Can This Graph Change Your Opinion About Income Inequality and Taxes?

A new study finds that bombarding people with stats and graphs on income inequality and taxes on the rich barely changes anybody's opinion on redistribution

Screen Shot 2013-05-01 at 9.08.39 AM.png

This graph tells you that the richest 1 percent has captured more than half of income growth during periods of economic expansion in the 21st century.

If more people knew that, would they support higher taxes on the rich?

That was the simple question behind a new study by Ilyana Kuziemko, MIchael I. Norton, Emmanuel Saez, and Stefanie Stantcheva, which found that even a crude survey of income inequality and tax data fails to persuade conservatives and low-income people to support more redistribution. I took the survey this morning and found that, if anything, the language and presentation slightly embellished the case for income inequality and encouraged participants to equate low taxes with lower growth. (You can see for yourself here.)

But it's the findings that will really surprise you.

First, the study asked me to enter my household income. At random, I picked $35,000. The study informed me that 52 percent of households earned less.

Screen Shot 2013-05-01 at 9.00.29 AM.png

Then the survey let me calculate how much a $35K household would make today if incomes had grown equally for everybody since 1980. The answer: $45,700, or 35% more.

Screen Shot 2013-05-01 at 9.01.55 AM.png

Then the survey offered a one-page explainer on the estate tax, which made a compelling (if one-sided) case that higher estate taxes "level the playingfield" between children of the wealthy and children of the middle class.

Screen Shot 2013-05-01 at 9.02.12 AM.png

Finally, I saw a graph presenting the last 100 years in three parts: (1) Low taxes and low growth between 1913 and 1933; (2) high taxes and high growth between 1933 and 1980; and (3) low taxes and uneven growth between 1980 and 2010. (Note that the graph associates, somewhat simplistically, high taxes with high growth and low taxes with low growth.)

Screen Shot 2013-05-01 at 9.02.43 AM.png
So far, I hope even liberals would admit that this survey has a viewpoint.* But the interesting thing about this study isn't what I've just showed you. It's what happened next.

The end of the survey asked the information-shocked participants a series of questions, including:

Is income inequality a problem? Indeed, participants were persuaded that it was. The survey's information closed "roughly 40 percent of the gap between self-identified liberals and conservatives on this question," the researchers found.

Do you support an expanded estate tax? Again, the participants said they did, causing the researchers to conclude that "providing information on the (small) share of estates subject to the tax dramatically increases support for the tax."

Would you support higher taxes on millionaires and higher minimum wages? Here, the participants barely budged. Support for taxing the rich increased slightly. But support for transfers to the poor did not. After the survey, low income/education respondents were *less likely* to support transfers to the poor.


Maybe it goes back to the survey's very first question: the one that tells you what percent of the country's households make less money than you. Low-income households might want to keep that percentage as high as possible. They'll oppose income redistribution policies that would hurt their standing, Kuziemko and Norton have written, out of a desire to punish those slightly below them to keep someone "beneath" them. This phenomenon is called "last-place aversion":

Last-place aversion predicts that those earning just above the minimum wage will be most likely to oppose minimum-wage increases as they would no longer have a lower-wage group beneath them, a prediction we confirm using survey data.

I would file this theory under: "Interesting but not entirely persuasive."

No matter the explanation, the survey's most compelling conclusion is that being subjected to an interactive/infographic polemic on income inequality and taxes on the rich barely changed anybody's opinion on redistribution. Commonly shared data helped more people agree on the problem. But it hardly got them closer to a common solution.

As somebody who writes about income inequality and taxes for a living, this study serves up a good morning's dosage of humility. Consider my expectations managed, forever.

*The economists admit as much themselves: "The goal of the information treatments were to provide a large shock to individuals' information about inequality and redistributive policies, rather than to provide a Ph.D-level, nuanced discussion about, say, the underlying causes of inequality or the trade-off between equality and efficiency."