AP Images

About the author: Jesse Rothstein is a professor of public policy and economics and Director of the Institute for Research on Labor and Employment at the University of California-Berkeley.

The economic security of the broad middle class in the United States has eroded in recent decades, with stagnating wages, vanishing job security, and necessities such as child care, higher education, and healthcare becoming less and less affordable. Restoring economic security should be the center of the next Administration’s domestic agenda.

One area that needs to be rethought is retirement. This was traditionally seen as a three-legged stool, with retirees relying in roughly equal parts on Social Security benefits, employer-provided pensions, and private savings. This was never the universal reality—even 30 years ago, barely half of workers nearing retirement participated in employer-based plans.1 But today, all three legs of the stool have been whittled away to almost nothing:

  • Defined-benefit pensions have largely disappeared. In 2010, only 22 percent of full-time, private-sector employees had a defined-benefit retirement plan,2 and public-sector pension funds (along with remaining private-sector funds) face serious financing shortfalls.
  • Less than 60 percent of near-retirement-age households have any retirement savings,3 and among those the median balance is only $91,000.4 This would purchase an annuity (at age 65) of only $5,000 per year.
  • Fewer and fewer Americans—about half in recent Gallup polling5—believe that Social Security will be there for them in retirement.

In this environment, only the wealthiest Americans can feel any sense of confidence about their retirement. Policy changes that strengthen the retirement system have tremendous promise to improve people’s feelings of security—not just retirees but also people in the prime of their lives.

Fortunately, the solution is simple. Our efforts should focus on strengthening and growing the part of the system that works well: Social Security. Fixing the financing shortfall and growing the system to provide real economic security in retirement, on its own, are affordable and feasible, and should be an important part of the middle class security agenda.

The problem:
Inadequate retirement security

Nearly every retiree receives Social Security benefits—about 40 million retired workers, plus millions more of their spouses, their other dependents, and survivors of those who die early.6 These benefits average under $1,350 per month,7 with 30 percent of retirees receiving less than $1,000 per month.8 This is not enough on its own to finance a decent standard of living.

Delivering equitable growth

Previous article: Early childhood development, Ariel Kalil
Next article: Unemployment insurance reform, Till von Wachter

Unfortunately, the other two legs of the three-legged stool, pensions and private savings, are no longer reliable. Half or less of workers nearing retirement age participate in employer-based retirement plans,9 and among full-time, private-sector workers with plans, most have only a defined-contribution plan, such as a 401(k); only about one-third have a defined-benefit plan (a traditional pension).10 As for private saving, less than 60 percent of near-retirement-age households have any savings in retirement accounts (including employer-based accounts).11 For those that do have savings, the median balance is only $91,000.12 If this were all converted to an annuity at age 65, it would provide less than $5,000 per year in retirement income. While a very small number of high-income retirees have substantial savings, this is very much not the norm.

Accordingly, Social Security makes up a very large share of most retirees’ incomes. For nearly half of married retirees and nearly three-quarters of unmarried retirees, Social Security benefits provide more than half of the household’s income.13 Over one-fifth of married retirees and nearly half of unmarried retirees rely on the program for fully 90 percent of their income.14 These benefits are not enough to support a comfortable retirement. Nine percent of seniors live in poverty.15 Using the U.S. Census Bureau’s new supplemental poverty measure that accounts for medical costs, geographic differences, and transfer payments, the proportion in poverty grows to 15.8 percent.16

No real prospect of restoring security
via pensions or private savings

Traditional, employer-provided pensions fit poorly with today’s labor market, where few workers can expect to remain at a single employer their whole careers. Moreover, in today’s financialized world employers find it too easy to renege on commitments to their past workers, leaving the government holding the bag and retirees without the pensions they were promised. Many private pension funds have gone broke; others have been closed to new enrollees. In the public sector, where it is harder for governments to slip out of past commitments, many public pension funds face serious gaps between promised benefits and funds with which to pay them, putting severe stress on state and local budgets.

