The census is a once-a-decade process of counting every person living in the United States, and it has been a regular occurrence since the nation’s beginning. While it may seem like an antiquated and even unnecessary process in a modern world where demographic data and predictive models are commonplace, the census continues to play an important role in politics and governing in the country.
According to the U.S. Census Bureau, the first count occurred in 1790, only a year after the adoption of the current Constitution and the beginning of George Washington’s presidency.
As mandated by the Constitution, the census was ordered to occur every 10 years for the purpose of apportioning each state’s seats in the U.S. House of Representatives. It called for counting all “free Persons,” all tax-paying Native Americans and “three-fifths of all other Persons,” meaning enslaved people.
According to the Census Bureau, it was made into a permanent office in 1902, operating even in the years between census counts.
It is important to note, however, that the scope of the census gathering operations and the use of data has expanded significantly from counting people for just the purpose of congressional representation in the nearly two and a half centuries since the original count.
Now, census data helps officials determine where to distribute billions, and even trillions, of dollars in federal funding, in addition to various private and public sector uses. However, as the census has played a greater role in governing, the importance placed on the count and the challenges related to it have likewise increased. While issues like the counting of minorities, students and rural residents have long raised concerns from census to census, the encroachment of politics and the all-encompassing COVID-19 pandemic made this year’s count particularly challenging and led to fears of a significant undercount.
Representation and federal funds
Today, census data factors into a wide range of demographic-related considerations for policymakers and other officials.
While the most visible purpose of the census remains democratic representation, it has expanded beyond just the seats apportioned in the U.S. House of Representatives. Alec Thomson, director of the Arizona Complete Count Committee, explained census data is used to apportion seats and draw districts for a range of elected offices.
“Arizona’s representation in Congress depends on the census count, and also our redistricting process is guided by 2020 census counts,” Thomson said. “So those local political boundaries, both at the city level, state legislative level and then congressional districts, all of those are guided by census numbers.”
But representation is far from the only way census statistics are applied. In addition to issues like redistricting and apportionment, census data is used by the federal government to distribute federal funds.
A study by Andrew Reamer of the George Washington Institute of Public Policy examined the ways in which census data is utilized by the federal government to apportion these important federal dollars to localities.
For context, the study explained that in 2017 alone, $1.5 trillion was distributed to state and local governments, as well as other private organizations and individuals through 316 federal spending programs.
According to the study, the federal government uses census data for funding in four main ways: eligibility criteria, geographical allocation, selection preferences and interest rates.
Eligibility criteria is how the data is used to identify who can receive funds in line with laws and program guidance. For instance, the data is used to determine the recipients of program funds intended to benefit “rural” or “low-income” households.
For geographical allocation, the information derived from the count is used to determine how many people live in each household, certain characteristics about them and as a result, how much money should be given to local programs to support the population.
Selection preferences are the use of census data to help the federal government make funding decisions for specific projects. The information helps guide decisions regarding the necessity of projects and their extent of need.
Lastly, the data is also used for setting the interest rates of federal loans. Data on the residents of various states and regions may serve as a basis for rates in different ways, such as a resident’s average or median income.
State and local budgets
Due to the massive amount of federal dollars handed out each year, a significant portion of the funds that make up state and local government budgets do not come directly from local residents’ tax dollars.
Thomson explained that based on census population figures, states receive about $3,000 per resident, per year. This means tens, and even hundreds, of billions of dollars are at stake for each state until the next count in 2030.
It also means the stakes of an accurate count are incredibly high for states. In the case of Arizona, an undercount of as little as 1% could lead the state to lose out on well over $2 billion over the next decade.
Because it is such a large amount of money, many programs and functions of local government have come to depend on federal dollars and would be affected by an undercount. Thomson described the types of programs affected and the importance of an accurate count to state officials.
“Programs ranging from education to health care, even transportation infrastructure, all of those tie back to the census count and the amount of money that goes to support those things in each state,” Thomson said. “We want to make sure our population count is accurately reflected so we’re getting the funds we need to support the services for every Arizonan.”
Greg Webb, the Coconino County Census event and volunteer coordinator, further explained the extent of the funding Arizona is entitled to and what the funds can be directed toward is based on census results.
