Census Estimates U.S. Population As High As 336 Million Ahead Of Actual Count

By Hansi Lo Wang

Published December 15, 2020 at 10:47 AM PST

Chris Worrell jokes with Teresa Jefferson while applying for a 2020 census job in Boston in February before the COVID-19 pandemic. Based on government records, the Census Bureau estimates the U.S. population has grown by as much as 8.7% since 2010.

Updated Wednesday at 12:12 a.m. ET

It's still not clear when the U.S. Census Bureau will release the first results from the 2020 census.

But when it does, the bureau estimates the count may show that the U.S. population has grown by as much as 8.7% since the 2010 census, which produced a count of 308.7 million people.

If the Demographic Analysis estimates released by the bureau on Tuesday are proved out in the official numbers, it would be the slowest rate of growth the U.S. has seen since the 1940 census. That national head count showed a 7.3% population increase during the 1930s, which was marked by lingering effects of the Great Depression.

The bureau estimates the number of people living in the country as of this year's Census Day, April 1, could fall within the range of 330.7 million to 335.5 million.

The bureau uses the estimates, as well as other quality indicators, to try to determine the accuracy of each census after the results are released. By putting out estimates now as benchmarks, the agency is trying to allow the public to "come to their own conclusions" about the bureau's work, Karen Battle, chief of the bureau's population division, explained during a press conference Tuesday.

The figures from Demographic Analysis are based primarily on birth and death certificates, Medicare enrollment files and other government records — unlike the census, which relies mainly on household responses to a questionnaire.

People who died from COVID-19 before April 1 were taken into account when the bureau worked on the estimates, said Eric Jensen, the bureau's senior technical expert for Demographic Analysis. The team that produced the estimates used monthly data about deaths from the National Center for Health Statistics.

"There weren't a lot of COVID deaths relative to what we see later in the year, but we did include COVID deaths in our estimates," Jensen said during Tuesday's briefing, noting that the bureau took extra steps to reflect the coronavirus pandemic's impact.

The 2020 census does not count U.S. residents who died this year before Census Day, but it does include people who died on or after April 1, according to the bureau's residence criteria.

These estimates do not affect each state's share of congressional seats, Electoral College votes or more than $1.5 trillion a year in federal money for public services.

But in a tumultuous year that saw the constitutionally mandated tally disrupted by the pandemic, historic hurricane and wildfire seasons and last-minute schedule changes by the Trump administration, any indicator of the count's precision is receiving intense scrutiny from census watchers.

One of the bureau's most pressing challenges right now is trying to resolve irregularities in this year's census records, which if left unfixed, could leave millions of people miscounted.

A particular concern among census watchers is whether the bureau has been able to count every resident once, only once, and in the right place, an often repeated goal of the bureau that carries major implications for the redrawing of voting districts, social science and public health research and policymaking that rely on census results.

The answer, however, won't be found in the bureau's Demographic Analysis estimates, which only provide a national-level look and do not give breakdowns by state or local community.

In fact, the Census Scientific Advisory Committee, one of the bureau's two panels of outside experts focusing on the 2020 census, warned in its recommendations to the agency last month that "the final 2020 Census results may be close to the Demographic Analysis numbers, and still many people could be counted in the wrong place, double counted, or not counted at all."

The bureau is currently conducting the Post-Enumeration Survey — a sort of mini-census involving fewer than 200,000 households — to determine how the 2020 census may have miscounted the country's population.

But the first results of that survey, also hindered by COVID-19, are not scheduled for release until November 2021.

In the meantime, a task force assembled by the American Statistical Association — which includes former Census Bureau directors and members of President-elect Joe Biden's transition agency review team for the Commerce Department, which oversees the Census Bureau — has been calling for the bureau to release additional quality indicators about this year's count.

In response, the bureau announced this month it's planning to release more quality metrics and allow independent experts to compare them with the census results in early 2021.

But exactly when that will happen in the new year remains an open question.

The timing of the results' release has become entangled with President Trump's push to make an unprecedented change that would leave unauthorized immigrants out of the numbers used to reallocate the 435 seats in the House of Representatives as well as votes in the next Electoral College. The Constitution requires the count to include the "whole number of persons in each state."

The Supreme Court has been reviewing a lower court ruling that has blocked that Trump memo, and despite the administration urging the justices to release a decision soon, the high court did not do so on Monday, its last scheduled day for releasing rulings this year.

After ending counting early, the administration has been pressuring the bureau to shorten the timeline for quality checks to try to deliver the first set of census results before the end of Trump's term on Jan. 20.

But the bureau has been on track to miss the Dec. 31 legal reporting deadline for months, and there's a chance that by the time it finishes putting together the numbers, Biden will be in the White House.