Could complicated cartography interrupt the nation’s quest to close the digital divide? It certainly seems that way.
Last year, Congress passed the Infrastructure Investment and Jobs Act (IIJA), which included $42.5 billion for states to spend on broadband access through the Broadband Equity, Access, and Deployment (BEAD) Program. There was widespread concern, though, that existing FCC broadband maps did not show coverage in a way that made it possible to target subsidies to unserved areas.
Why does this matter? Because if an area is considered “served,” it won’t qualify for federal subsidies, and people who lack broadband access get left out. Hence the heated debates about the maps.
Congress thus ordered the Federal Communications Commission (FCC) to create much more detailed maps so that the BEAD money could be deployed where it is most needed.
But Congress’s order has an inherent flaw: The assumption that any single map can precisely and accurately display up-the-minute broadband coverage. Specifically, the order reveals three problems.
First, all data has errors. It is unavoidable. Even a completed map that happened to be correct for a moment would become immediately obsolete: New broadband deployments take place by the day, and thus a street that was previously offline could be online by the time you’re done reading this column.
Second, creating a map from data requires making assumptions, and those assumptions have policy implications. Consider the fundamental problem with the current FCC maps: If an ISP serves a single home in a census block, the entire block shows up in the public data as “served,” causing the map to overstate coverage. In a dense city, the block-level estimate of coverage is probably about right since census blocks can be as small as a single apartment complex. If one apartment in a building is served, odds are all the others are, too. But not so for rural or remote areas, where census blocks can stretch for miles. One little house on the prairie with coverage does not mean the others in the block have it.
Congress essentially ordered the FCC to minimize problems created by assumptions needed to aggregate data by collecting connectivity information on every home and business in the country.
Much like pulling up a blanket that’s too short to keep your face warm, thereby exposing your feet, increasing the granularity will result in a lot more errors in the data.
To its credit, Congress seems to have anticipated the existence of measurement error and included a challenge process to the maps in the legislation. This allows anyone to dispute the FCC’s coverage map in order to find and fix these errors.
But the challenge process reveals the third problem: Because money follows the maps, they are inherently political. Not surprisingly, even before the maps debut, everyone with skin in the broadband game is gearing up to contest them.
States want the maps of their territory to show lots of unserved areas in order to, as one consultancy put it, “make sure they don’t miss out on their rightful share of funding.” Internet service providers, meanwhile, are wary of areas they already serve showing up as unserved because a flood of federal funding into those regions threatens their business.
No matter how much effort the FCC puts in -- and by all accounts it has its collective nose to the grindstone -- the one-map-to-rule-them-all effort is an attempt to achieve the impossible.
What’s the alternative? First, we need to drop the belief that 100% accuracy is possible and also recognize that we don’t need a “map,” per se, but data on connectivity. Once we start thinking of it that way it becomes easier to understand that we’re not trying to make pretty pictures. Error and uncertainty are inevitable, but knowing something about that error helps us address it.
Better and more up-to-date information can come from harmonizing existing data sets about internet access, updated whenever a given map has new information. Private sector entities, think tanks, other federal agencies, and states themselves have spent years mapping digital connectivity. Blending those maps together will yield a more accurate picture of America’s digital divide than the FCC’s go-it-alone effort.
Indeed, that’s how third parties will contest the FCC’s maps: By using their own versions and crowing about the discrepancies. But this presents the FCC with an opportunity: Get ahead of those disputes by incorporating as much of that existing information as possible -- and creating a more accurate description of America’s coverage in the process.
It is impossible to eliminate the political incentives, but a map derived from multiple sets of data will be better -- and less open to criticism -- than a single-source map built by one agency. Officially recognizing the benefits of third-party data and incorporating it in advance of the challenge process might help BEAD funding move forward and avoid years of unwinnable cartographic debates. ■
