Defining poverty

Measure by measure

The world’s richest country tries to count its poor

MOST people have an inherent sense of what it means to be poor. But choosing a definition is much trickier. Is poverty an absolute or relative condition? What is a decent standard of living? Such questions have dogged America’s social scientists for decades. This month the Census Bureau published a preliminary estimate of poverty, using a new definition. It was 16 years in the making. But it is not quite finished yet.
Poverty means different things in different countries. In Europe, the poor are those whose income falls below 60% of the median. Britain uses three measures: one relative, one absolute and a broader indicator of material deprivation, such as whether a child can celebrate his birthday. The concept of poverty becomes even more slippery when attempting international comparisons. The United Nations’ “human- development index” assesses countries across a range of indicators, such as schooling and life expectancy.
America’s official poverty measure is far simpler. Developed in the 1960s, the poverty threshold represents the basic cost of food for a household, multiplied by three. A family is judged to be poor if its pre-tax income falls below this threshold. But the official measure provides only a blurry picture. Food spending has become a flimsy reference point—in 2009 groceries accounted for just 7.8% of Americans’ spending. The poverty indicator does not account for programmes that help the poor, such as the earned-income tax credit, nor does it adjust for regional variations in the cost of living. In 1995 the National Academy of Sciences recommended changing the measure, but only now is a new one close to being established.

The “supplemental poverty measure” (SPM) will not replace the official one, which is used to determine eligibility for government programmes. Rather, census officials hope the new indicator will provide a better understanding of America’s poor, by measuring both the needs of families and the effect of government help. The SPM estimates the cost of food, clothing, shelter and utilities, then adds a further 20% for other expenses. This threshold is adjusted for the cost of living in different regions and for whether a family owns or rents its home. To assess a household’s ability to pay for basic expenses, the SPM counts cash income as well as food stamps, tax credits and other government support, minus tax payments, work expenses and out-of-pocket medical costs.

Final figures are due to be published in the autumn, but preliminary results were released this month. In 2009 15.7% of Americans were poor, compared with 14.5% in the official measure (see chart). The share of those in extreme poverty fell, relative to the official measure, thanks to the inclusion of government support. The poverty rate dropped in rural areas and rose in urban and suburban ones. It jumped in the north-east and the West, while staying almost level in the South and falling in the Midwest. The most dramatic rise was for the elderly—from 9.9% in the official measure to 16.1% in the SPM, in part because of their high medical expenses.

Timothy Smeeding of the University of Wisconsin, long a critic of the old measure, says that the SPM is a massive improvement. Some conservatives, however, are horrified. Most objectionable, according to Robert Rector of the Heritage Foundation, is that the new measure pegs household expenses at the 33rd percentile of American spending. This, he argues, makes the SPM a relative measure, rather than an absolute one. “It measures inequality,” Mr Rector insists, adding that it will help advance a misguided anti-poverty agenda.

For all the time spent developing the SPM, it is still a work in progress. Though first official numbers are supposed to be published in the autumn, even this is uncertain. Surveys must be expanded to collect additional data. The project needs about $7.5m, and a newly conservative Congress may be reluctant to provide it.

---