Child Poverty: Definition and Measurement

Child Poverty: Definition and Measurement

WHO IS POOR? Child Poverty: Definition and Measurement Kathleen S. Short, PhD From the Social, Economic, and Housing Statistics Division, US Census B...

206KB Sizes 0 Downloads 75 Views


Child Poverty: Definition and Measurement Kathleen S. Short, PhD From the Social, Economic, and Housing Statistics Division, US Census Bureau, Washington, DC The author reports no conflicts of interest. Address correspondence to Kathleen S. Short, PhD, Room 7H171, Social, Economic, and Housing Statistics Division, US Census Bureau, Washington, DC 20233 (e-mail: [email protected]).

ABSTRACT This article provides a discussion of what we mean when we refer to ‘child poverty.’ Many images come to mind when we discuss child poverty, but when we try to measure and quantify the extent of child poverty, we often use a very narrow concept. In this article a variety of poverty measures that are used in the United States are described and some of the differences between those measures are illustrated. In this article 3 measures are explored in detail: a relative measure of poverty that is used more often in an international context, the official US poverty measure, and a new supplemental poverty measure (SPM). The new measure differs from the other 2 because it takes into account noncash benefits that are provided to poor families. These include nutrition assistance such as food stamps, subsidized housing, and home energy assistance. The SPM also takes account of necessary expenses that families face, such as taxes and expenses related to work and health care. Comparing esti-


COUNTRIES AND many international statistical agencies calculate and publish measures of poverty. Policy makers use measures of poverty to understand who is poor, across groups and over time, to target scarce economic resources in the way of transfers and benefits as efficiently as possible. Thus, our poverty definitions and measures have important implications for targeting policies to improve economic well-being. Also, different measures reflect values in a society and might result in different kinds of policies to alleviate identified needs. Poverty represents a lack of necessary goods and services. All poverty measures consist of 2 parts: a measure of need, or poverty threshold, and a measure of the resources available to meet those needs. Families and individuals are poor if resources are not sufficient to meet needs. However, it is a complex task to specify what goods and services are necessary and to value available resources. There is a wide variety of poverty measures. In general, they fall into 2 broad groups on the basis of the measure of resources. The first group is referred to as income-based poverty measures. These measures specify a dollar amount (or other currency) that is considered to be the minimum amount required to buy necessary goods and services. For these types of measures, the amount needed might be

ACADEMIC PEDIATRICS Published by Elsevier Inc. on behalf of Academic Pediatric S46 Association

mates for 2012, the SPM showed lower poverty rates for children than the other 2 measures. Because noncash benefits help those in extreme poverty, there were also lower percentages of children in extreme poverty with resources below half the SPM threshold. These results suggest that 2 important measures of poverty, the relative measure used in international comparisons, and the official poverty measure, are not able to gauge the effect of government programs on the alleviation of poverty, and the SPM illustrates that noncash benefits do help families meet their basic needs.

KEYWORDS: poverty; poverty measurement; supplemental poverty measure


determined by what people spend for necessary items or it might represent a standard budget that lists the cost of necessary items as specified by experts. Another group of poverty measures uses nonmonetary resources and indicates material lack or want of necessary goods. These measures can represent the lack of specific items, such as shelter, clothing, furniture, transportation, or access to necessary services. They might also go beyond material lack to include human capabilities, such as skills and physical abilities, or self-respect in society. Often these measures take a multidimensional view of deprivation, with several indicators combined into 1 measure. A large literature shows that our perception of who is poor depends on which measures we use.1–3 Many researchers recognize that low income, consumption, or possession of goods might not exhibit a neatly specified relationship. Amartya Sen4 referred to the ‘direct’ method of measuring poverty, which meant observing the lack of basic needs, and the ‘income’ method, resulting in 2 alternative poverty concepts rather than 2 ways of measuring the same thing. The direct method identifies those whose actual consumption fails to meet the accepted conventions of

