A little over one year from now, the United States will participate in a democratic tradition that stretches back to the founding of the republic: the once-a-decade census of its population.
From 1790 (U.S. population: 3.9 million) to 2010 (U.S. population: 309 million), the decennial census has changed alongside the nation itself. From the territory it covers, to the questions it asks, to how it collects the information, the census has reflected evolution in technology, the role of the federal government, and the size of the country itself.
As we approach 2020, however, both technical and political changes in the census are introducing unprecedented new challenges. The stakes are high for cities and regions, which depend on a full and accurate count of their populations to ensure their fiscal health and political strength. Three areas of concern stand out.
For the first time in 2020, the U.S. Census Bureau plans to allow households to respond to the decennial census survey via the internet. This move could help reduce the costs of the census (e.g., fewer paper forms to collect and process) and improve response rates among a U.S. population that’s increasingly online, all the time.
Yet a digital decennial census could cause complications for participation, particularly in areas with limited broadband access. My colleague Adie Tomer and co-authors find that in 2015, almost one in four Americans (74 million people) lived in neighborhoods where fewer than 40 percent of households subscribed to broadband. It’s not only rural areas where broadband’s reach is limited; major metro areas with large Hispanic populations—and/or significant areas of concentrated poverty—exhibit troubling subscription gaps as well. As CityLab’s Kriston Capps reports, many cities are leaning on libraries to connect those populations to the online census, but tests thus far suggest the fix won’t be easy.
The massive online data collection effort also introduces significant data quality, privacy, and security challenges for the Census Bureau. Viruses, impersonator websites, data breaches, and service disruptions could threaten the integrity and accuracy of the count. While the Bureau is investing in addressing potential security weaknesses, the Government Accountability Office recently flagged significant challenges and risks the Bureau still faces to mounting a successful online 2020 census.
The 2020 census issue probably familiar to most readers concerns the Trump administration’s move to ask respondents, for the first time since 1950, whether they are U.S. citizens. Against the recommendations of his own Census Bureau, Commerce Secretary Wilbur Ross approved the addition of the question last March. He immediately became the defendant in multiple lawsuits filed by cities, states, and civil rights organizations to get the question removed. (NPR’s Hansi Lo Wang has published the definitive timeline of events that landed the citizenship question in federal court.) In January, U.S. District Judge Jesse M. Furman ordered the Trump administration to remove the citizenship question from the decennial survey, saying that Ross had violated a “veritable smorgasbord” of federal rules in approving the question.
My colleague Bill Frey describes how this controversy stems from analysis clearly indicating that including a citizenship question would deter participation in the census among minority and immigrant groups. Because they are more likely to live in households with noncitizens, Latinos, Asians, young people, and urban residents would be most negatively affected by this under-reporting. And that’s on top of the fact that the census already tends to undercount communities of color, particularly black, Hispanic, and American Indian and Alaskan Native populations. While the Trump administration maintains that the question would yield information important for enforcing voting rights laws, evidence abounds that the move was intended to reduce the political power of Democratic-leaning constituencies.
The challenge doesn’t end with Furman’s ruling, however; the case is likely to be heard by the Supreme Court in the coming months. And even if SCOTUS upholds the lower court ruling, or Congress acts to block the question in other ways, negative impacts are likely to linger. The Census Bureau’s research in 2017, before Ross officially put the question on the census, found that concerns about issues such as the “Muslim ban,” dissolution of Deferred Action for Childhood Arrivals (DACA), and stepped-up immigration enforcement were likely to deter census participation among Limited English Proficient and immigrant populations.
The Census Bureau has estimated the full “life cycle” cost of the 2020 census at $15.6 billion. Much of its ramp-up funding over the past few years came in below projected needs, leading to reduced testing of new operations and procedures. As Capps from CityLab details, the Bureau cut two of its planned dress rehearsals, conducting only one end-to-end test in Providence County, R.I. Moreover, the Census Bureau went without a permanent director for 18 months during this critical period, with Congress finally confirming federal statistical agency veteran Steven Dillingham to the position in early January.
Signs are looking up that Congress will provide adequate funding for the 2020 census in the coming year. Still, other preparation challenges confront the Bureau, including the need to hire more than 500,000 enumerators—the workers who go door-to-door to make sure that people who don’t respond online or by mail to the survey get counted. Last time around in 2010, the U.S. unemployment rate was hovering near 10 percent, and workers looking for a decent-paying temporary gig were relatively easy to find. Now with the rate south of 4 percent, it may be harder to find such workers, and as a result more difficult to ensure that harder-to-count communities are accurately captured in the census.
All of these looming concerns about the 2020 census could catch cities and regions in the crosshairs. According to the Congressional Research Service, census data are used in formulas that allocate more than $675 billion in federal funds annually to states and localities. Those data are also a key resource that cities and counties, businesses, and nonprofits use to guide local investments. And of course, census data are the critical building blocks for apportioning congressional and state legislative seats.
Fortunately, local leaders are not powerless in the face of these 2020 census challenges. Many are stepping up to ensure a complete and accurate count in their cities and regions, forging partnerships among elected, business, philanthropic, and other civic actors toward that end. Future posts in this series will examine the above challenges and promising bottom-up, cross-sector responses—in areas such as information and research, outreach and organizing, and funding—in greater detail. Local initiative will be critical to preserving the value of this unique 230-year-old institution.