Testimony | Retirement

The Wobbly Stool: Retirement (In)security in America

EPI Vice President Ross Eisenbrey’s testimony before the U.S. Senate Committee on Health, Education, Labor, and Pensions

Good morning, Chairman Harkin.  I am Ross Eisenbrey, Vice President of the Economic Policy Institute.  EPI is a non-partisan think tank with a long history of analyzing trends in employment, compensation, and income, as well as advocating for policies to ensure shared prosperity.  We are founding members of two important coalitions: Retirement USA—28 organizations advocating for a retirement system that delivers universal, secure, and adequate retirement income—and Strengthen Social Security—a coalition of more than 150 organizations who feel strongly that Social Security benefits should not be cut and the retirement age should not be raised. Today, however, I speak only for myself.
Polls show that Americans are scared about their retirement.  In a recent Gallup poll of people ages 44 to 75, more than 90% said we are facing a retirement crisis, and 61% said they fear depleting their assets more than they fear dying.  Unfortunately, they have good reason to be scared.
According to the Center for Retirement Research, American households ages 32 to 64 currently have a retirement income deficit of $6.6 trillion, a figure that dwarfs the federal deficit and casts a pall over hopes of them retiring in any kind of comfort.  That is how far behind they are in building sufficient pensions and private savings to maintain their standard of living in retirement. This sum comes to $90,000 per household, on average, which means these households have about half of what they need in retirement savings.
I have three main points to make in this testimony today:

1. Congress has made matters worse by focusing retirement policy on high-income households and neglecting low-income workers;
2. Congress and the Obama Administration will make matters even worse if they raise the Social Security retirement age; and
3. There are potential solutions to the retirement crisis, but tweaks and small changes at the margins won’t be enough.

1. Congress has made matters worse by focusing retirement policy on high income households and neglecting low-income workers.

The median household income of seniors in 2008 was less than $30,000, about half that of households under 65.

While nothing to tout, the financial situation of seniors today might be as good as it will ever get for the typical American.  Between declining pension coverage and Social Security cuts, it is possible that the next generation to retire will be the first to be worse off than its predecessor.
The surest vehicle for retirement savings (other than Social Security) has been the traditional defined-benefit pension, which is disappearing.  Almost from the day in 1978 that Congress created an alternative savings vehicle, the 401(k) plan, employers have been shifting employees out of pension plans and into these accounts that put all of the risk and more of the cost onto the backs of individual workers.  Only about one private sector employee in five is still covered by a real pension plan.
Traditional pension plans are pooled investments, managed by professionals, and spread risks over many years (even generations), while 401(k) participants must make their own investment decisions and bear the risk of adverse investment performance. But most 401(k) participants do not have the financial expertise to manage their investments. Many fail to diversify sufficiently and often make poor investment decisions. They tend to have an all-or-nothing approach to risk, and despite the lessons of Enron, many still have funds invested in employer stock.

Luck plays an oversized role in whether retirement savings in personal accounts will be adequate.  Even 401(k) participants who make relatively conservative investment allocation decisions over a long time horizon are subject to unacceptable risks. Gary Burtless of the Brookings Institution has estimated that 401(k) participants who contributed 4% of her wages over 40 years and invested the funds in a portfolio split equally between long-term government bonds and stocks would be able to replace a quarter of their pre-retirement earnings if they retired in 2008. This replacement rate is only half as much as a similar worker who retired in 1999, but much better than a worker who retired in 1974, who would have a dismal replacement rate of only 18%.
Another key risk—one the Gallup survey identified—is longevity.  A real pension guarantees a monthly payment for a lifetime, whereas retirees can and do outlive their 401(k) assets.
And finally, the fees associated with 401(k) plans can decimate long-term returns.  The Center for Retirement Research estimates that net investment returns were a full percentage point higher for defined-benefit pension plans than for 401(k)-type defined-contribution plans between 1988 and 2004, despite a lower concentration of funds invested in equities.  With compounding of the returns on the investment, this small-sounding difference can translate into a 30% larger nest egg at retirement.
The end result of the shift from secure pensions to insecure 401(k)s and Social Security cuts can be seen in the following chart, which presents the likelihood of inadequate retirement income for three successive generations, each with a smaller share of pension coverage than the generation before.

