Making Sense of Income Inequality
Friday, October 26, 2007
Former Labor Department chief economist DIANA FURCHTGOTT-ROTH explains why a snapshot view does not reflect Americans’ true well-being.
Income inequality has emerged as a major theme in the 2008 presidential campaign, particular among Democrats. John Edwards frequently talks about disparities between the “two Americas.” Just last week, a press release from Hillary Clinton declared, “The latest IRS data reveals an American economy that is leaving more and more middle class families behind. Widening income inequality has reached levels not seen since 1929.”
Has income inequality really increased? If so, does it matter? Or are there factors unrelated to income that determine a sense of well-being?
Between 1995 and 2005, inequality as measured by the Gini coefficient remained remarkably stable, according to data published by the U.S. Census Bureau. Yet some academic studies—most notably those by Thomas Piketty of the Paris School of Economics and Emmanuel Saez of the University of California, Berkeley—found that inequality was actually increasing, especially when analyzing the top 1 percent of income earners over a longer period of time. As with most economic comparisons, the years chosen as bookends make a crucial difference in the outcome.
Either way, it’s hard to get a reliable measure of inequality from income alone because income measures are crude. Moreover, they ignore other aspects of well-being. Stated incomes omit non-cash benefits such as healthcare, employer pension contributions, vacations, and sick leave, which have become an increasingly important part of worker compensation.
Income data also do not account for taxes paid, which reduce income more at the upper end of the distribution, or for transfer payments, which increase income at the bottom. Such transfers include Medicaid, Medicare, food stamps, housing vouchers, and the Earned Income Tax Credit.
Further, income-tax regulations have changed, and, as a result, so has tax reporting of income. After the 1986 Tax Reform Act became law, some business owners switched to filing under the newly-lowered individual tax rates, rather than under the corporate rates. Thus, what seems to be a post-1986 increase in taxable income among the upper brackets merely reflects a change in filing practices. This makes long-term comparisons of income inequality even more difficult and uncertain.
Our progressive tax system is designed to redistribute income from high earners to those with little or no income—and, in that regard, it is largely successful. The top half of income earners paid nearly 97 percent of all federal income taxes in 2005, which accounted for roughly 14 percent of their gross income. Meanwhile, the bottom half received most of the transfer payments and paid only 3 percent of their gross income in taxes. So when taxes and transfer payments are included in measurements of income, observed inequality is reduced.
It is reduced further when measured by how much people spend. Contrary to popular belief, Census data show that people are more equal in terms of consumption today than they were in 1986. For that matter, inequality is simply a way to assess well-being, and well-being also depends on such variables as good health, economic security, freedom of choice, and general happiness.
It’s hard to get a reliable measure of inequality from income data alone.
One reason that the top quintile of households collects more income is that these households tend to have more full-time earners. Census data show that the top quintile has two income earners per household, whereas the bottom quintile has about one earner for every two households. This means there are more than four times as many full-time workers in the top fifth of the income distribution as there are in the bottom fifth.
Women now earn well over half of all B.A. and M.A. degrees, plus half of all medical and law degrees. With higher numbers of well-educated women in the workforce, marriage often combines two medium-earning single-person households into one high-earning two-person household. Such demographic changes have increased both the number and relative wealth of two-earner couples.
Some 54 percent of households in the top quintile have two earners, and another 21 percent have three or more (usually when young adults live with their parents and work, or when teenagers get part-time jobs). By contrast, only 2.6 percent of households in the lowest quintile have multiple earners.
People move among income groups as they grow older and advance in their careers, so a snapshot view of income statistics does not reflect Americans’ true well-being. Given the reality of income mobility, some low-income households arguably are not poor. They may be students who have yet to enter the workforce and whose earnings will rise, or perhaps retired workers who are living off accumulated earnings. (The average age of those in the lowest income quintile is 54, which suggests many are retirees.) A full 30 percent of low-income households own their homes free of debt, compared with only 17 percent in the top quintile.
To be sure, some in the lowest quintile are genuinely poor. Some have less education. Others come from single-parent households with half as much income as some two-parent households. The rise in single-parent households has indeed amplified inequality, since single parents are more likely to be low-income. Others in the lowest quintile include newly-arrived immigrants, many of whom replace earlier immigrants who have moved up the income ladder.
Even when using only cash-income measures, over the past 25 years, more families have moved to upper-income brackets. In 1980, fewer than 40 percent of American families made over $50,000 per year in 2005 dollars. A quarter century later, 46 percent of families did. On average, households in all income brackets had more spending power in 2005 than they did in 1985. Households in the lowest quintile spent 8 percent more in real terms per person, while households at the top spent 10 percent more and households in the middle spent 6 percent more. This was especially visible in one highly discretionary category: entertainment. Households in the lowest quintile spent 22 percent more on entertainment in 2005 than they did in 1985. Those in the highest quintile spent 12 percent more on entertainment over the same period.
Today, countries with the fastest economic growth—notably the United States, Canada, and the Great Britain—have wider wage distributions, and hence more measured inequality, than countries with slower growth, such as Italy, Germany, and France. Some suggest that the United States could achieve more aggregate equality with higher taxes and more transfer payments. But this approach would slow economic growth and diminish job opportunities across the board—which would eventually reduce total income. As the 2008 presidential campaign heats up, and as the calls for addressing income inequality grow louder, Americans should keep that tradeoff in mind.
Diana Furchtgott-Roth is a senior fellow at the Hudson Institute and a former chief economist at the U.S. Department of Labor.
Image by Darren Wamboldt.