The unfixable part of the problem is demographic. The population aged 65 or older will increase from 37 million today to 75 million in 2035, when their average life expectancy will be 85. Because future beneficiaries will live longer, they will receive more benefits over their lifetimes. Meanwhile, the number of younger workers who pay for these retirees will barely rise. Economists Jagadeesh Gokhale and Kent Smetters note that today, “there are almost five people of working age (between 20 and 64) for each retiree age 65 and over. By 2030, the number of working-age people per retiree will decline to less than three; by 2080, the ratio will decline to about two.” Any burden shared by half as many taxpayers must, as a matter of simple arithmetic, become twice as heavy.
This demographic time bomb has been compounded by an arbitrary “wage indexing” formula. The level of benefits when people first start collecting Social Security happened, in the late 1970s, to be indexed to average growth of real wages rather than simply adjusted for inflation (as many experts proposed). As a result, benefits for new retirees become more generous by about 1 percent each year in real terms, which adds up fast. “The purchasing power of the average earner’s benefits at retirement is expected to nearly double between now and 2075,” notes the Congressional Budget Office, with the result that “45 percent of the rise in spending is due to a projected increase in the real value of Social Security benefit checks.” As long as initial benefits are indexed the way they are, faster economic growth will not help much — because it would result in faster wage growth and therefore larger Social Security benefits.
The reason future taxpayers cannot and will not pay rising real benefits to twice as many seniors is basic economics. Nobel laureate Edward Prescott found that lower income, payroll, and sales taxes fully explain why Americans work so much harder and longer than Europeans. In economic jargon, lifetime work effort is highly “elastic” (responsive) with respect to tax rates. “The large labor-supply elasticity,” Prescott concludes, “means that as populations age, promises of payments to the current and future old cannot be financed by increasing tax rates. These promises can be honored by reducing the effective marginal tax rate on labor and moving toward [different] retirement systems… . Requiring people to save for their retirement years [in personal accounts] is not a tax and does not reduce labor supply.”
A System and Its Canards
The Social Security problem, in short, is that attempting to provide rising real benefits to millions more seniors would require debilitating taxes on future workers — thus discouraging work, slowing the economy, and exacerbating the fiscal dilemma. That is why Prescott insists this problem “cannot” be fixed with higher taxes: The threat of higher taxes is the Social Security problem. While keeping the essence of this problem in mind, let’s examine a few erroneous suggestions that have been made about what to do about it.
One: “The system” must be saved. If balancing the books were all that mattered, this could easily be accomplished by, say, raising the Social Security retirement age to 100. What really matters is making more choices and opportunities available to younger people to help them plan for the fact that their “human capital” eventually wears out and will need to be augmented with some financial capital. Giving them the opportunity to divert — voluntarily — part of their payroll tax to a personal account is crucial.
Those who propose to “reform” Social Security without adding choice and ownership are actually just proposing various ways of saddling young people with higher taxes and lower benefits. For example, Peter Diamond and Peter Orszag in their book Saving Social Security rely on increased taxes on higher earners for 42 percent of their hoped-for reduction in unfunded debt — yet benefits to the top 15 percent would also be cut by a third. The other half of their savings comes from “a universal legacy charge on future workers and beneficiaries, roughly half in the form of benefit reductions for all beneficiaries becoming eligible in or after 2023, and the rest in the form of … increases in the payroll tax from 2023 onward.” “Onward” means forever: a tax increase every year. Taxpayers would pay more and more for their benefit reductions.
Peter Wehner, President Bush’s director of strategic initiatives, issued a memo saying, “If the goal is permanent solvency and sustainability … then [Personal Retirement Accounts], for all their virtues, are insufficient to that task.” Wehner’s emphasis was inverted. Young people do not need PRAs because such accounts solve Social Security’s financing problem; they need PRAs because all solutions to Social Security’s financing problem reveal that young people dare not rely on it for any substantial portion of their financial needs in old age.
Two: Social Security isn’t bankrupt. The president uses the term “bankrupt” to describe Social Security’s future; his critics respond by saying there is no immediate “crisis.” But these are quite different concepts. When giant corporations go bankrupt, that does not mean they have no income or assets; it means they don’t have enough income to pay their bills. So it will be with Social Security sometime after 2042, unless commitments are brought into line with resources. The trust fund is not a pile of cash. It is just a bookkeeping gimmick in which one part of the government promises money to another part of the government. That “fund” will not make it even one dollar easier to pay future benefits. Although Social Security is thereby authorized to pay benefits for 24 years after benefits begin to exceed Social Security receipts (in 2018), doing so would require adding some $200 billion to every budget deficit by 2027, and more thereafter.
