← 2010 April 1, 2020 2030  →

Seal of the U.S. Census Bureau

General information Country United States Topics Census topics

• People and population
• Race and ethnicity
• Families and living arrangements
• Business and economy
• Employment
• Housing
• Income and poverty Authority U.S. Census Bureau Website www .census .gov Results Total population 331,449,281 ( 7.4%) Most populous  state California (39,538,223) Least populous  state Wyoming (576,851)

The 2020 United States census represented the 24th iteration of the decennial United States census , a constitutionally mandated undertaking. The official reference date for this nationwide count, known as Census Day, was meticulously designated as April 1, 2020. This particular census marked a significant, albeit arguably overdue, shift in methodology. Beyond a limited, largely unpublicized pilot study that occurred during the 2000 United States census , this was the inaugural time the U.S. census genuinely embraced modern convenience, offering citizens the option to respond online or via telephone. Naturally, for those who prefer the tactile reassurance of paper or perhaps distrust digital interfaces, the traditional paper response form remained available, mirroring the methods employed in all preceding censuses.

Predictably, this grand civic exercise did not unfold without its own unique set of challenges. The enumeration period found itself squarely in the midst of the COVID-19 pandemic in the United States , an unforeseen global event that demonstrably complicated and disrupted its intricate administration. Despite these formidable obstacles, the census successfully recorded a resident population of 331,449,281 individuals across the 50 states and the nation’s capital, Washington, D.C. . This figure reflected a discernible increase of 7.4%, equating to an additional 22,703,743 people, since the preceding 2010 United States census . While the raw increase was substantial, the observed growth rate was, in fact, the second-lowest ever recorded in the nation’s census history, a rather unenthusiastic surge considering the decade. Paradoxically, the net increase itself still ranked as the sixth-highest in historical terms, a testament to the sheer scale of the American population. A notable milestone of the 2020 census was that for the first time, all ten of the most populous states each surpassed the ten million resident mark, and similarly, the ten most populous cities each exceeded one million residents.

The profound data collected during this census held significant ramifications, primarily determining the distribution of electoral votes for the hotly contested 2024 United States presidential election . However, the accuracy of this foundational data was later called into question. A subsequent review conducted by the esteemed United States Census Bureau itself unearthed significant miscounts, particularly noting undercounts in several minority populations and, rather inconveniently, in several states. One might wonder if precision is truly attainable in such a sprawling endeavor.

Background

A 2020 U.S. census yard sign in Columbus, Ohio . A rather quaint attempt at public engagement.

As unequivocally stipulated by the revered United States Constitution , the U.S. census has been a steadfast decennial ritual, diligently executed every 10 years since its inaugural iteration in 1790. The 2010 United States census served as the immediate predecessor to the 2020 count. For those residing within the United States aged 18 and older, participation in the census is not merely a suggestion, but a legal obligation, requiring truthful responses to all posed questions, as enshrined in Title 13 of the United States Code . The sanctity of personally identifiable information is, theoretically, paramount; the Census Bureau explicitly states it will never release such private data. However, the National Archives and Records Administration maintains the potential to unveil the original census returns in 2092, assuming, of course, that the enigmatic 72-year rule remains unaltered by the fickle whims of future legislation.

On the pivotal census reference day, April 1, 2020, the projected resident population of the United States (encompassing the 50 states and Washington, D.C. , but notably excluding overseas territories, military personnel, and civilian U.S. citizens residing abroad) stood at an estimated 329.59 million. This represented a 6.76% increase from the population recorded during the 2010 census , indicating a steady, if not exactly explosive, demographic expansion.

Purpose

One might assume the purpose of such an undertaking is self-evident, yet a closer look reveals layers of bureaucratic necessity and political consequence.

Reapportionment

See also: United States congressional apportionment

Allocation of U.S. congressional districts following the 2020 census. A rather complex dance of numbers, wouldn’t you say?

The most immediate and politically charged outcome of the 2020 census results is their direct influence on the composition of the U.S. government . These figures precisely dictate the number of seats each state is allocated in the House of Representatives . Consequently, this also determines the corresponding number of electors each state commands within the crucial Electoral College , a system that impacts elections from 2022 right through to 2032. The Census Bureau , with its characteristic gravitas, formally announced these critical apportionment figures on April 26, 2021.

The shift in political power, as reflected in congressional seats, was palpable: 13 states experienced alterations to their representation.

• California , New York , Illinois , Michigan , Ohio , Pennsylvania , and West Virginia each found themselves losing a single seat. A clear indication of demographic shifts, or perhaps, a failure to retain their populace.
• Conversely, Colorado , Florida , Montana , North Carolina , and Oregon each gained a seat, suggesting areas of more robust growth or, at least, better retention.
• Texas , ever the outlier, managed to secure an impressive two additional seats, a testament to its burgeoning population.

Interestingly, this reshuffling of seats proved to be less dramatic than what many independent analysts had initially forecast, suggesting that even the most meticulous predictions can fall short of reality.

Redistricting

See also: 2020 United States redistricting cycle

Beyond the federal stage, census data serves as the indispensable blueprint for state and local officials. It is the foundation upon which they redraw the boundaries for various electoral and administrative districts. This includes, but is certainly not limited to, congressional districts , state legislative districts, and even school districts. The integrity of these boundaries, and by extension, the fairness of representation, hinges entirely on accurate census data.

Federal funding distribution

One might overlook the fiscal implications, but census data is a veritable goldmine for dozens of federal programs. These programs rely heavily on the granular population data to direct funding effectively to state and local areas. The results of the census are instrumental in determining how over $675 billion in federal funding is allocated annually. These funds are critical for a vast array of public services, including the maintenance and development of roads, the establishment and support of schools, the operation of hospitals and health clinics, and the provision of essential emergency services, among countless other necessities. Without accurate counts, one can only imagine the ensuing chaos in resource distribution.

