INTRODUCTION

This report presents social and economic characteristics of the population below the poverty level in 1986 based on the March 1987 Current Population Survey (CPS). The poverty definition used here is that adopted for official Government use by the Office of Management and Budget and consists of a set of money income thresholds that vary by family size and composition. Families or individuals with income below their appropriate threshold are classified as below the poverty level. These poverty statistics exclude inmates of institutions, Armed Forces members in barracks, and unrelated individuals under 15 years of age. The poverty thresholds are updated every year to reflect changes in the Consumer Price Index. For example, the average poverty threshold for a family of four was $11,203 in 1986 and $10,989 in 1985. These thresholds are based on money income only and do not include the value of noncash benefits such as employerprovided health insurance, food stamps, or Medicaid. For further discussion, see the section entitled "Collection and Limitations of Poverty Data."

The data in the report consist of cross classifications of poverty status by such characteristics as age, race, sex, family relationship, educational attainment, work experience, and type of income received. Although the primary focus of these data is on the United States as a whole, some tables also show data by region and type of residence. The text under the heading "Poverty in 1986" was drawn from the advance report, Series P-60, No. 157, Money Income and Poverty Status of Families and Persons in the United States: 1986. In the text, the terms "poverty population," "poor," and "below the poverty level" are used interchangeably, as are the terms "nonpoor" and "above the poverty level." Characteristics such as age and marital status are as of the survey date (March 1987), while income, poverty, and work experience data refer to the whole previous calendar year (e.g., 1986). In the report, the reference year cited refers to the "income" year.

POVERTY IN 1986

The number of persons below the poverty level

percent.1 The number of poor in 1986 was not statistically different from the 1985 figure, but was below the recent peak of 35.3 million recorded in 1983. The 1986 poverty rate of 13.6 percent was lower than the 1985 rate of 14.0 percent at the 90-percent confidence level but not at the 95-percent level. The poverty rate was 15.2 percent in 1983.

Poverty data using the official Government definition were first tabulated for 1959. The poverty rate fell dramatically in the 1960's, decreasing from 22 percent to 12 percent in 1969. During the same period, the number of poor declined from 39.9 million to 24.1 million. During the 1970-77 period, the size of the poverty population fluctuated between 23 and 26 million, and the poverty rate was in the 11 to 13 percent range. From 1978 to 1983, the number of persons in poverty increased by 44 percent, from 24.5 to 35.3 million, and the poverty rate rose from 11 to 15 percent.

Whites represented 69 percent of the poor in 1986, Blacks represented 28 percent, and the remainder were persons of other races (principally Asians and American Indians). The poverty rate in 1986 was 11 percent for Whites, 31 percent for Blacks, and 16 percent for persons of other races. The poverty rate for persons of Hispanic origin (who may be of any race) was 27 percent in 1986. Neither the number of poor nor the poverty rate changed significantly for Whites, Blacks, or persons of other races between 1985 and 1986. Among Hispanics, the number in poverty did not change between 1985 and 1986, but their poverty rate declined from 29 percent to 27 percent.

The poverty rate for Whites has declined since 1983 (the rate in that year was 12 percent), but the 1986 rate was higher than the 1978 rate of 9 percent (1978 was a recent low-point in the poverty rate).2 Among Blacks, the 1986 poverty rate was lower than the 1983 figure of 36 percent and was at about the same level as the 1978 rate. The 1986 rate among persons of Hispanic origin was not statistically different from the 1983 rate but was well above the 1978 rate of 22 percent.

'At the 90-percent confidence level, the number of poor persons in 1986 could vary between 31.6 and 33.2 million. At the 90-percent confidence level, the poverty rate could vary between 13.3 and 13.9 percent.

2The poverty rate in 1978 was not statistically different from the

The number of poor and the poverty rates in the Northeast and Midwest Regions declined between 1985 and 1986, but there were no significant changes in the South or West. Each of the four regions had a 1986 poverty rate that was lower than the 1983 rate.

The number of poor families in 1986 was about 7 million, not significantly different from the 1985 figure. About 44 percent of poor families were composed of married-couple families, and 51 percent were composed of families with a female householder, no husband present. The number of poor married-couple families decreased from 3.4 to 3.1 million between 1985 and 1986, but the number of poor families maintained by a woman with no husband present did not change significantly.

