The Census Bureau's American Community Survey
MR. KOJO NNAMDI
Odds are good it's the most important survey you've never heard of and even if you've never filled out the American community survey, it's shaping your life in a variety of ways. The data gathered by the Census Bureau guides $400 billion in government spending each year, helping towns and counties decide where services are most needed. Countless organizations and businesses outside of the government rely on it too with retailers using it to decide where to put new stores.
MR. KOJO NNAMDI
But despite all that some say it's an intrusive waste of taxpayers' money and there's a movement afoot in congress to do away with it. So what will happen if it goes the way of the dodo? Some see nothing but huge savings while others predict pure chaos. Joining us in studio to discuss this is D'Vera Cohn, senior writer for the Pew Research Center's Social and Demographic Trends Project. D'Vera Cohn, good to see you again.
MS. D'VERA COHN
Nice to see you.
Andrew Biggs is a Resident Scholar at the American Enterprise Institute. Andrew Biggs, thank you so much for joining us.
MR. ANDREW BIGGS
Great to be with you.
It's a conversation you too can join by calling us at 800-433-8850. You can send email to email@example.com. Have you filled out an American Community Survey in the past? Were you at all worried about doing so? Do you think the American Community Survey should continue? Why or why not? 800-433-8850. D'Vera, before we talk about its potential demise we should establish exactly what the American Community Survey or ACS is. How is it different from the full scale decennial census and what purpose does it serve?
Well, the full scale decennial census has been cut down quite a bit in recent years. It now only asks about ten very basic questions. And what the American Community Survey does is fill in questions that used to be asked on what was called the census long form, which went to about one in six households and it asked dozens of questions from your immigration -- whether you're born in the U.S. to your commuting patterns to where you work, to your home value and so forth. So what it does is it fills in a lot of what we call the characteristics of the people who are counted in the census.
The ACS is relatively new. When and why did the switch from the long form census to this new method take place?
Well, there are a couple of reasons for that. One is that there was some concern about running a long form alongside a short form census. It was a lot of work for the Census Bureau to be able to manage both of those, especially because the long form had become increasingly controversial. People were resisting filling it out. And the second, another perhaps more important reason was that the ACS was designed to furnish data every year at the community level whereas the decennial census was only every ten years. So this was supposed to provide much more up-to-date recent local data than you could get from the census every ten years.
Andrew Biggs, when it was first introduced, the American Community Survey enjoyed bipartisan support. Did this move to end it surprise you?
A little bit. I can see both sides of the argument in the sense that I'm a Libertarian. I'm very concerned about privacy issues and, you know, a person out there in the country could say, why does the government need to know these things about me. Why does it want to know these things? On the other hand, as a policy analyst I want to know these sorts of things because I think without this kind of data it's very difficult to make good policy. I think the privacy issues have come up recently as certain members of congress have made it a priority to talk about them and that sort of gained momentum. But, yeah, it is a little bit surprising this has come about now.
If you rely on census data in your job, we'd like to hear from you about how you use it. Call us at 800-433-8850 or you can go to our website, kojoshow.org, and tell us there. Andrew, how far reaching would the implications be if this American Community Census survey was done away with?
Well, on one hand, if you're thinking nationally there are other datasets you can use like the Current Population Survey which give you similar sorts of data and can give you similar kinds of answers. The advantage of the American Community Survey is that it gives you much more detail, particularly at the local level. The current population survey asks the same kind of questions in many respects but it doesn't drill down in the sense that you really can get a feel for what's going on in different communities. The ACS gives you a lot more detail in that way and I think it's very helpful. It answers questions that can't otherwise be answered.
Please don your headphones because we have a caller on the line. Adam, who I think was anticipating my own next question. Adam, you're on the air. Go ahead, please.
Hi, Kojo. Thanks. As Mr. Biggs just mentioned, the Census Bureau is not the only federal agency that does these types of surveys. And I've looked at the ACS and I found many of the questions on that survey are almost identical to questions asked on other federal surveys. But the ACS is different because it requires the 3 million Americans a year that get it to respond. I'm wondering if your guests can give their thoughts on if that law that requires Americans to respond to not just the decennial census but the ACS is a good law.