Major design changes

The 2020 census , in a rare embrace of modernity, distinguished itself as the first U.S. census to fully operationalize an internet response option. Furthermore, it marked a significant departure from previous methodologies by extensively integrating technology for managing and executing fieldwork, moving away from the cumbersome reliance on paper.

Key design changes, for those interested in the minutiae of bureaucratic evolution, included:

• Three Response Options: Citizens were offered a choice: the efficiency of the Internet, the familiarity of paper, or the directness of a phone call. Ultimately, every household that did not initially respond online was subsequently sent a paper form, ensuring a safety net. Households situated in areas identified with demonstrably low internet access were, quite sensibly, provided with a paper form from the outset.
• Multiple Languages: Acknowledging the diverse linguistic tapestry of the United States , the 2020 census facilitated responses in 12 non-English languages both online and via phone. Additionally, a comprehensive suite of language guides, glossaries, and identification cards was made available in an impressive 59 non-English languages, an admirable, if challenging, effort to ensure inclusivity.
• In-Office Address Canvassing: A stark contrast to the laborious methods of the 2010 and earlier censuses, where census workers physically traversed every street in the United States to verify addresses. The 2020 census employed a more technologically advanced approach, utilizing satellite imagery and GPS to pinpoint areas where housing structures had undergone changes. Fieldworkers were then dispatched to verify only these specific addresses in person, a more targeted and efficient, if less romantic, method.
• Digital Case Management: Census takers were equipped with secure iPhone 8 smartphones , transforming their daily operations. These devices served as their primary tools for receiving assignments, navigating to interview locations, communicating with supervisors, and submitting timesheets. Bespoke software was developed specifically to optimize these assignments, streamline management processes, immediately flag any issues that arose, and, crucially, reduce the necessity of redundant follow-up visits, a small victory for efficiency.
• Streamlined Follow-up Visits Using Existing Data Sources: The 2020 census leveraged existing government and third-party data to refine its follow-up procedures. This included identifying vacant households, predicting the optimal time of day for a visit to a particular residence, and even, after multiple failed attempts, counting and providing demographic characteristics for household members by drawing upon high-quality data from trusted existing sources. A pragmatic approach, one might say, when direct contact proves elusive.

Questions and data uses

Copies of the 2020 census questionnaire. For those who enjoy a good interrogation.

As mandated by the Census Act , the U.S. Census Bureau dutifully submitted its list of proposed questions to Congress on March 29, 2018. It is crucial to reiterate that the U.S. census is legally prohibited by Title 13 United States code from sharing any participant’s information with any other government agency. This confidentiality, a cornerstone of public trust, has been challenged in the past, but the Supreme Court has consistently upheld Title 13 to safeguard the privacy and confidentiality of the information provided. Based on these approved questions and a subsequent executive order, the 2020 census inquired about:

• The number of people living or staying at the respondent’s home on April 1, 2020. This fundamental query is used to establish the total population count and to ensure that every individual is counted precisely once, and in the correct geographical location, based on their residence on Census Day .
• Whether the home is owned or rented. This question contributes to valuable statistics concerning homeownership and rental trends, serving as a vital economic indicator, informing housing programs, and guiding critical urban planning decisions.
• The sex of each person in the household. This data is utilized to produce statistics that underpin the planning and funding of government programs, as well as the enforcement of laws, regulations, and policies designed to combat discrimination.
• The age of each person in the household. Understanding the age distribution is paramount. This information allows agencies to better comprehend the size and characteristics of various age groups, enabling them to plan and fund government programs that specifically support different demographics, from children to the increasingly numerous older populations.
• The race of each person in the household. Federal agencies rely on this data to monitor compliance with anti-discrimination provisions, most notably those enshrined within the Voting Rights Act and the Civil Rights Act .
• Whether a person in the household is of Hispanic, Latino, or Spanish origin. Similar to race data, this information is crucial for federal agencies in their ongoing efforts to monitor compliance with anti-discrimination provisions, including those under the Voting Rights Act and the Civil Rights Act .
• The relationship of each person in the household to each other. This question helps to delineate household structures and is used to plan and fund government programs that offer support to families, including single-parent households and other configurations that may qualify for additional assistance.

It is worth noting that the United States Census Bureau had initially proposed, and then rather conspicuously withdrew, plans to introduce a new classification category specifically for Middle Eastern and North African peoples. This decision followed a protracted dispute regarding whether this demographic classification should be subsumed under the existing “white ethnicity” category or be recognized as a distinct racial group. Bureaucracy, it seems, struggles with nuance.

Timeline

Average annual population growth rate in each county of the 50 states , Washington, D.C. , and Puerto Rico between 2010 and 2020, according to the U.S. Census Bureau . A visual representation of demographic flux.