“They look at decennial census data and say ‘this is a baseline for how much money we think should go to this state, this community, this program or to this grant,’” Webb said. “That federal funding is huge. For the state of Arizona, it works out to just shy of $20 billion per year in funding.”
In addition to the state government, federal dollars also go directly to support city governments and local communities.
Jessica Drum, public affairs director for the City of Flagstaff, said the implications make the census incredibly important for communities like Flagstaff.
“It’s so, so phenomenally important for a document that took me five minutes to fill out for my entire family,” Drum said. “Those counts determine everything from the type of funding that our community, or any community, gets … everything from assistance for community development block grants that do funding for housing permanency, eviction prevention. I can’t overstate the importance and the widespread of programs that are impacted by census counts — impacted not just for a year but for a decade.”
Flagstaff gets millions of dollars in funding from the federal government. While it receives a smaller percentage of its funding from the federal government than the state, Flagstaff still received over $17 million in grants for the 2018-2019 fiscal year alone. For comparison, that same year, Flagstaff raised $23.5 million in sales and franchise taxes and $13.4 million in property taxes.
This money went toward programs including Flagstaff’s General Fund, which funds emergency services, governmental administrative costs, public works and more; the Housing and Community Services Fund, mainly for community development; the Metropolitan Planning Organization Fund; the Library Fund; and many more.
Additional data uses
In addition to representation and federal funds, census data is also used for a variety of public and private sector purposes.
Thomson explained that among many uses, census statistics are especially important in urban planning.
“I could go on for hours about the way census data is used,” Thomson said. “Knowing where we need things like new roads, new hospitals, all of those critical, essential infrastructure.”
As an example of how census figures are relied on, Thomson brought up the city of Maricopa, a community south of Phoenix not to be confused with Maricopa County. Located in Pinal County, it was only formally incorporated as a city in the last 20 years and has undergone significant population growth since.
U.S. Census Bureau statistics from the 2000 count show the city had less than 2,000 residents, but by the 2010 census, there were nearly 45,000. Additionally, the population is estimated to have grown by several thousand more in the decade since.
In order to properly account for the community’s development, Thomson explained the importance of census data to properly accommodate new infrastructure projects.
“Knowing the population in that city has increased so dramatically over the last decade, we need to build new roads, we need broadband infrastructure, we need potentially a new hospital out there, maybe a new fire station,” Thomson said. “All of those planning decisions, both from an infrastructure perspective and also from a policy perspective, are guided by the number of people living in a specific location.”
While governmental planning may arguably be one of the most important uses of census data, it is far from the only application. Thomson said the data is also used for a range of private sector considerations.
Thomson explained that businesses have access to census data and are able to use it to make decisions, such as where to pursue ventures like building a grocery store, how many people to hire, the availability of the workforce in a community and many more.
Although the main goal of the census is to provide an official tally of how many people are living in a given place at the turn of each decade, Thomson expressed his view that it can truly play a role in shaping communities and have significant implications for years to come.
“The breadth of what this data is used for really determines what a state looks like for the next decade and how it evolves, the services that are provided immediately, but also the way a state is going to grow over the next decade until the next census occurs,” Thomson said.
Completing the count and COVID-19
Because of how high the stakes of the census are for Flagstaff, Coconino County and other locations across Arizona, local governments are working in coordination with the state to ensure high response and overall enumeration rates, the census term for counting. Together they have formed the Arizona Complete Count Committee in order to assist the U.S. Census Bureau in awareness efforts and other operations related to the count.
Thomson explained it was formed back in 2019 by Gov. Doug Ducey with the goal of, as the name implies, ensuring a complete and accurate count in Arizona.
Thomson discussed the work of the committee in accomplishing this goal and how it leads the state’s census outreach efforts.
“I lead a group of 23 individuals from across the state working to build a grassroots campaign with community driven tactics to encourage Arizonans to respond,” Thomson said. “We’re also leading a paid media effort to get the word out to Arizonans about how important and safe it is to respond to the census.”
Committee coordinators, like Webb, follow similar goals and practices, just on a more local level. Webb’s main line of work involves outreach operations, advertising and census events.
However, despite the best efforts of various groups and years of preparation, the 2020 census is currently one of the more challenging counts in the long history of the census.
