Saturday, October 02, 2004

How to Save Social Security

The New York Times
October 2, 2004

How to Save Social Security

Rumors of the death of Social Security have been widely exaggerated. This year, the system's trustees reported that the fund is solvent until 2042, when it would still be able to pay about 70 percent of the promised benefits. That is not a crisis. To put the shortfall into perspective, consider that in the next 75 years - the span of time over which the trustees plan for solvency - President Bush's plan to lock in his tax cuts would cost about three times what it would take to fix Social Security. Now that's alarming.

Politicians have nearly four decades to phase in Social Security reforms. With so much time, the cost of updating Social Security can be shared by everyone, and mitigated for all.

The answer is not creating private investment accounts within Social Security - President Bush's chosen tack. (See our previous editorial, "How Not to Save Social Security" at And Senator John Kerry is not helping things any when he pledges never to cut benefits. Social Security has become such a third-rail political issue that few elected officials have the courage to be realistic about it in an election year. It's too bad, but not surprising, that neither Mr. Bush nor Mr. Kerry is choosing to present workable solutions.

Both men are right, however, in promising to protect current retirees and those who are close to retiring. Credible reform - both fair and adequate - should focus on workers who are still at least a decade away from retirement. It will require a combination of modest benefit cuts and tax increases.

What ails Social Security is well understood: there are increasingly fewer taxpaying workers to support each retiree, and retirees are living longer. But the onset of these destabilizing trends does not mean that Social Security is outdated. On the contrary, the system's adaptability is one of its great strengths.

Dozens of possible correctives - many of them eminently doable - have been proposed and endlessly debated. To us, the best reform packages are the fairest. They contain proposals that do not grasp for all of the needed revenue from high-income taxpayers, and are not overly reliant on benefit cuts for the elderly. Rather, they raise the necessary funds in ways that strive to reflect each group's fair share of Social Security's shortfall. All that is needed now is the political will to change, along the lines of the following:

Link benefits to life expectancy. The last major reform of Social Security, in 1983, increased the retirement age for full benefits to 66 from 65 starting next year, and to 67 starting in 2022.

But as people live longer, it is not feasible to simply keep pushing out the retirement age. The projected costs of increased life expectancy should be updated regularly, and the responsibility for them should be split between retirees and workers - through small automatic reductions in future benefits and modest increases in the payroll tax.

For an average 35-year-old today, for instance, the benefit cut in 2036 would be the equivalent of about $300 a year. For workers, the tax increase would start 10 years from now and rise each decade, so that by 2036, the rate would be 12.7 percent, up from 12.4 percent today, split equally between employees and employers. Taken together, those changes would close nearly one-third of Social Security's long-term financing gap.

Link life expectancy to income levels. The well-off live longer than the less affluent, and their lead is growing. That's bad news for Social Security. It means that those with high earnings not only draw the biggest retirement checks, but they also do so for a longer time, compared with everyone else. That, in turn, makes the system less progressive than it would be if life expectancy were roughly the same at all income levels. Currently, the highest-earning workers get 15 cents in benefits for every dollar of earnings at the top end of the benefit formula. Reducing that share to 10 cents over the next 25 years or so would affect only about the top 15 percent of retirees, would make the program more progressive and would close about 10 percent of Social Security's deficit.

Increase taxes - slowly. Since the 1983 reform, the rule specifying the amount of annual wages that is subject to the Social Security payroll tax - currently $87,900 - has not kept up with the income gains of the top earners. In 1983, only 10 percent of all wages escaped the payroll tax. Today, it is 15 percent. If the wage base was increased over 40 years so the amount of wages on which no payroll tax was paid was closer to the 1983 level, some 10 percent of the Social Security shortfall would disappear.

Wages above the cap should also be taxed, though not as much as the wages below the cutoff. The argument for not taxing wages above a certain level is that Social Security benefits rise only so high, no matter how much you make. But in a balanced system, the payroll tax wouldn't pay for benefits alone. It would also help pay off ongoing debt from previous generations, when retirees' benefits exceeded contributions to the system. Because high-income workers have a chunk of earnings that escape the payroll tax, they do not bear a proportional share of this burden. An additional tax of 3 to 4 percent on wages above the base, (split between employees and employers), imposed over 75 years, would make the system fairer and correct about one-third of Social Security's imbalance.

But if the well-off help pay off Social Security's generational debt, so should other workers. For starters, the four million state and local government employees who are not covered by Social Security should be brought into the system, just as federal employees were in 1983. Doing so would close 10 percent of the financing gap. The remaining 10 percent shortfall could be bridged by a tax increase that was spread among the approximately 150 million workers. The increase would come to 0.2 percentage points, if enacted immediately.

Clearly, it's possible to reform Social Security while preserving its essential character: a contract under which the young support the old via taxes, and the rich help the poor through a benefit formula that favors the neediest. It is also possible for the system to adapt while retaining its key provisions. We do not favor taking away full inflation protection because it is increasingly important over long lives and is unavailable in other retirement plans. We also do not endorse extending the length of time one must work to garner full advantage of the benefit formula, currently 35 years, because that could disproportionately harm women, who generally spend fewer years in the work force than men.

Any of the reforms we've recommended could be tweaked to cover the cost of enhancing benefits for vulnerable groups, like widows, the poor and the disabled. And if a particular proposal proved too contentious, it could be scaled back, slowed down or replaced with other measures. One intriguing idea is to keep the estate tax on the books after 2009, when it is set to expire, and dedicate the revenue to Social Security. That would affect a minuscule number of estates and would generate enough revenue to correct about a third of Social Security's imbalance.

It's time to stop fretting about Social Security. Everyone knows what needs to be done - and it's not all that drastic. We just need the will to do it.