The execution of a national census is a logistical behemoth, unfolding over a meticulously planned, if occasionally disrupted, timeline:

• January–March 2019: The U.S. Census Bureau commenced the establishment of 39 area census offices, laying the groundwork for the monumental task ahead.
• June–September 2019: The remaining 209 area census offices were opened. These offices served as critical hubs, providing essential support and management for the legions of census takers deployed across the nation to conduct the enumeration.
• August 2019: The Census Bureau executed its in-field address canvassing operation. Unlike previous, more exhaustive efforts, the 2020 census was the first modern census that did not necessitate in-person verification of every single address on the ground. Instead, a more sophisticated approach was adopted, leveraging satellite imagery , data from the U.S. Postal Service , and other current records to verify the vast majority of addresses. This allowed census workers to focus their in-person efforts exclusively on areas where changes in housing stock were identified.
• January 21, 2020: The official counting of the population began in the remote reaches of Alaska, with Toksook Bay earning the distinction of being the very first town to be enumerated. A rather symbolic start, one might say.
• April 1, 2020: Census Day was observed nationwide, marking the crucial reference date. By this point, households across the country should have received an invitation to participate in the 2020 census , with the now familiar three options for response: online, by mail, or via phone.
• April 2020: Census takers were scheduled to begin following up with households around selected colleges and universities, and also to conduct quality check interviews, though these specific operations faced initial delays.
• May 2020: The Census Bureau was slated to begin following up with households that had not yet responded (the NRFU, or Nonresponse Followup, operation). This, too, was delayed, ultimately commencing on August 11 and concluding on October 31. In a rather abrupt decision in August 2020, this three-month NRFU enumeration period was compressed to a mere two and a half months, abruptly ending on October 15, 2020.
• September 23–24: Individuals experiencing homelessness were systematically counted by officials who visited shelters, soup kitchens, mobile food vans, and non-sheltered, outdoor locations such as tent encampments.
• October 15: Self-response data collection officially concluded, with the Census Bureau reporting that an impressive 99.9% of households had either self-responded or been counted by census takers.
• October 16, 2020: The entire counting operation officially ceased.
• December 31, 2020: The Census Bureau was initially expected to deliver the critical apportionment counts to the U.S. president . However, this deadline was subsequently delayed to April 30, 2021, illustrating the cascading effects of the pandemic.
• April 1, 2021: The Census Bureau was scheduled to transmit redistricting counts to the states. This vital information is, of course, used by states to redraw legislative districts in accordance with population shifts. This, too, was delayed, ultimately to no earlier than September 30, 2021.
• April 26, 2021: The population results were finally released for the country as a whole and for each individual state.
• August 12, 2021: The Census Bureau began the phased release of more detailed data, including breakdowns by race, ethnicity, sex, and age, as well as population numbers for counties, cities, towns, and other smaller geographical areas.
• May 25, 2023: Demographic and housing data about local communities (DHC) was made available.
• August 31, 2023: Congressional district summary files were released, providing granular data for legislative analysis.
• September 21, 2023: File A of detailed demographic and housing data was released, further enriching the public dataset.
• August 1, 2024: File B of detailed demographic and housing data was released, concluding the extensive data dissemination process.

Response rates

According to the Census Bureau , a respectable 60.0% of all U.S. households had successfully submitted their census questionnaire by May 22, 2020, utilizing either the online platform, traditional mail, or the telephone option. The vast majority of U.S. households , exceeding 95% of the total, received their initial invitation letter to self-respond via mail between March 12 and 20. Prior to the pervasive coronavirus pandemic , the remaining 5% of U.S. households , predominantly situated in rural areas, were originally slated for in-person visits by census takers in April and May, who would drop off invitation letters directly to residents. This operation, like so many others, was regrettably delayed. However, most census offices gradually resumed work in mid-May. By July 14, 2020, the self-response rate had climbed to 62.1%, encompassing 91,800,000 households. For context, the self-response rate stood at 66.5% in 2010 and 67.4% in 2000, indicating a slight, though not catastrophic, decline in 2020.

In an update disseminated on October 19, 2020, the Census Bureau declared that an impressive 99.98% of addresses had been accounted for, with every single state, save one, surpassing a 99.9% completion rate. Paper responses bearing a postmark on or before October 15 were deemed valid for processing, provided they reached the processing center by October 22. A narrow window, but a window nonetheless.

Marketing and partnerships

As is customary with such large-scale public endeavors, the 2020 census strategically leveraged a vast network of trusted local voices across the nation. These partners were tasked with the crucial role of raising awareness, addressing public inquiries, and, most importantly, encouraging community members to actively participate. Hundreds of local “complete count committees” diligently dedicated their resources to these efforts nationwide, a testament to the decentralized nature of civic engagement in the United States .

VMLY&R, formerly known as Young & Rubicam , successfully secured the coveted Integrated Communications Contract for the 2020 census campaign in August 2016. As the primary agency of record for this substantial contract, VMLY&R assembled an integrated team, aptly named Team Y&R. This team comprised various subcontractors, each specializing in critical areas such as outreach to minority populations, digital media engagement, and earned media strategies, among others.

In March 2019, the campaign unveiled its official tagline for the 2020 census : “Shape your future. START HERE.” This slogan was meticulously crafted based on extensive research, which aimed to identify the most effective messaging strategies to reach and motivate all segments of the population, particularly those historically classified as “hard to count.” One might appreciate the effort, even if the sentiment borders on the saccharine.

Implementation problems

Chinese and Spanish language flyers in Brooklyn promoting the 2020 U.S. census . A necessary adaptation to a diverse populace.

No complex operation is without its share of unfortunate missteps, and the 2020 census was no exception. The printing company Cenveo was awarded a substantial $61 million contract in October 2017 to produce the myriad census forms and reminder notices. However, less than four months later, Cenveo declared bankruptcy, a rather inconvenient turn of events. The inspector general of the U.S. Government Publishing Office subsequently concluded that the agency had failed to adequately assess the company’s financial stability and had improperly permitted Cenveo to lower its bid after other bids had already been unsealed. A rather glaring oversight, one might observe.

