Last year, we manually checked availability of more than 11,000 addresses using Federal Communications Commission (FCC) Form 477 data as the “source of truth.” Based on the results, we estimated that 42 million Americans do not have the ability to purchase broadband internet.
In 2021, we expanded our study, manually checking availability of terrestrial broadband internet (wired or fixed wireless) for more than 58,000 addresses. In all, we checked more than 110,000 address-provider combinations using the FCC Form 477 data as the “source of truth”.
This study confirms our estimate that at least 42 million Americans do not have access to broadband. For the first time, we also estimate broadband availability for all 50 states.
In addition, the study shows that:
- All technologies, including DSL, fiber, cable and fixed wireless are over-reported.
- Over-reporting occurs in cities, rural towns and everywhere in between.
- For more than 12% of the addresses that did have broadband service, the technology reported by FCC data is incorrect.
The FCC’s 2019 Broadband Deployment Report stated that 21.3 million Americans lack access to broadband internet, including wired and fixed wireless connections. In its 2020 report, the FCC claimed that the number of Americans without access to broadband fell to 14.5 million.
The figures and estimates cited by the FCC rely upon semi-annual self-reporting by internet service providers (ISPs) using the FCC-mandated “Form 477.” However, there is a widely acknowledged flaw with Form 477 reporting: if an ISP offers service to at least one household in a census block, then the FCC counts the entire census block as covered by that provider.
The purpose of this report is to provide a better estimate of the number of Americans without access to terrestrial (wired or fixed wireless) broadband internet.
Based on new research conducted in 2021 on more than 58,000 addresses, we are confirming our estimate that at least 42 million Americans lack access to terrestrial broadband internet. Specifically, here’s what we did and how we arrived at this estimate:
- BroadbandNow Research selected 11 large ISPs with “check availability” tools (up from nine in our 2021 study). These tools allow users to enter an address and receive a response as to whether wired and/or fixed wireless service is available.
- From there, we selected 58,883 addresses (from a set of more than 1 million) where at least one of these 11 large ISPs offer service according to the FCC’s Form 477 data.
- We checked each address/provider combination, totaling 111,519 address-provider checks. We found that for 21.1 percent of combinations, the ISP’s tool indicates service is not available. (This is slightly higher than the rate found in 2020, which was 19.6 percent.)
- Some addresses are serviced by more than one of the nine large ISPs according to FCC data. For 11.1 percent of addresses, none of the ISPs’ tools indicate service is available (This is lower than the rate found in 2020, which was 13.6 percent).
- FCC data shows that approximately one-third of these addresses are also served by one or more local or hyper-local ISPs (the US has more than 2,000 ISPs). We were not able to check availability for local ISPs, and therefore conservatively assumed that residents of these addresses could receive broadband.
In total, we estimate that 42 million Americans do not have the ability to purchase broadband internet. This is an additional 6.5 percent of Americans beyond FCC estimates.
Estimating Broadband Availability By State
BroadbandNow Research checked more than 1,000 addresses manually in each state, except for Alaska and Hawaii, which we have excluded. Below is a table showing the FCC’s estimate of residents without broadband access, compared to BroadbandNow’s estimate.
Map of Over Reporting Rate by State
Below is a table of over-reporting by state, comparing FCC data to our manual checks across the country.
Unserved Americans by State
Based on our investigation of more than 55,000 addresses, below is BroadbandNow’s estimate of people without broadband access in all 50 states and the District of Columbia.
| State | Observed FCC Error Rate (1) | Urban Rate (2) | Population (3) | FCC Estimate – Population without Broadband Access (4) | BroadbandNow 2021 Estimate – Population without Broadband Access (5) |
|---|---|---|---|---|---|
| Alaska | Limited Data | 66% | 731,000 | 108,000 | 237,512 |
| Alabama | 28% | 59% | 4,902,000 | 608,000 | 1,202,976 |
| Arkansas | 22% | 56% | 3,018,000 | 574,000 | 1,010,599 |
| Arizona | 14% | 90% | 7,279,000 | 377,000 | 851,877 |
| California | 20% | 95% | 39,512,000 | 594,000 | 3,888,258 |
| Colorado | 21% | 86% | 5,758,000 | 161,000 | 674,433 |
| Connecticut | 22% | 88% | 3,565,000 | 27,000 | 386,786 |
| District of Columbia | 25% | 100% | 706,000 | 15,000 | 13,906 |
| Delaware | 16% | 83% | 974,000 | 22,000 | 43,665 |
| Florida | 24% | 91% | 21,477,000 | 804,000 | 2,373,981 |
| Georgia | 29% | 75% | 10,614,000 | 654,000 | 1,848,422 |
| Hawaii | Limited Data | 92% | 1,416,000 | 30,000 | 654,398 |
| Iowa | 25% | 64% | 3,155,000 | 127,000 | 387,344 |
| Idaho | 14% | 71% | 1,787,000 | 83,000 | 255,282 |
| Illinois | 23% | 89% | 12,672,000 | 259,000 | 1,226,709 |
| Indiana | 19% | 72% | 6,732,000 | 261,000 | 890,116 |
| Kansas | 18% | 74% | 2,913,000 | 125,000 | 341,908 |
| Kentucky | 25% | 58% | 4,468,000 | 257,000 | 832,791 |
| Louisiana | 25% | 73% | 4,649,000 | 538,000 | 1,152,783 |