Right now the survey is mandatory and before the effort to quash it completely there was a movement afoot to make it voluntary. What are the arguments for and against such a move?
Well, as I said earlier, I'm a little bit torn. I can see both sides of the argument. On the one hand, privacy matters. People are concerned about privacy, not simply at a philosophical level, you know, why does the government want to do this. They're also concerned about identity theft and things along those lines.
At the same time though, without this kind of data, policymakers in Washington are essentially flying blind. Instead of having data-based, factual-based policy discussions you have policymaking by anecdote where a member of congress or a president will say, I know of some person and I've heard of something and you don't know whether that example they give is representative of what's truly going on in the country. So there are costs and there are benefits. I think it's important to recognize both sides of them but I tend to think the benefits significantly outweigh the costs.
It should be noted the Census Bureau did a little bit of a study in 2003, what would happen if the American Community Survey were voluntary. And what they concluded was a percentage point drop of about 20 percentage points in the initial response rate. That is the percentage of households that mailed back their forms. As a result the census said that it would have to do more follow-up interviews in person, which are not only more expensive, but also more prone to error. So the cost of the taking of survey would rise significantly and the potential for error could increase.
But there is a bill in the Senate introduced by Senator Rand Paul that would make it voluntary.
And that was initially what the House was considering actually before it voted to eliminate the survey. And it could well be that that would end up becoming say a bargaining chip in conference or otherwise, you know, come up. And I think the Census Bureau has resisted it and it'll be interesting to see whether it prevails or not.
We're discussing the Census Bureau's American Community Survey with D'Vera Cohn, senior writer for the Pew Research Center Social and Demographic Trends Project and Andrew Biggs. He's a Resident Scholar at the American Enterprise Institute. If you'd like to join the conversation feel free to send us a Tweet at kojoshow or call 800-433-8850. Do you have concerns over the security of the information the Census Bureau collects? 800-433-8850. Adam, thank you for your call. Here is Allie in Atlanta, Va. (sic) Allie, you're on the air. Go ahead, please.
Hi. Yeah, I'm an urban planner in Atlanta, Ga. and we do a lot of work for local governments down here helping them write comprehensive plans. Those are long range vision for the future. And we rely on this data to give them a picture of where they are now and where they're going, where their strengths and weaknesses are. Without this we would really lose a wealth of information.
You know, D'Vera Cohn, the main purpose of the survey is to determine how government dollars, about 400 billion of them are spent but a lot of other groups make use of this data as well.
Indeed. And in fact, I was looking just this morning at a list of organizations that have signed a petition to Congress to save the American Community Survey. And I think there may have been more local groups than the national ones because a lot of local organizations don't have any other good data to rely on. As Andrew said, there are other surveys that ask some of the same questions as the ACS but there isn't a lot of really good local level data both to allow organizations to look at their own metropolitan areas but also to compare with other places around the country. And there isn't really any other good source of that sort of data right now.
And it's apparently used significantly in the private sector also. We have, and you'll find this on our website kojoshow.org, a video featuring Kate Whittington, director of Guest Insight for Target explaining how ACS data could have an effect on, well, what you wear.
MS. KATE WHITTINGTON
We started using the U.S. Census Bureau's data to notice the trend in urbanization, especially among our younger consumers and younger guests. So we've carried the merchandise that we think is most appealing to that guest. So a little bit more apparel and younger fashion, for example, as opposed to some of the more classic styles that we might carry in our suburban stores.
That I had not thought of, Andrew Biggs, that Target and Costco and others are using this to identify me and what I want to buy as their guest.
Well, I think this hits on one of D'Vera's points. There are different datasets which ask similar questions. But what the ACS does is it gives you answers at a very localized level. So if you're an urban planner thinking about how neighborhoods are built up or if you're running a Target store you want to know what sort of people live around here. It gives you information that will be otherwise very, very difficult to get a hold of.
So this may be an area -- and I'm skeptical of the government in a whole range of ways -- but this may be an area where the ACS is what you call public good, where it's something that the government can do much more easily, much more efficiently cost-wise than if every individual private sector actor had to put together their own survey to try to get this data.
On to Craig in Washington, D.C. Craig, your turn.