The number of poor families has decreased since 1983, when there were about 7.6 million, as has their poverty rate (12 percent in 1983; 11 percent in 1986). Similarly, the number of married-couple families below the poverty level decreased during the 1983-86 period, from 3.8 to 3.1 million and their poverty rate dropped from 8 to 6 percent. The number of poor families with a female householder and no spouse present was about 3.6 million in both 1983 and 1986, and their poverty rate was about 35 percent in both years.

Persons in families represented 76 percent of the poor in 1986 and persons living alone or with nonrelatives represented 21 percent (persons in unrelated subfamilies made up the remainder of the poverty population). Between 1985 and 1986, the number of family members in poverty declined by about 1 million but the number of poor unrelated individuals did not change significantly. Between 1983 and 1986, the number of poor family members declined by 3.2 million, but the number of poor unrelated individuals showed no statistically significant change.

The poverty rate in 1986 was about 20 percent for related children under 18 years old in families and 12 percent for persons 65 years and over. These rates were not statistically different from the 1985 rates, but were lower than the 1983 rates (22 percent and 14 percent, respectively).

The 1986 poverty rate was higher in central cities of metropolitan areas and in nonmetropolitan areas (both about 18 percent) than in suburban areas (about 8 percent). The poverty rate in central cities declined from 19 percent in 1985, while the rates for suburban and nonmetropolitan areas did not change significantly. The number of poor in central cities declined by about 900,000 between 1985 and 1986.

One-half of poor family householders worked at some time during 1986, and 17 percent worked yearround, full-time. Neither of these proportions changed between 1985 and 1986.

In 1986, the average amount of additional money needed to raise the incomes of poor families above

their respective poverty threshold was about $4,400. For unrelated individuals the comparable figure was about $2,500.

COLLECTION AND LIMITATIONS OF POVERTY DATA

This report presents findings from the Income Supplement of the Census Bureau's Current Population Survey of March 1987. The information was collected from a sample of persons representing the civilian noninstitutional population and members of the Armed Forces living off post or with their families on post in the United States. Because the collection of income data in the survey is restricted to persons 15 years old and over, unrelated individuals under 15 are excluded from this report. In addition, poverty numbers and rates exclude inmates of institutions, and Armed Forces members living in barracks.

The poverty definition used in this report is based on an index developed at the Social Security Administration (SSA) in 1964, and revised by Federal Interagency Committees in 1969 and 1981. A directive from the Office of Management and Budget, originally issued in 1969, established the Census Bureau's statistics on poverty as the standard data series to be used by all Federal agencies.3

The poverty concept is a statistical measure based on the Department of Agriculture's 1961 Economy Food Plan. It reflects the different consumption requirements of families in relation to their size and composition, and the age of the family householder. A ratio of food expenditures to income of one-third, based on the Department of Agriculture's 1955 Survey of Food Consumption, was used to derive the original poverty thresholds from the economy food plan. The poverty thresholds have been updated annually based on changes in the Consumer Price Index. For further details, see appendix A.

Data on income collected in the CPS are limited to money income received before payments for personal income taxes and deductions for Social Security, union dues, Medicare premiums, etc. Money income is the sum of the amounts received from earnings; Social Security and public assistance payments; dividends, interest, and rent; unemployment and worker's compensation; government and private employee pensions; and other periodic income. (Certain money receipts such as capital gains are not included.) Readers should be aware that for many different reasons there is a tendency in household

3Statistical Policy Handbook, Directive 14, Office of Federal Statistical Policy and Standards, U.S. Department of Commerce, May 1978.

surveys for respondents to underreport their income. For additional information on this topic, see Appendix C, "Underreporting of Income."

Money income does not reflect the fact that many families received part of their income in the form of nonmoney transfers. Many low-income families received benefits such as food stamps, health benefits, and subsidized housing; other families received employerprovided benefits such as health insurance coverage, contributions to retirement plans, and payment for education expenses.

Beginning with the March 1980 survey, the CPS income supplement has included a group of questions on the receipt of selected noncash benefits during the previous year. The means-tested benefits covered in the March 1987 survey were food stamps, free or reduced-price school lunches, public or subsidized housing, and Medicaid. Questions were also asked about the following nonmeans-tested benefits: Medicare, CHAMPUS or other military health care, employer-provided insurance plans, and employerprovided group health insurance plans. Data on such benefits for 1985 are contained in Current Population Reports, Series P-60, No. 155, Receipt of Selected Noncash Benefits: 1985. That report shows that about 59 percent of all poor households received at least one of the following means-tested noncash benefits in 1985: food stamps, free or reduced-price school lunches, Medicaid, and publicly owned or other subsidized housing.