Drum explained that just like most parts of society, the census experienced significant disruptions beginning in late March and early April as a result of the COVID-19 pandemic.
In a regular year, outreach personnel would have worked to spread awareness and encourage participation, while enumerators, the census employees working in the field, went from door to door to follow up on households that had yet to answer a census questionnaire.
This year, however, most census outreach and on-the-ground enumeration efforts were about to start, or had barely just begun, when the U.S. Census Bureau was forced to officially shut down in-person operations on April 1. The bureau was only able to begin a “phased restart” plan in June.
Drum stressed this negatively affected the census in significant ways.
“When I say this is a worst-case scenario, it’s an absolute [worst-case scenario],” Drum said. “It’s been a really, really challenging situation.”
A statement from the U.S. Census Bureau regarding its health and safety guidelines explained efforts were made to bolster counting practices that do not require human contact, including surveys conducted via phone, mail and online. When in-person counting was inevitably necessary, census workers were required to follow health and safety practices like wearing face coverings and social distancing.
In addition to the most immediate COVID-19 disruptions, challenges to on-the-ground efforts came from the employment of census workers themselves.
Each census, the bureau employs a large number of temporary workers — its target goal for 2020 was 57,000 hirees in Arizona and 500,000 nationwide — in order to complete the count. But Drum said many of the people who had previously been hired before the pandemic were no longer available or willing to work when the census was ready to resume operations.
“They’ve struggled significantly with even getting staff; hiring has been a huge challenge,” Drum said. “And then people are not super excited about going door to door in the middle of a pandemic.”
In addition to struggles with in-person enumeration, outreach and awareness operations, like what Webb and Thomson work on, have also been severely impacted.
Webb explained the earliest and most immediate impact on his work was the cancellation of all in-person events. Events that involved tabling and giving away promotional items in order to spread information and awareness about the census, a major part of local coordinators’ strategies, could not continue due to the pandemic.
“We were ready to get on the ground to communicate with people, have fliers, have events, have volunteers and literally the week after the census went live, everything went into shutdown,” Webb said.
However, more than just in-person events were affected. Webb explained that another major component of the outreach strategy, social media, was also disrupted.
Before the census fully got underway and the pandemic had become an issue locally, the outreach team planned to use established social media channels, such as the Coconino County and City of Flagstaff social media pages to spread information. However, Webb said due to the pandemic, those social media accounts needed to shift focus.
“When COVID-19 hit, our organizations were like, ‘We need to focus on COVID-19 messaging,’” Webb said. “There was a little bit of a debate internally, but there was the sense this is important enough that we need to basically focus all of our messaging that’s coming out of city and county social media on COVID-19.”
Webb said that having a major part of their strategy effectively sidelined significantly impacted their efforts in the early days of the pandemic.
“The bandwidth was gone and I think that impacted us, so from our end there were just these immediate things where it’s like, ‘Well what do we even do?’” Webb said. “We also actually suspended a lot of our census efforts for a month.”
Outreach and enumeration challenges
Even without the coronavirus, Webb said that advertising and media based outreach can be a great challenge in Coconino County.
Northern Arizona lacks local TV broadcasters and is instead dependent on stations located in Phoenix. Webb said this has limited media messaging specifically intended for a northern Arizona audience, however, they have still been able to utilize print, radio and social media advertising.
Outreach challenges aside, the fact that much of the county is quite rural and spread out makes the enumeration process difficult even in the best of years.
Webb explained how a combination of rural spread and a lack of home mailing services adds to enumerators’ challenges in the county.
“The area of the county is roughly the size of New Jersey and very spread out,” Webb said. “We have very isolated rural areas throughout the county, many of which don’t even get mail directly at their homes. One of the major ways the Census Bureau delivers information is through the mail. You don’t get anything from the Census Bureau in the mail if you don’t have a standard mailing address at your home.”
For residences that lack mail addresses, census workers must perform a significant amount of extra work in order to establish contact, mainly through distributing packages containing the census materials that would ordinarily be mailed out.
Webb explained that isolated communities and a significant lack of mailing addresses also makes it difficult to count many people living in the Navajo Nation.