The onset of the coronavirus pandemic inevitably introduced further complications, causing widespread delays to census field operations and significantly impacting the counts of individuals experiencing homelessness and those residing in group quarters. As of April 1, 2020, Census Day , the Census Bureau maintained its ambitious plan to complete the entire count by the end of the year, a goal that would prove increasingly difficult to achieve.

COVID-19 pandemic emergency

On March 18, 2020, the U.S. Census Bureau issued a press release from Director Steven Dillingham , announcing a two-week suspension of 2020 census field operations, effective until April 1, 2020, citing the escalating COVID-19 pandemic . This initial pause was merely the beginning. On March 27, 2020, the agency further announced the temporary suspension of in-person interviews for its ongoing surveys. The Census Bureau claimed that necessary staffing adjustments at its call centers, implemented to adhere to public health guidance, had resulted in “increases in call wait times, affecting different languages at different times.” Curiously, according to its own internal documentation, the U.S. Census Bureau continued to compensate 2020 census employees even during this period when field operations were officially suspended.

On March 28, 2020, another press release from the U.S. Census Bureau extended the suspension of 2020 census field operations for an additional two weeks, through April 15, 2020. Census Bureau officials communicated to the media that on March 27, 2020, an employee at the agency’s National Processing Center in Jeffersonville, Indiana, had tested positive for COVID-19 . Despite the broader suspension, this processing center remained operational, with the agency asserting a transition to “the minimum number of on-site staff necessary to continue operations.” On April 10, 2020, the agency announced steps to make “more employees available to respond to requests” at the beleaguered call centers.

A joint statement issued on April 13, 2020, by U.S. Department of Commerce Secretary Wilbur Ross and U.S. Census Bureau Director Steven Dillingham outlined further operational adjustments to the 2020 census , all ostensibly in response to COVID-19 health and safety concerns. The statement indicated that “steps [were] being taken to reactivate field offices beginning June 1, 2020,” and that “in-person activities, including all interaction with the public, enumeration, office work and processing activities, [would] incorporate the most current guidance to promote the health and safety of staff and the public.” This included the provision of “personal protective equipment (PPE) and social distancing practices.” Crucially, this release also stated, “in order to ensure the completeness and accuracy of the 2020 census , the Census Bureau is seeking statutory relief from Congress of 120 additional calendar days to deliver final apportionment counts.” Under this revised plan, the Census Bureau would extend the window for field data collection and self-response to October 31, 2020, allowing for apportionment counts to be delivered to the president by April 30, 2021, and redistricting data to be delivered to the states no later than September 30, 2021. A rather significant shift in deadlines, driven by circumstances beyond their control.

On April 15, 2020, U.S. Census Bureau Director Steven Dillingham responded in writing to Department of Commerce Inspector General Peggy E. Gustafson. This was in reply to a March 12, 2020, memo from the Office of the Inspector General requesting details on the Census Bureau’s plans to address the COVID-19 emergency by March 20, 2020. The inspector general’s memo specifically inquired about the Bureau’s strategy to ensure the safety of its staff and enumerators. Dillingham’s April 15 letter provided the following assurances:

The Census Bureau is closely coordinating the acquisition of needed PPE materials for field and office staff through the Department of Commerce’s Coronavirus Taskforce. Federal partners include the Department of Homeland Security and the Centers for Disease Control . We have generated and submitted estimates for equipment needs. On April 15, 2020, the Agency’s internal task force met and discussed our estimates for needed equipment, potential delivery dates, and budget implications. We continue to monitor the situation and make adjustments as necessary.

To ensure the completeness and accuracy of the 2020 census , the Census Bureau is seeking statutory relief from Congress of 120 additional calendar days to deliver final apportionment counts.

Under this plan, the Census Bureau would extend the window for field data collection and self-response to October 31, 2020, which will allow for apportionment counts to be delivered to the President by April 30, 2021, and redistricting data to be delivered to the states no later than September 30, 2021.

On April 24, 2020, Dillingham and other Census Bureau officials briefed the House Committee on Oversight and Reform on the agency’s response to the COVID-19 emergency. This briefing followed numerous requests from the committee since March 12, 2020, including a last-minute cancellation on April 20, 2020. During the briefing, Albert E. Fontenot Jr., the associate director for decennial census programs, clarified that the bureau was planning a “phased start to many of our census operations” rather than a nationwide resumption of field operations on June 1, 2020, as initially announced. He explained that operations would recommence at varying times in different regions of the country, contingent upon federal, state, and local public health guidance, as well as the availability of personal protective equipment . The strategy prioritized the reopening of mail processing centers and census offices, with the bureau pledging to notify Congress as operations restarted. This was, of course, despite the fact that the National Processing Center and Area Census Offices had, in fact, remained open throughout.

Beginning May 4, 2020, the U.S. Census Bureau commenced publishing dates as it claimed to initiate a “phased restart of some 2020 census field operations in select geographic areas.” They stated they had “ordered personal protective equipment (PPE) for all field staff, including those that work in a field office. These materials will be secured and provided to staff prior to restarting operations.” Publicly available procurement data, however, revealed that an award for non-medical, reusable face masks for area census offices, valued at $5,001,393.60, was signed on April 28, 2020, and granted to Industries for the Blind and Visually Impaired, Inc. . Around the same period, two contracts for hand sanitizer were awarded to Travis Association for the Blind : one signed on May 9, 2020, for $57,390.00, and another on May 13, 2020, for $557,251.20. Both contracts listed the principal place of performance as Jeffersonville, Indiana. The agency also determined that face shields were a necessary protective measure against COVID-19 exposure for employees, yet these were only provided to personnel at the headquarters and national processing centers. An OSHA complaint, lodged from Oklahoma City on May 1, 2020, alleged that employees were unable to maintain social distancing and were not supplied with adequate personal protective equipment such as gloves and masks. This complaint rather pointedly indicated that the office was operational before the Census Bureau’s officially published office restart date of May 4, 2020.