Yeah, hi. I work for an organization called OMB Watch here in Washington D.C. and we run a website that uses ACS data to look at transportation community needs and how well those needs are being met by federal dollars. The website's called egap, it's www.equitygap.net. And what we're doing is we're meshing together the federal spending information with neighborhood-level information to give local community advocates a chance to see whether or not particularly recovery dollars say from the Federal Transit Administration or from the Highway Administration is meeting the needs of the community.
So if you have a certain population -- and ACS will tell you this -- the number of workers in a household taking public transit to work or the number of households without vehicles. That gives you an indication of, I guess, where federal dollars should be targeted. And we take a look at that data and try and help local advocates, you know, advocate for more, you know, for more information.
And quite frankly, I don't know how federal decision makers can make federal program decisions without this kind of data. It's very useful and I would hope that congressmen and that federal agency officials are looking at this to improve federal programs and to better target federal programs.
Craig, thank you very much for your call. D'Vera Cohn, how do census data drive your work?
Well, we do a lot of work with census data and we use the American Community Survey in several ways. We publish profiles of the Hispanic and the foreign born population, which just are lists of tables that are useful to us and other people. We've done reports using ACS data on intermarriage, on cohabitation, on the economics of marriage. We've --
Indeed that's what you were here last to talk about, the economics of marriage.
That's right, that's right. And we're looking to -- we're considering trying to use the American Community Survey as the basis for some of the estimates we do on the unauthorized immigrant population. The Department of Homeland Security already uses the ACS as the basis for its own estimates of the size and distribution of unauthorized immigrants.
And in your work, Andrew Biggs?
Well, I can give you two examples of things that I do with the ACS that I can't do with any other dataset. I do a lot of work of public sector pay. Are we over or underpaying government employees? If you use other data it looks as if public school teachers are highly underpaid compared to other college-educated workers.
One issue with teachers though it's very geographically dispersed. They're in every community in the nation. If you don't have the tight geographic controls that the ACS contains you are effectively comparing people who live in a low-cost, low-salary rural environment to people who are living in suburbs and cities where the cost of living and salaries are higher. Similarly the ACS provides detail not on simply what level of education do you have, what degree do you have. It also asks what was your college major when you did your Bachelor's Degree.
We know now there's a lot of research that different college majors lead to different levels of earning in the economy. That's helpful for people making their own choices because they can say, I might want to major in engineering rather than in, you know, French poetry. But for public sector pay comparisons it's also useful as we find that people in the private sector tended to major in things where the earnings were higher. Public sector employees tended to major in things where the pay is a little bit lower.
So it gives you a more nuanced view of how we're compensating people in government. That's something where there's billions, perhaps hundreds of billions of dollars at stake in terms of making our government more cost effective. I couldn't do that kind of analysis using other datasets. Only using the ACS can I do that. So it's ironic that the push against the ACS is coming from many Republicans who want to save money and reduce government spending...
...which is what you'd like to do also.
...but this may be sort of pennywise pound foolish by cutting the ACS.
Here's Maryellen in Alexandria, Va. Maryellen, you're on the air. Go ahead, please.
Hi, Kojo. You and your guests have made a very good case for the current -- the value of the ACS in addressing all kinds of current needs, policymaking and commercial. But I want to at least have some -- you know, put up a kind of whine for the loss of the census in its longer form that everybody filled in. Because anybody who has done historical research or genealogical research knows how wonderful the older censuses were and what a wealth of information they gave you, not just about the individual you may have been searching, but the context in which they lived, the occupations that people had.
And depending on congress's interest at various decennial occasions it also told us where people came from. They didn't ask race by the way. They asked -- well, yes, they did. As a matter of fact it was a -- before the Civil War certainly they designated white or black. But I think that afterwards it was always just where did your parents come from? What language did they speak and all these very interesting ability to trace, you know, the story of Americans, particularly in such a very mobile country.
Well, you know, the ACS asks about everything from level of education to whether or not your home has a flush toilet. So it may be more efficient to ask, D'Vera, what areas of people's lives does the ACS not touch on?
