Why fill out your census form? There are roughly $447 billion in federal distributions at stake this year as Americans respond to the 2010 census. The results of the decennial headcount will determine future budgets, legislative redistricting and key decisions on highways, schools, health facilities and much more.
On Wednesday, March 24, Andrew Reamer, a Metropolitan Policy Program fellow and author of the report “Counting for Dollars,” was joined by POLITICO Senior Editor David Mark in an online discussion about the 2010 Census and the impact of its results.
The transcript of this chat follows.
12:29 David Mark: Hello, Andrew. Thanks for joining us. I just sent in my Census material and I’m interested in the difference between the long form and short form.
12:31 Andrew Reamer: Hi David. Well, there is no more long form. Once upon at time, from 1940 to 2000, one of six households got a detailed survey, about 50 or so questions, asking about things like occupation, journey to work, and income. This long form survey has been replaced by the monthly American Community Survey, same questions. So now, short form only, 10 questions.
12:32 [Comment From Erin: ] How much does the census itself cost? How many jobs are being created, and are they all short-term?
12:33 Andrew Reamer: The 2010 census costs about $14.8 billion, life cycle, that is over the decade. Jobs–3.8 million people recruited, 1.2 million field positions, 870K temp workers, and 635K for followup operations.
12:33 [Comment From Shaun Gallagher: ] Why can’t we submit our census information online?
12:34 Andrew Reamer: The Census Bureau says for security reasons. Back in 2000, there was a not particularly well publicized internet option. For 2010, that was withdrawn. However, the Census Bureau looks like it will develop an internet option for the ACS and says it will have one available for 2020. The econ data side of census collects data online from businesses now, through Census Taker.
12:34 [Comment From Shawn: ] What are the most common sectors where Census funds are used?
12:36 Andrew Reamer: Census-related data are used to guide the distribution of about a half trillion dollars annually across the U.S. 61% of the funds go for health care, primarily Medicaid, 11% for transportation, 12% for income security, like housing vouchers.
12:36 [Comment From Fred: ] I’ve read that some 15 percent of the federal budget is allocated using the Census. Is that right?
12:36 Andrew Reamer: Yup, that’s right.
12:36 [Comment From Laurie : ] Why is it important for Americans to fill out the Census?
12:38 Andrew Reamer: Three reasons–democracy–congressional apportionment, Electoral college votes by state, redistricting for federal, state, and local legislative offices.
Public policy–distributing all that money, enforcing anti-discrimination laws, figuring out where to put schools, highways, health clinics, attracting new businesses.
Economy–businesses use the data to figure out where to locate stores and operations, what goods and services to sell to the community.
12:39 [Comment From Eric: ] How accurate is the census? Are there any groups that the Census fails to reach or that have typically low response rates?
12:40 Andrew Reamer: The census has been getting more accurate. In 1990, the estimated undercount was 1.6 percent. In 2000, there was a 0.5% overcount. Five percent of blacks and Hispanics were missed in 1990, but that undercount dropped in 2000 to 1.8 percent for blacks and 0.7 percent for Hispanics. Hopefully, the 2010 Census will do even better.
12:41 [Comment From Mark: ] What will happen in December when the Census Bureau gives the information to the president for apportionment?
12:42 Andrew Reamer: By law, the Census Bureau is required to submit the population figures by state in December so that the number of seats in the House of Representatives can be apportioned. Next March, a data package goes to the states so that they can redistrict those congressional seats.
12:42 [Comment From Suzie: ] Do you know what the response rate is for the census forms? Do they have some kind of calculation that makes up for people not responding?
12:44 Andrew Reamer: I believe that the mail response rate for census form was in the 60s in 2000. If someone doesn’t’ t fill out their form, the household will get a visit from an enumerator to ask the questions in person. Multiple visits are made. The Census Bureau makes every effort to get information on a household to get as complete a count as possible.
That said, there is no adjusting for any undercount. What we see is what we get.
12:44 [Comment From Paul: ] It seems like the Census Bureau is spending a lot on advertising. Why are they doing this in such tough economic times?
12:46 Andrew Reamer: I think the ad budget is about $133 million, so less than 1 percent of the overall census budget. The advertising was found very helpful in 2000 in boosting participation, in other words it can save money. This year, each 1 percent increase in the mail-back rate saves 85 million bucks, so if the ads boost mailback by 1.5 percent, they pay for themselves.