Volume 16, Number 3S April 2016


minimum needs, while the income method is after spotting those who do not have the ability to meet these needs within the behavioural constraints typical in that community.4 This article focuses on income-based poverty measures, the most commonly used measures of poverty.5,6 For these measures, the measure of need and the resources available to meet those needs are expressed in monetary terms. Three poverty measures are examined herein in detail; a relative poverty measure that is often used in international comparisons, the official poverty measure of the United States that is most commonly cited, and a new supplemental poverty measure (SPM) that has been designed to address some of the criticisms of the official poverty measure. Note that other types of measures not examined herein would provide yet another perspective. One of these important sets of measures are basic needs budgets. A basic needs budget approach, or expert budget, prescribes a set of goods and services that families need and assigns a cost to each of these. Summing these items yields a threshold that is used to determine whether or not a family has adequate resources. There are several studies that illustrate the practice of constructing basic needs budgets.7–9 Often, however, these basic needs budgets are drawn to represent ‘self-sufficiency,’ an amount that families would need to meet their needs on their own, without the aid of government benefits. Many of the studies that develop family budgets find thresholds that are approximately twice the official poverty thresholds, suggesting that the poverty measures studied herein describe a level of living that is below a selfsufficient standard.10 Although not explicitly examined herein, these studies illustrate the difficulty inherent in drawing appropriate poverty thresholds. International comparisons of poverty most often use an income-based poverty measure referred to as a ‘relative income’ poverty measure.11 This measure is typically used in developed countries and in reports that compare poverty rates in the United States with those in other countries. This measure uses information about the distribution of after-tax income and counts as poor those individuals with household income below some percentage of the median of that distribution. The poverty threshold for this measure, then, represents the central tendency of the resource distribution, and poverty rates on the basis of this measure provide information about the shape and size of the lower tail of that distribution. Figures from the Organization for Economic Cooperation and Development, using 50% of median equivalized income as the poverty threshold, compare child poverty rates across 34 member countries for 2010 and show that children in the United States do not fare well compared with other countries. According to these calculations, on average, 13% of all children were poor in 2010. However, there is wide variation across countries. Child poverty rates were below 9% in Austria and the Nordic countries, but they exceeded 20% in Chile, Israel, Mexico, Spain, Turkey, and the United States.12 Although



these figures and comparisons are often reported in the United States, this is not the poverty measure used for official purposes. In the United States, there is a designated ‘official’ measure of poverty.13 The current official poverty measure was developed in the early 1960s with only a few minor changes implemented since that time.14 The official measure consists of a set of thresholds for families of different sizes and compositions that are compared with before-tax family income to determine poverty status. At the time they were developed, the official poverty thresholds represented the cost of a minimum diet multiplied by 3 (to allow for expenditures on other goods and services). In general, most estimates of poverty prevalence and studies of the characteristics of the poverty population for the United States use the official measure of poverty. Over time, concerns about the adequacy of the official measure developed, culminating in a Congressional appropriation in 1990 for an independent scientific study of the concepts, measurement methods, and information needed for a poverty measure.15 In response, the National Academy of Sciences established the Panel on Poverty and Family Assistance, which released its report, titled Measuring Poverty: A New Approach, in the spring of 1995.6 This report recommended a new official poverty measure for the United States. The SPM generally follows the major recommendations of this expert group. In March of 2010, an Interagency Technical Working Group on Developing a Supplemental Poverty Measure listed a set of initial starting points to permit the Census Bureau, in cooperation with the Bureau of Labor Statistics, to produce a new measure of poverty.16 Their suggestions included the following: (1) The ‘SPM thresholds’ should represent a dollar amount spent on a basic set of goods that includes food, clothing, shelter, and utilities and a small additional amount to allow for other needs (eg, household supplies, personal care, nonwork-related transportation). This threshold should be adjusted to reflect the needs of different family types and geographic differences in housing costs. The threshold should be set at the mean of expenditures between the 30th and 36th percentiles of the distribution of spending on basic needs. (2) The ‘SPM resources’ should be defined as the value of cash income from all sources, plus the value of noncash benefits that are available to buy the basic bundle of goods minus necessary expenses for critical goods and services not included in the thresholds. Noncash benefits include nutrition assistance, subsidized housing, and home energy assistance. Necessary expenses that must be subtracted include income taxes, payroll taxes, childcare and other work-related expenses, child support payments to another household, and contributions toward the cost of medical care and health insurance premiums. The interagency group followed recommendations of the National Academy of Sciences Panel that medical benefits, such as Medicare and Medicaid, should not be added to income, reflecting that medical needs are not included in the thresholds.