Well, it does not ask, for example, about religion. The Census Bureau doesn't ask about religion. It doesn't ask about matters that might be interesting to some of us, but not to federal policy, like, for example, pet ownership or what lawn products you use. And there are some other areas that it doesn't ask about because all of its questions are supposed to be targeted to federal uses. That is, to guide the allocation of federal money or the eligibility criteria for federal programs or the design of federal programs.
Got to take a short break. When we come back, we'll be continuing this conversation on the Census Bureau's American Community Survey. If you have comments or questions, call us at 800-433-8850. If you filled out the survey or the long form census before it, were you surprised by the level of detail in the questions? 800-433-8850, or you can send email to firstname.lastname@example.org. I'm Kojo Nnamdi.
Welcome back to our conversation with Andrew Biggs about the Census's American Community Survey. Andrew Biggs is a resident scholar at the American Enterprise Institute. He's joined in studio by D'Vera Cohn, senior right for the Pugh Research Center's social and demographic trends project. We're taking your calls at 800-433-8850. If you rely on census data in your job, we want to hear from you about how you use it, or do you think the American Community Survey continue, why or why not? 800-433-8850.
Legal requirements for things like providing non-English language ballots for voters rely on this data as well. How would those measures be affected if the ACS is done away with, D'Vera Cohn?
Yeah. That's a good question, and I'm glad you raised it. There are some civil rights implications or uses of ACS data. One of them is if a certain percentage of voters or potential voters in a jurisdiction speak a language other than English, that jurisdiction is required to provide ballots in that language, as well as explanatory materials. The Census Bureau puts out that data for those jurisdictions for that purpose.
There's also a usage of ACS data to look at it used to be called the old EEOC file. Maybe it's still called that, which is you look at who holds which jobs by race and gender and that's used in some ways to assess progress and achievements of different groups as well as disparities. So those uses, if the ACS were done away with, and assuming that these laws were continued and not struck down in any way, they would have to find an alternative source of data.
Onto Adam in Arlington, Va. Adam, you're on the air. Go ahead, please.
Hello again, Kojo. Your guests have said they think the ACS strikes the right balance between the costs and benefits, but the Census Bureau doesn't really do a good job on the ACS or in the decennial census of explaining to people that 72 years after the information is collected, it's all made public. This was done with the 1940 census records just a few weeks ago.
Which is how we found out that African Americans were undercounted by about a million back in 1940.
Exactly. But the same information might be used to hack into someone's bank accounts because with this you can figure out what their mother's maiden name was, or any other information. I'm wondering if the guests can talk about the privacy concerns of this going forward, and what this means for the data that the Census Bureau already has.
Well, thank you very much for your call, Adam, and if you don't mind I'm going to put you on hold while we go to Gordon in Shepherdstown, West Va. who seems to have a similar concern. Gordon, your turn.
Well, as a matter of fact it is a similar concern. Good afternoon. I worked with or dealt with three different government agencies who all require personal identification information, and they all promised, oh, we will never lose your information, and all three of them did, Department of Vets, the Veteran Administration, Department of Commerce, and an independent government agency. And, I mean, right, nobody's gonna have any concern about privacy here.
Well, here's Andrew Biggs.
Well, I'm sympathetic, but, I mean, there are laws against disclosing this information. If a census employee discloses the information, they can be fined or jailed. I've gone through census training and they're very strict about putting down what the rules are regarding this, and I don't think with census data there have been very many, or at least none that I know of, instances in which data has been leaked and people had their identities stolen or things along those lines.
So I'm very sympathetic. I think Census needs to do everything it can, particularly going forward because we're in a different age today than we might have been in the past. But at the end of the day, it's weighing costs and benefits, and, you know, if we don't gather this data, yes, there is a slightly smaller risk of somebody having their identity stolen, or the privacy violated, but there is a much, much larger risk of policymakers in Washington making policy that doesn't make sense because they don't know what's going on. So you have to weigh the two of them together.
Gordon, thank you very much for your call. D'Vera Cohn, you wanted to address this?
Well, I wanted to talk about the 72-year rule.
That's something that has been in place since about the 1950s, and it roughly related to the lifespan at the time of an ordinary person. It was felt that by the time 72 years had gone by, the person might be gone, or that the information wouldn't be relevant to a lot of people living today. Of course, lifespan is now longer, concerns about privacy are greater. We have electronic means of disseminating and analyzing data that weren't available then.