Additional “restart” dates, commencing May 18, were published on May 15, 2020, for other geographic areas spanning eleven states. On that very same day, an OSHA complaint was recorded from St. Louis , detailing concerns about desks remaining closely packed without physical dividers, improper sanitation practices, and a lack of remote work options for high-risk employees. The published restart date for the St. Louis Area Census Office was May 11, 2020, again suggesting a discrepancy between official announcements and on-the-ground reality.

On May 21, 2020, procurement information for two more contracts was entered into the Federal Procurement Data System . One contract, for $1,502,928.00, was awarded to Industries for the Blind and Visually Impaired, Inc. for hand sanitizer, while another, for $7,053,569.85, for four-ounce (118 ml) hand sanitizers, went to NewView Oklahoma, Inc. Both contracts, predictably, listed Jeffersonville, Indiana as the principal place of performance.

May 22, 2020, brought two further contracts: a disinfectant wipes contract for $3,137,533.00, awarded once more to Industries for the Blind and Visually Impaired, Inc. , and a contract for $2,107,000.00 to NewView Oklahoma for blue nitrile gloves. Both, again, specified Jeffersonville, Indiana, as the principal performance location.

A press release on May 22, 2020, announced May 25 “restart” dates for ten more states. An OSHA complaint, however, had been filed from Concord, California , on April 3, 2020, alleging at least two confirmed cases of COVID-19 that were not recorded on OSHA 300 logs, and that employees were working in close quarters without disinfection of shared equipment such as headsets, laptops, and tablets. The published restart date for the Concord, California, Area Census Office was May 25, 2020, again raising questions about the actual timeline of operations.

Offices were also reopened in the areas of “American Samoa , the Commonwealth of the Northern Mariana Islands , Guam and the U.S. Virgin Islands in preparation for resuming operations for the 2020 Island Areas Censuses” on May 22, 2020.

On May 29, 2020, a press release was published announcing the “restart” of operations in seven additional states and the Washington, D.C. , area, commencing from the week of June 1. Yet another OSHA complaint had been lodged from Austin, Texas , on May 27, 2020, asserting that CDC guidelines were not being adhered to, employees were unable to practice social distancing , and individuals exhibiting flu-like symptoms and positive COVID-19 test results continued to report for work. This clearly indicated that the office was active before the Census Bureau’s officially published office restart date of June 1, 2020.

In a June 5, 2020, press release , the U.S. Census Bureau announced that additional area census offices (ACOs) would “restart” on June 8. They declared that with “these additions, field activities have restarted in 247 of 248 area census offices stateside, all ACOs in Puerto Rico and the island areas, and 98.9% of the nation’s update leave workload will have resumed.” The June 5 press release was subsequently reissued on June 9, 2020, with the notable addition of a June 11 “restart” at the Window Rock, Arizona, Area Census Office. Days later, perhaps predictably, the Navajo Nation began reinstating lockdown restrictions and curfews due to an alarming surge in new cases, highlighting the inherent risks of resuming operations too soon.

A June 12, 2020, press release further shared that the update leave (UL) operation had resumed, alongside the fingerprinting of selected applicants. The agency declared that the update enumerate (UE) operation would restart on June 14 “in remote parts of northern Maine and southeast Alaska ,” where employees would update the Census Bureau’s address list and interview households for the 2020 census . They confidently asserted that “all census takers have been trained on social distancing protocols, and will be issued personal protective equipment (PPE) and will follow local guidelines for their use.” The June 12 press release also indicated that the communications campaign had been adapted due to the pandemic and would continue through October, “the end of 2020 census data collection operations,” with additional paid media planned for July, August, and September. However, a July 15 list of media vendors only showed plans extending through the end of July, a minor inconsistency in a sea of them.

On August 3, 2020, the Census Bureau announced a rather abrupt decision: field collection would now conclude on September 30, rather than the previously planned October 31. This compression of the timeline sparked considerable concern. A leaked internal document revealed that Census Bureau career officials themselves had determined that initiating Nonresponse Followup Operations under this revised plan would, in fact, jeopardize the health and safety of employees, explicitly stating, “These ACOs will have to deploy staff regardless of the COVID-19 risk in those areas to open on these dates.” On September 8, 2020, Mark H. Zabarsky, Principal Assistant Inspector General for Audit and Evaluation, issued an alert on behalf of the Department of Commerce Office of Inspector General , stating that the number of COVID-19 -related safety issues reported via hotline complaints had tripled between July 1 and August 21. A clear indication that the rush to complete the count was not without its human cost.

State rankings

Population change 2010–2020. A rather dry accounting of demographic shifts.

City rankings

See also: List of United States cities by population

The demographic landscape of the United States is, of course, heavily influenced by its urban centers. The 2020 census provided a fresh snapshot of these bustling hubs, revealing their populations, land areas, and densities, categorized by their respective Census Regions .

Citizenship question debate

The U.S. decennial census is, at its core, a tool for allocating federal funds, grants, and support to states. A seemingly straightforward administrative function, until the perennial “citizenship question” inevitably rears its head. This particular query had been a standard feature of the census until 1950, after which it was removed, though a question regarding place of birth persisted. In a January 2018 memo, an initial evaluation by Census Bureau officials rather sensibly advised against reintroducing such a question, arguing that compiling citizenship data from existing administrative records was demonstrably more accurate and significantly less expensive. However, Wilbur Ross , the then-Secretary of the United States Department of Commerce , the very entity overseeing the Census Bureau , unilaterally decided that the administrative approach alone would not suffice.

