New Poverty Figures Offer Lessons for Grant Makers

The United States Census Bureau last month unveiled its long-awaited new approach to measuring poverty. The result of two decades’ worth of analysis and discussion, the new measure seeks to provide a more realistic estimate of how many Americans are poor by taking into account additional types of income and expenses.

Although the change in approach did not produce much difference in the estimate of the number of people in poverty, it does alter our understanding in important ways of who is poor. Grant makers and nonprofit groups seeking to help the needy ought to consider how the new findings affect their work. In particular, they will see that they might achieve the best results by helping people who face temporary setbacks, such as the loss of a job or an illness.

The new Census Bureau measure is meant to supplement (though not replace) the “official” poverty line, developed in the early 1960s by Mollie Orshansky, a federal statistician, who based her approach on what it costs to buy enough food for a minimally nutritious diet. Using a 1955 Agriculture Department survey that showed families at the time spent one-third of their budgets on food, Ms. Orshansky tripled the cost of the bare-bones plan to arrive at an estimate of how much money people needed for basic necessities.

It was not long before criticisms of the measure began. Although it was adjusted for family size and composition, some people believed it ought to reflect other sorts of differences as well, especially the wide range in prices in urban and rural areas. While the poverty line was adjusted to account for inflation, it did not reflect changes in particular kinds of expenses, such as housing or medical care, which often rose more rapidly. Nor did it count new items, such as automobiles or television sets, which were not as likely to be part of family budgets in the 1950s. Instead, the Orshansky poverty line effectively locked into place the consumption patterns from more than five decades ago to set the threshold for being poor.

It also relied on a definition of resources that quickly became unrealistic as well: cash income before tax payments or credits.

Soon after the poverty measure was introduced, the federal government started to expand financial assistance to the needy. Programs that provided noncash benefits, such as food stamps, became more generous, and others, such as Medicare and Medicaid, were enacted. Eventually, the taxes low-income workers paid, such as Social Security, rose too. Early in the 1970s, the earned-income tax credit became law, giving aid through the tax system to families with low earnings.

As a result, the actual amounts Americans in poverty had to live on increasingly differed from how much cash they had, especially after tax payments and credits were figured in.

The Census Bureau’s new measure tries to respond to the criticisms of the old one. It takes into account new surveys of spending and geographic variations in the cost of necessities, such as housing.

It also includes noncash benefits, tax payments, and tax credits, as well as estimates of work expenses and out-of-pocket medical bills, to determine the real resources available to low-income people. Not least important, it broadens the definition of a “family” to include both related and unrelated children and adults who share the same address.

For 2010, these changes raised the poverty line for a household with two adults and two children to just above $24,000, a $2,000 increase over the figure calculated the old way. But because of the more realistic measure of income used, the percentage of the population classified as poor rose by less than one percentage point, from 15.2 to 16 percent. Over 49 million Americans were counted as poor in 2010, no small number but just three million more than the number arrived at for the same year using the lower Orshansky threshold.

However, the new method did produce some noteworthy differences in who is considered poor. Under the old measure, 22.5 percent of children under 18 and 9 percent of adults over 65 were in poverty. Using the new one, the poverty rate for children declined to 18.2 percent but rose for the elderly to 15 percent. Blacks, Midwesterners, and Southerners were also less likely to be poor than they were with the Orshansky poverty line.

In addition, the Census Bureau’s new measure shows that the poor are less poor than previously thought. For most demographic groups, in fact, the number in “extreme poverty”—with resources below half of the poverty line—fell. On the other hand, it found far more people just below the poverty line or just above it.

Benefits from the earned-income tax credit, food stamps, and other federal aid were chiefly responsible for helping Americans avoid “extreme poverty.” But taxes, work expenses, and out-of-pocket medical costs, especially for the elderly, had the opposite effect, placing more Americans in poverty or at risk of becoming poor.

These findings should not come as a surprise to anyone familiar with other federal studies.

Since the 1980s, for example, the Survey of Income and Program Participation has been collecting data on individual, family, and household resources, including those from government-assistance programs. It has found that a large percentage of Americans move in and out of poverty regularly. Only a relatively small proportion— 2.2 percent in the most recent study—remained consistently poor from 2004 to 2007.

For grant makers and nonprofit groups trying to reduce poverty, the implication of this finding—reinforced by the new approach to counting the poor— is that identifying ways to assist those on the margins of poverty could make a difference.

Although government programs to provide income to the poor have succeeded in lowering the number of people in deep poverty, they have been less effective at offsetting the economic, social, and personal factors that cause people to become temporarily poor. Some contend that these programs even create barriers to upward mobility.

By paying more attention to people who have had at least a little success, philanthropy may be able to help bring down the unacceptably high rates of poverty that have long existed in the United States.

Such a strategy need not mean ignoring the plight of the very poor. But with a new measure of poverty should come a new understanding of it, one that recognizes how the poor really live, not how they measure up to a standard established nearly 50 years ago.