California is spending $154 million on ads and education to persuade all its residents to participate in the upcoming census because of the high stakes involved — billions in federal grants and congressional seats.

“California is not leaving its fate to the federal government,” said Dita Katague, head of California’s Complete Count Committee. “I believe other states should not either.”

So how much is Utah spending to make sure it is not shortchanged in the once-every-decade count?

Zero.

Instead, it will depend on volunteers for help, and will rely on the Census Bureau’s own preparation work.

“Without funding, we kind of have to move to a plan B,” said Evan Curtis, a planning coordinator in the Governor’s Office of Management and Budget who is serving as co-chairman of Utah’s Complete Count Committee. “We’re definitely going to have to rely more on getting the message out through our partners.”

His committee recruited more than 50 organizations to help reach and build trust with communities that are hard to count, whether it is because they are difficult to find, distrust the government or are just wary of providing personal information. They range from people experiencing homelessness to immigrants, refugees, American Indian tribes and fast-growing areas with plenty of new addresses.

The upcoming census comes with new hurdles. It will be the first time the Census Bureau will ask most people to respond via the internet after a postcard notice. “And an estimated 73,000 residents in the state are without internet, so that creates a challenge,” Curtis said.

Also, the U.S. Supreme Court will hear arguments on April 23 — and should decide by June — about whether the Trump administration will be allowed to ask respondents if they are U.S. citizens. Many undocumented immigrants fear this information could be used to deport them despite assurances by the Census Bureau that responses are confidential.

Census officials have testified in court that the citizenship question may decrease initial response rates by 5 percent nationally — and the damage of scaring groups enough that they will avoid the count may have already been done whether or not the question is ever asked.

No state funding

So why didn’t the Utah Legislature approve some extra funds for the state’s Census efforts?

Gov. Gary Herbert requested $70,000 for some targeted online ads for hard-to-count groups, and Rep. Karen Kwan, D-Murray, pushed for an additional $500,000. Neither got a dime.

“In 2010, every county reported an undercount” in post-census studies, Kwan said, adding it likely cost the state in lost federal aid. Data show that only 75 percent of Utahns responded to that census initially, meaning counters were sent to a quarter of the population, increasing costs and forcing some “imputation” — inputting missing information by inferring what it might be — to estimate how many people lived in homes where no one ever responded.

She worries lawmakers are penny-wise and pound-foolish by not funding efforts to encourage participation now and believes the importance of an accurate count "was lost in all the discussions this year about tax reform and the budget.”

Kwan hopes the state will add some money next year, but that could come when census counts are already underway. They begin March 12.

Why the census matters

State and federal officials plan to stress in ads and outreach efforts the benefits that residents could win — or lose — by answering (or ignoring) the census.

“Money, power and planning are kind of the big three reasons to participate,” Curtis said.

Kaile Bower, a communications coordinator for the Census Bureau, said $675 billion in federal funding is divided annually based on formulas that use census population numbers.

In Utah, “Over $5.7 billion of our $18.5 billion [state government] budget is federal funds. That is over 27 percent, and a lot of those funding streams depend on census numbers,” Curtis said.

That comes out to $1,086 per Utahn, said Shannon Simonsen, a governor’s office employee who is co-chairwoman of the state’s Complete Count Committee. She adds that a 1 percent undercount in the census would cost the state an estimated $14 million a year.

The federal money goes for programs from highways to school lunches; from Head Start and transit to public housing; and from business loans to children’s health insurance.

Census numbers also determine how many seats each state receives in the U.S. House of Representatives. This time, Washington, Colorado, Texas, North Carolina and Florida are expected to gain seats, and Minnesota, Illinois, Michigan, West Virginia, Pennsylvania and New York are expected to lose them.

In the 2000 census, Utah just missed an extra U.S. House of Representatives seat by 80 people (which instead went to North Carolina). It was another decade before the Beehive State gained its fourth House seat.

