Counting Every Bean

With a new president soon taking office and the country in the throes of a recession, it is no secret that government welfare programs may be on the chopping block. But this is not the first time they have faced scrutiny.

The latest official poverty rate of 12.5 percent— based on the Census Bureau’s annual income surveys—is indistinguishable from the 1970 rate of 12.6 percent. This lack of change has led politicians to question the effectiveness of resources devoted towards these programs. However, in a soon-to-be-released study, the Harris School’s Bruce D. Meyer and coauthor James X. Sullivan (University of Notre Dame) conclude that poorest Americans are actually better off today than they were 35 years ago.

“We find that it really does matter whether you use the conventional measure that the Census Bureau publishes or alternative measures,” said Meyer. “You get a very different picture of how poverty has changed over time, what the effects of government programs have been, and how poverty differs across population groups.”

Income poverty measurement is riddled with well-known flaws, including a narrow definition of income and an inflation adjustment that overstates cost of living increases. The official income measure does not account for tax credits or transfers from social support programs like Medicaid and housing assistance. In previous work, Meyer and Sullivan argue that consumption

more accurately measures poverty than pre-tax income because it reflects the resources available to a household, accounting for assets, non-monetary resources like homes and cars, and government support.

In this paper, Meyer and Sullivan examine consumption data between 1972 and 2005. Additionally, they look at poverty among specific population groups defined by age, marital status, and presence of children. “The main picture here is that things have gotten a lot better over time,” said Meyer. “So that says that some combination of growth in the economy and our government programs are working.”

Overall, consumption-based poverty measures show large declines in poverty compared to income poverty indicators. Furthermore, people living below the poverty line have seen qualityof- life improvements since 1972. But patterns do differ between demographic groups, with poverty falling faster for the elderly, but more slowly for married couples with children. With this information, Meyer said, “We might want to think differently about which groups we target for poverty reduction.”

Last summer Meyer testified at a U.S. House Ways and Means subcommittee hearing on establishing a modern poverty measure. “There’s near universal agreement in the research community that things should be changed,” he said. Given that strong consensus, Meyer believes it is “conceivable” that lawmakers might seriously consider modifying the current measure, but much will depend on the priorities of a new Congress.

“The longer term pattern is one of improvement, and that’s what we’re emphasizing rather than changes over the last couple of years,” he explained. “You can think of our research as emphasizing how people are doing relative to what their parents experienced.”

Bruce D. Meyer and James X. Sullivan, “Three Decades of Consumption and Income Poverty” (working paper), August 2008.