Missing corporate income and the question of U.S. tax progressivity
The New York Times earlier this week published a column by Steven Rattner, a Wall Street financier and former Obama administration official, that looks at all the ways in which inequality as commonly understood might be understated in the United States. Rattner analyzes a variety of measures that show the relatively low U.S. tax progressivity compared to other rich economies. He cites data from the accounting firm KPMG, but other data sources including the Congressional Budget Office show the U.S. tax system remains progressive but less so than is generally assumed.
Yet a recently released paper argues that widely accepted economic models of our tax system—models like CBO’s that policymakers will rely on when making any legislative changes to our tax code— actually overstate U.S. tax progressivity. What’s causing analysts to overstate this progressivity? According to Patrick Driessen of Bloomberg Government, the culprit is the treatment of corporate income in economists’ favored models.
Thes models used by the Congressional Budget Office and the Joint Committee on Taxation were originally built to understand the distributional effects of changes in tax policy at the individual level. CBO and JCT built these models with an eye on the need to evaluate reductions or increases in individual income taxes. This means the corporate income tax became the “neglected-but-younger-and-complicated sibling,” as Driessen puts it.
The standard models—a type known as capital-gain-realization, or CGR models—measure corporate income tax by looking at how much capital gains were realized by individuals and then imputing those gains back to corporations. But Driessen argues this method has big flaws because it misses a number of sources of corporate income, most of all deferred foreign earnings.
Because the United States doesn’t tax corporate income earned overseas until it returns to the United States, corporations have increasingly kept earnings oversees. According to Driessen, foreign deferral was about $50 billion 25 years ago. Now it’s $400 billion. The CGR models don’t capture this income and therefore doesn’t account for it in the calculation of tax rates.
Driessen presents another way of calculating corporate incomes—one that accounts for foreign deferral and other holes in the CGR models. His result is this—the U.S. corporate income tax base is increased by $1 trillion and the corporate tax rate falls to 14 percent. In contrast, the CGR models calculate a corporate tax rate of 41 percent from a corporate income tax base of about $500 billion. This 14 percent tax rate is close to other recent estimates of the effective tax rate paid by corporations.
The result of all these adjustments by Driessen is that the tax rate on U.S. corporate income is much lower than the standard distributional models would have you believe. The progressivity of the corporate income tax depends on part on who bears most of its cost: capital or labor. This is a contested issue within economics. Some economists and analysts argue that labor actually pays most of any corporate tax because corporations pass on the cost to labor in the form of lower compensation. But, at least in the range he considers, Driessen finds that the incidence doesn’t affect progressivity much. And regardless of your view of the effects the corporate income tax on society, we can all agree that it would be best to account for all the income being taxed in any distributional model. On that front, we all owe Driessen thanks.