I know that this is an issue that I didn't see really come up as much when the 1930 census results were released ten years ago. You're just beginning to hear more about it now I think as part of the larger conversation about privacy.
Adam, thank you very much. Did we answer your question?
Well, Mr. Biggs mentioned he didn't think there had ever been a situation where this information was leaked in a harmful fashion. That's actually not true. Evidence was finally brought to light just in 2007 that in 1942, only two years after that 1940 census data was collected, the names and addresses of Japanese living in America was given to the military, and that's how they rounded up the Japanese and put them in internment camps. I can't think of a more harmful violation of privacy than that.
I actually was aware of that instance. It's a little bit difficult to judge a lot of decisions which are made in wartime, and so I think we may look back in the future on decisions that were made during the War on Terror and say these were the wrongs things to do. In retrospect, that was the wrong thing to do. On the other hand, we had just been attacked. So I was thinking more of the context of financial privacy and the issues of identity theft that people are dealing with today, and I don't know of any instances in which anybody has had that happen to them.
I'm not saying it's impossible, but there are safeguards in place, and there's also laws against it such that people that can be prosecuted, and my own opinion of the Census Bureau in this area is pretty high.
Thank you very much for your call, Adam. Do you have concerns over the security of the information the Census Bureau collects? You can call us at 800-433-8850. We're talking with D'Vera Cohn, senior writer for the Pugh Research Center's Social and Demographic Trends Project, and Andrew Biggs, resident scholar at the American Enterprise Institute. Here is Richard in Reston, Va. Hi, Richard.
Hi, Kojo. This is -- yes, as you said, Dick, and I love your show. I've called in once before. I retired from the federal government. I used to manage the Community Development Block Grant Program, which is one of the federal programs that this data is essential for. It is critical in distributing at least -- well, it used to be $4 billion, it's probably been cut. It used to be four, four-and-a-half billion dollars to 1200 cities in the 50 states based on need.
And you need this kind of data that we used to get from the long form, but then the data got out of date within year by year until the next census, whereas the rolling data for this American Community Survey is going to give us the capability of having much more current data, and that's just one example of a critical federal program that the politicians ought to recognize that whether you're conservative or liberal is not relevant. This is the basic efficiency of government doing what the Congress wants it to do.
So the idea of privacy's been well-discussed and defended. I mean, when someone has to reach back to World War II to pick up an example of when it was misused, I think that puts it in perspective. And I'd like to try to answer any questions or...
I don't know. Let me see if Andrew Biggs or D'Vera Cohn have any comment or questions for you? Oh, no, indeed. Well, Richard, thank you very much for your call. We got an email from another individual named Richard who says, "I am an epidemiologist by training and profession. These government studies are essential for public health planning." And I guess that was one of the points that we made earlier in the discussion about how much these are used for the distribution of the $400 billion or so that goes to these state and local communities.
Back to the issue of privacy, D'Vera Cohn, there have long been privacy concerns surrounding the census. With people sharing every detail of their lives today on social media, one wonders if that's starting to change. We got a tweet from Stephanie who says "The American Community Survey is an essential part of the census. Google and Facebook collect information invisibly. This information is free to all." D'Vera Cohn?
Yes. I do wonder whether some of what's going on with social media is being mixed in with some of the issues about the government. But I think, you know, you could make an argument that the government is requiring you to answer your American Community Survey or your census form, where as in theory it's voluntary whether you go onto Facebook or Google, and I think that it's…
In theory indeed. But go ahead.
Yeah. But it's the mandatory aspect I think that sticks with some people.
On to Allison in Fairfax, Va. Allison, you're on the air. Go ahead, please.
Yes. I'm calling because this particular data is so crucial to future historians. As an historian, I'm currently using census data from 1940 up through 2000, and if we don't do the American Community Survey, we're not gonna have this information in the future.
And what did you say you were using specifically for?
I'm working on a dissertation looking at central heating in American homes.
Hence the question about how much fuel we're using in our homes. Andrew Biggs, you don't know how some of these things will come into use researchers.
