Social Security by Shutterstock.com
Even with the fiscal cliff agreement behind us, we can expect more budget negotiations as we inch closer to the debt ceiling. In exchange for increasing the debt limit, congressional Republicans, who failed to include any significant spending cuts in the deal from January 1, should demand that entitlement reforms be put on the table to stabilize the country’s fiscal path while also honoring its financial obligations. Rising entitlement outlays are the major driver of the long-term fiscal gap, and cuts to future entitlement benefits are more enforceable than other spending cuts, helping address conservative concerns that future Congresses will not follow up the new tax increases with agreed-upon spending cuts.
Of the three major entitlements, Social Security is the most likely to be on the table, as the available policy options are well-understood and House Speaker John Boehner (R-OH) and President Barack Obama already have touched on the program in the early fiscal cliff negotiations. But any Social Security reforms proposed by congressional Democrats will involve tax increases. Republican negotiators should bear in mind the economic effects of raising payroll taxes, which run contrary to common-sense policies in an aging society. If Social Security reform must involve tax increases, perhaps the best approach might be the simplest: raise the 12.4 percent Social Security payroll tax rate. At the least, this approach would ensure that everyone receiving benefits would be involved with paying for them.
Raising the Payroll Tax Is Not the Best Answer
Social Security levies a flat 12.4 percent payroll tax on earnings below $110,100 and pays a progressive benefit based on the same taxable earnings. The generosity of benefits has changed little over time, with a rising retirement age effectively compensating for longer life spans. The real change has been to taxes: what began as a 2 percent tax in 1935 has evolved into a 12.4 percent tax on earned income. In 1936, the Social Security Administration (SSA) promised that a 6 percent total tax, to be implemented in 1949, would be “the most you will ever pay.” In an ironic sense, SSA’s statement was correct: today, the Social Security tax is the biggest tax that most workers pay.
Plenty of reasons exist not to raise Social Security payroll taxes. Consider this logic: an aging population requires that a smaller workforce support growing numbers of retirees. The only way for population aging not to devolve into a zero-sum fight between the young and the old is to grow the economy. Simply put, a larger economic pie is the only way to avoid a fight over the size of the slices.
To grow the economy, public policy should encourage Americans to do three things: work more, meaning more hours per day and more days per year; save more, in practice meaning greater participation in 401(k)s and other pension plans; and retire later, meaning extending work lives from today’s common retirement age of 62 to 65 or even 70. Together, these steps could meaningfully increase economic output and ease the financial burden of a graying America.
Now, consider how raising the Social Security payroll tax rate would impact these goals. First, like other taxes, it would discourage work, lowering the net return paid to employees in exchange for their time and effort. For instance, a 2005 study by Wilbert van der Klaauw of the University of North Carolina and Kenneth I. Wolpin of the University of Pennsylvania found that raising the payroll tax from 12.4 to 15.0 percent would reduce annual hours worked by around 6 percent for middle-aged individuals and even more for near-retirees. This would mean lower incomes and tax revenues.
Second, raising the payroll tax would discourage personal saving, because higher taxes would both facilitate higher future Social Security benefits and leave less take-home pay available to save. A literature review by the Congressional Budget Office concluded that one dollar of future Social Security benefits crowds out between zero and 50 cents of private saving. More recent research using better data indicates that, for middle and high earners, roughly two-thirds of Social Security benefits could be offset against private saving. This implies that high earners would mostly compensate for benefit reductions by saving more on their own, while tax increases to raise payable benefits to these groups would result in lower saving. Lower saving would weaken economic output.
And third, higher taxes would encourage early retirement, because the after-tax replacement rate offered by Social Security—that is, retirement benefits as a percentage of after-tax preretirement earnings—would increase. This makes retirement appear more attractive than continued work.
All told, raising Social Security taxes likely means we will have a smaller economic pie to fight over.
Discouraging Saving in Middle-Income Households
One could argue that despite all this, Social Security benefits are so important that raising taxes yet again is worthwhile. It is not simply that truly low-income Americans depend highly on Social Security in retirement. Middle- and even higher-income households also derive a significant share of their retirement income from the program. For instance, in a 2008 study for the Social Security Administration, Glenn Springstead and I found that for middle-income retirees Social Security constituted 41 percent of total household income. Even for the top fifth of retiree households, Social Security made up 29 percent of income.
But what many progressives count as a success may fairly be seen as a failure. Although dependence on Social Security by low earners is to be expected and the cost of that support is accepted by both political parties, middle and high earners could, should, and would save more on their own in the absence of generous Social Security benefits. One study by economists Jagadeesh Gokhale, John Sabelhaus, and Lawrence Kotlikoff attributes most of the postwar decline in personal saving to the size and structure of government entitlement programs.
As I have noted, research shows that middle and high earners treat Social Security benefits and personal saving as substitutes. Social Security currently owes roughly $22 trillion in accrued benefits, a value that rivals the nonresidential private capital stock of the United States. Had part of that value been in the form of real saving, US productivity and economic output would have been considerably higher. Real saving raises the capital stock, making resources available for investment in factories, tools, computers, and other assets that make workers and the economy more productive. While little can be done today to convert existing Social Security obligations into real capital, it makes sense to boost saving going forward.
Finding Ways Forward
The bottom line is that Social Security needs to be fixed. But if Social Security reform enters the budget negotiations, Democrats will certainly demand tax increases to fix Social Security. And this set of tax hikes will fall on the precisely same group targeted in the fiscal cliff deal: high earners. For most Democrats, the preferred Social Security reform by far is to increase or eliminate the $110,000 cap on which payroll taxes are levied. On paper, eliminating that cap and applying payroll taxes to all earnings would keep the program funded for almost 75 years, if you take the trust fund to be a real store of value rather than a mere accounting measure.