I hope this committee will recognize that the retirement income deficit we are leaving for the Gen Xers is at least as serious as the “burden of debt for our grandchildren” that gets so much attention in the media and in political debate.
How can it be that after 32 years and trillions in tax subsidies, 401(k)s have worsened—rather than improved—retirement security?  First and foremost, the design of the 401(k) ensures that its tax subsidies go disproportionately to high-income earners who least need the government’s help in saving, while providing little or nothing to low-income earners, many of whom struggle to meet their daily expenses, let alone save for a distant retirement.
The Urban-Brookings Tax Policy Center estimates that 80% of the tax subsidies for retirement savings go to the top 20% of earners.  This is government welfare stood on its head.  There is no rationale for providing a larger tax break to a millionaire than to a Wal-Mart cashier for the same dollar contribution to a 401(k) plan (and nothing at all if the cashier owes payroll but not income tax).  Similarly, high earners receive more help from employers, who contribute 5% of earnings, on average, to the retirement accounts of households in the 75th percentile, compared with less than 2% for those at the 25th percentile, according to the Congressional Research Service.
Rather than continue to make this situation worse by increasing the 401(k) contribution limits, which benefits only the highest earners, Congress should re-structure the tax subsidies to ensure that they help everyone save for retirement and provide no greater aid to the upper class than to the working class.  One common sense improvement would be to change the current system of deductions into tax credits and make them refundable.  But bolder steps are called for.

The system of relying on tax subsidies to expand the employer-based retirement system has proven a failure. Only about half of all private sector workers in the Uni
ted States between the ages of 25 and 64 participate in a retirement plan—and participation is much lower for blacks and Hispanics.  Despite rising enroll ment in 401(k)s, this figure has remained essentially unchanged for 30 years because employers have simply replaced traditional pensions with 401(k) plans.

2. Congress and the Obama Administration will make matters worse for most Americans if they raise the Social Security retirement age.
Knowing that retirement insecurity is growing and that the coming generations are even less well prepared than those nearing retirement now, how can Congress consider raising the retirement age, which is exactly the same as a benefit cut?
Social Security is the one part of retirement income working people can count on.  It isn’t adequate—it replaces only 39% of pre-retirement income for the average retiree—but it is universal and secure.
Unfortunately, we are already weakening this foundation of our retirement system, and some are proposing further cuts. Taking into account the increase in the normal retirement age from 65 to 67 as well as Medicare deductions and income taxes paid on benefits, the net replacement rate for the average earner retiring at 65 is already scheduled to drop from 39% to 28% in two decades.

