Eakinomics: Progressive Policies and the Facts

It is the most pervasive and important narrative of our time: Over the past five decades, incomes of the lower and middle classes have stagnated, while the share of the economic pie going to the affluent has steadily expanded. It led President Obama to declare inequality as the defining issue of our time. It undergirds the aggressive tax plans of presidential candidates such as Bernie Sanders and Elizabeth Warren, who target the income and wealth of the richest of America’s rich. And it is the key empirical evidence behind the notion that the rich have captured the federal government and turned it into an agent of their ascendance.

This notion is largely founded on the work of Thomas Piketty and Emmanuel Saez (PS) (2003) and Thomas Piketty, Emmanuel Saez, and Gabriel Zucman (PSZ) (2018). PS found that the share of income going to the top 1 percent more than doubled, from 9 percent to 20 percent, between 1962 and 2015, while PSZ estimated that the real, pre-tax incomes of the bottom 50 percent has stagnated since 1979.

But what if it wasn’t true?

In important research, Gerald Auten and David Splinter demonstrate that “Top income share estimates based only on individual tax returns, such as Piketty and Saez (2003), are biased by tax-base changes, major social changes, and missing income sources. Addressing these issues requires numerous assumptions, especially for broadening income beyond that reported on tax returns. This paper shows the effects of adjusting for technical tax issues and the sensitivity to alternative assumptions for distributing missing income sources. Our results suggest that top income shares are lower than other tax-based estimates, and since the early 1960s, increasing government transfers and tax progressivity resulted in little change in after-tax top income shares.”

Similarly, “In contrast with the PSZ estimate that average real pre-tax incomes of the bottom 50 percent remained virtually unchanged, we estimate that they increased by nearly one-third. For pre-tax/after-transfer income (which includes Social Security benefits) and after-tax income, we estimate a real increase for the bottom half of the distribution of nearly two-thirds.”

How do Auten and Splinter arrive at such different conclusions? As it turns out the only way to get a really large sample of the incomes of the affluent is to use tax data. And while it may feel like a strip search to fill out your taxes, the reality is that tax data do not contain a lot of important information. As the authors put it, “Tax rules and incentives for reporting income have changed over time as the result of tax legislation. Declining marriage rates and changing household structures can lead to biased results when tax units are the unit of observation. While many adults do not file tax returns, many returns are filed by individuals under age 20, other dependents, and non- residents. Important sources of income are missing in tax data, including government transfer payments and non-taxable employer-provided benefits. The share of income missing in tax data has increased over time, such that market income on tax returns accounts for only about 60 percent of national income in recent years. In addition, there are many technical issues with respect to differences between what is reported on tax returns and what economists regard as current-year economic income. Prior studies may have been misleading as a result of failure to adequately account for these challenges.”

Objectively, the paper is dense as a brick and as dull as dishwater. But its findings shine a bright light on one of the central issues in the domestic policy debate.