12:46 [Comment From Megan: ] How are the Census-determined funds distributed? For instance, are the funds allocated for transportation a separate pot of money that is distributed independently, or are they tacked onto other funding mechanisms, like the transportation bill?
12:49 Andrew Reamer: We found 215 programs that rely on census-related data to distribute funds. Each one is unique. The large majority of programs use census data in the allocation formula, e.g., each state gets funds based on its share of total population, of poor kids, of poor kids with disabilities, of people in small towns. The formulas are in the authorizing legislation, typically, rather than the appropriations. About a quarter of the programs use the data to determine eligibility for the $, e.g., a rural area, an urban area.
12:49 [Comment From Ines: ] How do I know that the information I fill out on the Census and ACS is secure?
12:50 Andrew Reamer: The Census Bureau and all its employees by law are required, under penalty of jail time and a fine, to keep the data confidential. They take this charge very seriously as they know that confidentiality is necessary to get a complete count. There has been no instance since WWII of the census data being used inappropriately, as far as I know.
12:51 [Comment From Peter: ] Which comes first, the USA Patriot act or the US Census being anonymous?
12:51 Andrew Reamer: The ruling came down last week that Title 13, the Census Act requiring confidentiality, trumps the Patriot Act, no question.
12:51 [Comment From Matthew: ] It kind of seems like Republicans don’t like the census. Is that because of redistricting?
12:54 Andrew Reamer: I’ll answer in two ways. One, I think some people are using the Census as a symbol of big intrusive government, seeking to stoke fear and paranoia about government in general and the Democrats in particular. As I mentioned, the census data are safe; in the meantime, we’re not hearing much hew and cry about giving data to businesses, e.g., credit card firms, who sell the data to other users.
Two, straight up political reality is that Republicans benefit from an undercount of nonwhites, who tend to vote Democratic. Democrats are the beneficiaries of a low undercount.
12:55 [Comment From Mark: ] Can you say more about the American Community Survey? I got one of those, and it’s a lot of information. How will it be used?
12:58 Andrew Reamer: The ACS goes out to about 250,000 households a month, or 3 million a year. The idea is that over 5 years the Census Bureau will have enough of a sample to publish five-year averages for neighborhoods, updating this average annually. So the first five year data come out in late 2010 for 2005-2009.
The data are used in multiple ways. One is for state and local government and nonprofit planning for schools, travel, health delivery, and, very importantly, disaster planning. the 2000 long form data were important in understanding who got impacted by Katrina in New Orleans. The data are used to drive federal funding and by businesses for site location, as well.
12:58 [Comment From Matthew: ] I don’t think that cities usually get a fair shake when it comes to funding from the national government. It seems like city infrastructure is crumbling while highways are constantly being repaved. Will the census help with that disparity at all?
1:00 Andrew Reamer: Well, yes, sort of. The bulk of the census-driven funds, like for transportation, go to state governments. The states then figure out where to spend the money in-state. And they use census data to help figure that out, so an accurate count will help your community. But the states have flexibility to use other factors as well. Each federal program has its own rules about how the states should divvy up the money in-state, the highway program’s a very flexible one.
1:00 [Comment From Tom: ] What happens if I don’t fill out my Census form? And how does anyone know that I haven’t filled it out?
1:02 Andrew Reamer: This year, each census form has a bar code, so when you mail it back, it gets registered. The Census Bureau has a daily updated map of return rates across the U.S., on the web. At some point soon, the Census Bureau will start looking to see who hasn’t sent back their form, and if it’s you, you get a visit from one of your neighbors working for the Census Bureau.
1:02 [Comment From Ray: ] Do you know if the Census questionnaires were mailed on recycled paper? It seems like they should have been!
1:03 Andrew Reamer: Hey, good question! Back in the 1990s, the federal government issued guidelines for encouraging or mandating the purchase of recycled paper, I assume that applies to the census forms.
1:03 [Comment From Ragen: ] Is there any penalty for not filling out the census form?
1:04 Andrew Reamer: Filling out the form is mandated by law, and so there is the potential for a penalty for not filling it out, but I haven’t heard of anyone being fined for not doing so. The Census Bureau is much more about carrots than sticks, I think.
1:05 David Mark: What kind of people are more difficult to count than others? Does this break down along race, ethnicity, income level, etc.?