Three: Transition costs would be prohibitive. Critics of partial privatization claim it would cost a trillion dollars or more to set up such accounts and that diverting taxes to such accounts would make Social Security less solvent. Both statements are based on an illusion, resulting from the government’s backward-looking accounting. Gokhale and Smetters ask us to suppose that Mr. Smith is allowed to carve out $1 from his Social Security tax and deposit that $1 in a personal account. In exchange, Mr. Smith accepts an actuarially fair $1 reduction in his future Social Security benefits. Bookkeeping aside, this transaction would have zero net effect on the fiscal imbalances of either Social Security or the rest of the government. Social Security’s unfunded debt would not change in the slightest, nor would the overall long-term government budget.
Four: It would be fiscally wise to raise the cap on earnings subjected to Social Security taxes. There has been considerable Washington chatter about lifting this earnings cap from the current $90,000 to $140,000. If this happens, the maximum yearly tax would jump from $11,160 to $17,360. Under current rules, however, paying larger taxes would entitle taxpayers to larger benefits, leaving little net effect on Social Security’s solvency. This scheme also looks like a slippery first step toward eliminating the cap altogether, as happened with Medicare. Only 6 percent of Americans have salaries above $90,000, so soaking the rich would not raise much revenue; yet adding this tax hike would push the overall burden on professionals and entrepreneurs above the highest tax rates in Europe, with devastating economic effects. This terrible idea is being peddled to Republicans as a compromise. But there is a difference between compromising and being compromised. For me, this is a deal-breaker.
Five: We can simply “refinance” un-financed promises. Some supporters of a specific proposal by Rep. Paul Ryan and Sen. John Sununu say the entire $12 trillion unfunded debt would be no problem if we simply “refinanced” it. But you can’t refinance something that was never financed in the first place — which is what “unfunded” means. The main problem with unfunded future Social Security promises is that they imply a horrific tax burden after 2030 or so. Trying to pay those bills by issuing debt would not fix that problem. On the contrary, it would convert a formidable threat of horrific future taxes into an absolute certainty, because future taxpayers would then be obligated to repay the extra debt or pay interest on it. To lock in a huge real increase in benefits, thus committing future taxpayers, is not “refinancing”; it is a recipe for disaster.
Six: Chile’s privatized retirement system doesn’t work. Larry Rohter of the New York Times recently concocted a story about “the bitter legacy” of Chile’s privatized pensions. “For all the program’s success in economic terms,” he wrote, “the government continues to direct billions of dollars to a safety net for those whose contributions were not large enough to ensure even a minimum pension approaching $140 a month. Many others, because they earned much of their income in the underground economy … or work only seasonally, remain outside the system altogether.”
Chile’s guaranteed minimum pension, roughly similar to our SSI program, costs $70 million a year — not billions, as Rohter would have it. And Americans who work seasonally or in the underground economy are “outside” our Social Security system too, but Chile’s underground economy actually shrank with privatization. The worst case Rohter could find was a man who “earns just under $950 a month … [but] his nearly 24 years of contributions will finance a 20-year annuity paying only $315 a month.” This Chilean’s inflation-protected annuity will replace a third of his final salary after fewer than 24 years, whereas our Social Security now replaces only 28 percent of higher salaries after 35 years and will be unable to replace 30 percent of middling salaries after 2042.
Rohter also writes that “critics respond that the privatized system has been less successful in ensuring a dignified retirement for the elderly.” Less successful than what? The Chilean economy was a basket case before privatization and an “economic miracle” ever since. The average payroll tax on Chilean firms fell from 30 percent to 5 percent in six years, according to Jonathan Gruber of MIT, and annual economic growth jumped from about 3 percent to 7 percent a year.
Seven: Social Security is, for most people, a good deal. “For workers with low lifetime household earnings,” says the CBO, “total Social Security benefits received over a lifetime exceed dedicated taxes paid over a lifetime, on average. The opposite is true for workers with average and above-average earnings.” Except for those with low lifetime earnings (which often means working the fewest years rather than working at lower pay), Social Security is a very bad deal indeed. At the end of last year, the average monthly retirement benefit was $911 a month and the maximum benefit was only twice that amount. Two million aged people who never paid enough Social Security tax to qualify were collecting $428 a month from the Supplemental Security Income program. In other words, if you paid the maximum Social Security tax for 40 years or more, your benefit will be only about four times larger than if you never paid a dime.
Between all the misconceptions and the outright disinformation, reforming Social Security will be an uphill fight. But future workers and retirees deserve our best effort to stay focused on the facts.