: Brookings Institution Press, 2004); Bridgitte Madrian, James Choi, & David Laibson, “$100 Bills on the Sidewalk,” Center for Retirement Research.

Jobs and older workers: Richard Johnson, “Job Demand Among Older Workers.” Monthly Labor Review (July 2004): 48–56; The Onion, “More Companies Phasing Out Retirement Option,” January 25, 2006, http://www.theonion.com/content/node/44679

Pension reform: General Accountability Office, Private Pensions: Recent Experiences of Large Defined Benefit Plans Illustrate Weaknesses in Funding Rules, GAO-05-294 (Washington, D.C.: GAO, 2005); General Accountability Office, Private Pensions: Information on Cash Balance Pension Plans, GAO-06-42 (Washington, D.C.: GAO, 2005); Teresa Ghilarducci, “Delinking Employee Benefits from a Single Employer: Alternative Multiemployer Models,” in Benefits for the Workplace of the Future, ed. David S. Blitzstein et al., (Philadelphia: University of Pennsylvania Press, 2003); Richard W Johnson & Cori E. Uccello, “Cash Balance Plans: What Do They Mean for Retirement Security?” National Tax Journal 57, no. 2 (June 2004): 315–28.