- Table of
- What's New
- Research &
- ASA Home
Diana M. Pearce, University of Washington
Sociologists now have choices regarding poverty measures: the official measure, the Supplementary Poverty Measure (SPM), and “basic needs” budgets. Why and how did we get these different approaches to measuring poverty?
Created almost a half-century ago, the official federal poverty measure (called the Federal Poverty Line, or FPL) was criticized almost from the start, even by its author, Molly Orshansky. These critiques culminated in the congressionally mandated 1995 National Academy of Sciences (NAS) Report, Measuring Poverty. The Census Bureau’s reports on “experimental” measures were based on its recommendations; other groups also made NAS-type calculations, such as New York City’s Council on Economic Opportunity. Rep. Jim McDermott and Sen. Chris Dodd introduced the Measuring American Poverty Act, which proposed an NAS-type measure, a “medical risk” measure, and a “basic needs” budget measure. In 2010, the Obama Administration appointed the Interagency Technical Working Group, which recommended an NAS/Census-type measure, with some modest improvements. This became the SPM, now in its third year.
A major change is how the SPM thresholds are calculated. While the FPL is based on the cost of a minimum food budget, updated only for inflation, the SPM is pegged to expenditures of families at the 33rd percentile on core essentials (housing, utilities, food, and clothing, plus a small amount for miscellaneous). Thus, the SPM thresholds rise or fall as living standards rise or fall. For this reason, the SPM is deemed to be a “quasi-relative” measure. (It is “quasi” because core essentials expenditures do not rise or fall as much as total expenditures.) Over time, however, while the FPL thresholds will not rise in real terms, the thresholds will rise as living standards rise.
Arguably, however, the SPM’s most important change is the measure of resources. With the FPL, family resources were simply pretax cash income. However, today noncash benefits clearly affect well-being, but do not count as “income.” By not reflecting the impact of these programs, many felt that the FPL “overestimated” poverty. Thus the SPM, unlike the FPL, counts the cash equivalent of benefits such as food stamps as if they were cash (but only benefits that offset the costs included in threshold, i.e., food, housing, utilities and clothing). Other “necessary” expenditures, such as health care and child care, are deducted from income, but only to the extent of actual expenditures.
The resulting SPM thresholds are modestly higher than FPL thresholds, which raises the count of the poor. However the broader definition of resources, by counting such benefits as food stamps as income (thus raising family incomes, some above the thresholds), lowers the count of the poor. The net result is only a minimal difference in the count of the poor: In 2012, the FPL counts 47 million (a 15.1 percent poverty rate) as poor compared with 49.7 million (16 percent) using the SPM.
The minimal difference was no accident. Partly this was methodological, for if the thresholds were raised substantially and changes in the definition of resources were instituted, it would be impossible to determine which change increased the poverty count. Nevertheless the choice to keep the thresholds relatively low was mainly strategic and political. It was even the subject of a West Wing episode, a fictional White House TV show, where it came down to the rhetorical question: “Do we want to be the Administration that ‘doubled’ the number of people in poverty?” In other words, to be politically acceptable, any revision of the FPL could not result in a substantial increase in the poverty count. Pegging the threshold to expenditures at the 33rd percentile achieved that goal. The SPM will rise or fall with living standards and thus not continue to fall as with the FPL, but it will also never increase significantly either.
While the SPM does not greatly change the overall count of the poor, it does change the picture of poverty among subgroups. Using the SPM, the poverty rate of the very poorest, those below 50 percent of SPM thresholds, is reduced (5.4 percent compared with 6.8 percent for the FPL), while the poverty rate of those with income/resources between 100 and 200 percent of the thresholds is higher than that with the FPL (31.8 percent vs. 18.8 percent). That is, noncash benefits are affecting the well-being of the very poorest. This is still an underestimate of these programs’ impact, however, as benefits that reduce the costs of needs not among the core essentials, such as Medicaid or child care assistance, cannot be counted because these items are not in the thresholds.
The SPM also changes the demographics of poverty. The poverty rates of the elderly increase from 8.7 percent using the FPL to 15.1 percent for the SPM (reflecting high levels of health care expenditures beyond Medicare), while it decreases for children (from 22.3 to 18.1 percent), reflecting the SPM’s counting of in-kind benefits. Reflecting geographical differences in housing costs, SPM poverty rates are higher in the Northeast and West than under the FPL and lower in the Midwest, the South, and rural areas.
The FPL is widely used not only to measure the extent of poverty, but to determine eligibility or need for programs in a given community. However, because the SPM counts the value of noncash benefits in calculating poverty rates, it cannot be used this way. It would not make sense to use the SPM to determine need when some of those needs have already been met. For this reason, the SPM will never be a replacement of the federal poverty measure but will remain, as its name suggests, supplementary.
Though in many ways an improvement, the SPM has left the most widespread critique of the FPL, that it is unrealistically low, unaddressed. In fact the FPL has fallen from 50 percent of median income (which is the commonly used relative poverty measure in Europe and elsewhere) at its inception, to less than 30 percent of median income in 2012. The SPM, being pegged at a similar level, is likewise at a very low level. Both have been deemed by some observers to now be measures of “deprivation.”
While neither the FPL nor the SPM provide realistic thresholds, a third alternative does. Following in a long tradition of family budgets, a “basic needs” measure was first proposed by economist Trudi Renwick, and its most developed form is that of the Self-Sufficiency Standard (SSS). In contrast to the FPL and the SPM, the SSS is built from the ground up, with the costs of each basic need—housing, utilities, food, child care, health care, transportation, taxes and tax credits, plus miscellaneous—determined independently and based on government sources, such as Fair Market Rents for housing (see www.selfsufficiencystandard.org.)
While the FPL is “frozen,” and cannot take account of any “new” costs such as child care or health care, and the SPM does so in a very limited way (by deducting only the actual expenditures on these items from income), only the basic needs budget measures include the costs of all basic needs, not just core essentials. This also means that the resource measure can take account of all benefits that reduce the cost of basic needs, not just food and shelter as with the SPM.
Another difference between these poverty measures is that the FPL has no geographical variation (within the continental United States), the SPM incorporates geographical variation in thresholds but only for housing costs, while basic needs budgets have by far the most detailed geographical variation and for all costs.
In short, each measure has its strengths and weaknesses—the FPL has a long history (50 years), the SPM improves and modernizes the resource measure and incorporates some geographical differences in the cost of living, and basic needs budgets provide a modern, detailed estimate of the minimum cost of living varied by place and family composition, pegged at a minimally adequate level.
For those teaching about poverty, the challenge is to help students, especially undergraduates, understand that there is a gap between what we often mean by poverty, and what we measure as poverty. A budget exercise can be an eye-opener and discussion starter. For researchers, the challenge is not to simply acknowledge the shortcomings of the FPL or SPM, and then go ahead and use a measure, but to try to address these shortcomings by using an alternative or by discussing the impact of each measure’s limitations on findings. For theorists, the quandary presented by competing poverty measures is a core sociological issue in the fields of stratification and inequality, epitomized by such questions as: What do we really mean by “poverty”? (For example, if one can afford food and shelter but not health care and child care, is one poor?)