Thinking about inequality before taxes and transfers

Trying to disentangle the causes and the current state of income inequality in the United States can be a tricky enterprise. Take what economists refer to as the distribution of disposable income, or income after accounting for the effects of taxes and government transfer programs such as unemployment insurance and government pensions. By that measure, the United States has the highest level of income inequality among a group of 19 high-income countries, including Germany, the United Kingdom, and Norway.

But does the United States have such high disposable income inequality because its distribution of market income, or income earned through a job or from holding a financial asset, was much higher or because taxes and government transfer programs were doing less than other countries?

For a while, the evidence indicated that the second answer was correct. Data from LIS, an organization that collects international data on household income, show that the distribution of market income in the United States is in the same neighborhood as other high-income countries. The United States lags behind in efforts to alleviate income inequality through taxes and transfers.

But a new reevaluation of that data actually shows the United States should look more toward the market distribution of income. Janet Gornick and Branko Milanovic of the LIS Center looked at the data again, but made one important adjustment. They only looked at the distribution of income (both market and disposable) for households with workers under the age of 60. In the United States, older workers stay in the labor force longer and draw much more of their income from labor earnings than older workers in other high-income countries. A 62 year old in the United States has to work longer and earn more labor income than a pensioner living in a European country who can rely upon government programs. Older households in European countries end up looking much poorer in terms of market income and that in turn makes the overall market income inequality in those countries looks much higher.

As a result, when Gornick and Milanovic restrict their analysis to households with people under the age of 60, the United States becomes a starker leader in the inequality in terms of market incomes. What’s more, the two researchers find that the inequality reduction due to tax and transfer redistribution in the United States is in line with the rest of the high-income countries.

What does this mean going forward? As Paul Krugman points out, the policy implications are quite strong. Policymakers in the United States should focus more on policies concerned with the market distribution of income, or what Yale political scientist Jacob Hacker has called “predistribution.” Some of these potential polices were outlined in a column by Eduardo Porter in The New York Times earlier this week. He cites several ideas, including those floated in a new book by Oxford University economist Anthony B. Atkinson. They include higher minimum wages, stronger labor unions, and a focus on antitrust policy.

But Atkinson also suggests raising tax rates on those at the very top of the income ladder. The goal of these rate increases wouldn’t simply be higher levels of government revenue, but rather to reduce the incentives for corporate executives to bargain for higher compensation. Economists Thomas Piketty at the Paris School of Economics, Emmanuel Saez at the University of California-Berkeley, and Stephanie Stantcheva at Harvard University have made a similar argument.

While there is a distinction between market income and disposable income, we have to remember that tax-and-transfer programs also have an effect on market incomes. And not just at the top of the income ladder as Atkinson and Piketty, Saez, and Stantcheva point out. The more generous retirement programs in European economies are responsible for fewer oldsters remaining in the workforce and that, as discussed above, shapes the market income distribution.

The paper by Gornick and Milanovic and other evidence points us toward thinking more about the distribution of market income than redistribution, but we must recognize that policies aimed at the latter can affect the former as well.