Table 1. Poverty Measures: Official, Supplemental, and Relative Official Poverty Measure

Supplemental Poverty Measure

Measurement unit

Families and unrelated individuals

Resource measure

Gross before-tax money income

Poverty threshold

Cost of minimum food diet in 1963

Threshold adjustments

Vary by family size and composition

Updating thresholds

Consumer Price Index: all items

Families, including any coresident unrelated children who are cared for by the family (such as foster children) and any cohabitaters and their relatives, or unrelated, noncohabitating individuals Sum of cash income, plus noncash benefits that families can use to meet food, clothing, shelter and utility needs, minus taxes (or plus tax credits), minus work expenses, out-of-pocket medical expenses, and child support paid to another household The mean of expenditures on food, clothing, shelter, and utilities over all 2-child consumer units in the 30th–36th percentile range multiplied by 1.2 Three parameter equivalence scale to adjust for family size and adjusted for geographic differences in housing costs Five year moving average of expenditures on food, clothing, shelter and utilities

Relative Poverty Household

After-tax income

50% Median equivalized after-tax income (comparable to OECD calculations)

Square root of household size

Current median income

OECD indicates Organization for Economic Co-operation and Development.

The SPM extends the official poverty measure by taking account of many of the government programs designed to assist low-income families and individuals that are not included in the current official poverty measure.17 In this article the general perception of childhood poverty on the basis of the official measure is examined and compared with that of the SPM and a relative income measure of poverty. Table 1 lists the important differences in the calculation of these 3 measures. In comparing them, we can understand more fully what we mean by childhood poverty in the United States.

POVERTY ESTIMATES FOR 2012 The official ‘Orshansky’ thresholds are used for the official measure presented in this article (Table 2). The SPM thresholds used in this study are on the basis of out-ofpocket spending on food, clothing, shelter, and utilities. Separate thresholds are calculated for each of 3 housing status groups: owners with mortgages, owners without mortgages, and renters. The ‘relative income’ poverty

thresholds are on the basis of disposable income; cash income minus taxes paid. Using income concepts defined by the Canberra Group for disposable income, noncash benefits are not included as income, however, tax credits, such as the Earned Income Tax Credit are included. That threshold is $15,103 per adult equivalent for 2012 or $31,060 for a household consisting of 2 adults and 2 children. The Figure shows poverty rates for the 3 measures for the total population and 3 age groups. The percent of the population that was poor according to the official measure for 2012 was 15.1%. The SPM yields a rate of 16.0% for 2012. The poverty rate using the relative measure was 18.5%. In examining each of the age groups, there are differences in poverty rates across the measures.

Table 2. Two-Adult, 2-Child Poverty Thresholds: 2012 Measure


Official Measure Relative Income Measure Supplemental Poverty Measure Owners with a mortgage Owners without a mortgage Renters

$23,283 $31,060

Data from

$25,784 $21,400 $25,105

Figure. Poverty rates according to 3 measures: total and according to age group, 2012. SPM indicates supplemental poverty measure. *Includes unrelated individuals under the age of 15 who are not included in published official poverty statistics.


