Prospective Social Security reformers on both the Left and the Right commonly argue that a reformed Social Security program should be more progressive than the current system. The two sides differ, of course, on how to accomplish this. Yet it is an open question whether it should happen at all.

Most on the Left argue for increasing the cap on wages subject to payroll taxes, which is currently $106,800. Some argue for simply raising the cap, often to around $170,000, while others argue for eliminating it outright. Either option would increase marginal tax rates for affected individuals by over 10 percentage points. President Obama has his own plan for taxing high earners: he would apply a payroll surtax of 2 percent to 4 percent on earnings above $250,000, with no additional benefits paid in return. While less economically harmful than plans to lift the “tax max,” Obama’s proposal would fix only around 15 percent of the long-term deficit, meaning more pain in the future.

Policy makers need to consider Social Security reform as if they were designing a system from scratch: how would they decide to spread the burdens and benefits?

The Right would also make high earners pay most reform costs, only through lower benefits rather than higher taxes. President Bush’s plans for “progressive indexing” of future retirement benefits protected the bottom third of earners from any benefit changes but made significant reductions versus scheduled benefits for high earners.

But those proposing these tax and benefit changes never ask how progressive the Social Security program should be in the first place. While this is a difficult question to answer with precision, we can glean useful information by comparing Social Security’s progressivity to those of national pension systems around the world. The chart below, based on data from the Organization for Economic Cooperation and Development’s (OECD) invaluable “Pensions at a Glance” compendium, illustrates.

Each line represents the benefits paid by a national pension system to individuals by earnings levels. In general, the lines slope upward at less than a 45-degree angle; this signifies that programs are progressive, as benefits do not grow commensurately with earnings. The “flatter” a country’s line the more progressive its benefit structure, as it denotes that benefits rise more slowly as the individual’s earnings increase.

Now consider where the United States—the bold line—compares with other countries. While there are several countries with more progressive benefit structures than the United States—interestingly, most are of Anglo origin, including the United Kingdom, New Zealand, Ireland, and Canada—the majority of national pension plans are less progressive than the U.S. Social Security program.

In addition to varying benefit progressivity, countries tax different amounts of income. During his presidential campaign, then-Senator Obama argued for new taxes on high earners to ensure they pay their “fair share.” But another cross-country comparison shows that the United States already imposes pension taxes at higher levels than most of our foreign competitors. The chart below, also drawn from OECD data, shows the maximum earnings level upon which pension taxes are applied. In the United States, the current “tax max” of $106,800 is equal to around 290 percent—or 2.9 times—the average wage.

Only two countries, the Czech Republic and Portugal, apply pension taxes to all earnings, as many on the left desire. (For instance, see this testimony from the Economic Policy Institute’s John Irons before the Senate Aging Committee.) Most large industrial countries levy payroll taxes on a significantly lower earnings level than the United States: in Canada, taxes are applied only up to 96 percent of the average wage; in France, to 101 percent; Germany, 151 percent; Japan, 150 percent; and the United Kingdom, 115 percent.

Those proposing these tax and benefit changes never ask how progressive the Social Security program should be in the first place.

The broader point here is that policy makers need to think beyond merely filling Social Security’s multi-trillion-dollar shortfalls, and particularly beyond doing so merely by sticking the bill to high earners. Rather, they should consider reform as if they were designing a system from scratch: if so, how would they decide to spread the burdens and benefits of a national pension program over citizens of various earnings levels? If too progressive, the program begins to resemble “welfare,” with the associated stigma and resentment that implies. If not progressive enough, the program may fail to accomplish its social insurance goals. In either case, though, policy makers need to consider facts and data, not simply overstated claims that high earners are not paying their fair share.

The Social Security program has grown more expensive over time, and its costs are projected to rise further as the baby-boom generation retires and the population ages. However, the system has not gotten more progressive over time—the general tilt of the program toward low earners instituted during the Roosevelt administration has more or less been retained. Separating concerns about system solvency from structural reforms would lead policy makers to a more coherent and manageable system.

Andrew G. Biggs is a resident scholar at the American Enterprise Institute.

FURTHER READING: Biggs has also discussed immigration in "New Blood for Social Security," how New Jersey faces "No Garden-Variety Public Pension Crisis," and whether we should "Tax to the Max?" to pay for Social Security. Elsewhere, he and Jason Richwine say "Federal Employees Are Not Underpaid 22 Percent," Biggs writes that "Personal Accounts Are No Cure-All," and he offers "A Conservative's Take on Social Security Reform."

Image by Darren Wamboldt/Bergman Group.