Measuring Poverty

The Official federal Poverty Measure (OPM), often known as the Federal Poverty Level (FPL), was developed five decades ago and today has become increasingly problematic and outdated as a measure of income adequacy. Indeed, the Census Bureau itself states, “the official poverty measure should be interpreted as a statistical yardstick rather than as a complete description of what people and families need to live.” Despite the many limitations of the OPM, it is still used to calculate eligibility for a number of poverty and work support programs.

The most significant shortcoming of the OPM is that for most families, in most places, the poverty level is simply too low. While the Standard changes by family type to account for the increase in costs specific to the type of family member—whether this person is an adult or child, and for children, by age—the Federal Poverty Guideline (FPG) increases by a constant $4,160 for each additional family member and therefore does not adequately account for the real costs of meeting basic needs.

The Self-Sufficiency Standard shows that the income needed to meet basic needs is often far above the OPM, indicating that families can have incomes above the federal poverty measure and yet lack sufficient resources to adequately meet their basic needs. For this reason, most assistance programs use a multiple of the federal poverty measure to determine need. For instance, children’s health insurance with low-cost premiums is available through Colorado Child Health Plan Plus program for families with incomes up to 260% of the FPG.

However, simply raising the poverty level, or using a multiple of the FPG, cannot solve the structural problems inherent in the official poverty measure. In addition to the fundamental problem of being too low, there are five basic methodological problems with the federal poverty measure.

First, the measure is based on the cost of a single item—food—rather than a “market basket” of all basic needs.

Over five decades ago, when the OPM was first developed by Mollie Orshansky, food was the only budget item for which the cost of meeting a minimal standard, in this case nutrition, was known. (The Department of Agriculture had determined household food budgets based on nutritional standards.) Knowing that the average American family spent a third of their budget on food, Orshansky reasoned that multiplying the food budget by three would yield an estimate of the amount needed to meet other basic needs, and thus this became the basis of the OPM.

Second, the poverty measure’s methodology is “frozen,” not allowing for changes in the relative cost of food or non-food items, nor the addition of new necessary costs.

Since it was developed, the poverty level has only been updated annually using the Consumer Price Index. As a result, the percentage of the household budget devoted to food has remained at one-third of the OPM even though American families now spend an average of only 13% of their income on food. At the same time, other costs have risen much faster—such as health care, housing, and more recently, and energy—and new costs have arisen, such as child care and taxes. None of these changes are, or can be, reflected in the federal poverty measure based on a “frozen” methodology.

Third, the poverty measure is dated, implicitly using the demographic model of a two-parent family with a “stay-at-home” wife, or implicitly assumes she is not employed.

This family demographic no longer reflects the reality of the majority of American families today. According to the U.S. Bureau of Labor Statistics, both parents were employed in 59% of two-parent families with children in 2013. Likewise, 68% of single mothers with children were employed and 81% of single fathers with children were employed in 2013. Thus paid employment with its associated costs such as child care, transportation, and taxes is the norm for the majority of families today rather than the exception. Moreover, when the poverty measure was first developed, these employment-related items were not a significant expense for most families: taxes were relatively low and child care for families with young children was not common. However, today these expenses are substantial, and borne by most families, and thus these costs should be included in a modern poverty measure.

Fourth, the poverty measure does not vary by geographic location.

That is, the official poverty measure is the same whether one lives in Louisiana or in the San Francisco Bay Area of California (with Alaska and Hawaii the only exceptions to the rule). However, housing in the most expensive areas of the United States costs nearly four times as much as in the least expensive areas. Using the 2015 Fair Market Rents, the cost of housing (including utilities) at the 40th percentile for a two-bedroom unit in the most expensive place—the San Francisco metropolitan area—is $2,062 per month. This is nearly four times as much as the least expensive housing in the country, found in most counties in Kentucky, where two-bedroom units cost $558 per month. Even within states, costs vary considerably: in Colorado, the cost of a three-bedroom housing rental in Bent County is $801 per month, while in Park County a three-bedroom unit is $2,307 per month.

Finally, the poverty measure provides no information or means to track changes in specific costs, nor the impact of subsidies, taxes, and tax credits that reduce (or increase) these costs.

The OPM does not allow for determining how specific costs (such as housing, child care, etc.) rise or fall over time. Likewise, when assessing the impact of subsidies, taxes, and tax credits, poverty measures cannot trace the impact they have on net costs unless they are explicitly included in the measure itself.

For these and other reasons, many researchers and analysts have proposed revising the federal poverty measure. Suggested changes would reflect twenty-first century needs, incorporate geographically based differences in costs, and respond to changes over time. One such effort is the Supplemental Poverty Measure (SPM). Read more about the SPM and how it differs from the Standard.