It seems so reasonable, particularly to many Democrats. To solve the long-term shortfall in Social Security, why not tax all earnings instead of just the first $97,500? Wouldn't taxing pay above that level be the economically progressive and fiscally responsible way to solve Social Security's problems?
Prodded by Tim Russert at the Democratic presidential debate in New Hampshire on September 26 -- and again in Philadelphia on October 30 -- several of the candidates agreed that it was the thing to do.
Joe Biden endorsed the proposal without qualification. Barack Obama said lifting the cap was probably the best step. John Edwards also supported the idea, while saying he would apply the increase only to those making more than $200,000.
None of those candidates challenged the premises of Russert's question that Social Security faces so dire a prognosis that only a big tax increase or cut in benefits can solve the problem. And when Hillary Clinton refused to endorse any specific remedy -- insisting that she wanted a bipartisan commission to address Social Security's long-term future -- the other candidates and the media pounced on her for being evasive.
Let's think carefully about this. Do Social Security's problems need to be addressed now by a tax increase? How many tax increases can the next president expect to get through Congress? And what would be the impact of taxing all earnings on the long-term political viability of Social Security?
Social Security does not face an urgent crisis. It will be solvent through 2041 even under the dismal 1.8 percent economic growth rate assumed by the Social Security actuaries under the most-cited "intermediate" projection for the program. According to that scenario, Social Security will have a deficit equal to 2 percent of payroll (or .7 percent of gross national product) over the full 75-year period that the actuaries attempt to forecast.
These forecasts are not facts in the usual sense of that word. They depend on guesses not just about economic growth, but about inflation, birth and death rates, immigration, and many other unknowns -- all the way out to the year 2082. We're not certain what most of these variables will be next year. But a lot of people have become persuaded that the very height of responsibility is to promise to raise taxes next year on the basis of speculative projections of what will happen decades from now.
Let us suppose that we do nothing, the forecasts come true, and Social Security starts to run short in the 2040s. During the following decades, the federal government would have to dedicate 13 percent of its general revenue to pay Social Security benefits. This is not our most pressing worry.
Given the political resistance to any proposals for increasing taxes, a new president will have to consider carefully whether the first priority for new revenue in 2009 ought to be solving the speculative budgetary problems of the 2040s. A new administration might want, for example, to do something about the 47 million Americans who now don't have any health insurance. Or it might want to make productive investments in education, research, and infrastructure that could help keep the economy growing at more than 1.8 percent a year.
And consider this: Eliminating the cap on taxable earnings in Social Security will change the relationship to the program of people in the income brackets over $97,500, most of whom are not to be confused with plutocrats. Social Security remains a good deal for most of them because it provides income in old age that, unlike private pensions, is indexed for inflation. Even for affluent seniors, this is a form of inflation-risk protection that serves a valuable role in a portfolio of retirement income.
But take off the cap on taxable earnings, and many upper-middle-income people will decide the program is no longer in their interest. If you want to revive the movement toward privatization, eliminating the cap on taxable earnings would be just about the best thing you could do.
From the beginning, Social Security has been built on an ingenious political compromise that's been the secret of its durability. Overall -- that is, considering both how it takes in revenue and how it pays out benefits -- the program has a strongly progressive effect on the distribution of income. No other social policy has been so successful in reducing poverty. But because taxable earnings are capped, the program has avoided generating a tax revolt.
Since Franklin Delano Roosevelt got the program enacted, most Democratic leaders have understood that the cap was crucial for Social Security's political support and survival. Apparently, however, Biden, Obama, and Edwards don't get it.
I'm not suggesting there couldn't be some tinkering with the cap, but we already have a method for annually adjusting it. In fact, the cap went up from $94,200 to $97,500 in the past year under a formula established in the 1980s that sets the limit on taxable earnings on the basis of changes in average wages.
One of the reasons for the long-term projected shortfall in Social Security revenue is that this formula no longer works as intended. When it was adopted, the aim was to set the cap on taxable earnings so that the tax would apply to 90 percent of total earnings in the United States. But because earnings have become increasingly concentrated at the top, the share subject to Social Security taxes has fallen to 83 percent.
In other words, one way to solve Social Security's long-term problems is not only to grow the economy faster than 1.8 percent, but to adopt policies that ensure more of the growth goes to people in the middle class.
But suppose the distribution of income continues to be sharply unequal. Former Social Security Commissioner Robert Ball proposes to change the formula -- slowly, over a period of years -- so that the cap would rise and again apply to 90 percent of total earnings. That would close about one-third of the long-term shortfall in payroll tax revenues. Ball would close the other two-thirds of the gap by earmarking a reduced estate tax for Social Security and investing a portion of the Social Security trust funds in equities rather than Treasury bonds.
The difficulty with Ball's proposal, however, is that getting the cap up to the level where Social Security taxes would apply to 90 percent of earnings is going to hit a lot of upper-middle-income earners pretty hard. As Ball acknowledges, there's also a danger here of overshooting the mark and over-funding Social Security if, for example, the economy grows at more than 1.8 percent annually (which is well below historical levels).
My own view is that there is no need to close 100 percent of the 75-year actuarial gap. Rather than raise payroll taxes, we could adopt the latter two of Ball's proposals (earmarking the estate tax and investing some of the Trust Funds collectively in equities), which would cut the projected deficit by two-thirds. A new administration ought to save any proposed tax increases for more urgent problems.
Senator Clinton seems to understand that endorsing a payroll tax increase wouldn't just be a problem in the general election. It would be a problem if she got elected. A new Democratic administration is sure to get into trouble if it starts raising taxes for purposes that show no benefit to the voters. And it makes sense to insist that any proposal for addressing Social Security's long-term future come out of a bipartisan commission. That's how we've dealt with the issue in the past. It's not brave for the Democrats to put themselves on record for a payroll tax increase -- it's just plain dumb.