Table 3. Percent of People According to Ratio of Income/Resources to Poverty Threshold, 2012 < 50% of Threshold

100%-149% of Threshold

150% - 199% of Threshold

$400% of Threshold

200%-399% of Threshold


90% CI*


90% CI*


90% CI*


90% CI*


90% CI*


90% CI*













10.3 6.2 2.7

0.4 0.2 0.2

12.0 7.4 6.4

0.5 0.2 0.4

11.5 8.5 11.8

0.4 0.2 0.5

10.4 8.6 12.8

0.4 0.2 0.6

29.0 29.5 33.7

0.5 0.4 0.8

26.9 39.7 32.6

0.5 0.4 0.7













4.7 5.4 4.7

0.2 0.2 0.3

13.3 10.1 10.1

0.4 0.3 0.4

21.4 15.1 18.0

0.5 0.3 0.6

16.3 13.4 14.3

0.5 0.3 0.6

32.7 35.7 33.1

0.6 0.4 0.8

11.7 20.3 19.7

0.4 0.3 0.7












8.7 5.8 4.0

0.3 0.1 0.2

14.7 10.8 15.0

0.3 0.2 0.3

19.9 14.6 20.6

0.3 0.2 0.4

15.2 14.3 15.9

0.2 0.3 0.2 0.3 0.2

31.7 37.9 32.1

0.3 0.3 0.4

9.7 16.6 12.6

0.2 0.2 0.3

Est indicates estimate; CI, confidence interval; and SPM, supplemental poverty measure. Includes unrelated individuals younger than 15 years of age. *A 90 percent confidence interval is a measure of an estimate’s variability. Data from the US Census Bureau, Current Population Survey, 2013 Annual Social and Economic Supplement. For information on confidentiality protection, sampling error, nonsampling error, and definitions, see Details might not sum to totals because of rounding.


Official All people Age Younger than 18 years 18 to 64 Years 65 Years and older SPM All people Age Younger than 18 years 18 to 64 Years 65 Years and older Relative All People Age Younger than 18 years 18 to 64 Years 65 Years and older

50%-99% of Threshold




Comparing the 3 poverty measures for children, the SPM rate is lowest at 18.0%. The official poverty rate for children was 22.3% and the relative measure was 23.5%. This finding results from the inclusion of noncash benefits in the SPM resource measure, benefits often aimed specifically at families with children and not included as income in the other 2 measures. These findings suggest that neither the relative poverty measure, used in international comparisons, nor the official poverty measure, is a useful gauge for understanding the effect of many policies aimed at low-income families. The SPM, however, shows clearly that government programs help them meet their basic needs. It is important to note, however, that for each of the 3 measures, children have the highest rates of poverty across the 3 age groups shown herein. For the other 2 age groups the relative measure is the highest, followed by the SPM, with the official rate the lowest. Note the high poverty rates for the 65 years and older age group according to the relative measure, as well as the SPM, compared with the official. This partially reflects that the official thresholds are set lower for elderly households and the other 2 thresholds do not vary according to age. In general, we find that the official poverty thresholds are lower than Social Security benefits on average, so that many older individuals, who rely solely on Social Security, are not poor, but have incomes just above the official poverty line. If we move the poverty line higher, as with the relative measure, we find a large number of these individuals to be now classified as poor, explaining the high relative poverty rate for this group. The SPM thresholds are not as high as the relative thresholds, but the SPM subtracts medical out-of-pocket costs from income. Because this group has very high medical out-of-pocket costs, SPM rates exceed official rates for this group. Comparing the distribution of income allows an examination of the effects of taxes and transfers across the 3 measures and across the distribution of resources overall, not just for the poor. To make this comparison we construct income-to-poverty threshold ratios, dividing income by each measure’s respective poverty threshold. For example, a poor family has an income-to-poverty threshold ratio below 1.00. Individuals with income below half, or 0.5, of the poverty threshold can be considered to be in extreme poverty. Table 3 shows the distribution of income-topoverty threshold ratios for the 3 measures. For most groups, including the value of targeted noncash benefits in the SPM reduces the percentage of the population in the lowest category. This is especially true for children because, as noted herein, many of these benefits are aimed specifically at families with children. Using the official measure, 10.3% of children are in extreme poverty compared with 8.7% using the relative income measure, and only 4.7% using the SPM. Recall that noncash benefits are only included in the SPM and refundable tax credits are counted in the SPM and the relative poverty measure. These programs seem to be well targeted to families with children who are most in need.