But that is true only on paper. A 2006 study by Harvard’s Jeffrey Liebman and the University of California’s Emmanuel Saez—the former a Social Security expert and high-ranking Obama budget official, the latter the author of numerous tax studies embraced by the left—concluded that under conservative assumptions regarding economic and budgetary feedbacks, eliminating the payroll tax cap would raise less than 60 percent of the net revenue projected under static analysis. This would come at the price of increasing effective top marginal tax rates by roughly 10 percentage points. Using more aggressive, but still reasonable, assumptions regarding taxes and economic behavior, such as those generated by Harvard economists Martin Feldstein and Raj Chetty, net revenues likely would not rise at all. In other words, the policy alternatives embraced by the left are not much of an alternative.
Thus, if increasing Social Security taxes is the only way to achieve Social Security reform, the most sensible approach might be to do it the old-fashioned way: by raising the payroll tax rate. To make the system permanently sustainable through only tax increases would require an immediate and permanent increase in the rate from 12.4 percent to 16.3 percent. As I have noted, raising the payroll tax rate carries significant economic penalties. Yet spreading the tax increase among all workers, rather than simply the top several percent, may lessen, though obviously not eliminate, the aforementioned negative economic effects.
Raising taxes is not a solution to the overall entitlement problem. The tax increases necessary to fund rising Social Security, Medicare, and Medicaid outlays would harm the economy. Moreover, Social Security is the entitlement for which tax increases are least appropriate: while it is difficult for individuals to make up for lower Medicare and Medicaid benefits through their own saving, middle and high earners would react to lower promised Social Security benefits by saving more and delaying retirement. These steps could ameliorate Social Security benefit cuts and strengthen the economy’s ability to support larger numbers of retirees with a relatively smaller population of workers.
But if tax increases for Social Security are to be on the table, raising the payroll tax rate would force all Americans to pay for the entitlement programs that all Americans benefit from. Progressives are fond of citing polls stating that Americans love Social Security and are willing to raise taxes to fund it. But what an individual tells a pollster may differ from what he or she actually would support. Congressional Democrats appear to understand that: most of the reform proposals sponsored by liberals, such as a recent plan introduced by Senator Mark Begich (D-AK), involve raising taxes only on high earners, not across-the-board rate increases. At least discussing an overall payroll tax rate increase would put Americans on notice that if they want a growing entitlement state, they must be willing to pay for it.
1. See Social Security Online, “Social Security Numbers: The 1936 Government Pamphlet on Social Security,” www.ssa.gov/oact/tr/2012/tr2012.pdf.
2. Wilbert van der Klaauw and Kenneth I. Wolpin, “Social Security and the Retirement and Savings Behavior of Low Income Households” (PIER Working Paper Archive 05-020, Penn Institute for Economic Research, Department of Economics, University of Pennsylvania, 2005).
3. Orazio P. Attanasio and Susann Rohwedder, “Pension Wealth and Household Saving: Evidence from Pension Reforms in the United Kingdom,” American Economic Review, 93, no. 5 (2003): 1499–1521.
4. Glenn Springstead and Andrew G. Biggs, “Alternate Measures of Replacement Rates for Social Security Benefits and Retirement Income,” Social Security Bulletin, 68, no. 2 (2008).
5. Jagadeesh Gokhale, Laurence J. Kotlikoff, and John Sabelhaus, “Understanding the Postwar Decline in U.S. Saving: A Cohort Analysis” (NBER Working Paper No. w5571, May 1996).
6. Jason Schultz and Daniel Nickerson, “Unfunded Obligation and Transition Cost for the OASDI Program,” Social Security Administration, Office of the Chief Actuary, Actuarial Note, no. 2010.1 (September 2010). 2010 figures are increased to 2012 levels based on historical growth rates. Capital stock figures from the National Income and Product Accounts.
7. Jeffrey Liebman and Emmanuel Saez, “Earnings Responses to Increases in Payroll Taxes” (working paper, University of California–Berkeley, 2006).
8. The employee payroll tax rate is 6.2 percent, which at first glance would imply a smaller increase in marginal taxes. However, the standard assumption is that employers would reduce wages by the amount necessary to pay their own 6.2 percent tax, implying that after-tax wages would nevertheless decline by around 12.4 percentage points. However, assuming that individuals would continue to accrue benefits based on those higher taxable earnings, the new tax rate net of future benefits received would be roughly 10 percent.
9. See Martin S. Feldstein, “Effects of Taxes on Economic Behavior” (NBER Working Paper No. w13745, January 2008); and Raj Chetty, “Bounds on Elasticities with Optimization
Frictions: A Synthesis of Micro and Macro Evidence on Labor Supply,” Econometrica 80, no. 3 (2012): 969–1018.
10. Feldstein calculates an elasticity of taxable income of around 0.5 for high earners; Liebman and Saez show a range of values including 0.0, 0.2, 0.5, and 0.8. The figures cited in this paragraph for Liebman and Saez assume an elasticity of 0.2.
11. Board of Trustees, Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds, The 2012 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds, April 25, 2012, www.ssa.gov/oact/tr/2012/tr2012.pdf.
12. This derives from the rule in public finance that the “dead-weight loss” of a tax rises with the square of the tax rate; thus, spreading a given tax increase over a larger number of taxpayers would have smaller negative economic effects than imposing it on a smaller group of taxpayers.
13. Begich’s proposal would eliminate the payroll tax ceiling and also base cost-of-living adjustments on a Consumer Price Index for the elderly, an experimental inflation measure based on the purchasing habits of individuals over age 65. See “Begich Proposes Changes to Strengthen Social Security,” November 13, 2012, http://begich.senate.gov/public/index.cfm/pressreleases?ID=4b784c46-38c0-41f9-be08-6bc45db6ab62.