Conventional, one-dimensional policies will not reverse U.S. income inequality growth
A new paper from Brookings highlights why the magnitude of inequality makes even a big move on taxes seem small
A recent paper from The Brookings Institution examines the effect that raising the top marginal U.S. personal income tax rate to 50 percent, and some variations on that theme, would have on income inequality. The authors, economists Bill Gale of Brookings, Melissa Kearney of the University of Maryland, and Peter Orszag, a non-resident fellow at Brookings and a vice chairman at Citigroup Inc., conclude that the effect on inequality would be “exceedingly modest.” They end their paper by suggesting that such results mean that we need to look outside tax policy if we want to address inequality. I take a different lesson from their analysis. What I see is that if one really wants to reverse the rise in inequality then the economic interventions will have to be very large, whether the lever is taxes or any other area.
Gale, Kearney, and Orszag don’t publish the change in the effective rate on the top 1 percent of their simulated tax change, but it is certainly in the single digits. In other words, it reduces the after-tax income for this top group by a few percent. Given that the gap in average income between the best-off 1 percent and the middle 20 percent has risen by over 300 percent since 1979, it isn’t surprising that taking 4 or 5 percent of the top group’s income through a tax hike isn’t going to put much of a dent in inequality. Gale, Kearney and Orszag characterize their modeled tax increase as “significant.” But that begs the question: “significant relative to what?”
The problem here isn’t necessarily the size of the dent, it’s the size of what they’re trying to put a dent in. Income inequality has gotten to the point where getting back to the distribution of income that existed in 1979 would require transferring more than $1 trillion from the top income groups to the rest of the income spectrum (a fact the authors allude to in a follow-up piece). To accomplish this just through the tax system would be a significant undertaking. The effective rate on the top 1 percent would need to rise by 40 percentage points—a cool tripling of their income tax burden—with all that revenue redistributed to the bottom 95 percent. Even getting back to the 1989 income distribution would require a 25 point increase in the top 1 percent effective rate. Getting to 1999 would require a 13-percentage point increase.
It’s easy to dismiss these tax increases as unrealistic. But using any other single policy faces the same hurdle. It may be more difficult to calculate what would be needed in terms of improved education, or building infrastructure, or profit sharing incentives, or Wall Street regulation, or job training, or apprenticeship programs to substantially reverse the rise in inequality over recent decades. But if the brute force method of redistributing through the tax system would require gargantuan change, then none of these other more indirect methods are likely to achieve results any easier.
Of course, there may be non-tax approaches that re-jigger the U.S. economy in a way to target the benefits of growth differently that would build on themselves over time more so than just after-tax redistribution—any of the list above might qualify. But given the magnitude of the undertaking, it’s hard to see how that could be successful with anything but a set of interventions that are, in total, very dramatic.
Bottom line? I’ve long thought that the solutions being offered to address income inequality in the United States are not at the scale needed to actually reverse the trends of recent decades. This little exercise convinces me of it.
—Michael Ettlinger is the director of the University of New Hampshire’s Carsey School of Public Policy and previously a senior director at The Pew Charitable Trusts and vice president of economic policy at the Center for American Progress.
A note on methodology
My calculations estimate what would be required to redistribute the 2007 distribution of income to match that of 1979, 1989 and 1999. I used the 2007 distribution because the Congressional Budget Office data only go to 2011, when the very top income earners were still suffering from the impact of the Great Recession. The Dow didn’t return to its pre-recession levels until 2013 and is now 20 percent higher. Also, other preliminary analyses have us approaching 2007 levels of inequality as of 2014. So 2007 is a good proxy for now. It’s worth noting that even as of 2011, the gap between the top 1 percent and the middle 20 percent of income earners had tripled since 1979, an effective tax rate increase of 26 percent on the top 1 percent would be required to get back to the 1979 distribution, which would result in $500 billion being redistributed.