Lies, Damned Lies, Poverty Statistics

David Brady. Photo: Jon Gardiner

Each August, we Americans tell ourselves a lie. That's when the U.S. Census Bureau releases the official poverty rates for the year. The rates for 2004, made public on August 30, prompted pundits, politicians, the press, and the president to engage in their annual rehearsal of empty remarks on why poverty was higher than last year. As usual, they attributed this failure to things that really have nothing to do with poverty's true causes. In my research and in a freshman seminar on poverty, I explore the nature of poverty in affluent democracies. Central to this inquiry is a deep skepticism for the flawed official poverty statistics. My students and I have scrutinized the official statistics and concluded that the entire annual ritual of announcing and discussing them is profoundly dishonest.

This dishonesty is not because U.S. poverty is insignificant. While 12.7 percent of the U.S., or nearly 37 million people, are "officially" poor," better estimates put those numbers closer to 18 percent, or 50 million people. As part of my seminar, I explain that U.S. poverty is nearly twice that of Canada and the U.K and about three times that of many European countries. My students are surprised to learn that the richest country in the world has the most poverty of any industrialized democracy.

The dishonesty is not the fault of government statisticians like Molly Orshansky, who unwittingly constructed the formula for the official measure in 1963. With data from 1955, Orshansky multiplied the Department of Agriculture's "low-cost food budget" by three, assuming food amounted to one-third of a family's expenses. She developed this poverty line purely for research purposes, never intended it to form a basis for formulating policy, and quickly repudiated it.

President Lyndon B. Johnson's administration substituted the "economy food plan," which was about 25 percent lower, and made it the official measure. The line was purposely set low, so that the administration could "win" the War on Poverty. The measure neglects taxes and government assistance, has been adjusted only for inflation, and, as a result, ignores the enormous changes in families since 1955. For one thing, food only amounts to about one-sixth of a family's budget today. For another, the official poverty line underestimates the cost of necessities like health care and health insurance, child care, housing, and transportation--some of which have risen in real cost (after inflation) much faster than food.

The Census Bureau, aware of these problems, presents alternatives and pushes Congress to revise the official measure. Unfortunately, it has never been revised--though one such attempt occurred in an episode of The West Wing.

In my seminar, students learn that most international poverty researchers use a relative measure--a person is poor relative to the living standards and customs of a time and place. A student in my class may raise the argument that has been brought up again and again in debates about poverty: "The U.S. poor are rich compared with people in developing countries."

But the American poor don't live in developing countries. Heck, if one plays with comparisons to previous centuries or Africa, it is easy to say that there are no poor people in the U.S. The poor are poor relative to what it takes to make ends meet and participate as citizens in contemporary U.S. communities. I try to help my students understand this point by referencing Michael Harrington, who wrote, "To have one bowl of rice in a society where all other people have half a bowl may well be a sign of achievement and intelligence.... To have five bowls of rice in a society where the majority have a decent, balanced diet is a tragedy."

As part of my seminar, I teach students how to construct state-of-the-art poverty measures, which embrace relative poverty. First, we estimate the median income after considering all taxes and assistance. Then, we define poverty as living in a household with less than 50 percent of that median. With this measure, the Luxembourg Income Study, the most sophisticated international poverty research outfit, estimates that 17 percent of the U.S. would have been poor in 2000--not the official 11.3 percent.

To see how this affects policy, the students examine the debate surrounding President George W. Bush's push to privatize Social Security. They analyze data from the Luxembourg Income Study and reconsider the rhetoric that "the U.S. has conquered elderly poverty." Sure, the elderly live more securely than in the 1960s, and Social Security certainly has reduced elderly poverty. But, according to the Luxembourg Income Study, 24.7 percent of the elderly were poor in 2000, not the official 9.9 percent.

The dishonesty of official poverty is not entirely the responsibility of politicians. Although many are guilty of an unwillingness to revise the official measure, I've never heard any politician intentionally misrepresent poverty statistics. In contrast to budget debates, for example, political commentary on poverty statistics seems quaintly sincere.

The dishonesty is really the fault of us, the American people. Only on rare occasions, as in the weeks following the devastation from Hurricane Katrina, do we even bother to acknowledge the poor. Most of the time, we contentedly believe that only a few people are poor, and those undeserving poor have themselves to blame.

This, of course, is a lie. But when we are told a lie many, many times, and obvious evidence, if we bothered to look at it, shows it is false, we are equally responsible for perpetuating the dishonesty.

— Brady is an assistant professor of sociology.