With America’s top earners enjoying bigger and bigger gains each year relative to the rest of the population, most voters would probably see their post-tax-and-transfer income go up (at least in the short term) if the United States implemented a European-style welfare state. So why haven’t voters demanded more redistribution? Why, after forty years of rising economic inequality, does the American political consensus remain so market-oriented, at least relative to other industrialized countries?Two explanations are popular on the left. The first explanation—popular among Bernie Sanders-style economic populists—posits that voters actually do want more redistribution, but their democratically expressed preferences are blocked by a billionaire class that has the political class in its pocket. The second explanation—popular among elite social liberals—is that many socially conservative voters are duped into voting against their economic self-interest by Republicans’ exploitation of their latent prejudices and resentments on social issues. But the data presented in a new National Bureau of Economic Research working paper from researchers at Yale and Princeton casts serious doubt on both of these explanations.First, consider the Sanders theory. The thrust of the paper (written by Vivekinan Ashok, Ilyana Kuziemko, and Ebonya Washington) is that Americans’ support for economic redistribution has remained constant or declined since inequality began rising in the 1970s. Public support for the statement “government should reduce income differences between the rich and the poor” has trended gradually downward for the past forty years, and public support for the statements “government should do more to solve the nation’s problems” and “government should do more to help the poor” has trended sharply downward. (Support for the statement “government should ensure that everyone has a decent standard of living” has remained flat or risen modestly). So while it may be that elites do push public policy in a less redistributive direction than the voters would like, the data suggest that the decline of New Deal/Great Society liberalism over the past generation is largely the product of changing public opinion rather than the secret machinations of plutocrats.Now consider the “elite social liberal” theory that many low-to-middle-income Americans (especially working class whites) ignorantly vote against their economic self-interest because of their primitive views on race and culture wars issues. It is conventional wisdom on the left that many poor whites moved to the right economically in the post-Civil Rights era because of racial resentment. Like all matters of conventional wisdom, this theory probably has an element of truth—but it is complicated by the fact that blacks moved even more dramatically against redistribution than whites did over the forty year period addressed by the study. “While there has been no significant movement on the issue by whites, in both datasets, blacks, who have a much higher desire for redistribution on average, have significantly decreased their support,” the authors write.The authors also more-or-less explicitly tested the Thomas Frank hypothesis that cunning Republican demagoguery on God, guns, and gays leads low-information voters to support a plutocratic agenda. They control for views “on certain ‘hot-button’ issues—abortion, homosexuality and gun control” and find that “these single issues explain less than 10 percent of our trends in redistributive views by age and race.” In other words, it appears that public opinion on economic redistribution is mostly independent of public opinion on social issues.Ultimately, while the data in this study don’t provide any concrete explanations for one of the political-economic paradoxes of the past generation—that Americans have not moved in favor of wealth redistribution even as the economy delivers more and more unequal returns—they do suggest that this trend is more complicated than many on the left would like to believe. They also highlight the perils of a materialistic model of politics that assumes that people’s preferences are based on clear incentives, rather than a complex matrix of cultural and historical factors.That said, economic libertarians shouldn’t be too overjoyed about this study. It only uses data through 2012, and the fact that the public hasn’t yet responded to rising inequality by demanding more redistribution doesn’t mean that it never will. Indeed, given the Democrats’ sharp leftward shift and the rise of a less classically liberal right in Donald Trump, this seems like an increasingly distinct possibility.