Inequality: Making the Point

In the essay that Marshall Steinbaum criticizes, I discuss findings by Professors Thomas Piketty and Emmanuel Saez showing that market income inequality has returned to the peak last seen in the 1920s. In my essay I do not criticize those findings. In fact, in previous writings I have strongly defended Piketty and Saez’s landmark research against criticisms by conservative commentators.

Instead, my recent article makes two simple points. First, inspired by the recent publication of Piketty’s best-selling book, a number of people have described in a misleading way the implications of that book for the long-term trend in U.S. inequality. Many writers appear to misunderstand the limited concept of “inequality” that Piketty and Saez have estimated. Under a comprehensive income definition, inequality today is certainly below, and probably far below, its level in the 1920s. Second, contrary to a common claim advanced in many discussions of the Piketty-Saez estimates, real incomes of Americans in the middle and at the bottom of the income distribution have increased over the past 35 years. Although the growth in their incomes has fallen far short of the income gains enjoyed by the top 1%, it is simply wrong to say that all or nearly all U.S. income gains have been obtained by the top 1%.

Steinbaum is correct to highlight the main argument offered in my essay. I noted that the definition of income estimated by Piketty and Saez excludes many income items, such as government transfers, that supplement the market incomes earned by tens of millions of American families. I pointed out that government transfers constituted less than 1% of personal income in 1929 and 17% of personal income by 2012. Whereas government transfers now represent 280% of the market incomes received by Americans in the bottom one-fifth of the income distribution, they represent 0.7% of the market income received by the top 1%. Families in the middle one-fifth of the distribution receive transfers equal to about 20% of their market incomes. Clearly, government transfers go a long way toward closing the gap between the after-tax, after-transfer incomes of rich, middle-income, and poor Americans. Since transfers are many times more important in determining incomes today than was the case in the 1920s, it follows that even if market income inequality is the same today as it was in the 1920s, the inequality of the income we actually get to spend is considerably lower today.

Since the logic of this reasoning is hard to fault, Steinbaum argues that it is “missing the point on income inequality.” A fair reading of his argument is that government redistribution policy, through changes in tax schedules and public transfers, has become less redistributive than it was in the early 1980s. Even if true, this is perfectly consistent with the two main points of my essay: Overall inequality is not as high today as it was in the 1920s, contrary to a common misinterpretation of the Piketty-Saez estimates, and real incomes have improved at the bottom and in the middle of the income distribution, contrary to a common claim of many writers.

I agree with Steinbaum that tax and transfer policy can be improved to make the final distribution of income more equitable. But it is worth remembering that many of the redistribution policies both Steinbaum and I support tend to make the market distribution of income less equal. Social Security is one of the crowning achievements of the New Deal, and arguably the most popular. It has played an enormous role in closing the income gap between the nation’s elderly and nonelderly. Medicare has been equally important in making good health care affordable to the aged and disabled. Is there any doubt that these crucial pillars of the social safety net make market incomes less equal? When the aged and disabled were solely dependent on their earnings and private savings for support, a higher proportion of them worked, no matter how meager the wages they earned. When Social Security and Medicare made it feasible for aged and disabled workers to retire with dignity, many took advantage of the opportunity and stopped working. A moment’s reflection will show that the resulting trend toward earlier retirement made market incomes less equal. A rising percentage of families now support themselves with little or no market income.

Steinbaum and I agree that inequality has increased since the early 1980s, regardless of whether inequality is measured with market income alone or with income measured under a comprehensive definition that includes taxes and transfers. We also probably agree on many policies that would improve the distribution of net incomes. It seems to me foolish, however, to pretend that the only or the best measure of income is one that ignores the crucial contribution of government redistribution to family incomes. Public redistribution has succeeded in greatly reducing the gap between rich, middle class, and poor, even if our best estimates of market income inequality suggest otherwise. Steinbaum’s point seems to be that government policy could and should do more. I agree. But it is not “missing the point” to remind Americans that inequality has fallen substantially over the past century because public policy has produced a final income distribution that differs significantly from the one we would see based on market incomes alone.