The report, Income, Poverty, and Health Insurance Coverage in the
United States: 2003, is available on the Internet at <http://www.census.gov/hhes/www/income.html>.
The report’s data were compiled from information collected in the
2004 Annual Social and Economic Supplement (ASEC) to the Current Population
Also released today were tabulations from
the 2003 American Community Survey (ACS). The survey is the largest household
survey in the United States (800,000 housing units per year during the
test phase). Like the decennial census long form it is designed to replace,
the ACS provides information on money income and poverty, as well as a
range of other social and economic indicators. ACS data for 2003 are shown
for 116 metropolitan areas, 233 counties and 68 cities, all with populations
of 250,000 or more. Starting in 2006, the Census Bureau expects data will
be available for all areas with populations of 65,000 or more. And by
2010, data will be available down to the census tract and block group
The fact sheet, Differences Between
the Income and Poverty Estimates From the American Community Survey and
the Current Population Survey Annual Social and Economic Supplement,
provides information on the differences in concepts and purposes of the
ACS and the CPS.
- Real median income for the nation remained unchanged between 2002
and 2003 for all types of family and nonfamily households.
Race and Hispanic Origin
- Real median income did not change between 2002 and 2003 for non-Hispanic
white households (about $48,000), black households (about $30,000) or
Asian households (about $55,500).
- Households with Hispanic householders (who can be of any race) experienced
a real decline in median income of 2.6 percent between 2002 and 2003.
- Comparison of two-year moving averages (2001-2002 and 2002-2003)
showed that the real median income for households with householders
who reported American Indian and Alaska native, regardless of whether
they reported any other races, increased by 4.0 percent to $35,441.
There was no change for those who chose the single race of American
Indian and Alaska native ($32,866).
- Real median household income remained unchanged between 2002 and
2003 in three of the four census regions — Northeast ($46,742),
Midwest ($44,732) and West ($46,820). The exception was the South, where
income declined 1.5 percent. The South continued to have the lowest
median household income of all four regions ($39,823). The difference
between median household incomes in the Northeast and West was not statistically
- Native households had a real median income in 2003 ($44,347), not
different from that in 2002. Foreign-born households experienced a real
decline of 3.5 percent to $37,499.
- Real median earnings of men age 15 and older who worked full-time,
year-round in 2003 ($40,668) remained unchanged from 2002. Women with
similar work experience saw their earnings decline — 0.6 percent
to $30,724 — their first annual decline since 1995. As a result,
the ratio of female-to-male earnings for full-time, year-round workers
was 76 cents for every dollar in 2003, down from 77 cents for every
dollar in 2002.
- Income inequality showed no change between 2002 and 2003 when measured
by the Gini index. The share of aggregate income received by the lowest
household income quintile (20 percent of households) declined from 3.5
percent to 3.4 percent, while remaining unchanged for the other quintiles.
- The number of people below the official poverty thresholds numbered
35.9 million in 2003, or 1.3 million more than in 2002, for a 2003 poverty
rate of 12.5 percent. Although up from 2002, this rate is below the
average of the 1980s and 1990s.
- The poverty rate and number of families in poverty increased from
9.6 percent and 7.2 million in 2002 to 10.0 percent and 7.6 million
in 2003. The corresponding numbers for unrelated individuals in poverty
in 2003 were 20.4 percent and 9.7 million (not different from 2002).
- As defined by the Office of Management and Budget and updated for
inflation using the Consumer Price Index, the average poverty threshold
for a family of four in 2003 was $18,810; for a family of three, $14,680;
for a family of two, $12,015; and for unrelated individuals, $9,393.
Race and Hispanic Origin
- In 2003, among people who reported a single race, the poverty rate
for non-Hispanic whites was 8.2 percent, unchanged from 2002. Although
non-Hispanic whites had a lower poverty rate than other racial groups,
they accounted for 44 percent of the people in poverty.
- For blacks, neither the poverty rate nor the number in poverty changed
between 2002 and 2003. People who reported black as their only race,
for example, had a poverty rate of 24.4 percent in 2003.
- Among those who indicated Asian as their only race, 11.8 percent
were in poverty in 2003, up from 10.1 percent in 2002. The number in
poverty also rose, from 1.2 million to 1.4 million. For the population
that reported Asian, regardless of whether they also reported another
race, the rate and the number increased to 11.8 percent and 1.5 million.
- Among Hispanics, the poverty rate remained unchanged, at 22.5 percent
in 2003, while the number in poverty increased from 8.6 million in 2002
to 9.1 million in 2003.
- The poverty rate of American Indians and Alaska natives did not change
when comparing two-year averages for 2001-2002 and 2002-2003.
- The three-year average poverty rate for people who reported American
Indian and Alaska native as their only race (23.2 percent) was not different
from the rates for blacks or Hispanics. It was higher than the rate
for non-Hispanic whites who reported only one race. The three-year average
poverty rate for people who reported American Indian and Alaska native,
regardless of whether they also reported another race (20.0 percent),
was lower than the rates for blacks or Hispanics and higher than the
rate for non-Hispanic whites who reported only one race.
- For all children under 18, the poverty rate increased from 16.7 percent
in 2002 to 17.6 percent in 2003. The number in poverty rose, from 12.1
million to 12.9 million.
- Neither people 18 to 64 years old nor those age 65 and over experienced
a change in their poverty rate, 10.8 percent and 10.2 percent in 2003,
- The poverty rate for Arkansas (18.5 percent) — although not
different from the rates for New Mexico, Mississippi, Louisiana, West
Virginia and the District of Columbia — was higher than the rates
for the other 45 states when comparing three-year average poverty rates
for 2001 to 2003. Conversely, New Hampshire’s rate (6.0 percent)
— though not different from the rate for Minnesota — was
lower than those of the other 48 states and the District of Columbia.