Efforts to shore up the old pension system have had little success. Coverage rates, especially in the private sector, have plummeted. Recent efforts have focused on designing portable pensions that are not tied to specific employers. These efforts, while worthy, face fundamental challenges of financial viability, administrative complexity, and legal constraints. They are unlikely to reach many workers in the short run.

Efforts to promote retirement savings in defined-contribution plans, including IRAs and 401(k)s, have by and large been unsuccessful. Tax subsidies for retirement saving largely serve to induce households that would have saved anyway to shift from unsubsidized to subsidized accounts, with little impact on overall savings.17 While some “nudge” programs have been found to have larger effects,18 decades of policy and research work have failed to dent the problems of low participation rates and low balances among those who save.

For those who do manage to save, private accounts are a poor solution. Savers pay high fees on their investments, and often make bad investment decisions (in many cases under instruction from advisors facing serious conflicts of interest). Even wise investment plans expose savers to substantial market risk, as many Baby Boomers discovered in 2008.

Moreover, private savings plans have not solved the fundamental challenge of retirement savings—the need to insure the individual against uncertainty over the length of the retirement period and over investment returns. While retirees are often advised to buy annuities, these are very expensive and few retirees purchase them. Absent annuities, the only way individuals can be confident that they will not outlive their savings is to save much more than they will likely need, and to limit their drawdown, to preserve money that they will likely never spend. While some wealthy individuals may want to leave substantial bequests for their heirs, most would be better off under a retirement system that allowed them to spend down their savings during their lives, confident that they would be protected if they lived longer than expected.

The solution is simple:
Expand Social Security

Fortunately, it is possible to dramatically expand retirement security without fixing the problems with employer-provided pensions and private savings. Social Security has worked for decades and can easily be scaled up to fill in the holes in the retirement puzzle.

As a social insurance program, Social Security builds in protection against longevity and earnings risk, and shields retirees—most of whom have no interest in or capacity to participate in financial markets—from bad markets and unscrupulous or incompetent investment advisers. Its funding is not tied to the continued viability of any single employer, and while occasional adjustments are needed as demographic and economic realities diverge from earlier projections, these are small relative to the volatility faced by pension funds or private savings.

Policymakers should therefore pursue a major expansion of Social Security. There are a number of attractive ways to do this. I articulate three principles that should guide such an expansion, then describe three options that would satisfy the following principles:

  • Social Security benefits should be enough for retirees to live on, on their own
  • The features that have made the program such a success to date, including its universality and its dedicated, untouchable financing stream, should be preserved
  • The existing program’s long-term financing shortfall should be closed via new revenues rather than benefit cuts, with the pain of this offset by real, visible benefits in the form of higher payments to retirees

The following three policy options are consistent with these principles

Three policy changes
to expand Social Security

Expand the program via proportionate increases to tax rates and to all recipients’ benefits

A modest increase in the payroll tax rate of less than one percentage point—from 6.2 percent to 7.2 percent, respecting the current taxable earnings cap—on both workers and employers would finance a 15 percent benefit increase for all current and future retirees as well as survivors and disabled workers. This could (and should) be scaled to provide larger increases—for example, a 2 percentage point increase in the tax rate would finance a 30 percent increase. If disabled workers’ benefits were excluded from the expansion, the benefit increase would be larger, 18 percent per percentage point on the tax rate.

Eliminate the current taxable earnings cap, and split the revenues among increasing benefits for those who pay higher taxes, closing the financing gap, and increasing benefits for the lowest earners

Earnings above $118,500 per year are exempt from Social Security taxes, specifically the Federal Insurance Contributions Act, or FICA, taxes. Only 6 percent of workers earn more than this amount, but the share of wages above the threshold has grown as these income earners have pulled away from the rest of the workforce.19 Eliminating the cap, without increasing benefits, would yield more than enough revenues to close the 75-year projected finance shortfall for Social Security and Social Security Disability Insurance.20

More consistent with the principles above would be to use some of the revenues to increase benefits for high earners in order to preserve the program’s universality while imposing a reasonable cap on the size of the individual benefit. Another portion should go toward increasing benefits for lower-earning recipients. This would help promote retirement security and would create a constituency to support the cap elimination. The remainder could still make a substantial down-payment toward eliminating the financing gap.