Consequently, the Census Bureau announced in March 2018 its intention to add a question directly related to citizenship for the 2020 census : “Is this person a citizen of the United States ?” Secretary Ross informed Congress that, for the 2020 census , these citizenship numbers were deemed necessary to enforce the Voting Rights Act’s crucial protections against voting discrimination. This justification, however, quickly drew fire, with Democrats in Congress accusing Secretary Ross of outright lying, claiming the citizenship question had been requested by the Justice Department and then approved by him.

Upon the bureau’s announcement, a chorus of criticism erupted from various state and city officials. Their concerns centered on the very real risk of discouraging participation from immigrant communities, which would inevitably lead to an undercount of the population. They also, quite reasonably, questioned the underlying motives of Secretary Ross in reintroducing the question. This controversy swiftly escalated into three simultaneous, separate federal lawsuits, initiated in the district courts of New York , Maryland , and California . Amidst this legal maelstrom, the Census Bureau conducted a test census in June 2019, involving approximately 480,000 households. The purpose was to assess the potential impact of the citizenship question on participation rates and to prepare the bureau’s staffing and counting methodologies for the anticipated reduction in responses.

During the discovery phase of these trials, a rather damning revelation emerged in May 2019: documents indicated that the late Thomas B. Hofeller , a Republican strategist widely recognized as an architect of gerrymandering , had concluded that adding the census question could facilitate the creation of electoral maps that “would be advantageous to Republicans and non-Hispanic whites.” Hofeller subsequently drafted the DOJ letter that provided the ostensible justification for the policy, claiming it was essential for enforcing the 1965 Voting Rights Act . Following this rather inconvenient discovery, the United States House Committee on Oversight and Reform issued subpoenas, demanding that the Department of Justice provide all materials related to the census question and seeking testimony from both Commerce Secretary Wilbur Ross and United States Attorney General William Barr , with a view to judging them in contempt. The Trump administration, on June 12, 2019, asserted executive privilege over certain portions of the requested documents. As a direct consequence, the House committee promptly voted along party lines to hold both Ross and Barr in contempt that very same day. The full House followed suit, voting to hold Ross and Barr in contempt on July 17, 2019, in a 230–198 vote that, predictably, also fell along party lines.

New York District Court and subsequent Supreme Court case

See also: Department of Commerce v. New York

A significant lawsuit, spearheaded by New York state’s attorney general Barbara Underwood and supported by seventeen other states, fifteen cities, and various civil rights organizations, was filed in the United States District Court for the Southern District of New York . During the discovery phase of this trial, new and rather inconvenient information surfaced: Secretary Ross had engaged in prior discussions with Steve Bannon before March 2018 regarding the intention to add the citizenship question. This directly contradicted statements Ross had previously made to Congress in March. This revelation prompted District Judge Jesse M. Furman in September 2018 to request that Ross allocate a day in his schedule to provide a deposition to the court concerning the addition of the census question, prior to the planned commencement of the trial in November.

The Trump administration responded by filing a writ of mandamus with the United States Supreme Court , seeking to postpone the trial and defer any direct involvement with Ross until the trial’s official start. The Supreme Court issued an order that permitted the trial, United States Census Bureau v. State of New York, to proceed, but it did agree to postpone Ross’s deposition until after the trial had begun. Furthermore, the Supreme Court consented to treat the writ of mandamus as a writ of petition and granted certiorari to review the question raised by the government: specifically, whether a district court possesses the authority to request a deposition from a high-ranking executive branch official on a matter related to a trial before any evidence has even been presented.

Judge Furman ultimately ruled in January 2019 that the addition of the citizenship question to the census was unlawful. He asserted that “the decision to add a citizenship question to the 2020 census  – even if it did not violate the Constitution itself – was unlawful for a multitude of independent reasons and must be set aside.” The Justice Department subsequently filed a petition for writ of certiorari before judgment to have the case heard directly by the Supreme Court , thereby bypassing the standard appeal process that would have seen it heard by the Second Circuit . This expedited request was driven by the looming deadline of June 2019 for the publication of the census forms. The Supreme Court accepted the petition related to Furman’s ruling on February 15, 2019, a matter distinct from the question of Ross’s deposition, and oral arguments for the case were heard on April 23, 2019.

The Supreme Court delivered its decision on June 27, 2019, effectively rejecting the Trump administration’s stated rationale for including the question. While the Court majority concurred that the question was permissible under the Enumeration Act , they also upheld the District Court’s authority to demand further explanation from the Department of Commerce regarding the question, as per the Administrative Procedures Act (APA). Crucially, they also agreed that the explanations Commerce had provided thus far appeared to be “contrived” and merely pretextual, leaving open the possibility that Commerce could offer a more convincing rationale. The case was remanded back to the District Court, allowing Commerce an opportunity to provide a more satisfactory explanation for the question’s inclusion. The District Court would then determine if this revised rationale was sufficient before permitting the question to appear on the census. This entire process, of course, hinged on these steps being completed before the self-imposed deadline for printing the census forms. On July 7, the DOJ announced its intention to replace its entire legal team handling the question. However, on July 9, Furman rejected the DOJ ’s action, stating that reasons must be provided for the withdrawal of each attorney and pointing out that the administration had been insisting for months that the matter needed to be resolved by July 1.

The American Civil Liberties Union (ACLU) has since taken steps to introduce the aforementioned Hofeller evidence into the New York case, though it was not expected to be heard until late 2019, after the census forms were slated for publication.