Sure. The work I've done with the ACS was probably not what the designers of the ACS thought it would be used for. Having the data out there and freely available to researchers and to scholars is important because it allows people's creative inputs to be applied to the data, and you don't know exactly what you're gonna get from it. But I think to go back to D'Vera's points, I mean, the privacy issue is important, but the questions are not about things such as what is your religion, what political party are you in, things that we think are sort of fundamental to our rights as Americans.
You know, you can question the government needs to know how many flush toilets you have, but at the same time, that's a different sort of question than the more personal things that people may be concerned about.
Allison, thank you very much for your call. I think Joe in Manassas, Va. would like to address some of the issues that you just raised, Andrew. Joe, you're on the air. Go ahead, please.
Yeah. Thank you. First of all, I never answer the census, and if I do, it's only -- I will only put down how many people are there because that's all the Constitution requires, and all the Constitution allows. It shouldn't be any bit of the government's business how many toilets we have, what our race is, or anything like that. Like your guest said before, this information is used by policymakers. Well, it's used by policymakers to impose more regulations, to try to do, you know, to try to have, you know, to design our society even more. And yes, I know you're gonna talk about maybe sickle cell anemia and those types of things for public policy, but where is the authority for the government to know how many toilets we have, or where is the authority by the government to demand to know what our race is?
What are -- what are your concerns about the government demanding to know this information, Joe? Are you concerned about how it is likely to be used?
Well, why do they ask it? Why do they care what race somebody is in the house? I mean, you're flipping it in your head. Your assumption is, oh, if there's nothing wrong, we can just tell the government. Well, why does the government need to know what race I am? Why do they need to know how many toilets I have unless they're going to try to regulate toilets or something like that?
Well, again, I sympathize, but, I mean, in my own policy work, I'm often accused of being a person who wants to gut the federal government. I want to eviscerate it. My argument on this is that you don't want to start with the eyes and the ears of the federal government, the parts of the federal government that enable us to know what's going on so that we can design better policy.
I mean, I've come up with certain conclusions on public employee pay based on my analysis of the ACS. People can dispute those things by looking at the data, looking at the statistical methods I've used to analyze things. Without that, it would simply be a back and forth of anecdotes and insults, and so this is not something that indisputably pushes you towards larger more intrusive government. If you think the facts are on your side in these policy debates, knowing more facts, I think, is an important advantage.
And D'Vera Cohn, you can talk about how these questions get there in the first place, because it's my understanding Congress approves these questions.
That's absolutely right, and really, the caller's question goes to the heart of what government is all about. All of these questions are there because in one form or another they're deemed to support programs that Congress has funded, or include questions that Congress has specifically asked for, and thus are a reflection of the kind of government that we have.
Before it's very existence was being called into question, the buzz surrounding the ACS was that it's moving online. What's the upside to that?
Yes. They've been -- the Census Bureau has been under a lot of pressure to go online. Why is it people ask that the IRS can have you file your tax returns online, but the Census Bureau can't seem to get you to file your census forms online? So they have been putting a fair amount of study into this. They released a couple of research reports on it recently, and starting in January, anybody who receives the ACS will be encouraged to respond online, the thinking being, first of all, that it will lead to quicker responses.
It might lead to more accurate responses because it's easier to tap the data into a computer than to try to scan it off a paper form, and, of course, that might reduce costs as well.
D'Vera Cohn is senior writer for the Pugh Research Center's social and demographic trends project. Thank you so much for joining us.
Andrew Biggs is a resident scholar at the American Enterprise Institute. Andrew, thank you for joining us.
Good to be with you.
Thank you all for listening. I'm Kojo Nnamdi.
Transcripts of WAMU programs are available for personal use. Transcripts are provided "As Is" without warranties of any kind, either express or implied. WAMU does not warrant that the transcript is error-free. For all WAMU programs, the broadcast audio should be considered the authoritative version. Transcripts are owned by WAMU 88.5 FM American University Radio and are protected by laws in both the United States and international law. You may not sell or modify transcripts or reproduce, display, distribute, or otherwise use the transcript, in whole or in part, in any way for any public or commercial purpose without the express written permission of WAMU. All requests for uses beyond personal and noncommercial use should be referred to (202) 885-1200.