1:08 Andrew Reamer: The Census Bureau has studied the characteristics of the hard-to-count populations. Characteristics include people who live in crowded housing, are in poverty, not a H.S. grad, unemployed, don’t speak English, and a few others. Of course, undocumented residents are difficult as well. A key aim of the advertising and the local complete count committees is to reach people with HTC characteristics.
1:08 David Mark: What types of federal assistance are premised on the Census count? And how do Census figures play into congressional reapportionment?
1:12 Andrew Reamer: There are four kinds of federal assistance that are driven by census stats. The biggest by far is grants, $420B. Next is direct loans, $4.8B, guaranteed or insured loans, $10.4B, and direct payments (e.g., housing vouchers), $11.6 billion.
The Census figures are used to determine how many seats in the House of Representatives each state gets for the 2012-2020 elections. There’s a somewhat elaborate procedure, but it comes down to taking the total population counted in the 2010 Census, dividing by 435 seats, and then dealing with the fractions one way or the other in each state. After the 2000 census, Utah missed getting a fourth seat by under 1,000 residents, and promptly sued the federal government in court to get permission to count overseas Mormon missionaries from Utah. Utah lost.
1:12 [Comment From Greg: ] Why is the Census only every 10 years?
1:13 Andrew Reamer: A census every ten years is called for by the Constitution, in Article I, section II. The government, of course, could do one more frequently, if Congress so chose.
1:14 [Comment From Kenneth: ] How do homeless people figure in to this equation? Can they use shelters as a home address? What if they are living on the street?
1:15 Andrew Reamer: Good question. Yes, the Census Bureau does the best it can to count the homeless. It visits shelters across the U.S. to get a count of those residents. I’m not sure how it goes about counting homeless people not in shelters, but I do know that they try.
1:15 [Comment From Nate: ] Are immigrant populations underreported? It seems like some individuals in the US illegally might be pretty hesitant to fill out the Census forms.
1:17 Andrew Reamer: Yes, immigrants not in the U.S. legally are undercounted out of hesitancy to fill out a form. A good part of the “get out the count” effort at the local level is to have trusted intermediaries, e.g., churches, aid organizations, tell the undocumented that they are safe if they fill out the form and filling it out will help their community.
1:17 [Comment From Matthew: ] Do states get involved in getting people to participate in the census? It seems like they have a lot to gain!
1:18 Andrew Reamer: Another good question. States do get involved in promoting the census. An interesting finding of our study on census-driven funding is that states get over 80% of the money and that overall 20% of state budgets come from federal grants. That said, some states are less active this year than others due to budget constraints, which I think is shortsighted given the multiple billions at stake.
1:19 [Comment From Matthew: ] That comment about the Mormons is interesting. Do U.S. citizens abroad get counted in any way?
1:20 Andrew Reamer: I’m not sure of the rules for counting people overseas, we’ll see if we can find the answer.
1:21 David Mark: Is there much disparity among states on how much money they receive from Washington based on Census results?
1:25 Andrew Reamer: Yes, a big difference on a per capita basis. DC is at the top of the list at $4,656 per person and Nevada takes up the rear at $742. The national average is $1,469. The major reason, but not only, for the difference is the nature of each state’s Medicaid program and their federal reimbursement rate. Medicaid accounts for 58% of the census-driven $. States have a lot of flexibility in designing the program, so Nevada’s is conservative and Vermont’s and DC’s are generous. The richer the state, the lower the reimbursement rate, that makes a difference as well. Rural states get slightly more $ per person than urban states. States with large income inequality tend to have higher per capita fund flows, because they are wealthy enough to have a generous Medicaid program and they have poor people who use it. NY state fits this profile.
1:26 [Comment From Kenneth: ] To follow up on the question about homeless people. Oftentimes homeless people are couch surfing at friends’ and relatives’. Should they get reported on those household’s numbers?
1:27 Andrew Reamer: Yes. People who don’t have a permanent address should be counted where they are on April 1.
Back to the overseas question–only overseas Armed Forces personnel are counted.
1:28 David Mark: Thanks for joining us, everybody. Enjoy the afternoon.
Reflections on progress: Essays on the global political economy
The productivity puzzle: How can we speed up the growth of the economy?
Disrupting ISIS recruitment online
[Mitt Romney] may be the last 'white' presidential candidate—in the sense of a candidate appealing to only white voters.