Skip to main content

How Census gets it wrong on poverty

By Bruce D. Meyer and James X. Sullivan, Special to CNN
  • Two economists say new Census report paints grim picture of poverty in U.S.
  • But they say its way of measuring gives flawed view of who's suffering
  • They want to see things like Medicaid, tax credits, food stamps factored in
  • Writers: Method must be updated in order to design programs effective at reducing poverty

Editor's note: Bruce D. Meyer is McCormick Foundation professor at the Harris School, University of Chicago. James X. Sullivan is an associate professor of economics at the University of Notre Dame. Their joint research can be found on their Poverty Measures website.

(CNN) -- When the Census Bureau released its annual U.S. poverty report last week, the news looked grim. Poverty had risen to 14.3 percent in 2009 from 13.2 percent in 2008 -- the largest single-year increase since 1980. And there is no end in sight for those struggling to make ends meet, as unemployment has remained high throughout this year.

The general message of these latest numbers is clear: The most serious recession in a generation is taking a considerable toll on our nation's poorest families.

But in several important ways, the Census gets it wrong.

Census poverty figures are based on a narrow measure of income that often doesn't accurately reflect an individual's true economic circumstances.

That is, an individual's income may put that person below the poverty line, according to the Census measure. But that measure ignores the on-the-ground effects of some of the most critical anti-poverty weapons, most notably the Earned Income Tax Credit, Medicaid, food stamps, and housing subsidies. Recently, for example, the EITC alone provided more than $40 billion annually to the working poor.

By excluding these benefits, the Census poverty figures fail to recognize that these programs lift many people out of actual poverty.

These income-based numbers also overlook the struggles of countless others who are tightening their belts because their wealth has deteriorated. It says nothing about those who are worried about losing their jobs or facing foreclosure, and nothing about those who devote a large chunk of their paycheck to paying off medical bills or old debts. The standard of living for these folks is lower than their income would suggest.

Simply put, these poverty figures do not accurately identify those who are the worst off and fail to capture considerable long-term progress in our fight against poverty.

A better way to determine who is suffering from the current recession is to look at people's spending, which includes things like housing, food, and other goods they are able to enjoy. Preliminary data from the Bureau of Labor Statistics for consumption in 2009, like the Census figures, also indicate a rise in poverty, but tell a very different story about who is suffering most from the current recession.

Since 2007, for example, poverty measured by consumption has risen for the elderly. The Census numbers based on income, on the other hand, show no change for those 65 and older in 2008 and a sharp 8 percent dip in 2009. These numbers ignore the current strain that many elderly face as their retirement savings disappear or their home values fall.

Over the longer term, the trend is more encouraging for the elderly. In fact, there has been real progress in the war on poverty for many age groups over the past 40 years. But here, too, the Census numbers don't tell the whole story.

The official Census numbers point to a poverty rate today that is higher than it was in 1970, suggesting that low-income households have not gained at all from the economic growth of the past four decades -- a period during which incomes more than doubled, when adjusted for inflation.

There are two main reasons the official measure of poverty doesn't reflect the progress we've made. First, the Census does not count the benefits of anti-poverty programs, which have expanded sharply over the past 40 years.

Second, it accounts for inflation using the Consumer Price Index, a benchmark that is slow to incorporate new consumer products (it took 15 years to include cell phones), misses changes in the quality of goods, and doesn't fully reflect the low prices at big-box stores such as Wal-Mart.

Although these lower prices should make it easier to get out of actual poverty, their absence from the CPI causes the Census to overstate inflation and keep official poverty thresholds higher than they should be.

Correcting these flaws in poverty measurement will have only a small effect on the changes in poverty between 2008 and 2009. But small year-to-year differences can add up over time.

Since 1980, the poverty rate according to our calculations -- based on a more comprehensive measure of income, that uses a more accurate price index and that includes the Earned Income Tax Credit, food stamps, and other benefits -- has fallen by 3 percentage points more than the official measure. And poverty based on consumption fell by more than 5 percentage points during this same period, which translates into more than 15 million people moved out of poverty.

To be sure, there are still far too many families struggling to make ends meet. An accurate measure of poverty is crucial for designing programs that can reach these families and that are effective at reducing poverty.

The Census is currently revising the poverty measure, incorporating important improvements such as the inclusion of tax credits, food stamps, and other anti-poverty programs. But as an income-based measure of well-being, this still falls short.

Keeping track of those in need will be particularly important as the fallout from the economic crisis adds to the ranks of the impoverished.

The opinions expressed in this commentary are solely those of Bruce D. Meyer and James X. Sullivan.