The percent in the highest income category also differs across measures. The official measure had the highest percent in this category; 35.7%. The SPM and relative measures, because they are after-tax measures, have lower percentages in the category, 18.2% and 14.4%, respectively. Examining income-to-poverty threshold ratios in this way shows the effect of taxes and transfers on the distribution of income, moving families from the extremes to the middle of the income distribution.

CONCLUSIONS A wide variety of poverty measures are used to identify poor children and it is always important to understand the measures that are used for this purpose. In this article 3 measures of poverty for the year 2012 are examined. Results showed poverty rates for the official poverty measure, the SPM, and a relative measure of poverty. The SPM resulted in slightly higher poverty rates than the official measure and the relative poverty rates were the highest for the total population. The SPM showed lower poverty rates for children than the other 2 measures. Because noncash benefits and refundable tax credits help those in extreme poverty, there were lower percentages of individuals with resources below half the SPM threshold in extreme poverty and this was especially true for children. Because the SPM takes account of more government programs designed to help families meet basic needs, it is the better measure to use to understand the effects of those programs on childhood poverty and to quantify how well scarce resources are being targeted to those most in need. ACKNOWLEDGMENTS This article reports the results of research and analysis undertaken by Census Bureau staff. It has undergone a more limited review than official Census Bureau publications. This report is released to inform interested parties of research and to encourage discussion. Any views expressed are those of the authors and not necessarily those of the US Census Bureau.

REFERENCES 1. Nolan B, Whelan C. Measuring poverty using income and deprivation indicators: alternative approaches. J Eur Soc Policy. 1996;6:225–240. 2. Rector RE, Johnson KA, Youssef SE. The extent of material hardship and poverty in the United States. Rev Soc Econ. 1999;57:351–358. 3. Short K. Material and financial hardship and income-based poverty measures in the USA. J Soc Policy. 2005;34:21–38. 4. Sen AK. Issues in the measurement of poverty. Scand J Econ. 1979; 81:285–307. 5. Atkinson A, Cantillon B, Marlier E, et al. Social Indicators: The EU and Social Inclusion. Oxford: Oxford University Press; 2000. 6. Citro CF, Michael RT, eds. Measuring Poverty: A New Approach. Washington, DC: National Academy Press; 1995. 7. Dinan KA. Budgeting for Basic Needs – A Struggle for Working Families. New York, NY: National Center for Children in Poverty, Columbia University, Mailman School of Public Policy; 2009. 8. Economic Policy Institute. Gould E, Wething H, Sabadish N, Finio N. What families need to get by: The 2013 update of EPI’s family budget calculator. Economic Policy Institute Issue Brief #368, July 3, 2013. Available at: Accessed December 18, 2015.

ACADEMIC PEDIATRICS 9. Chauna Brocht J, Spade-Aguilar M. How Much Is Enough? Basic Family Budgets for Working Families. Washington, DC: Economic Policy Institute; 2000. 10. National Center for Children in Poverty. Columbia University, Mailman School of Public Health. Measuring Income and Poverty in the United States. Available at: text_707.pdf. Accessed December 18, 2015. 11. UNECE. Canberra Group Handbook on Household Income Statistics, Second Edition. Available at: id¼28894. Accessed December 18, 2015. 12. OECD. OECD Family Database. Available at: family/database. Accessed December 18, 2015. 13. DeNavas-Walt B, Proctor D, Smith JC. Income, Poverty, and Health Insurance Coverage in the United States: 2012. US Census Bureau,


14. 15.




Current Population Reports. Washington, DC: US Government Printing Office; 2013:60–245. Orshansky M. Who’s who among the poor: a demographic view of poverty. Soc Secur Bull. 1965;28:3–32. Ruggles P. Drawing the Line–Alternative Poverty Measures and Their Implications for Public Policy. Washington, DC: Urban Institute Press; 1990. US Census Bureau. Observations from the Interagency Technical Working Group on Developing a Supplemental Poverty Measure. Available at: TWGObservations.pdf. Accessed December 18, 2015. US Census Bureau. Short K. The Research Supplemental Poverty Measure: 2012. Available at: 2013pubs/p60-247.pdf. Accessed December 18, 2015.