California District Court case

The second lawsuit challenging the census question emerged in the United States District Court for the Northern District of California . This suit was initiated by the state of California itself, alongside several cities within its borders, and was overseen by Judge Richard Seeborg . In March 2019, Seeborg reached a conclusion remarkably similar to that of Judge Furman in New York , finding that the addition of the census question was unconstitutional and subsequently issuing an injunction to block its implementation. The government predictably appealed this decision to the Ninth Circuit before the Supreme Court remanded the case.

Maryland District Court case

A similar legal challenge, focusing on the intent behind the question, was brought by several immigrant-rights groups in the United States District Court for the District of Maryland . This case was presided over by Judge George J. Hazel . Hazel ruled in favor of the pro-immigration groups in April 2019, determining that the inclusion of an immigration question on the census was unconstitutional. The government, as expected, appealed his decision to the Fourth Circuit Court of Appeals .

The newly unearthed Hofeller evidence was presented to Judge Hazel while the case was under appeal during June 2019 at the Fourth Circuit . Hazel acknowledged that the new evidence “raises a substantial issue.” On June 25, 2019, the Fourth Circuit remanded the case back to Hazel’s District Court, instructing him to review the additional evidence and determine if it demonstrated discriminatory intent. Should Hazel find such intent, he would possess the authority to issue an injunction against the inclusion of the census question during a new discovery phase, irrespective of the Supreme Court’s decision in Department of Commerce v. New York. This potential action would effectively render the question moot, as the census forms would, by that point, need to be published without the citizenship question to meet mailing deadlines.

Subsequent actions

Following the Supreme Court’s decision in Department of Commerce, President Trump publicly stated his intention to find a way to delay the census for as long as possible, hoping to resolve the judicial matter. On July 2, 2019, the Department of Justice (DOJ) announced that the citizenship question would not be included in the census, and the Commerce Department commenced printing census forms without it. However, the very next day, Trump rather defiantly insisted that his administration was “absolutely moving forward” with the citizenship question, and the Justice Department subsequently confirmed in court that it had been instructed to find a legal avenue to include it in the census. A clear demonstration of conflicting directives.

In response to an order from Judge Hazel , the Justice Department affirmed on July 5, 2019, its intention to pursue a route to add the citizenship question to the census, though at that specific time, it did not specify which route it would take. Hazel had requested this response to begin coordinating a schedule for further proceedings and discovery in both the New York and Maryland lawsuits with Judge Furman . On July 7, the DOJ announced its intention to replace its entire legal team on the case. However, Furman permitted the DOJ to dismiss only two of its eleven attorneys, writing in his July 9 rejection that the DOJ had “provide[d] no reasons, let alone ‘satisfactory reasons’, for the substitution of counsel.” Furman pointed out that the case had already exceeded the DOJ’s own previously requested deadline of July 1, and replacing counsel would only introduce further, unnecessary delays.

Separate from the judicial wrangling, Trump publicly declared he was considering using an executive order to mandate the inclusion of the citizenship question on the census. However, on July 11, he issued Executive Order 13880 , which instead directed the Department of Commerce to obtain citizenship data from other federal agencies, rather than directly via the census questionnaire. He asserted that “we are not backing down in our effort to determine the citizenship status of the United States population” and that data gleaned from other federal agencies would be “far more accurate” than a census question. A spokesperson for the Department of Justice stated that while the DOJ had agreed with Ross’s original plan to include the question, “Today’s executive order represents an alternative path to collecting the best citizenship data now available, which is vital for informed policymaking and numerous other reasons. Accordingly, the department will promptly inform the courts that the government will not include a citizenship question on the 2020 decennial census .” In addition to federal agencies, the Department of Commerce also began obtaining citizenship data from various state records.

Joe Biden , on his very first day in the presidency, January 20, 2021, issued an executive order that effectively revoked both Trump’s July 11 executive order and his July 21 memo. This action ensured that the census would revert to its standard practice of including the counts of undocumented immigrants within the final population numbers. The order, however, notably did not rescind a directive for the Census Bureau to utilize government records to produce block-level citizenship data.

Apportionment challenges

Alabama lawsuit

While the citizenship question itself was embroiled in litigation, the state of Alabama , along with one of its congressional representatives, Mo Brooks , initiated a separate lawsuit against the Department of Commerce and the Census Bureau in May 2018. Filed in the United States District Court for the Northern District of Alabama , this suit contended that the framers of the Constitution never intended for illegal immigrants to be included in the census count or, crucially, the apportionment base. Alabama believed that including such individuals would result in the state losing a congressional seat to other states that had experienced increased numbers of immigrants over the preceding decade. The Mexican American Legal Defense and Educational Fund sought to intervene on behalf of Latino voters, joined by the city of San Jose , California , Santa Clara County, California , and King County, Washington . These entities argued that the exclusion of illegal immigrants would negatively impact federal funding directed to their cities and counties. The motion to intervene was granted by the end of 2018.

As the citizenship question case continued its winding path through the courts, the Census Bureau began discussing alternative methods for obtaining immigration data. Attorney General Barr , referencing the Alabama suit, remarked that “for example, there is a current dispute over whether illegal aliens can be included for apportionment purposes. Depending on the resolution of that dispute, this data may be relevant to those considerations. We will be studying this issue.” Spurred by Barr’s comments, which suggested the government might not vigorously defend itself in the Alabama case, a coalition of fifteen states and other groups also moved to intervene. Their motion was granted by September 2019.

