Skip Main Navigation Skip To Navigation Content

Current Population Survey (CPS)

Skip top of page navigation
You are here: Census.gov People and Households Current Population Survey (CPS) MainData › Poverty Thresholds

Poverty Thresholds

Poverty status is determined by dividing family income by a poverty threshold. If a family's income is less than 100% of the family's poverty threshold, it is in poverty. The Poverty Status - Alternative gives you the ability to define family income (see Income Definition), select from a variety of poverty thresholds, and determine what income-to-poverty percent to use to divide families into groups below this percent and those at or above this percent.

Disclaimer:

While the results of tabulations based on alternative assumptions about poverty thresholds and income definitions may be conceptually the same as published estimates, in most cases they will not exactly match published estimates. The reason is that the Table Creator is based on the CPS public use file, which lacks some of the detailed income information, topcodes several categories of income, and supresses some geographic identifiers in order to protect survey confidentiality.

This page describes the poverty threshold options and the income-to-poverty ratio percent cutoff. In addition to the official poverty thresholds, there are two thresholds based on the Consumer Expenditure Survey using the recommendations of the 1995 National Academy of Science Panel on Poverty and Family Assistance and several options for defining poverty thresholds based on relative income.

  • Official Poverty Threshold
    These are the thresholds that are used to define poverty in the official U.S. poverty statistics published annually by the Census Bureau. See http://www.census.gov/hhes/www/poverty/threshld.html for poverty thresholds over time.
  • Three-Parameter Poverty Threshold
    In several reports on alternative income and poverty measures, the Census Bureau has utilized a modified set of poverty thresholds using a 3-parameter scale to adjust for differences in family size. The advantage of the 3-parameter scale is that it utilizes a somewhat more systematic relationship between the thresholds for families and individuals by characteristic (poverty thresholds vary by size of family, number of children, and in the case of individuals and 2-person families, by age of householder). See Appendix B of http://www.census.gov/prod/2007pubs/p60-232.pdf for more details on the three-parameter scale and a comparison of between 2005 official thresholds and those based on the three-parameter scale.
  • CE-Based Threshold
    In its report on improving the official U.S. poverty measure, the National Academy of Sciences (NAS) recommended a new way of deriving poverty thresholds (see http://www.nap.edu/readingroom/books/poverty/summary.html  [an error occurred while processing this directive], Recommendation 2.1). The panel recommended using Consumer Expenditure Survey data to define thresholds. Specifically, they recommended using a specified percentage of median spending on food, clothing, and shelter, along with a specified multiplier to cover other needs. For information on the Census Bureau's derivation of these thresholds, see http://www.census.gov/prod/2001pubs/p60-216.pdf. It should be noted that the panel's report highlighted the importance between consistency between needs (as defined by thresholds) and resources to be compared to those needs to define poverty status. Thus, care should be taken when using these thresholds to combine them with an income definition that is conceptually consistent. To replicate Census Bureau published tables, this threshold should be used with the predefined “NAS income measure which subtracts medical out-of-pocket expenditures”. See http://www.census.gov/prod/2001pubs/p60-216.pdffor background and discussion of these topics.
  • CE-Based Threshold with Medical Out-of-Pocket Expenditures in the Threshold
    An alternative threshold based on the NAS recommendations uses data from the Consumer Expenditure survey to define thresholds for food, clothing, shelter and miscellaneous expenditures but adds to this amount an imputed estimate of expected medical out-of-pocket expenditures (MOOP). This approach increases the poverty threshold to take a family's potential MOOP expenses into account, instead of subtracting their actual expenses from income. Using data from the 1996 Medical Expenditures Panel Survey, the Census Bureau estimated expected MOOP based on the family's size, the presence of elderly family members, the self-reported health status of the family members, and differences in health insurance coverage across families. To replicate Census Bureau published tables, this threshold should be compared to the pre-defined NAS income measure.
  • Relative Income Thresholds
    The table creator allows users to select one of three equivalence scales for use with relative income poverty measures. For more information on the equivalence scales described below, see http://www.oecd.org/dataoecd/61/52/35411111.pdf  [an error occurred while processing this directive].
  • OECD Equivalence Scale (1) Threshold
    Compute equivalence scale factor as sum across all household members of the following:
    • 1.0 (for the householder of any age)
    • 0.5 (for each additional member age 14 and over)
    • 0.3 (for each additional member under age 14)

    This scale is sometimes referred to as the "OECD-modified scale" and was adopted for use by the Statistical Office of the European Union (EUROSTAT) in the late 1990's.
  • OECD Equivalence Scale (2) Threshold
    Compute equivalence scale factor as sum across all household members of the following:
    • 1.0 (for the householder of any age)
    • 0.7 (for each additional member age 18 and over)
    • 0.5 (for each additional member under age 18)

    This scale, also called the "Oxford Scale," was used by EUROSTAT in the late 1890's and early 1990's.
  • Luxembourg Income Study Equivalence Scale Threshold
    Compute equivalence scale factor as the square root of the number of household members. This implies, for example, that a household of four persons has needs twice as large as one composed of a single person.

There are two price adjustment options which apply to either the official poverty threshold or the three-parameter poverty threshold(these do NOT apply to the relative income thresholds or the CE-based thresholds )

  • CPI-U
    Every year, official poverty thresholds are updated based on changes to the Consumer Price Index for Urban Consumers, or CPI-U. For more information, see http://www.bls.gov/cpi/.
  • CPI-U-RS
    Over time, the Bureau of Labor Statistics (BLS) has modified and updated its various methods of constructing the CPI-U. The CPI-U-RS represents the best attempt by BLS to present a time series that reflects current methodology, including improvements BLS has made to the CPI-U methodology over time. The net result of these changes is to lower the measured rate of increases in the cost of living over time, which in turn results in lower poverty thresholds. See page 14 in http://www.census.gov/prod/2007pubs/p60-232.pdf for a discussion of the CPI-U and CPI-U-RS as they relate to measuring poverty. For a detailed discussion of the CPI-U-RS, see http://www.bls.gov/cpi/cpirsdc.htm.

There are two geographic price difference adjustment options (these do NOT apply to relative income thresholds)

  • No geographic price adjustments
    Official poverty thresholds are not adjusted to reflect differences in the cost of living throughout the country.
  • Geographic price adjustments
    The NAS panel recommended that poverty thresholds reflect differences in the cost of living across geographic areas. This option uses an adjustment methodology based on relative housing costs. For more information on the adjustment methodology, see http://www.census.gov/prod/2001pubs/p60-216.pdf.

Income Base for Relative Income Thresholds: the choices are "median" or "mean". Though the median is typically used for relative income poverty calculations (for example, OECD uses 50 percent of the equivalized median as a basis for its relative poverty definition), the mean is included as an alternate choice. The default proportion of the income-to-poverty ratio percent cutoff is 100%. You can enter any other integer from zero to 999. For relative poverty computations, this field determines the percent of the median (or mean) to be used as the relative poverty cut-off. For example, choosing "40" and "median" will result in statistics showing the number or percent under 40 percent of the equivalized median household income.

The default proportion of the income-to-poverty ratio percent cutoff is 100%. You can enter any other integer from zero to 999. For relative income poverty computations, this field determines the percent of the median (or mean) to be used as the relative income poverty cut-off. For example, choosing "40" and "median" will result in statistics showing the number or percent under 40 percent of the equivalized median household income.


[an error occurred while processing this directive] [an error occurred while processing this directive]
Source: U.S. Census Bureau | Current Population Survey (CPS) |  Last Revised: