Middle class better off than previously thought

Long portrayed as stagnant, the income growth of the U.S. middle class may be more than 10 times greater than previously suggested by some economists, according to a new study at Cornell.

"When we use traditional methods to measure the income of the median American [the person whose income places him/her midway in the distribution of Americans lined up from lowest to highest by reported income] we find that it rose by as much as 37 percent between 1979 and 2007," says Richard Burkhauser, Cornell's Sarah Gibson Blanding Professor of Policy Analysis. His study is published in the March issue of National Tax Journal.

But when Burkhauser and his colleagues use the methodology developed by such economists as Thomas Piketty of the Paris School of Economics and Emmanuel Saez of the University of California-Berkeley, they find that the income of an American living in the median tax unit (the person whose 1090 form, if sent to Uncle Sam each April 15, places him or her midway in the distribution of incomes reported on those forms) increased by only 3.2 percent between 1979 and 2007 -- a seeming contradiction.

Since much of the political debate about inequality, income stagnation and economic stratification -- the 1 percent versus the 99 percent -- depends on numbers developed using Piketty- and Saez-type measures, the dramatic difference in how these alternative ways to describe what is happening to the middle American is quite disturbing, Burkhauser says. Which number is right? Both, he says, depending on what the question is.

Burkhauser says that Piketty, Saez and their colleagues use Internal Revenue Service data to measure the before-tax market income (such as labor earnings, rents, interest and dividends) of tax units to reach their income-growth conclusions; whereas he and his colleagues analyzed both this measure and alternatives using the U.S. Census Bureau's Current Population Survey.

Burkhauser's analysis shows how much the growth in median income changes when federal taxes are taken into consideration. Taxes redistribute market income from higher to lower income households. For instance, the tax reforms of the 1980s and the broadening of the Earned Income Tax Credit in the 1990s resulted in a growing share of lower-income people not paying any federal income tax and some even get returns above what they paid in taxes. Incomes below the top 1 percent have also been disproportionally boosted by the growth in government benefits such as Social Security and Unemployment Insurance. He and his fellow researchers also adjusted for household size and the value of health insurance -- factors, Burkhauser says, that more accurately reflect financial resources available to all Americans.

"When we broaden our measure to include government taxes and transfers and look at households adjusted for size, the gains of middle-class Americans are 10 times larger than previously thought," Burkhauser says. "The gains are even more when we include the value of in-kind income such as employer- and government-provided health insurance."

The study shows how estimates of how the average American has fared over the last 30 years varies depending on what researchers decide to include when measuring income. Such insight can have implications for public-policy choices targeted at specific income classes, says Burkhauser.

Burkhauser's co-authors are Jeff Larrimore of the Joint Committee of Taxation and Kosali Simon, professor of public and environmental affairs at Indiana University.