On Wednesday, President Obama declared that “increasing inequality is most pronounced in our country, and it challenges the very essence of who we are as a people.”

But measuring inequality is not simple. Should income be measured before the government removes taxes, or after? Should income include government transfers such as food stamps, Medicare, Medicaid, unemployment benefits and housing supplements? How about wealth?

One measure of well-being is spending, and government spending data shows that the ratio of spending between the top and bottom 20 percent of society has hardly changed over the past quarter century. In terms of total spending, inequality is at the same level as 1987.

Why do other measures show increasing inequality? Some studies use measures of income before taxes are paid and before transfers, such as food stamps, Medicaid and housing allowances. Including these transfers and taxes reduces inequality.

Most studies do not take into account changes in the composition of households over the past 25 years. America has more two-earner households at the top, and more one-person households at the bottom.

Some believe the increase in measured inequality since the 1970s is due to the Tax Reform Act of 1986, which lowered top individual income tax rates from 50 percent to 28 percent, below the 35 percent corporate tax rate. After 1987, it appeared that incomes rose, but some of this was small businesses filing taxes as individuals rather than as corporations.

Quintiles differ in the number of people per household and the number of earners per household, so comparisons of quintiles are misleading. In 2012, households in the lowest fifth had an average of 1.7 people, and in half these households there were no earners. The highest fifth had 3.1 persons per household, with 2 earners.

One reason for income inequality is women moving into the workforce in record numbers, resulting in more two-earner households. If there were only one-earner households, the distribution of income would be far more even.

Another change is the shrinkage in household size at the bottom of the income scale, due to the increased longevity of today’s seniors and to the higher numbers of divorced people and single-parent households.

Spending power shows how individuals are doing over time relative to those in other income groups, and spending has not noticeably changed.

On a per-person basis, households in the top fifth of the income distribution in 2012 spent $32,054, which is 2.5 times the amount spent by the bottom quintile. That was about the same as 25 years ago. There is no increase in inequality. In addition, the overall level of inequality is remarkably small. A person moving from the bottom quintile to the top quintile can expect to increase spending by only 146 percent.

Compared with 1987, the big winners are the lowest-income group, whose expenditures increased by 12.1 percent in constant dollars. In contrast, the highest group spending per person increased by only 9.2 percent. This shows that even though the distribution of income might be wider, those at the bottom are doing better than they did 25 years ago because they have greater spending power, after adjusting for inflation.

From 1987 to 2007, all groups did better. But over the past five years, expenditures per person have declined in all groups as the recession took its toll.

Much concern over inequality can be traced to problems of measurement and changes in demographic patterns over the past quarter century. Spending, a better measure of well-being, shows remarkable stability over the past 25 years and, if anything, a narrowing, rather than an expansion, of inequality.

Examiner Columnist Diana Furchtgott-Roth (dfr@manhattan-institute.org), former chief economist at the U.S. Department of Labor, is a senior fellow and director of Economics21 at the Manhattan Institute for Policy Research.