By John Besl
In my October post, I reported on the latest national findings on income and poverty from the Current Population Survey. The nation’s official poverty rate increased from 14.3% in 2009 to 15.1% in 2010, while the number of people in poverty grew by 2.6 million. As grim as those numbers may sound, many researchers believe that the official poverty measure, which was developed in the 1960s, understates the true extent of deprivation in the United States.
The current measure employs a set of income thresholds stratified by family size and the age composition of family members. A family’s before-tax money income is compared to the appropriate threshold to determine that family’s poverty status - If income is below the threshold, the family and all its members are counted as poor. The official poverty thresholds represent the cost of a minimum food plan, multiplied by three, under the assumption that food expenditures represent one third of a family’s overall budget.
Momentum has been building over the past 20 years for the federal government to develop a new measure of poverty that would address a number of inadequacies in the official measure, better reflecting the costs of basic living expenses and the resources that people have to pay for them. In November, the Census Bureau delivered the goods with a new metric that’s being called the Supplemental Poverty Measure (SPM). The SPM will not replace the official poverty measure, and will not be used for program eligibility. The new measure factors in additional data elements such as taxes and tax credits, non-cash government benefits (e.g., food stamps, housing subsidies), and cost-of-living adjustments for different geographic areas.
One of the fundamental concerns with the measurement of poverty revolves around the definition of income or other family resources that can be used to meet basic living expenses. Both the official measure and SPM count the following government cash benefits as income: Social Security, Temporary Assistance to Needy Families (TANF, a.k.a., welfare payments), Supplemental Security Income, and unemployment insurance. The official measure, however, has never accounted for in-kind government benefits such as food stamps, school lunch, housing subsidies, or home energy assistance. These “near-money” benefits free up family income to spend on other needs, and their value is now being added to family resources in the SPM calculations.
Another frequently cited weakness of the official measure is the failure to account for the effect of taxes on disposable income. The SPM addresses this weakness by subtracting income tax and payroll tax from family resources, and by adding tax credits like the Earned Income Tax Credit (EITC), which reduce the amount of tax owed and can even provide refund payments to qualifying low-wage workers. Besides taxes, the SPM also factors in the impact of other nondiscretionary expenses such as child support, out-of-pocket expenditures for medical care, and work-related expenses (specifically, child care and commuting). Money that families spend for these nondiscretionary items is subtracted from gross income in SPM calculations.
A second fundamental concern in any system of poverty measurement is to define the economic unit of analysis. Under the official measure, the family unit consists of all related individuals who live at the same address. The new metric expands on this definition to include unrelated children who are under the care of the family (e.g., foster children). Cohabiting couples were a rare household type back in 1965 when the official poverty measure was instituted; they were viewed as economically independent individuals who did not pool their resources to meet household expenses. The SPM, however, makes the more reasonable assumption that cohabiting partners will pool their incomes and share expenses, and as a result, fewer of these households are counted as poor in the SPM.
Determining poverty thresholds is one of the absolutely key issues in poverty measurement, and the SPM introduces several enhancements over the official measure to expand the scope of basic living expenses and to adjust for geographic cost-of-living differences. The official measure focuses on a minimum food basket, but the new metric looks at four categories of basic needs: food, clothing, shelter, and utilities (FCSU expenses). The SPM employs five years of data on FSCU expenses from the Consumer Expenditure Survey, a recurring federal survey, to establish three sets of thresholds based on housing status. The three SPM housing categories are renters, owners with a mortgage, and owners without a mortgage.
Another huge weakness of the official poverty measure is the fact that it fails to account for geographic differences in housing costs and other essential costs of living. There is one set of poverty thresholds that’s applied to everyone, regardless of where you happen to live. The researchers who developed the SPM acknowledge that living expenses are higher in New York, Boston, and San Francisco than in the Cincinnati metropolitan area or a rural area of West Virginia. Geographic adjustments based on American Community Survey data on rent paid make the SPM a more realistic metric of poverty.
Incorporating the geographic cost-of-living adjustments and other poverty threshold changes outlined above, the new thresholds under SPM are generally higher than the official 2010 income limits. For a family of two adults and two children, the official 2010 poverty threshold is $22,113, compared to the overall SPM figure, $24,343 (not accounting for housing status). The comparable SPM thresholds for owners with mortgages, owners without mortgages, and renters are $25,018, $20,590, and $24,391, respectively.
Poverty Rate Comparisons
The Census Bureau recently published poverty rates using the official and SPM definitions for the nation as a whole and for various demographic groups. In order to standardize the comparisons somewhat, the official rates presented below will differ marginally from published rates, since unrelated individuals under age 15 are included in the universe here. Four charts with bullet points are presented next, followed by a brief summary discussion. In general, SPM rates are higher, overall and for most groups.
Figure 1. Poverty Rates by Age Group: 2010
* Includes unrelated children under age 15
Source: Current Population Survey, 2011 Annual Social and Economic Supplement
- Child poverty rates are substantially lower with the SPM, mostly due to the inclusion of tax credits and in-kind government benefits such as food stamps, school lunch, and housing subsidies that are targeted at families with children.
- Poverty rates for the older population are markedly higher under the SPM, primarily due to medical out-of-pocket expenditures that are subtracted from family income.
Figure 2. Poverty Rates by Race and Ethnicity: 2010
* Includes unrelated children under age 15
Source: Current Population Survey, 2011 Annual Social and Economic Supplement
- Using the SPM definition, Hispanics have the highest poverty rate, while poverty is highest among blacks with the official measure. This may be attributable to higher participation in government in-kind benefit programs by the black population.
Figure 3. Poverty Rates by Residence: 2010
* Includes unrelated children under age 15
Source: Current Population Survey, 2011 Annual Social and Economic Supplement
- The impact of geographic cost-of-living adjustments can be seen in the reduced poverty rate for non-metropolitan areas when SPM is used in place of the official measure.
Figure 4. Poverty Rates by Health Insurance Coverage: 2010
* Includes unrelated children under age 15
Source: Current Population Survey, 2011 Annual Social and Economic Supplement
- Poverty rates for the uninsured and the privately insured are higher using the SPM definition, since medical out-of-pocket expenses and contributions toward health insurance premiums are subtracted from family income under SPM guidelines.
Using the SPM, the number of people in poverty in the United States in 2010 was 49.1 million, an increase from the 46.6 million using the official poverty definition with a modified universe. SPM poverty rates are higher for most groups, with some notable exceptions: children, renters, blacks, and cohabiting couples, as well as people in non-metropolitan areas and those with publicly funded health insurance. Research on alternative poverty measurement is supposed to continue, and the Census Bureau plans to report both the official measure and SPM for the foreseeable future.
John, I really appreciate your insight into this matter.
Posted by: John Barth | January 03, 2012 at 01:06 PM
Very useful information specially for policy makers and stake holders! Nice job. Glen Chabi
Posted by: Glen Chabi, PhD | January 04, 2012 at 07:56 AM