“On the other side of the county, we have areas that overlap with the Navajo Nation where the obstacle is even things like having a phone, having cell service and poverty,” Webb said. “There’s no standard addresses whatsoever and there’s a whole different set of challenges in trying to count those areas.”
The extra effort involved is not the only challenge in enumerating people without home mailing addresses.
Webb explained there have been issues in distributing the census packages. Although many homes without addresses exist together in small communities or on a single property, the census still identifies them as separate households and a packet is supposed to be given to each. However, Webb said there have been reports of those being improperly distributed, further complicating the enumeration process.
“There’s really supposed to be an individual package for each household,” Webb said. “We were hearing stories that basically they would go out and find six or seven packages just hung on a tree. The Census Bureau people on the ground hadn’t bothered to go to every single door and drop off the individual packages to the individual doors. That made things confusing for those folks because there’s supposed to be a specific ID number with each package, but whose number was whose, they were just hung on a tree.”
However, even when census workers follow protocol and properly distribute materials, Webb explained that environmental factors like wind have caused packets to get blown away and lost. He described an incident where a packet was blown from one household to a neighboring home where residents mistakenly filled it out.
The combination of the COVID-19 pandemic and longstanding challenges unique to northern Arizona unsurprisingly led to a rocky restart of operations.
Restarting the census
As for the specific experiences of local workers currently enumerating during the pandemic, officials at the Flagstaff Census Bureau office have been less than forthcoming. When attempting to interview enumerators about their experiences, officials first intervened to limit the scope of the discussion and then prevented interviewees from participating altogether, less than a day before a scheduled interview.
On the topic of how the enumeration process has gone since resuming after the initial COVID-19 induced pause, Webb and Thomson’s views diverge significantly.
Webb pointed to the issues already explored and several others as evidence the census’ return to operations has been problematic. He succinctly summarized his thoughts on the matter.
“I guess I would say poorly,” Webb said. “That is the one word answer.”
Thomson, on the other hand, expressed a great amount of optimism related to the state’s high level of households that have been enumerated and pointed to as successes in the face of many challenges.
“I think we’re doing well,” Thomson said. “Arizona has the best response rate this year compared to the last two decades, and I think we achieved that in a year where there was a global pandemic speaks very highly of both the work of the Census Bureau and our local partners … cumulatively and collaboratively we were able to build a really strong campaign that drove response.”
Webb’s worries are directed toward Coconino County in particular. Both he and Drum’s concerns about a potential undercount are serious enough that they both suggested a special enumeration may be necessary.
Drum explained that a special enumeration is where the Census Bureau can return to a community to perform a recount after the regular census has concluded. Communities may request a special enumeration if a significant undercount occurred.
However, Drum also explained there are challenges associated with actually performing a special enumeration. Recounts can be expensive, and it falls on the community to prove an undercount occurred and that it was the fault of the Census Bureau for those costs to be subsidized. Otherwise, the community must front the costs itself, which Drum explained can cost tens of thousands of dollars for a community the size of Flagstaff.
An undercount among minorities
Adan Chavez, deputy director for the National Association of Latino Elected and Appointed Officials National Census Program (NALEO), explained that many minorities, such as Native Americans in northern Arizona and Latinx folks in various parts of the state, are at a particularly high risk of being undercounted.
Chavez said his organization is focused on ensuring higher rates of Latinx participation in the census, with a particular focus on Latinx children.
“We’re known for the work we do around the census every 10 years to make sure there’s a complete and accurate count of our community nationwide,” Chavez said. “I have been supporting and co-leading both of our national campaigns, one that gets a full and complete count of Latinos and the other that gets to making sure we have a full and complete count of Latino children. This is because in 2010 there were about one million who were completely missed on their census forms and about 40% of those, four hundred thousand, were Latino children.”
In the current census, Chavez said he has serious concerns about lagging response rates among minority communities nationwide.
“We do know it just happens to be that counties that are majority Latinos are lagging and have been lagging behind from anything between 10 and 12 points compared to non-Latino counties,” Chavez said. “And Black majority counties have also been lagging behind.”
Chavez said Arizona is no different and has in fact been a state of particular concern. He said internal research at the NALEO found Arizona to be one of the states with the lowest levels of census participation nationwide. Official census data also shows Arizona lagged behind the majority of states, ranking No. 32 out of 50 in terms of response rates.