The trust fund has more than $2 trillion and will be able to pay 100% of promised benefits for another 27 years.  Even then, Social Security will not “go broke” but will be able to pay 78% of promised benefits.
So the question isn’t how to “save” the program; it will survive without any change.  The problem is how to get more money into the trust fund so full benefits can be paid in perpetuity.  The goal is, or ought to be, to preserve full benefits and to maximize the retirement income of the tens of millions of households that depend on Social Security.
Yet the Peterson Foundation and a host of other mostly well-off “experts” have managed to convince much of the media and many Washington policy makers that the way to save Social Security benefits is to cut them.  Working people can see through this, however, and every poll shows large majorities that reject cuts in benefits, reject raising the retirement age, and support higher taxes to pay for promised benefits.
Despite what we have heard from your former colleague, Alan Simpson, the average Social Security recipient isn’t living in a gated community.  The average benefit is about $14,000—less than a minimum wage income—and Social Security provides more than half the income for 55% of seniors.  Cutting such modest benefits means reducing the consumption and living standards of tens of millions of households.
The cuts that Simpson, Alice Rivlin, and others call for would come on top of major cuts Congress imposed in 1983, which are still taking effect. I know that you, Mr. Chairman, and Sen. Sanders understand that raising the retirement age is not a fair way to deal with longer life expectancies.  You should both be commended for introducing S. Res. 664, your Sense of the Senate Resolution opposing any benefit cuts.
Over the past quarter century, life expectancy at age 65 has increased by one year for lower-income men, compared to five years for upper-income men. Men in the lower half of the earnings distribution have not even caught up to where upper-income men were in 1982. In the case of women, although life expectancy has grown slowly overall, lower-income women are actually seeing declines and upper-income women are seeing only modest improvements. The general pattern appears to hold with older women as well.
Second, many workers in physically demanding jobs are already unable to work to the full retirement age.  They retire before 66 because a lifetime of working on their feet as cashiers, or doing construction, or lifting patients in a nursing home, have worn them out and left them hurting.  It is easy for a Member of Congress, an economist, or a lawyer to imagine working until 70, but it is much harder to imagine a truck driver or factory worker doing so. Research by Hye Jin Rho of the Center for Economic and Policy Research found that 45% of older workers last year were employed in physically demanding jobs or jobs with difficult working conditions. These are jobs most likely to be held by less educated workers who are more likely to find themselves out of work late in life.
Third, raising the retirement age disproportionately hurts low-income Americans who rely on Social Security the most. A two-year increase in the retirement age is equivalent to a 13% cut in benefits for someone who retires at 65. For seniors in the bottom fourth of the income distribution, this translates into an 11% cut in much-needed income, because these seniors rely on Social Security for 84% of their total income. For seniors in the top fourth of the income distribution, however, this would amount to a 2.6% percent cut in total income. This is not to suggest that we should shrink Social Security by targeting cuts at the top, however, because Social Security’s strength is its universality. The fact that even high-income earners have an important stake in Social Security is why the program has remained almost unscathed for 75 years, while other parts of our safety net are in tatters.
As Social Security Chief Actuary Stephen Goss has pointed out, the main pressure on the cost side isn’t rising life expectancy, but rather declining birth rates. Revenues, however, are also declining due to stagnant wages, growing wage inequality, and rising health care costs.
The Greenspan Commission predicted the Baby Boom and rising longevity and took them into account when they balanced benefit cuts and increased revenues.  What the Commission didn’t anticipate was the enormous growth in inequality, that the top 1% would get 55% of all income growth over the last 30 years, while the bottom 90% would get only 16%.  Rising inequality has meant that much more income growth has occurred above the taxable income cap than below it, shrinking the program’s revenue dramatically.
As the earnings of most workers have stagnated and earnings of those at the top have skyrocketed, the system’s revenues have suffered because earnings above the taxable earnings cap—currently set at $106,800—are not subject to the Social Security payroll tax. Though the cap is indexed to average wages, these wages have not grown as fast as earnings at the top, leading to an erosion of Social Security’s tax base. As a result, the share of untaxed earnings grew from 10% in 1983 to around 16% in 2008.
The problem has been compounded by health care cost inflation, which increases the share of compensation going to untaxed fringe benefits. The Social Security actuaries estimate that the recent health care overhaul will somewhat mitigate this problem, but health care cost inflation remains a problem for Social Security and the economy as a whole.
Most Americans don’t realize that someone with a salary of $300,000 or even $30 million a year pays no more in Social Security taxes than someone earning roughly $107,000.  When they do realize this, they don’t like it.  A poll commissioned by the Rockefeller Foundation and the National Academy of Social Insurance (NASI) found that 83% of respondents support lifting the Social Security tax cap so that all workers pay the same payroll tax rate, regardless of income.  

In prior congressional testimony, EPI Research and Policy Director John Irons recommended a variation on elimination of the cap: eliminating it for employers while retaining but raising the cap on high-income em
ployees.  With earnings up to the employee cap credited for benefit purposes, this change would reduce the long-term funding shortfall by about three-fourths.