If earnings were subject to taxes without a cap but non-wage income remained exempt, many high earners might reclassify their earnings as capital gains or business income. To prevent evasion while being consistent with the overall structure of the program, the FICA tax might be applied to all personal income above the taxable earnings cap, regardless of whether that income comes as earnings or in another form. This would raise dramatically more money, more than enough to close the entire financing gap and support larger benefit increases for low-income retirees. While a few very high-income individuals would face sharply increased tax bills under this proposal, dedicating the revenues to the Social Security trust fund, just as FICA taxes are today, would promote the tax increase’s political viability and protect it from attack.

Create an optional extra tier of Social Security, into which workers could contribute in order to purchase extra benefits when they retire

Some higher-income workers may want more retirement consumption than even an expanded Social Security benefit can support. As discussed above, pervasive market failures make private retirement saving very inefficient. The government can solve this problem by allowing savers to use the existing Social Security infrastructure, with its built-in mechanisms, to pool and share risks at much lower “load” than any private savings vehicle.

Pricing of optional add-on benefits would need to account for a modest degree of adverse selection, as those with longer life expectancies would be somewhat more likely to participate. It would also be more expensive to administer than the current system, though likely much less so than existing private annuity products given the existing infrastructure to track contributions and benefit eligibility. But even with pricing that accounts for these extra costs—set to ensure that the new tier of optional add-on benefits is entirely self-financing—this would be a very attractive option relative to the private annuity market. Participants would contribute throughout their careers, with no need to make investment decisions, pay fees, or face market risk. In return, they or their survivors would receive higher benefits when they retired, became disabled, or died.

Other proposed changes that are similarly consistent with the principles enumerated above are also worthy of consideration. It is noteworthy, however, that recent discussions have moved away from the ideas of pre-funding Social Security benefits and investing trust funds in equities. These do nothing to address the above principles, and only expose the program (and the retirees it supports) to political and market risk.

Countering the objections
to expanding Social Security

There are three primary objections to the above proposals. None are compelling.

The first is that Social Security would be too expensive. It is undeniable that larger Social Security benefits would require correspondingly more revenue. But there is no way to support higher consumption for retirees—no way to address the retirement security problem—without in some way directing additional resources to these retirees. Any alternative would impose equal or higher costs, if not as higher payroll taxes than as higher diversion of before- or after-tax income into private savings or pension funds. It is much more efficient to route the needed resources through Social Security. Moreover, taxpayers see Social Security taxes as worth paying, tied as they are to eventual benefits.

The second objection is that higher tax rates would reduce the incentive to work. Again, Social Security expansions would not dramatically change the effective “wedge” between earnings and consumption—the money that would be collected through higher payroll taxes either represents new additions to the taxpayer’s consumption possibilities in retirement or merely displaces private savings with, if anything, a positive impact on retirement consumption. Thus any labor supply impacts would be very small.

The above proposals to uncap income subject to FICA taxes, while capping benefits, would expand the wedge for the highest-income workers. There is little evidence that these workers’ labor supply is very responsive to taxes, and the scope for evasion (which is more responsive) could be minimized by assuming that tax increases apply to a broad income base. In any case, this objection applies equally to any proposal to increase top income tax rates, a move that is likely inevitable as the share of national income accruing to the top 1 percent continues to grow.

