The Measure of Our Poverty

How poor have we become?

On Tuesday, the Census Bureau revealed that the poverty rate remained stuck at 15 percent last year, the same as the year before and some 2.5 percentage points higher than in 2007 - before the housing bubble burst and sent the United States careening into recession.

Critics pounced: how could a measure based on how much people spent on food in the early 1960s â€" one that ignored entirely the effect of government anti-poverty programs â€" have any relevance today?

The charge is correct, of course. But it is wrong to suggest that there are fewer poor Americans than the official count shows, or that poverty has improved much over time.

Using modern, sophisticated measures that take into account government programs as well as unavoidable expenses, like child care and out-of-pocket medical costs, the poverty rate has actually been higher over the last three years than the official statistic suggests.

The debate over the proper way to measure our deprivation points to a fundamental question: What is poverty? What do we measure it for? Are we measuring the right thing?

In the 1960s, when the government first came up with the concept, it decided you were poor when you had to spend more than a third of your cash income on what it considered the minimally acceptable diet. It made sense, sort of. Americans spent a much bigger share of their incomes on food and a lot less on other things like health care.

But this is hardly the only way to do it. Other advanced nations go about it in an entirely different way. In Europe, for instance, few countries measure a poverty line in the American sense of the minimum income needed to avoid extreme hardship.

Less tolerant of inequity than the United States, they prefer to measure deprivation in terms of people’s relative position: the poor are those who earn less than half the income of a typical citizen in the middle of the income distribution.

“They think of social isolation and cultural poverty,” said Timothy M. Smeeding, who heads the Institute for Research on Poverty at the University of Wisconsin. “If you get further and further from the median, you get socially isolated.”

By this measure, the United States doesn’t look that good: from 1995 to 2010, the poverty rate increased to 17.4 percent, from 16.7 percent. And the poverty rate here is higher than in any other advanced nation.

Source: O.E.C.D. Income Distribution Database

Indeed, comparing the two measures suggests how the notion of poverty in the United States has diverged from that in the rest of the world. In the early 1960s, the poverty line was roughly at half the median income, but as incomes have gone up, the poverty line has remained behind: today, the poverty line is at about 30 percent of the median.

Source: Timothy M. Smeeding, Jeffrey P. Thompson, Asaf Levanon and Esra Burak, “Poverty and Income Inequality Over the Early Stages of the Great Recession”

The Census Bureau is unlikely to embrace Europe’s notion of poverty anytime soon. It has acknowledged, however, that the food budget is a pretty poor marker of poverty today, when Americans spend a much smaller share of their budgets on food and more on things like child care and medical expenses.

The official poverty measure not only fails to include the effect of the earned-income tax credit or food stamps â€" the biggest anti-poverty tools the government has. It doesn’t take into account people’s housing status and housing costs, or how many people live in a household, pooling incomes.

In 2009, the census tried to fix all that. Since then it has been using all these numbers to produce a Supplemental Poverty Measure. The 2012 edition will be published in November.

The changes do make a big difference in the composition of poverty. Under the Supplemental Poverty Measure, there are far fewer children in poverty than using the official rate â€" mostly because they benefit disproportionately from government anti-poverty programs. And there are more elderly poor.

But perhaps the most interesting impact of the new measure is that after adding all the new variables, once the dust is clear we are even poorer than when we measured poverty with food. In 2012, the supplemental measure put the share of people in poverty at 16.1 percent â€" quite a bit more than the official rate of 15 percent.