The United States does less to redistribute income than virtually any other economically “advanced” democracy. So why does class conflict loom so much larger in U.S. public opinion about government spending than in other affluent democracies? The answer may have something to do with our peculiar system of taxation.

The claim that America is riven by class conflict may come as a surprise to people who like to think that “There are no classes in America,” as Rick Santorum put it during his 2012 presidential campaign. But the fact is that rich and poor Americans disagree about government spending to an extent virtually unmatched elsewhere in the world.

In 2006, just before the onset of the Great Recession, the International Social Survey Programme (ISSP) asked people in 33 countries about “some things the government might do for the economy.” In the United States, 63 percent of the respondents favored (or strongly favored) “cuts in government spending” to boost the economy, while only 13 percent opposed (or strongly opposed) such cuts. But that was not so unusual; in 15 other affluent democracies, an average of 57 percent of the respondents favored cuts in government spending.

What is much more remarkable about the pattern of opinion in the United States is the extent to which it was polarized along class lines. In the other affluent democracies, net support for spending cuts was virtually constant across income groups, from the very poor to the very affluent. In the United States , however, poor people were only slightly more likely to favor than to oppose spending cuts, while affluent people were vastly more likely to favor spending cuts. No other rich country even came close to matching this level of class polarization in budget-cutting preferences. Among the poorer countries included in the ISSP survey, only one displayed a larger division of opinion between rich and poor — South Africa.

What accounts for the remarkable enthusiasm for government budget-cutting among affluent Americans? Presumably not the sheer magnitude of redistribution in the United States, which is modest by world standards. And presumably not a traditional aversion to government in American political culture, since less affluent Americans are exposed to the same political culture as those who are more prosperous. A more likely suspect is the entanglement of class and race in America, which magnifies aversion to redistribution among many affluent white Americans. Another is the “hidden” nature of the American welfare state, which funnels subsidies to affluent people indirectly through tax breaks on mortgages and health insurance rather than providing them with public housing and free clinics.

The U.S. tax system is also quite different from most affluent countries’ in its heavy reliance on progressive income taxes. The political implications of this difference are magnified by the remarkable salience of income taxes in Americans’ thinking about taxes and government. Matthew Yglesias chides the Heritage Foundation for harping on the steep progressivity of federal income taxes while ignoring payroll taxes and state and local taxes, which make the overall U.S. tax system much less progressive. But it’s not just the Heritage Foundation.

Income taxes seem to dominate public discussion of taxes and tax policy. For example, years of dramatic political confrontation culminated in a grudging agreement to shave a few percentage points off the Bush tax cuts for incomes over $400,000 per year; meanwhile, a major reduction in the payroll taxes paid by millions of ordinary working Americans expired with barely a whimper. Thinking about the cost of government solely in terms of federal income taxes makes it tempting for affluent Americans to see politics as a battle between “makers” and “takers” — between them and Mitt Romney’s famous “47 percent … who are dependent on government.”

Some readers of Martin Gilens and Benjamin Page’s bombshell research on economic elite domination have questioned whether the outsized political influence of affluent people really matters, since they agree with ordinary Americans on most political issues anyway. Most issues, but by no means all—as Gilens’s previous work handsomely demonstrates. The remarkable enthusiasm of affluent Americans for government budget-cutting is a prime example of what is at stake in economic elite domination of the U.S. political system.

(The affluent democracies included in my comparison were those with GDP per capita in excess of $25,000 in 2006 — Australia, Canada, Denmark, Finland, France, Germany, Great Britain, Ireland, Japan, the Netherlands, New Zealand, Norway, Spain, Sweden, and Switzerland. The other countries in the 2006 ISSP study included Chile, Croatia, the Czech Republic, the Dominican Republic, Hungary, Israel, Latvia, the Philippines, Poland, Portugal, Russia, Slovenia, South Africa, South Korea, Taiwan, Uruguay, and Venezuela. The levels of net support for government spending cuts shown in the figures are calibrated to range from −100 for strong opposition to +100 for strong support; thus, the positive values indicate net support for spending cuts all across the income spectrum in the U.S. and elsewhere. The levels of net unmet demand for social welfare spending are calibrated to range from −100 for respondents who want to spend much less on each program to +100 for those who want to spend much more; so positive values indicate net support for spending increases.)