- Seven states — Illinois, Michigan, Nevada, North Carolina,
South Dakota, Texas and Virginia — showed increases in their poverty
rates based on two-year moving averages (2001-2002 and 2002-2003), while
two states — Mississippi and North Dakota — showed decreases.
- The native population had increases in their poverty rate (from 11.5
percent in 2002 to 11.8 percent in 2003) and their number in poverty
(from 29.0 million in 2002 to 30.0 million in 2003). Poverty rates remained
unchanged for foreign-born naturalized citizens (10.0 percent) and for
foreign-born noncitizens (21.7 percent). Although the number for foreign-born
naturalized citizens in poverty (1.3 million) did not change from 2002,
the number of foreign-born noncitizens in poverty increased (to 4.6
million in 2003 from 4.3 million in 2002).
American Community Survey
- In the 2003 ACS, Somerset County, N.J., while not different from
Howard County, Md., or Prince William County, Va., had the highest median
household income ($89,289) of the 233 counties with populations of 250,000
or more in the sample.
- The median household income of Hidalgo County, Texas ($24,926), while
not different from Cameron County, Texas; Bronx County, N.Y.; or Lubbock
County, Texas, was lower than those of the remaining 229 counties.
- Somerset County, N. J., while not different from Waukesha County,
Wis.; Anne Arundel County, Md.; Howard County, Md.; Prince William County,
Va.; or Anoka County, Minn., had a poverty rate (1.7 percent) that was
lower than those of any of the other counties with a population of 250,000
- Hidalgo County, Texas (38.0 percent), and Cameron County, Texas (36.5
percent), had poverty rates higher than those of the other 231 counties,
though not different from one another.
Children Under 18 Years Old
- Somerset County, N. J., while not different from 17 other counties,
had a child poverty rate (2.0 percent) that was lower than any of the
remaining counties of 250,000 or more in the 2003 ACS.
- Hidalgo County, Texas, while not different from Cameron County, Texas,
had a child poverty rate (48.6 percent) that was higher than those of
the other counties of 250,000 or more.
- The number of people with health insurance coverage rose from 242.4
million in 2002 to 243.3 million in 2003. Nonetheless, the percentage
with coverage dropped from 84.8 percent to 84.4 percent, mirroring a
drop in the percentage of people covered by employment-based health
insurance (61.3 percent in 2002 to 60.4 percent in 2003). This decline
in employment-based health insurance coverage essentially explains the
drop in total private health insurance coverage, from 69.6 percent in
2002 to 68.6 percent in 2003.
- The percentage of people covered by government health insurance programs
rose in 2003, from 25.7 percent to 26.6 percent, largely as the result
of increases in Medicaid and Medicare coverage. Medicaid coverage rose
0.7 percentage points to 12.4 percent in 2003, and Medicare coverage
increased 0.2 percentage points to 13.7 percent.
- The proportion of uninsured children did not change in 2003, remaining
at 11.4 percent of all children, or 8.4 million.
Race and Hispanic Origin
- The uninsured rate did not change for blacks (about 19.5 percent)
or Asians (about 18.7 percent) between 2002 and 2003. (The health insurance
coverage rates of blacks and Asians were not different in 2003.) Non-Hispanics
who reported white as their only race saw their uninsured rate increase
from 10.7 percent to 11.1 percent.
- The uninsured rate for Hispanics, who may be of any race, was 32.7
percent in 2003 — unchanged from 2002.
- Based on a three-year average (2001-2003), 27.5 percent of people
who reported American Indian and Alaska native as their only race were
without coverage, lower than the uninsured rate for Hispanics (32.8
percent) but higher than that of the other race groups. Comparisons
of two-year moving averages (2001-2002 and 2002-2003) showed that the
uninsured rate for American Indians and Alaska natives did not change.
- The proportion of the foreign-born population without health insurance
(34.5 percent) was about two-and-a-half times that of the native population
(13.0 percent) in 2003.
- The South was the only region to show an increase in its uninsured
rate in 2003, up from 17.5 percent in 2002 to 18.0 percent. The health
insurance coverage rates of people in the South and in the West (17.6
percent) were not different in 2003. The percentages for the Northeast
and Midwest were 12.9 percent and 12.0 percent, respectively.
The estimates in the income, poverty and
health insurance report are based on the 2002, 2003 and 2004 Annual Social
and Economic Supplements to the Current Population Survey (CPS ASEC),
which is conducted in February, March and April at about 100,000 addresses
nationwide. The CPS is a labor force survey conducted monthly by the Census
Bureau for the Bureau of Labor Statistics using Computer-Assisted Telephone
Interviewing (CATI) and Computer-Assisted Personal Interviewing (CAPI).
The American Community Survey (ACS) is
an integral part of the plan to redesign the decennial census and will
replace the “long form.” During the 2000-2004 testing program,
the ACS has been collecting data from a sample of about 800,000 addresses
per year. These estimates are collected on a rolling basis every month.
The ACS uses the Census 2000 self-response mail-out/mail-back methodology,
followed by CATI, followed by CAPI.
Estimates from the CPS ASEC may not match
the estimates from the ACS because of differences in the questionnaires,
data collection methodology, reference period, processing procedures,
etc. As both are surveys, they are subject to sampling and nonsampling
errors. All comparisons made in the report have been tested and found
to be statistically significant at the 90-percent confidence level, unless
For additional information on the CPS
For additional information on ACS data, visit <http://www.census.gov/acs/www/UseData/Accuracy/Accuracy1.htm>.