The third objection is that expanding Social Security would crowd out private saving and reduce national savings. This is a prediction of standard economic models. But these models do not reflect reality. In fact, most workers have very little savings to be crowded out, and those who do save are not primarily saving for retirement (though they may store their savings in retirement accounts in order to obtain tax benefits).

Regardless, the national savings rate is simply not a first-order policy concern today. Finance is increasingly global, and profitable U.S. investments can be financed by savers elsewhere. Moreover, concerns about promoting saving are yesterday’s problem. Today, the evidence indicates that the nation and the world face a serious savings glut, not a shortfall. In today’s world, there is little if any reason to prefer prepaid to pay-as-you-go retirement systems.

  1. Monique Morrissey and Natalie Sabadish, “Retirement Inequality Chartbook,” (Washington, DC: Economic Policy Institute, 2013): 11, http://www.epi.org/publication/retirement-inequality-chartbook/
  2. Ibid, 12.
  3. Ibid, 16.
  4. Ibid, 20.
  5. Frank Newport, “Many Americans Doubt They Will Get Social Security Benefits,” Gallup, August 15, 2015, http://www.gallup.com/poll/184580/americans-doubt-social-security-benefits.aspx?version=print
  6. Social Security Administration, “The 2016 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds,” (Washington, DC: Social Security Administration , 2016), 132.
  7. Social Security Administration, “Fact Sheet: Social Security.” Undated. Retrieved from https://www.ssa.gov/news/press/factsheets/basicfact-alt.pdf on September 25, 2016.
  8. Social Security Administration, “Retired worker beneficiaries in current payment status at the end of June 2016, distributed by benefit level, sex, and age group.” Undated. Retrieved from https://www.ssa.gov/OACT/ProgData/benefits/ra_mbc201606.html on September 25, 2016.
  9. Morrissey and Sabadish, “Retirement Inequality Chartbook,” 10.
  10. Ibid, 10.
  11. Ibid, 16.
  12. Ibid, 20.
  13. Social Security Administration, “Fact Sheet: Social Security.” Undated.
  14. Ibid.
  15. Loraine A. West, Samantha Cole, Daniel Goodkind, and Wan He., “65+ in the United States: 2010.” (Washington, DC: U.S. Census Bureau, 2014), 79.
  16. Ibid, 82.
  17. Eric M. Engen, William G. Gale, and John Karl Scholz, “The illusory effects of saving incentives on saving.” The Journal of Economic Perspectives 10, no. 4 (1996): 113-138. See also: Raj Chetty, John N. Friedman, Søren Leth-Petersen, Torben Heien Nielsen, and Tore Olsen, “Active vs. Passive Decisions and Crowd-Out in Retirement Savings Accounts: Evidence from Denmark,” The Quarterly Journal of Economics 129, no. 3 (2014): 1141-1219; Esther Duflo, William Gale, Jeffrey Liebman, Peter Orszag, and Emmanuel Saez, “Savings Incentives for Low‐ and Moderate‐Income Families in the United States: Why Is The Saver’s Credit Not More Effective?,” Journal of the European Economic Association 5, no. 2‐3 (2007): 647-661.
  18. Emmanuel Saez, “Details matter: The impact of presentation and information on the take-up of financial incentives for retirement saving,” American Economic Journal: Economic Policy 1, no. 1 (2009): 204-228.
  19. Kevin Whitman and Dave Shoffner, “The Evolution of Social Security’s Taxable Maximum,” Social Security Administration Policy Brief 2011-02 (Washington, DC: Social Security Administration, 2011), 5.
  20. The Social Security Trustee’s Report (Social Security Administration 2016) states that Medicare taxable payroll (which has no taxable maximum) is about 25 percemt larger than Social Security taxable payroll (p. 5), implying that the eliminating the cap would raise revenue by 25 percent * 12.6 percent = 3.15 percent of (current) taxable payroll. By contrast, the projected 75-year actuarial deficit for the Old Age and Survivors Insurance and Disability Insurance trust funds is 2.66 percent of taxable payroll.