America prides itself on being the “Land of Opportunity,” a place where people are defined by their hard work and achievements, not held back by where they started. In the 21st century, Americans need internet access to compete in education and the modern labor market. Currently, internet inequality is a problem that creates a serious opportunity gap between Americans. According to the Federal Communications Commission (FCC), fixed broadband is unavailable to ~15 million people, or about 5% of the nation's population. This statistic is extremely hard to compute accurately, but other estimates tend to be significantly higher; for example, the White House estimates that ~30 million people do not have access to fixed broadband. This analysis will use estimates consistent with FCC data throughout the following pages, but actual numbers are likely even larger. Even among those who could theoretically have access to acceptable broadband, ~120 million more people do not use the Internet at “broadband speeds” (at least 25 Mbps). In the United States, two private companies (Comcast and Charter) provide broadband access to the vast majority of Americans. This private system has proven to be economically efficient but has proven to be extremely inequitable. While it functions well for most Americans, others have been left behind. Unfortunately, Black areas are less well-serviced than the nation as a whole, even after controlling for other important factors.
Access to fixed broadband internet helps Americans compete in the modern economy. The Internet has played a core role in the increase of productivity of American workers since 2000, but a “bottleneck” is emerging as a proportion of the American labor force cannot access high-quality Internet services despite their needs. Wealth inequality is already higher in the United States than in most developed countries, and the problem is worsening as various factors continue contributing to this growing inequality. It is challenging for policymakers to solve many catalysts of this inequality, including the relative weakness of housing markets and the growing “diploma divide.” However, asymmetric access to broadband services could be more easily fixed through government intervention.
In 2021, 93% of US adults said they use the Internet, and 77% said they have a broadband connection at home. These statistics increased during the COVID-19 pandemic, and recent survey data has shown that internet usage and a home broadband connection are vital for school and work. 71% of Americans say they would be unable to do their job without a strong internet connection at home, a cohort of Americans that has grown by 7% since the start of the pandemic. While the Internet is effectively mandatory for students who have seen their schools move to a more hybrid learning environment since the start of the pandemic, students are generally left on their own to find a home internet connection. While 8% of the poorest students (income less than $25,000) were provided Internet paid for by their school district during the pandemic, more than twice as many were left without access (NCES). Although the Internet became more ingrained in Americans’ academic and professional lives well before 2020, the pandemic accelerated this change.
Analysts should be most worried about the ~30 million people without any potential access to fixed broadband at their homes. Satellite internet is inferior to cable, it is much slower and suffers from higher latency. In October 2020, Microsoft’s Airband Initiative conducted a nationwide study on broadband access and usage. The Airband Initiative aims to close the rural broadband gap and estimated that 11.3 million rural Americans lack broadband access. The author conducted an original statistical analysis of this data, combining the Airband Initiative’s dataset with 2020 and 2021 US Census and American Community Survey data from data.census.gov. Understanding what types of Americans are most affected by this shortcoming is essential to determining what public policy interventions can solve the problem and which political actors and groups are most likely to support them.
Which Areas Tend to be Affected by the Broadband Access Gap?
Americans generally associate broadband accessibility problems with extremely rural areas, generally in the Western portion of the United States. The Western United States is much more rural than the rest of the country and contains most counties in the 10th percentile or lower in population density.
Modeling indicates a strong correlation between rurality and lack of broadband access across America. While the densest counties have nearly universal internet access, more rural counties have much more variance in their quality of internet coverage. A deeper analysis, however, shows that most counties with the very worst internet access are found in the South. The region is much denser than the West but significantly poorer.
The Broadband Access Gap Hurts Black Americans
Unfortunately, counties with the worst broadband access in the United States tend to be disproportionately Black. Even after controlling for income and density, the proportion of residents who are African-American is still a significant negative predictor of broadband coverage (Pr(>|t|)<2e-16). Beta regression modeling reveals an unexpectedly profound disparity in broadband coverage between Black and non-Black counties, even after accounting for county income and density.
Although these statistics are extremely problematic in their own right, the problem could be much deeper than county-wide analysis suggests. There is no robust data regarding broadband access at a more detailed census tract level on a national scope. Still, some counties have recently released plans for broadband expansion since the American Rescue Plan Act was passed in 2022. Even in counties with generally strong broadband availability, neighborhoods with high proportions of Black Americans often lag behind the rest of the county, even other poorer areas. Check out these images for comparison in Cleveland.
Although there are poor areas around the periphery of downtown Cleveland, those with Tier 1 Broadband Access (below 60% availability) tend to be found in the nearly exclusively Black portions of the city to the East of downtown Cleveland. The analysis outlined above omits the disparities seen within a single county, exacerbating the challenges already highlighted.
This leads us to ask a simple question: why? Broadband companies generally don’t build infrastructure in areas where they don’t feel it will be profitable to provide internet access. Two primary attributes would change the calculus of internet companies to make a community less desirable to invest in, low population density and low income. But even after controlling for both factors, Americans living in more Black areas have significantly worse broadband access. I’m not sure there is any acceptable answer I can think of for why this phenomenon exists other than some weird long-term reverberation of systemic racism in the late 1900s. It’s hard to imagine that internet companies intentionally discriminate against Black communities, choosing to build broadband infrastructure in less profitable, predominantly White communities rather than in more profitable, more Black communities. But it does not matter how we got here, and it’s incredibly disappointing that inequalities this strong still exist in 2023.
I redid the analysis using the CDC's six-level urban-rural classification instead of population per unit area, and it removed race as a significant factor. The R-squared of the model also goes up when using the CDC's classification (0.31 -> 0.41).