| Massachusetts | 13% | 92% | 6,892,000 | 140,000 | 179,089 |
| Maryland | 15% | 87% | 6,046,000 | 152,000 | 222,923 |
| Maine | 21% | 39% | 1,344,000 | 46,000 | 297,567 |
| Michigan | 20% | 75% | 9,986,000 | 421,000 | 1,317,805 |
| Minnesota | 18% | 73% | 5,639,000 | 139,000 | 880,011 |
| Missouri | 18% | 70% | 6,136,000 | 422,000 | 1,058,308 |
| Mississippi | 30% | 49% | 2,975,000 | 587,000 | 1,175,639 |
| Montana | 19% | 56% | 1,069,000 | 142,000 | 258,878 |
| North Carolina | 24% | 66% | 10,488,000 | 472,000 | 1,567,091 |
| North Dakota | 15% | 60% | 762,000 | 24,000 | 122,583 |
| Nebraska | 20% | 73% | 1,934,000 | 71,000 | 184,638 |
| New Hampshire | 18% | 60% | 1,360,000 | 44,000 | 247,022 |
| New Jersey | 18% | 95% | 8,882,000 | 129,000 | 413,500 |
| New Mexico | 18% | 77% | 2,096,000 | 270,000 | 482,345 |
| Nevada | 18% | 94% | 3,080,000 | 88,000 | 139,261 |
| New York | 20% | 88% | 19,454,000 | 250,000 | 1,258,600 |
| Ohio | 19% | 78% | 11,689,000 | 328,000 | 1,404,448 |
| Oklahoma | 22% | 66% | 3,954,000 | 481,000 | 931,800 |
| Oregon | 17% | 81% | 4,218,000 | 216,000 | 686,854 |
| Pennsylvania | 15% | 79% | 12,802,000 | 525,000 | 1,224,298 |
| Rhode Island | 15% | 91% | 1,059,000 | 15,000 | 32,438 |
| South Carolina | 27% | 66% | 5,149,000 | 451,000 | 1,192,700 |
| South Dakota | 20% | 57% | 885,000 | 45,000 | 143,124 |
| Tennessee | 24% | 66% | 6,829,000 | 433,000 | 1,270,431 |
| Texas | 28% | 85% | 28,977,000 | 1,230,000 | 4,396,820 |
| Utah | 20% | 91% | 3,206,000 | 138,000 | 221,454 |
| Virginia | 16% | 76% | 8,541,000 | 498,000 | 936,208 |
| Vermont | 28% | 39% | 624,000 | 43,000 | 182,028 |
| Washington | 19% | 84% | 7,614,000 | 283,000 | 1,285,107 |
| Wisconsin | 17% | 70% | 5,822,000 | 394,000 | 670,592 |
| West Virginia | 36% | 49% | 1,792,000 | 319,000 | 900,010 |
| Wyoming | 19% | 65% | 579,000 | 42,000 | 101,252 |
| Total | 21% | 81% | 328,211,000 | 14,462,000 | 43,658,570 |
(1) Percentage of address-provider combinations where FCC reports service and provider check availability tools indicate service is not available.
(2) Percentage of population living in urban areas, 2010, from Iowa State University
(3) Population (millions) from FCC’s 14th Broadband Deployment Report
(4) Population (millions) unserved by terrestrial broadband internet from FCC’s 14th Broadband Deployment Report.
(5) Population (millions) unserved by terrestrial broadband internet according to BroadbandNow Research 2021 Study.
* for these states, data was not statistically significant
FCC Reporting Errors by Technology
Over-reporting occurs for every technology. Below is the rate of over-reporting for address-provider combinations by technology.
| Technology | % Over-reported |
|---|---|
| DSL | 18% |
| Cable | 17% |
| Fiber | 23% |
| Fixed Wireless | 35% |
In addition, sometimes an address will have broadband access, but the technology reported isn’t available. This occurred 12 percent of the time.
Where Do We Go From Here?
The FCC can take several specific actions to improve our current data collection and mapping practices. The single most important step that can be taken would be overhauling the Form 477 census block qualifier.
Drilling down to address-level granularity
The FCC has relied on census blocks to generalize where service is and isn’t available in the U.S. for over a decade. Acting Chairwoman Jessica Rosenworcel has established a Broadband Data Task Force to look into the current reporting methodology and make recommendations for how to improve it. This task force should recommend removing the census block reporting and replacing it with address-level granularity as the national requirement for provider self-reporting.
By doing so, we would eliminate the gaps created by counting hundreds – and sometimes thousands – of Americans as one entity. This is especially true within the context of massive federal funding initiatives like the Rural Digital Opportunity Fund, where accuracy will be imperative in order for funding to be effective.
Increasing the minimum speed limit
The current FCC broadband definition requires providers to provide at least 25 Mbps download speeds and 3 Mbps upload. In addition to improving our mapping practices, a new threshold for broadband needs to be adopted in order to better reflect the needs of Americans in 2021 and beyond. BroadbandNow advocates for a new standard of at least 100 Mbps download and 50 Mbps upload.
About This Study
BroadbandNow Research manually checked provider availability at 11 ISP web sites that offer online “check availability” tools. We checked at least one provider for 58,883 addresses from a dataset of more than 1 million addresses. In total, there were 111,519 provider-address combinations checked, of which a sample was re-checked for errors. We also verified the existence of a sample of addresses using real-estate sites. Based on the results of the study, we estimate a 21.1 percent error rate for provider-address combinations across the U.S.
For addresses with multiple providers checked, 11.1 percent of the time none of the providers offered service. Many small and mid-sized providers do not offer online “check availability” tools and so could not be checked manually. Approximately one-third of the addresses in our sample did not have a local ISP alternative (according to FCC Form 477). When we only assume that these addresses are unserved, the total unserved population in the US is more than 42 million.
Written by
Written by
