http://www.census.gov/Press-Release/www/2002/cb02-124.htmlPoverty Rate Rises, Household Income Declines, Census Bureau Reports
After falling for four straight years, the nation's poverty rate rosefrom 11.3 percent in 2000 to 11.7 percent in 2001. Median household incomedeclined 2.2 percent in real terms from its 2000 level to $42,228 in 2001,according to reports released today by the Commerce Department's CensusBureau.
"Like the last year-to-year increase in poverty in 1991-1992 and thelast decrease in real household income in 1990-1991, these changescoincided with a recession," said Daniel Weinberg, chief of the CensusBureau's Housing and Household Economic Statistics Division.
The poverty rate and the number of poor increased among several population groups between 2000 and 2001, including all families,married-couple families, unrelated individuals, non-Hispanic Whites,people 18-to-64 years old and the native population.
"The real median earnings of women age 15 and older who worked full time, year-round increased for the fifth consecutive year, rising to$29,215 -- a 3.5 percent increase between 2000 and 2001," said Weinberg.Men with similar work experience did not experience a statistical changein earnings ($38,275). As a result, the female-to-male earnings ratioreached an all-time high of 0.76. The previous high was 0.74, firstrecorded in 1996.
The reports, Poverty in the United States: 2001 and Money Income in the United States: 2001, are available on the Internet. The data were gathered in the 2002 Annual Demographic Supplement to the Current Population Survey. In addition, the reports discuss experimental measures of income and poverty that account for noncash benefits (such as food stamps) and taxes (such as the Earned Income Credit) in income.
Poverty
According to the poverty report, about 1.3 million more people were poorin 2001 than in 2000 -- 32.9 million versus 31.6 million. The number ofpoor families increased from 6.4 million in 2000 (or 8.7 percent of allfamilies, a record low rate) to 6.8 million (or 9.2 percent) in 2001.
For non-Hispanic Whites, the poverty rate rose from 7.4 percent in 2000to 7.8 percent in 2001. But poverty remained at historic lows for AfricanAmericans (22.7 percent), Hispanics (21.4 percent) and Asians and PacificIslanders (10.2 percent). Among these groups, only non-Hispanic Whites (up905,000 to 15.3 million) and Hispanics (up 250,000 to 8.0 million) saw anincrease in the number of poor.
The three-year-average (1999-2001) poverty rate for American Indians andAlaska Natives was 24.5 percent, with an estimated 800,000 living inpoverty. American Indians and Alaska Natives were the only group to show adecline in their poverty rate when the two-year 2000-2001 average wascompared with the two-year 1999-2000 average. (The average was usedbecause the American Indian and Alaska Native population is relativelysmall and multiyear averages provide more reliable estimates.)
The poverty rate for the population age 18 to 64 rose from 9.6 percentin 2000 to 10.1 percent in 2001. Children under 18 continued to have ahigher poverty rate (16.3 percent) than people 18 to 64 or 65 and over; itwas unchanged from 2000.
Increases in poverty were concentrated in metropolitan areas particularly outside central cities) and in the South. The poverty ratefor people living in the suburbs rose from 7.8 percent in 2000 to 8.2percent in 2001, but did not change for those in central cities (16.5percent) or in nonmetropolitan areas (14.2 percent). The South was theonly region to have an increase in its poverty rate from 2000 to 2001. Itsrate of 13.5 percent was the highest among all regions.
Averaging 1999 to 2001, poverty rates ranged from 6.2 percent in NewHampshire (whose rate was not statistically different from Minnesota,Maryland, Connecticut and Iowa) to 18.8 percent in New Mexico (whose ratedid not differ from those in Arkansas, Mississippi and Louisiana). Basedon comparisons of 1999-2000 and 2000-2001, two states (South Carolina andUtah) showed increases in poverty, while four states -- California,Delaware, Massachusetts and Nevada --experienced declines.
The average poverty threshold for a family of four in 2001 was $18,104in annual income; compared with $14,128 for a family of three; $11,569 fora family of two; and $9,039 for unrelated individuals.
Income
"Like the increase in poverty, the decline in real median household income between 2000 and 2001 coincided with the recession that started inMarch 2001," said Weinberg. "The decline was widespread. With theexception of the Northeast, where income was unchanged, all regionsexperienced a decline, as did each of the racial groups."
For non-Hispanic Whites, median household income declined 1.3 percent,in real terms, between 2000 and 2001 to $46,305. For African Americans andAsians and Pacific Islanders, the drops were 3.4 percent (a loss of$1,025) to $29,470 and 6.4 percent (a loss of $3,678) to $53,635,respectively. The percentage decline in median household income of AfricanAmericans did not differ from that of non- ispanic Whites and Asian andPacific Islanders. The real median income of Hispanics, however, remainedunchanged at $33,565. This was the first annual decline for non-HispanicWhites and Asians and Pacific Islanders since 1990-1991 and the first forAfrican Americans since 1980-1981.