Technical Paper 57, Estimates of Poverty Including the Value of Noncash Benefits: 1986, focuses on five major noncash benefits: food stamps, free or reducedprice school lunches, Medicaid, Medicare, and publicly owned or subsidized housing. The paper presents two different methods for valuing these benefits: (1) market value, (2) recipient value. It is the fifth in a series of technical papers updating the estimates to 1986; Technical Papers 50, 51, 52, 55, and 56 contain estimates for the period 1979 through 1985. Since the same poverty thresholds were used as in the official definition, the resulting poverty estimates are all lower than the official estimates. The report for 1986 shows that if selected benefits had been counted as income, the official estimate of poverty would have been reduced by 8 to 34 percent, depending on the range of benefits valued and the method used.

It should be noted that there is no general agreement on how to value these benefits, which benefits should be valued or whether any noncash benefits should be valued without adjusting the poverty thresholds.4

*The Census Bureau's income concept does not assign a value to any noncash benefit, regardless of the persons' or family's income level. Benefits such as employer-provided health and life insurance, pension plans, stock options, and use of company

See Proceedings of the Conference on the Measurement of Noncash Benefits, Dec. 12-14, 1985, Vol. 1, and also Harold W. Watts, "Have Our Measures of Poverty Become Poorer?" Focus, Volume 9, No. 2, Summer 1986, University of Wisconsin-Madison, Institute for Research on Poverty.

RELATED REPORTS

Current Population Survey. A detailed explanation of the poverty concept appears in "The Measure of Poverty, A Report to Congress as Mandated by the Education Amendments of 1974," prepared by a Poverty Studies Task Force under the auspices of the Department of Health, Education, and Welfare. This report also discusses limitations of the poverty definition and presents alternative methods for measuring poverty.

A detailed description of the original poverty index developed by the Social Security Administration appears in the following articles by Mollie Orshansky published in the Social Security Bulletin, "Counting the Poor: Another Look at the Poverty Profile," January 1965, and "Who's Who Among the Poor: A Demographic View of Poverty," July 1965.

Data on persons in poverty based on the original poverty definition appear in Current Population Reports, Series P-60, No. 54, The Extent of Poverty in the United States: 1959 to 1966. Data based on the revised definition of poverty appear in later reports in the P-60 series. Poverty data are also included in some reports in the P-20 and P-23 series of the Current Population Reports. All relevant titles are listed on the inside of the back cover of this report. Data based on the Current Population Survey showing the distribution of families and persons by income levels also appear in the P-60 series; the latest report is P-60, No. 156, Money Income of Households, Families and Persons in the United States: 1985.

Detailed data on the receipt of noncash benefits in 1986, including the poverty status of recipients appear in Current Population Reports, Series P-60, No. 155, Receipt of Selected Noncash Benefits: 1985. Several approaches to the valuation of these benefits are explored in Technical Paper 57, Estimates of Poverty Including the Value of Noncash Benefits: 1986.

Decennial census. Poverty data collected in the 1980 Census are available for all States on computer files (referred to as summary tape files, or STF's) as well as in published reports. Summary Tape File 3 contains poverty data cross classified by race, age, type of family, and presence of children, for States, standard metropolitan statistical areas (SMSA's), counties, county

Various subfiles of STF3 present information for neighborhoods, Congressional districts, school districts, and zip code areas. There are three report series based on STF3: Series PHC80-3, Summary Characteristics for Governmental Units and Standard Metropolitan Statistical Areas, Series PHC80-4, Congressional Districts of the 98th Congress, and Series PHC80-S2, Advance Estimates of Social, Economic, and Housing Characteristics.

Summary Tape File 4 provides somewhat more detailed information on the age and family structure of the poverty population, and additional characteristics includ-ing receipt of Social Security and public assistance income, work experience, and ratio of income to the poverty level. All items in STF4 are repeated by race and Hispanic origin. Data are shown for States,

SMSA's, counties, county subdivisions, places, and tracts. The three report series based on STF4 are Series PC80-1-C, General Social and Economic Characteristics, Series PHC80-2, Census Tracts, and Series HC80-1-B, Detailed Housing Characteristics.

More detailed poverty data for fewer geographic areas (States, SMSA's, counties, and cities of 50,000 or more inhabitants) are presented in STF5. This file is the basis for Series PC80-1-D, Detailed Population Characteristics, and Series HC80-2, Metropolitan Housing Characteristics. Inquiries concerning availability of printed reports, computer tapes, and microfiche may be addressed to Data User Services Division, Customer Services, Bureau of the Census, Washington, DC 20233.