NEWS CONFERENCE ON
2004 INCOME AND POVERTY ESTIMATES
FROM THE AMERICAN COMMUNITY SURVEY
Preston Jay Waite
Associate Director for Decennial Census
August 30, 2005
The data I am discussing today are the income, poverty, and earnings data from the 2004 American Community Survey.
The American Community Survey is a powerful new tool designed to produce current local data. In the past, users of Census Bureau data had to choose between current national data and older local data. Now planners and local officials don’t have to make that choice. Now they have current local data.
Data from the American Community Survey are available online at www.census.gov. There you’ll find local data on subject areas such as:
In January 2005 the American Community Survey was fully implemented after years of testing that began in 1996. Every month now American Community Survey questionnaires are mailed to about 250,000 addresses throughout the United States and Puerto Rico, or about 2.5% of addresses annually.
The data we are releasing today are from the 2004 ACS. This includes data for the United States, all states and the District of Columbia, and sub-state areas with populations of 250,000 or more (236 counties and 70 places).
This is the last year we’ll be talking about data from the developmental phase of ACS. Next year at about this time you will see the first data from the fully implemented 2005 American Community Survey. Those data will be available for about 8,000 areas with populations of 65,000 or more. Later in the decade, American Community Survey data will be available for even smaller geographic areas.
That is a future we can all look forward to.
The American Community Survey is the only source of current and timely survey data for counties and places. For the purposes of my presentation all of the national, state, and local data are taken from the American Community Survey.
Median Household Income for Counties with populations of 250,000 or more. Three of the counties with the highest estimates of median household incomes are in the Washington, DC, metropolitan area – Fairfax County, Virginia; Montgomery County, Maryland; and Howard County, Maryland. Three of the counties with the lowest estimates of median household incomes are on the Texas-Mexico border (Hidalgo, Cameron and El Paso Counties).
Median Household Income for Places, specifically cities with populations of 250,000 or more. Three of the cities with the highest estimates of median household incomes are in California (San Jose, San Francisco, and San Diego).
Other cities with high estimates of median household incomes include Anchorage, Alaska, and Virginia Beach, Virginia. Some of the cities with the lowest estimates of median household incomes are Miami, Florida; Newark, New Jersey; Cleveland, Ohio; Detroit, Michigan; and Buffalo, New York.
An important feature of the American Community Survey is that it allows you to compare counties and places within a state to the state and national data.
For example, let’s look at Florida, with a median household income of $41,200. As you can see, the estimates of median household incomes in Seminole and Lee Counties and Jacksonville city are high relative to the state of Florida. Estimates of median household incomes in Lake and Polk Counties and the cities of Miami, Tampa, and St. Petersburg are relatively low.
In 2006 and following years, the American Community Survey will enable us to “drill down” to small geographies within all of the states—small counties and cities, towns, and eventually to the neighborhood level (census tracts and block groups). This will make the American Community Survey a more and more valuable planning resource.
I have been talking about levels of
median household income. Now I would like to turn your attention to changes
in median household income. For the time being, we can look at changes for
the 37 counties and the 9 places with populations of one million or more.
Among very large counties, Fairfax County, Virginia,
and Orange County, California, had increases in the estimates of median household
income between 2003 and 2004. King County, Washington; Palm Beach County,
Florida; and Philadelphia County, Pennsylvania; had decreases. Philadelphia
is the only large city to experience a change in its estimate of median household
income—it decreased.
Now I am going to talk about levels of poverty. Among counties with populations of 250,000 or more, the three counties in Texas with high estimates of poverty are the same ones where we saw low estimates of median household incomes—Hidalgo, Cameron, and El Paso Counties.
Conversely, three of the places with the lowest estimates
of poverty are the same as places with the some of the highest estimates of
median household incomes—Anchorage, Alaska; Virginia Beach, Virginia;
and San Francisco, California.
Here we can learn from the capability to drill down
to local areas with the American Community Survey. The State of California
has an estimate of poverty similar to that of the United States as a whole.
Looking at counties and places within the state, we see differences. The two counties of 250,000 or more with the highest estimates of poverty, around 20 percent, are Tulare and Kern. Two other California counties have estimates of poverty of about 5 percent—San Mateo and Placer. The two cities with the highest estimates of poverty are Long Beach and Fresno. The two with the lowest estimates of poverty are San Jose and San Francisco.
Turning now to changes in poverty level, you can see that several counties with populations of one million or more experienced increases in their estimates of poverty between 2003 and 2004:
Two others had decreases in their estimates of poverty:
The only city of one million or more that had a change in its estimate of poverty between 2003 and 2004. The estimate of poverty in New York City increased from 19.0 to 20.3 percent.
I have been talking about the poverty rates of people of all ages. Now I am going to talk about child poverty. Many of the counties and places of 250,000 or more with high poverty rates also have high rates of child poverty.
Hidalgo, Cameron and El Paso Counties, Texas, have estimates of child poverty close to 50 percent. In contrast to these, several counties have estimates of child poverty less than 5 percent—Morris and Monmouth Counties in New Jersey, Bucks County, Pennsylvania, DuPage County, Illinois, and Waukesha County, Wisconsin.
Several cities of 250,000 or more have estimates of child poverty between 40 and 50 percent:
Other large cities have estimates of child poverty between 10 and 13 percent:
Using American Community Survey data to drill down to counties and places in Texas, we can see a big difference between the estimates of child poverty in Hidalgo and Cameron Counties (around 50 percent) on one hand, and Collin and Denton Counties (around 5 percent) on the other. There also is a difference, but a smaller one, between the estimates of child poverty in El Paso and Dallas compared to Austin and Arlington.
Today’s data release also includes interesting information about the earnings of men and women in different occupations.
Men in legal occupations have the highest estimates of median earnings by far. Occupations where women have the highest estimates of earnings are:
Let’s now look at occupations with the lowest estimates of median earnings. For both men and women, four occupations are near the bottom of the earnings list:
The occupational category in which estimates of women’s earnings are least compared to those of men is legal occupations. Women earn less than half as much as their male counterparts.
In general, women earn less than men, even if they are in the same occupation. This slide shows that estimates of women’s earnings approach those of men in several relatively low-paying occupations:
One well-paying occupational category in which women earn 85.6 percent as much as their male counterparts is computer and mathematical occupations.
I have shown you just some of the data available from the 2004 American Community Survey. The best is yet to come. Starting next year and every year thereafter, the Census Bureau will release American Community Survey data for—
As the saying goes, knowledge is power, and knowledge of how things are is even more powerful than how things were. Beginning in 2010 and every year thereafter, the American Community Survey will provide vast amounts of information for all communities, regardless of size.
This powerful new Census Bureau tool will arm local officials, such as mayors, local planners, and school board members, with the current knowledge they need to plan for services like roads, schools, senior centers and affordable housing. We at the Census Bureau are excited about the American Community Survey!! It helps us do our job better and is a great new tool.
