By James Kwak
As the AARP says that it is open to modest cuts in Social Security benefits, it’s worthwhile asking a more fundamental question: are Social Security and Medicare programs that benefit the elderly?
The answer may seem obvious. After all, the bulk of Social Security Old Age and Survivors Insurance benefits go to people over 62, and almost all Medicare beneficiaries are over 65. So it’s often observed in passing that our long-range budget issues are the product of transfers to the elderly. For example, in Restoring Fiscal Sanity 2005, Alice Rivlin and Isabel Sawhill write, “These big programs, which benefit primarily the elderly, will drive increases in federal spending in the longer run” (p. 36). Other commentators have occasionally argued that the problem is that the elderly have become too powerful and therefore claim too large a share of government spending, especially compared to the very young.* When you add to that the frequent complaint that, by running budget deficits, we are imposing burdens on our grandchildren, this age-based inequity seems even greater.
But the problem with this framing is that “the elderly” change every year. There’s nothing inherently wrong or unfair with a program in which you pay insurance premiums while you work and collect benefits when you retire. Saying such a program benefits the elderly is like saying that life insurance doesn’t benefit the insured, only the beneficiaries: it’s true in a trivial sense, but people still want and buy life insurance anyway.
The more interesting question for any program in which you make contributions during your working years and collect benefits in retirement is whether it’s good for participants over their entire lifetimes. Currently, the Social Security payroll tax is 12.4 percent** of wage earnings up to $106,800 (that number is indexed). The benefit formula is progressive, so the more you pay in payroll taxes, the more you get back in absolute dollars, but your marginal benefit per dollar contributed declines. For middle-quintile earners born around 1975 and retiring around 2040, Social Security benefits will be about 55 percent of their immediate preretirement earnings*** (see Table 7 in this paper by Andrew Biggs and Glenn Springstead). For low earners, the replacement rate will be about 91 percent; for high earners, it will be about 38 percent.
So the real question is whether you would pay 12.4 percent of your wages for roughly 40-45 years in order to get back that distribution of replacement rates for about 20-25 years.**** There are a few different ways to think about this.
In aggregate, it is true in some sense that the people in the workforce today, plus all the people who will join the workforce in the future, will be net losers in Social Security. This is true because the first generations of beneficiaries were net winners; they received benefits that could not have been funded by their contributions. But that doesn’t mean that the median earner is worse off with Social Security. Most obviously, there is no way to replicate Social Security for yourself. You can’t find an insurance company that will sell you an annuity on Social Security’s terms in exchange for 12.4 percent of lifetime earnings, and that’s not just because Social Security exists. No insurance company could afford to offer such a product without mandatory participation, because of adverse selection; and if any insurance company does offer you such a product, you shouldn’t believe that they will still be around when it comes time to collect your benefits.
More importantly, you don’t know at age 21 whether you will be a low earner, a median earner, or a high earner, although you may have some idea. The progressive benefit formula gives you insurance against your career not working out as well as you might have hoped. Sure, if you’re a 45-year-old corporate executive making half a million dollars a year, you will be a net loser from Social Security, but that’s not the question; you can’t decide whether to buy insurance after you find out if the insurable event occurs. There are certainly some people who would opt out at age 21 if they could — notably, the scions of the rich, who don’t need old age insurance — but rationally speaking, it’s not just the people with below-median earnings potential who benefit from the existence of Social Security, it’s also a lot of the people with above-median earnings potential.
Of course, this analysis is predicated on the assumption that Social Security benefits will be there when your retire. If not, then Social Security is definitely a program that benefits the current elderly. And here it is necessary to point out that Social Security is projected to run a deficit starting around 2018.
But the key point here is that the deficit is not really that big. The 75-year deficit is equivalent to 2.22 percent of taxable payroll, which means that the program could be brought into balance through 2085 by increasing payroll taxes by 2.22 percentage points — from 12.4 percent to 14.6 percent. Alternatively, balance could be roughly achieved by eliminating the cap on earnings subject to the payroll tax and counted by the benefit formula.*****
But even without eliminating the cap on earnings, the question is: are Social Security benefits worth a 14.6 percent payroll tax? I think the answer is yes: (a) for median earners, because of the ability to use premiums during working years to pay for insurance against living too long; (b) for most people at age 21, because of the insurance against not making much money; and (c) for everyone behind the Rawlsian veil of ignorance. You might have a different answer to that question. But that’s the question: not whether we can afford to maintain benefits for “the elderly.”
(Medicare is the same issue, plus uncertainty about health care cost inflation. Because of the additional risk, the value of insurance is even higher.)
* Poverty among the elderly was an important historical factor behind the creation of both Social Security and Medicare. Since then, the programs have done a pretty good job of reducing old-age poverty, although they have by no means eliminated it.
** Actually, it’s 10.4 percent, but only for this year and next year, because of the December 2010 tax cut.
*** Average earnings over the five preretirement years. The full retirement age will be 67 by then under current law, so you can get slightly higher benefits by working for two more years.
**** Life expectancy at age 65 is currently around 19 years (see Table 22 in this CDC report), and will only be higher in 2040.
***** Today, only 83 percent of wages are subject to the payroll tax, so eliminating the cap would increase the tax base by over 20 percent — equivalent to a 2.5 percentage point increase in the payroll tax rate. However, increasing the cap would also increase future benefit payouts, although by much less because of the progressivity of the benefit formula.