There are several advantages to this approach. It would eliminate most of the long-term shortfall, while maintaining a link between contributions and benefits. It would not lead to extremely large benefits for millionaires, which could be a concern if all earnings were credited for benefit calculations. Finally, self-employed taxpayers, who are responsible for both the employer and employee contributions, would not face as large an increase in payroll taxes as a full elimination of the cap.
Furthermore, this option would have a modest impact on the standard of living of upper-income taxpayers. On the employee side, this would mean an increase in tax payments of, at most, 2.6 percent of income. If income growth for the top 5% of households continues as it has for the past 20 years, and assuming that all 6.2% of the employer tax were passed on to employees in the form of lower wages, this additional tax obligation would be recouped by these households in less than four years. Affected taxpayers would also recoup some of these higher taxes in the form of higher benefits.

3. There are potential solutions to the retirement crisis, but tweaks and small changes at the margins won’t be enough.
Given the $6.6 trillion retirement income deficit, strengthening Social Security, rather than further weakening it by reducing benefits, is a necessary but insufficient first step to restoring retirement security.  As we said at Retirement USA’s inaugural conference last year:
“We need a comprehensive solution that addresses interrelated problems. For example, a system that places most of the burden for retirement saving on individuals will always have to wrestle with the problem of pre-retirement loans and withdrawals (simply plugging these leaks will not work, because many workers would stop contributing to the system). A system that relies on tax incentives to promote individual retirement savings will necessarily tend to favor high-income workers who can afford to save more and who benefit the most from these tax breaks. Conversely, a truly universal system would need to shield low-income workers from out-of-pocket costs or wage cuts.”
EPI has published and advocated what we feel would be an excellent national supplemental retirement plan, the Guaranteed Retirement Account, which was authored by Prof. Teresa Ghilarducci, Director of the Schwartz Center for Economic Policy Analysis at the New School for Social Research.  In a nutshell, the GRA would mandate employer and employee contributions to a federally administered cash balance plan.  The combined 5% of payroll contributions would be invested by a Thrift Savings Plan-like entity in the bond and stock markets, with a guaranteed minimum return of 3% beyond inflation.  A $600 tax credit would cover the entire 2.5% contribution for workers earning $24,000 or less, and greatly reduce the effective contribution rate for other lower-paid workers. We calculate that at the end of a normal working life, the average worker would accumulate, along with Social Security, enough to assure a 70% replacement rate of pre-retirement income.
Retirement USA has not endorsed the GRA, except to affirm that it meets all of the 12 principles the coalition set out as essential to deliver retirement income that is universal, secure, and adequate. Our coalition has asked the public for other model reform plans that meet our principles and have received more than two dozen that satisfy most or all of them.  It is clear to the Retirement USA coalition that any successful model will have certain common elements:
• All jobs must come with benefits that provide a steady retirement income for life. As currently structured, Social Security is not enough. Relying primarily on tax incentives to encourage employers to provide benefits or individuals to save is ineffective and helps those who least need it.
• Investment and longevity risks must be spread, not just shifted from employers to workers. Here too, government can play a role, and so can multiple-employer plans.
• Responsibilities must be shared. A do-it-yourself system does not work, but neither does a system that places the entire burden on employers. Government must also be involved, especially to offset the cost of contributions for lower-income workers.
• Finally, the key to achieving adequacy is maintaining steady contributions and preserving funds for retirement by preventing pre-retirement loans and withdrawals and by limiting fees.
The most interesting plans we received include the Variable Defined Benefit Plan conceived by Gene Kalwarski, CEO of Cheiron, Inc., the Retirement USA-Plus presented by Nancy Altman, Chairman of the Board of the Pension Rights Center, and Glenn Beamer’s Guaranteed Pension and Community Investment Plan, all of which are summarized, with others, on the Retirement-USA website (www.retirement-usa.org).

See related work on Retirement

See more work by Ross Eisenbrey