July 2020 memo

Main article: Trump v. New York

On July 21, 2020, President Trump signed a rather controversial memo directed to the Department of Commerce , titled “Memorandum on Excluding Illegal Aliens from the Apportionment Base Following the 2020 Census.” This memo contained explicit instructions not to include illegal immigrants in the census totals when calculating apportionment figures. The memo argued that the Constitution does not explicitly define which “persons” must be included in the apportionment base and cited past instances where some legal immigrants temporarily residing in the country were excluded, thereby attempting to justify this significant change. Legal and census experts immediately criticized this as an invalid interpretation, pointing to established case law that had historically supported the inclusion of “whole persons,” including illegal immigrants. The ACLU wasted no time, promptly announcing its intention to file a lawsuit against the administration over the memo. Common Cause , the city of Atlanta , and other groups and individuals filed the first suit seeking an injunction to prevent the government from implementing the memo, just a week after its signing, in the United States District Court for the District of Columbia .

On September 10, 2020, a three-judge panel of the United States District Court for the Southern District of New York unanimously rejected the order, ruling that it was so unequivocally illegal that a lawsuit challenging it was, in fact, superfluous. Eight days later, the Trump administration filed notice that it would appeal this decision directly to the Supreme Court , bypassing the standard circuit court appeals process. The Supreme Court accepted the petition on October 16, 2020, and scheduled expedited oral arguments for the case on November 30, 2020. The Court ultimately issued a per curiam decision on December 18, 2020, determining that the case was premature due to a lack of standing and ripeness . Crucially, the Court did not rule on any of the underlying constitutional challenges at that time. The decision vacated the District Court’s ruling and remanded the case back to the District Court with instructions for it to be dismissed.

Early completion of count

The Trump administration, in its characteristic fashion, sought to accelerate the completion of the census count beyond its original schedule. In September 2020, federal district court judge Lucy Koh issued a preliminary injunction against the plan to conclude counting on September 30, rather than the originally scheduled October 31. She deemed that the Commerce Department had “never articulated a satisfactory explanation” for this abrupt change. She also blocked a plan to deliver the final count results to the White House by December 31, a date which, conveniently, would have predated Trump potentially leaving office in April 2021. On the very next business day, Commerce Secretary Wilbur Ross announced that the count would instead end on October 5, as the administration appealed Koh’s decision to the Ninth Circuit . Koh then ordered the government to produce documents substantiating the Commerce Department’s reasoning. The appeals court ultimately upheld Koh’s ruling, and the Census Bureau announced on October 2 that the count would indeed continue until October 31. Also on October 2, Koh rather sternly threatened to hold Ross in contempt for repeated violations of her court order.

The Ninth Circuit decision was subsequently appealed to the Supreme Court . On October 13, in a 7–1 ruling, the Court issued an unsigned order granting the request to end the count early. Justice Sonia Sotomayor stood as the lone dissenter, eloquently stating that “meeting the deadline at the expense of the accuracy of the census is not a cost worth paying, especially when the Government has failed to show why it could not bear the lesser cost of expending more resources to meet the deadline or continuing its prior efforts to seek an extension from Congress .” The count ultimately concluded at 5:59 a.m. Eastern Time on October 16, 2020, a premature end to a critical national undertaking.

Biden changes

As one of his very first official acts in office, President Joe Biden signed Executive Order 13986 on January 20, 2021. This order effectively discontinued citizenship tabulations at the city-block level that relied on 2020 census data combined with administrative records. He also revoked a previous Trump directive that would have excluded individuals residing in the country illegally from the population figures used for apportioning congressional seats among the states. A swift reversal of policy, one might note, underscoring the political nature of even seemingly objective data collection.

Differential privacy

Researchers across various disciplines rather widely criticized the Census Bureau for intentionally introducing inaccuracies into block-level data through the implementation of differential privacy . In an effort to purportedly prevent the identification of individuals based on their age, gender, race, household relationships, or homeownership status, “disclosure avoidance noise” was deliberately injected into the data. This process involved shifting individuals between blocks, towns, or other geographical units. The consequence of this methodological choice, however, could be substantial discrepancies, particularly affecting minority populations and the reported sizes of small places. For instance, Monowi, Nebraska , famously known as the smallest incorporated municipality in the country, was erroneously reported to have two residents instead of its actual single inhabitant, a rather stark illustration of the potential for distortion. Furthermore, this approach risked corrupting redistricting data, thereby making the creation of equally sized districts and, more critically, majority-minority districts , significantly more challenging. One might question whether the pursuit of absolute privacy, in this instance, inadvertently undermined the very accuracy that the census is meant to uphold.

Accuracy

On March 10, 2022, the Census Bureau released its estimates regarding total overcounts and undercounts, broken down by various demographic characteristics. The results revealed that the total Hispanic population had likely been undercounted by a significant 4.99%, the Black population by 3.3%, and those identifying as “Some other race” by 4.34%. Conversely, Asians were estimated to have been overcounted by 2.62%, Non-Hispanic Whites by 1.64%, and Pacific Islanders by 1.28%. Native Americans were estimated to have been undercounted by 0.91%; however, a more granular analysis showed that those residing on reservations experienced a substantial undercount of 5.64%, while those living elsewhere were, curiously, overcounted by 3.06%. A complex tapestry of inaccuracies, indeed.

Additional data, released on May 19, 2022, further elucidated the discrepancies, finding that six states (Arkansas , Florida , Illinois , Mississippi , Tennessee , and Texas ) had experienced significant undercounts of their populations. Conversely, eight states (Delaware , Hawaii , Massachusetts , Minnesota , New York , Ohio , Rhode Island , and Utah ) were found to have had significant overcounts. The quest for perfect enumeration, it seems, remains an elusive one.

See also

• Race and ethnicity in the United States census