The availability of broadband is a known problem for rural areas, despite several federal programs attempting to help. However, past research into this area has suggested that it is not simply the availability of broadband that matters for rural economies, but rather how many people are actively using it.
These studies largely found that adoption rates matter more than availability for rural economic outcomes such as income growth or job gains. This body of evidence also hypothesized that how people were using the Internet also mattered, but a good measure of “use” has evaded researchers to date.
A new dataset from GoDaddy (a large retailer of website domain names and hosting services) provides some new insight into this topic. The data focuses on “ventures” which are defined as individual domain names with an active website. Using data from over 20 million websites, GoDaddy has developed an intriguing measure of broadband use by assessing how many ventures exist per 100 people. GoDaddy estimates that about 75% of the active websites are business-oriented (as opposed to nonprofit or personal sites).
The data is publicly available and is provided at the county level, with plans for monthly and quarterly updates. Recent estimates put GoDaddy’s market share for web hosts at around 15-20% – nearly twice as large as the next biggest provider. Extrapolating to other hosts could change the results, but given the size of GoDaddy’s presence, their data is a good place to start.
As the resulting map shows, non-metro counties are pretty diverse in the intensity of these ventures. Many counties in the South have less than oner venture per 100 people, while more typical values in the Midwest range from one to three. Hotspots of counties with more than six ventures for every 100 people exist in non-metro counties in the Northeast and Colorado but also pop up sporadically across the country.
An initial analysis of the data by professors at the University of Iowa and Arizona State University found that the rate of small-scale entrepreneurs running these businesses had meaningful impacts on their local communities. The lead researcher, Dr. Karen Mossberger, has written many pieces about the importance of “digital citizenship” and argues that this new density-of-websites measure is a good indicator of the extent of online activity in an area.
A simple correlation between the number of ventures per 100 people and median household income shows a clear positive relationship in nonmetro counties (below). What’s more, ventures defined as “highly active” have an even stronger relationship – each additional one per 100 people is associated with an increase of over $4,700 in income. Similar relationships hold for a prosperity index capturing seven indicators such as poverty, job growth, and education.
Relationship between Venture Density and Income – Nonmetropolitan Counties only
Source: Nov. 2018 GoDaddy VentureForward data; ACS 2013-17 Median Household Income
Of course, a lot of other factors are at play in determining these economic outcomes. But Dr. Mossberger and her team conducted a more thorough analysis that controls for other influential variables like race, age, and local industry mix and still found that each additional highly active venture per 100 people added $331 to each county’s median household income. They also found evidence that these ventures are even more important in places with low rates of broadband adoption and in places that were slow to recover from the 2008 recession (both characteristics of rural areas).
This new data has already been featured in a New York Times article and in a Harvard Business Review write-up. These outlets have noted the rural component, but largely in passing.
The most recent estimates from the U.S. Census suggest that 75% of households in rural, non-core counties have a broadband connection, which is more than 10 percentage points behind households in metropolitan areas. The results from this new data suggest that more effort should be put into encouraging broadband adoption and use in rural locations – particularly for small businesses that could benefit from expanding their customer base. This is particularly noteworthy given a recent report that nearly 2/3 of existing main street businesses do not currently sell online. Some university-based community engagement programs are already doing exactly this type of work. Expanding on these efforts will be crucial – as will continuing to find new ways to measure Internet “use” by rural communities and residents.