Chavez also explained that certain census response rate figures can be somewhat misleading. While high rates may accurately show that a certain state or region is performing well overall, it may obscure significant undercounting within certain communities.
This same idea applies to response rates in the Navajo Nation. Webb explained that when Arizona households were estimated to be 99% enumerated, the Navajo Nation had not quite reached 80%.
As for Arizona’s Latinx community and Chavez’s concerns that they are at risk of being undercounted, even Latinx folks in urban areas, he pointed to the impact of COVID-19 as a significant factor.
“I would be remiss to not mention that our community has been hardest hit by COVID,” Chavez said. “Many of them are getting sick at much faster rates and it just so happens to be that many of them are essential workers, and they are on the front lines of providing services and other relief during this moment. Right now, what’s top of the mind for our community, oftentimes is being able to have food on the table, a roof over their heads, having access to a good-paying job, so the census is not necessarily a priority.”
Chavez also pointed to the controversy surrounding a citizenship question on the census as a major factor behind lagging Latinx participation.
He explained that the question originally pushed by the Trump administration would have inquired about respondents’ citizenship status. This created a significant political controversy in terms of whether or not such a question would be necessary, what the information would be used for and the impact on individuals who fear harassment and/or legal action concerning their citizenship or immigration status.
While the question was ultimately never included within the census, Chavez said the damage had already been done.
“Still to this point when we conducted our research testing right after the debate and right after the court and Trump administration decided not to move forward with the citizenship question, we saw more than half of Latinos we polled thought there was going to be a citizenship question,” Chavez said. “So that is a challenge that has been particularly affecting Latino communities across the state.”
An undercount among students
Aside from minority communities, college students are also a great concern for potentially being undercounted. This could have significant ramifications for a college town like Flagstaff where a large portion of the population are students.
Webb explained there is often a lot of misinformation regarding where college students should be counted.
“We have a lot of college students and college students often think ‘Oh, I’m just counted on my parents’ census and that’s not the way that’s supposed to work,” Webb said. “You’re supposed to be counted where you live most of the year, and if you’re going to school, you’re spending nine months out of the year in this community.”
Webb said the coronavirus only made matters worse as the census intends for people to report where their residence was on April 1. This was right around the time the NAU campus shut down and many students left Flagstaff for the remainder of the spring semester.
While Webb stressed that in this case most students should have been counted at their Flagstaff address, there was a great amount of confusion regarding this and likely significant undercounting.
Drum further explained that students are often confused about where they are considered residents and students’ parents often count them when they fill out the census.
“Historically, what has been really challenging with university students is parents typically get the census and then they count their kids,” Drum said. “If you go to college out of state or out of the town you live in, sometimes kids change their voting, sometimes kids move full time and a lot of kids go back and forth.”
Drum said this is also a problem that applies more to students living off campus than on campus.
“NAU submitted data to the Census Bureau for everyone who lived on campus, so students who lived on campus in the spring, whether they were there past spring break, they still got submitted to the Census Bureau as they live here,” Drum said.
Drum also stressed off-campus students still account for a significant portion of the student population, and as previously discussed, many were likely not counted.
Drum said this puts Flagstaff in a difficult situation as many students were only gone temporarily and currently live in the city without having been counted.
“Students use services, they use our roads, they use our health care system,” Drum said. “Maybe you weren’t here, but next year there will be those 10,000 students living off campus in our community for the next ten years. Students being undercounted is a chronic problem made worse by COVID-19.”
Outreach efforts were expanded to the student population, but Drum said it remains difficult to know how successful they were.
Politics and the fluctuating end date
The various local, regional and on-the-ground challenges in the enumeration came as a result of the COVID-19 pandemic and other persistent factors are not the only things to have complicated the process. Politics and constantly shifting guidance from Washington have been a major factor in the current census, particularly in relation to the end date.
Prior to the COVID-19 pandemic, the Census Bureau planned to have both the self-response phase, where residents can respond to the census on their own via mail, online and over the phone, and the nonresponse follow up — door-to-door and in-person enumeration — run from the middle of March until July 31, with final results due to the office of the president by Dec. 31.