The three-year-average (1999-2001) median household income estimate forAmerican Indians and Alaska Natives was $32,116. As with the poverty data,averages were used because the American Indian and Alaska Nativepopulation is relatively small and multiyear averages provide morereliable estimates. Based on comparisons of two-year-average medians(1999-2000 versus 2000-2001), the real median household income of AmericanIndians and Alaska Natives did not change statistically.
The real median incomes of family households and of nonfamily householdsdeclined between 2000 and 2001. Family household income dropped 1.7percent (from $53,155 to $52,275) and nonfamily household income declined1.5 percent ($26,012 to $25,631). These percentage declines are notstatistically different.
Real median household income did not change in the Northeast between2000 and 2001, remaining at $45,716. It did, however, decline by 3.7percent in the Midwest (to $43,834); 2.3 percent in the West (to$45,087); and 1.4 percent in the South (to $38,904). The South has thelowest median household income of all four regions. The percentage changein household income for the West was not statistically different fromthose for the South and the Midwest. The difference between the 2001median household incomes for the Northeast and the West was notstatistically significant.
Real median household income declined by 1.6 percent for households inmetropolitan areas, falling to $45,219 between 2000 and 2001. Both thoseinside and outside the central cities of metropolitan areas experienced adecline. Households outside metropolitan areas did not experience a changebetween 2000 and 2001, remaining at $33,601.
Real per capita income was unchanged between 2000 and 2001 for the overall population ($22,851), each of the race groups and Hispanics. Itwas $26,134 for non-Hispanic Whites; $14,953 for African Americans; $24,277 for Asians and Pacific Islanders; and $13,003 for Hispanics.
Averaging 1999, 2000 and 2001, real median household income in Alaska,although not statistically different from the median incomes for Maryland,Connecticut and Minnesota, was higher than the values for the remaining 46states and the District of Columbia. Conversely, median household incomefor West Virginia, although not statistically different from the medianfor Arkansas, was lower than the incomes of the remaining 48 states andthe District of Columbia.
Based on comparisons of two-year-average medians (comparing 1999- 000with 2000-2001), real median household income rose in three states(Arizona, Massachusetts and Pennsylvania) and declined in 12 (Illinois,Indiana, Iowa, Michigan and Wisconsin in the Midwest;Alabama, Florida,Mississippi and Tennessee in the South; Maine and Vermont in theNortheast; and Washington in the West).
Various measures of income inequality differ on whether inequality changed between 2000 and 2001. For example, using the Gini Index, overallhousehold income inequality did not change on a year-to-year basis for theeighth consecutive year. However, the share of income going to the poorestfifth of households declined, from 3.6 percent to 3.5 percent. Almost allthe measures examined in the report show inequality to be above its 1999and earlier levels.
Experimental measures
The Census Bureau also produces a series of experimental estimates onhow much noncash benefits and taxes -- which are not considered in theofficial measures -- affect income and poverty. The income report shows 14experimental definitions of income.
Valuing noncash benefits and subtracting taxes also affects the estimated poverty rate. In response to a 1995 report issued by a panelfrom the National Academy of Sciences, the Census Bureau developed sixexperimental poverty measures, each of which accounts for benefits andtaxes in income but differs by how they account for health care costs andthe effect of geographic differences in the cost of living. Four of thesix experimental measures showed a significant increase in poverty between2000 and 2001, while two showed no change.
All of the alternative measures also present a different picture of whois poor than the official measure presents. Among the groups with lowerexperimental poverty rates are people in families with a femalehouseholder with no spouse present, children under 18 and AfricanAmericans.
Among the groups with higher poverty rates using the alternative measures are non-Hispanic Whites, people in married-couple families, people in male householder families, those age 65 and over and, when theeffect of geographic differences on the cost of living are taken intoaccount, people who live in areas with high housing costs (such asHispanics).
The estimates in these reports are based on the 2000, 2001 and 2002Current Population Survey Annual Demographic Supplements. These income andpoverty estimates, the first to use population estimates based on Census2000 results, also include the results of a sample expansion of 28,000households. The larger sample was designed to improve the reliability ofnational and state estimates.
Because results presented in these reports were recalculated based onthe expanded sample and Census 2000 results, they may differ frompreviously released estimates. All statements in the reports haveundergone statistical testing, and all comparisons cited are statistically significant.