EDITOR’S NOTE: This article was published last week by the Census Bureau. The article uses the Census definition of rural, as opposed to the metropolitan/nonmetropolitan frequently used in Daily Yonder articles. Also, the study defines broadband as any speed faster than dialup. Dialup speed is approximately 56 kilobits per second. The Federal Communications Commission defines broadband as 25 megabits per second download and 3 megabits per second upload.)
For the first time, U.S. Census Bureau statistics show the impact that a county’s rural or urban geography and income level can have on residents’ rates of subscription to the internet.
While the Census Bureau has reported national estimates of internet subscription for each of the past four years, the new data for all counties in the United States allow us to look at trends across smaller areas, including less populated rural areas.
The 2013-2017 American Community Survey (ACS) estimates released December 6 show overall widespread subscription to the internet throughout the country. Nationally, 78 percent of households subscribe to the internet, but households in both rural and lower-income counties trail the national average by 13 points.
In addition, the data show that Americans connect to the internet in a variety of ways, including through high-speed wired connections at home and cellular data plans while on the go. We define broadband internet subscriptions as any service that is capable of delivering faster speeds than “dial up” — no longer used by most, but still used by less than 1 percent of households nationally.
Focusing on counties with at least 10,000 population (for which statistics are more reliable), Douglas County, Colo., had the highest rate of subscription at 94.6 percent. Loudoun County, Va.; Howard County, Md.; Fairfax City, Va.; Kendall County, Ill.; and Fairfax County, Va., also all have subscription rates of 92 percent or higher.
Counties with lower rates of subscription are prominent throughout the South, especially in the Mississippi River basin, as well as in a number of counties in the Midwest and West.
Telfair County, Ga., has the lowest rate of subscription at 24.9 percent out of counties with populations of 10,000 or more. Holmes County, Miss.; Kemper County, Miss.; Apache County, Ariz.; Monroe County, Ala.; McKinley County, N.M.; and Leflore County, Miss., all also have subscription rates at or below 40 percent.
When categorizing counties by the share of the population that lives in rural areas, rural counties tend to fare worse than urban counties when it comes to subscribing to broadband internet.
Rural areas are defined as all population, housing and territory not included within an urbanized area or urban cluster. (EDITOR’S NOTE: This article, written by a researcher at the Census Bureau, uses the Census definition of rural. That’s different from the metropolitan/nonmetropolitan system the Daily Yonder frequently uses. For more information on various classification systems, visit the USDA Economic Research Service website.) Census blocks are identified as urban if they have a density of 1,000 people per square mile. These blocks are then aggregated to define urbanized areas that contain 50,000 or more people and urban clusters are areas with at least 2,500 but fewer than 50,000 people.
“Completely rural” counties have 100 percent of the population living in rural areas, while “mostly rural” counties have at least 50 percent of the population living in rural areas. Lastly, “mostly urban” counties have populations where more than 50 percent of the population lives in urban areas.
In the average “mostly urban” county, over 75 percent of households have a subscription to broadband internet.
The average county classified as “mostly rural” has a household subscription rate of approximately 67 percent, while the average “completely rural” county has a subscription rate of 65 percent.
Despite a higher average subscription rate, some “mostly urban” counties are not overwhelmingly connected as well. As seen in the map at the top of the page on the right, a number of “mostly urban” counties in Texas, New Mexico and along the Mississippi River have subscription rates below 55 percent.
“Mostly rural” counties, shown on the left map at the top of the page, tend to have lower subscription rates, with many such counties located across the South. The Northeast, Midwest and parts of the West contain a number of “mostly rural” counties with subscription rates above the national average, although there are also a few in the South.
Like “mostly rural” counties, “completely rural” counties (Figure 4, above) tend to have broadband subscription rates below the national average, and many of these are found in the South. A handful of “completely rural” counties are well connected, such as Daggett County, Utah (90.5 percent).
As shown in the latest ACS data, higher-income counties tend to have higher rates of internet subscription than lower-income counties.
An average of over 77 percent of households have a broadband subscription in counties with median household incomes equal to or above $50,000. In counties with median incomes below $50,000, the average rate of subscription falls to roughly 65 percent.
Not surprisingly, higher income counties that are also “mostly urban” have the highest rate of internet subscription, with an average of 79.9 percent of households subscribing to some broadband service (Figure 5, above).
“Completely rural,” lower income counties fare the worst, with an average broadband subscription rate of only 61.7 percent.
Despite differences in level of income and rurality, households in “mostly urban,” lower income counties are within 2 percentage points of “completely rural,” higher income counties in their rate of subscription. An average of 69.7 percent of households subscribe to broadband internet in “mostly urban,” lower-income counties compared to 71.4 percent in “completely rural,” higher-income counties.
Perhaps most interestingly, the five-year American Community Survey estimates also allow for the exploration of patterns of internet subscription on an even smaller, close-up scale.
The Memphis, Tenn., metropolitan area provides an example of what can be done using tract level data (Figure 6, below). Memphis was chosen because it contains a set of core urban tracts while still extending into a larger rural space.
In the immediate downtown area, subscription rates are mixed, but many tracts fall below 55 percent. While in many cases urban environments may lead to higher rates of subscription, the disconnected urban core in Memphis demonstrates that this may not always be the case.
The American Community Survey does not ask why people do or do not subscribe to the internet, but decisions could be shaped by services that are available. In some rural areas, the options might be limited to slower speed services. On the other hand, urban areas may lag behind because of higher cost of broadband services, particularly in lower income urban counties.
Additional information about computer and internet use in America can be found with the December release of American Community Survey estimates.
Michael J. R. Martin is a statistician in the Education and Social Stratification Branch in the Census Bureau’s Social, Economic, and Housing Statistics Division.