As previously discussed, self-response rates were impacted by a variety of factors and on-the-ground efforts were temporarily halted altogether for health and safety reasons.
To remedy this Census Bureau director Steven Dillingham and commerce secretary Wilbur Ross, who supervises the census, announced a plan in a joint statement on April 13 to extend the self-response period and field operations to Oct. 31 and have the final results to the president by April 30, 2021.
However, The New York Times reported July 28 the Trump administration and Senate Republicans were signaling their desire to cut the counting short. Within a few days, the Census Bureau announced it was moving the end date up a month to Sept. 30.
Tim Olson, the associate director of field operations for the Census Bureau, was quoted in an email to another official as having given a strongly worded statement on the plausibility of meeting the December 2020 final deadline.
“It is ludicrous to think we can complete 100% of the nation’s data collection earlier than [Oct. 31] and any thinking person who would believe we can deliver apportionment by [Dec. 31] has either a mental deficiency or a political motivation,” Olson said in the email.
Similarly, Christa Jones, the chief of staff to the director of the Census Bureau, was quoted in an email as indicating that the order to move the date came from outside the bureau.
However, these changes were far from the only to occur to the end date of counting operations. As the planned Sept. 30 counting deadline drew closer, a string of events saw the end date fluctuate several times in September and early October.
On Sept. 26 a U.S. District Court judge ruled in favor of a lawsuit to have the deadline once again extended to Oct. 31. However, the Census Bureau proceeded to announce a self-imposed deadline of Oct. 5.
Once the Oct. 5 deadline was shot down in court, the end date moved back to Oct. 31 until the U.S. Supreme Court took on an appeal to the decision, overruling the lower court and moving the end date to Oct. 15. Finally, on Oct. 13, the bureau put out a statement affirming it would comply with the deadlines imposed by the high court, and officially halted in-person enumeration operations and stopped receiving self-responses Oct. 15.
Supreme Court Associate Justice Sonia Sotomayor, who opposed the Supreme Court’s order to impose an Oct. 15 deadline and wrote the dissenting opinion in the case, was critical of both the Census Bureau’s conduct and the court’s order. She pointed out the sudden reversal and contradictory nature of the bureau first extending operations and the deadline after claiming the original December deadline could not be met “under any circumstances,” but then reestablishing the December deadline and cutting the already shortened enumeration process even further.
Sotomayor maintained the reversal was “arbitrary and capricious” and that conforming to the December 2020 final deadline “at the expense of the accuracy of the census is not a cost worth paying,” particularly when the bureau could not, in her eyes, articulate a valid reason for the move. In her criticism of the majority opinion, which was written by her colleagues on the court, Sotomayor suggested that rulings with “such a painfully disproportionate balance of harms” are rarely handed down by the court.
Sotomayor suggested the sudden reversal of the bureau, and by extension the Trump administration, was likely politically motivated and closely related to President Trump’s desire to exclude undocumented immigrants from the count during his current term.
The fluctuating end date predictably impacted local operations. Within the key weeks in September and early November, Webb explained the Complete Count Committee had to take it day by day, not knowing if their operations would suddenly be halted or continue on for several more weeks.
Webb said that specific parts of the strategy, like outreach advertising, were likewise affected. Coordinating with media outlets to purchase ads can be costly and need to be arranged ahead of time, and when there was uncertainty about how late in October the counting would go, Webb explained that the Coconino County Complete Count Committee canceled ad buys planned for later in the month. Generally, Webb described the uncertainty as being left in a “limbo” and only adding to the census’s challenges.
“That sent all of us scrambling to figure out what we were going to do,” Webb said.
The census is, no doubt, an incredibly important process, oftentimes playing a large role in elections at multiple levels, factoring heavily in the distribution of federal funds and shaping communities for the decade to come.
But a mix of the COVID-19 pandemic, politics and longstanding enumeration difficulties have made the 2020 census one of the more challenging counts in recent memory, and have raised the potential for an undercount in the eyes of many who are closely involved.
As it becomes clearer just how accurate the count really is and to what extent it impacts society, the 2020 census may indeed be remembered as either a well-executed process under the most challenging of circumstances or yet another disaster in such a pivotal year in American history.