By: Scott Winship
April 30, 2013
Since the financial crisis, income inequality has been a topic of obsession in many journalistic and advocacy quarters. There is some irony in this, because the crash brought about the first reversal of inequality between "the 99 percent" and richer Americans in years, and there was much less concern about the subject during the period in which inequality was rising steadily. A very visible group of academic, policy, and media elites is convinced that rising inequality is the problem of our time, and many of them struggle to understand why the issue has not inspired an outcry from the broad middle class and poor.
The head-scratching and excuse-making was recently put on display in an unintentionally revealing way in a recent op-ed attempting to explain Americans' views toward economic inequality. In it, the authors, Ilyana Kuziemko and Stefanie Stantcheva, laid out what they believe to be a paradox: Americans care about inequality but do not want government to address it. Kuziemko and Stantcheva go on to describe research they have conducted with famed inequality scholars Emmanuel Saez and Michael Norton that they believe explains the "complicated" views of Americans. Their conclusion: rising inequality may have weakened faith in government.
However, a fair read of polling on inequality refutes the idea that Americans are self-contradictory. There is little evidence that Americans are particularly bothered by inequality; therefore, it is unsurprising that they do not want government to reduce it. On the other hand, there is ample evidence that Kuziemko, Stantcheva, Saez, and Norton believe Americans should be more bothered by inequality. But you can't always get what you want.
Kuziemko and Stantcheva (K & S) cite a December 2011 Pew Research Center poll finding that 66 percent of Americans agreed there was strong conflict between rich and poor Americans. This result is regularly interpreted as indicating the public is concerned about inequality but indicates no such thing. The question measures respondents' sense of the existence of conflict, not whether that conflict is worrisome. You may recall that in late 2011, Occupy Wall St. had finally made inroads with the general population, as the economy sputtered. Respondents to that poll accurately perceived that there was more conflict than there had been even a few months earlier. Indeed, upper-income respondents were no less likely than poor respondents to agree there was strong conflict.
The increase in perceived conflict the poll found between 2009 and 2011 looks worrisome, but 2009 was a historical low point. The level of perceived conflict at the end of 2011 was no higher than in 1992, and it dropped to 58 percent in late 2012. That level was the same as in 1987 and just a bit higher than in 2000. K & S do not show that concern for inequality is high, let alone rising.
K & S cite research by political scientist Larry Bartels to argue that most people understand inequality is rising and that most of those believing it has risen disapprove. That is true, but it is worth emphasizing that 45 percent of those who believed inequality has increased did not call it "a bad thing," and even among those who did believe it was bad, we don't know how many felt particularly strongly in their view.
In early 2011 Gallup polling that asked for an open-ended response to the question of what is America's most important problem, just one percent said inequality, well below pressing issues like "lack of respect for each other" and "foreign aid," to name just two that outranked inequality. Viewed against this evidence, it is unsurprising that the Gallup poll they cite from later in the year indicated only a minority of Americans thought it was important for the federal government to reduce inequality.
In short, there is little evidence that "Americans are deeply troubled about the current level of income inequality," as K & S put it. In fact, one suspects that they are recent converts to this talking point. You can envision them, along with Saez and Norton, initially hypothesizing that it must be that Americans don't understand how much inequality we have and how harmful it is. If we could educate them, they would surely demand more government intervention. Hence, the research study they designed that randomly gives a "tutorial" on income inequality to some participants but not others and that then considers how policy preferences differ between the two groups.
It turned out that the tutorial had little effect on policy preferences toward redistribution, despite it being designed to make inequality look more harmful and redistribution less so than the evidence has established. But the tutorial did reduce respondents' trust in government. From this narrow result, they conclude that rising inequality may have reduced faith in government. So Americans' preferences for government inaction on inequality, it turns out, are not because they are ill-informed about the subject. And if they are well-informed, it can't be that they simply don't care about inequality, right? So Americans, they conclude, are "troubled by" inequality but they need to be convinced "that their government is up to the task of addressing it."
There is more to be said about the study itself, which I will take up in a second essay.