These numbers also are used to draw voting maps for legislative districts and school boards. "It really has a lot of impact on how we govern ourselves as a state,” Curtis said, and how much power each neighborhood receives.

Finally, the numbers guide public and private planning. “It guides how we plan everything from our roads and other public projects, but also our business are going to use this to determine where they locate,” Curtis said.

Internet advantages, hurdles

Census officials say that using the internet as the primary response method next year has benefits and problems. People in most of Utah will receive a postcard next March telling them how and when they can log in and complete the questionnaire.

Bower said collecting information by internet will be cheaper. The method also allows people to respond anywhere at any time on such devices as smartphones. She envisions crowds at sports arenas or church gatherings even being asked to all pull out their smartphones and answer the census at the same time.

However, many people have limited access to the internet. So much of rural Utah, for example, won’t receive an invitation to use the internet. Instead, they will be initially sent a traditional by-mail questionnaire.

Also, Bower said some people — such as older folk — may not feel comfortable with the internet. So for the first time, people will be offered the option to do questionnaires over the phone. Postcards will list numbers people can call for 12 main languages nationally.

Some American Indian reservations, where residences may lack street addresses, will have counters sent directly to homes. Bower said the census has worked with individual tribes about whether they feel address lists are adequate for mailings, or whether they prefer local counters to be deployed.

The citizenship question

Census officials acknowledge that proposals to ask about U.S. citizenship concerns undocumented immigrants and others — and could generate undercounts.

In a test of census methods in Providence, R.I., last year, “We saw growing concerns about data privacy. What was new this time, though, is we’ve got fear of repercussions for participating,” such as deportation, Bower said.

Actually, something like that happened once.

Margo Anderson, an emeritus University of Wisconsin professor who has written histories about the census, said it occurred during World War II — when Congress opened up census address data used in part to put people of Japanese descent into concentration camps.

But after the war, the law was changed back to keep all responses confidential by law for 72 years — and she said no other breaches occurred during later wars or even after 9/11. “It’s been pretty impregnable since then,” she said.

Bower stressed that federal law prevents the census from providing personal data to anyone or any other agency, and that its employees sign lifetime vows not to share it.

Michael C. Cook, chief public information officer for the Census Bureau, said it hopes to get that message out by having it delivered by trusted voices inside immigrant or other groups that may not trust the government. “We would really like somebody to first hear about the census from a trusted voice,” he said.

Curtis added that is why Utah’s Complete Count Committee is seeking help with immigrant and other groups to find trusted people to deliver that message, and to help figure out what messages would be most effective.

Other challenges

The upcoming census has several other potentially challenging aspects that could affect Utah and its count, including:

• Officials say one of the groups suffering the biggest undercounts each census are children under 5 years old — and Utah is known for its many children. Bower said officials are unsure why that is. Theories include that young parents are so overwhelmed they forget to respond, or children sometimes are split between grandparents or divorced parents who think someone else is supposed to include them in responses.

• This will be the first census to count same-sex married couples, since they are now legal nationwide. Officials say it will likely end what they think were too-high estimates of that population in the past, in part because unmarried, same-sex couples were likely included.

• This will be the first Census to exclude the word “negro” and instead use “black.”

• LDS missionaries living abroad are not included in the census, thanks in part to decisions from lawsuits Utah waged when it just missed getting an extra U.S. House seat in 2000. Missionaries serving in other states are counted there. Military and federal officials who are deployed abroad, however, are counted in their home home base.

• Also because of rulings from Utah lawsuits after the 2000 census, officials are able to “impute” data at homes where no one responds. That means they can copy data from what they believe are similar homes nearby, or use administrative records such as from Medicare or Social Security to fill in data. Still, law bans using statistical sampling for the census.

People looking for a part-time or temporary job should keep an eye on the website 2020Census.gov/jobs. The Census is already hiring, and plans by this fall to hire 400,000 to 450,000 people nationwide — most at $17 to $21 an hour — to help count people door-to-door who do not respond initially.