Do not cut the American Community Survey: an editorial.

The House Republicans recently voted to remove funding from the US Census. According to news reports, this action was motivated by a mix of Tea-Party symbolism and the legacy of a long-standing fight with the Census.

This post will present a short editorial. While I have sympathy for part of the motivation for this action – namely, the desire of every household to be left alone – it seems overwhelmed by everything in the other direction.

Let me put it this way. Though I tend to be a man of moderate language, removing funding looks very stupid. In this case I can bring many professional and personal observations to the topic.

A recap

In case you have stopped keeping track, let’s state more precisely what happened. The House voted to remove funding for the American Community Survey (ACS). The ACS was initiated some time ago as a less expensive way of collecting information about US citizenry than using the long form during the US Census.

Let’s remember why the long form was invented. Every ten years the entire population needs to be counted. That is an expensive and arduous undertaking. Why not learn about more than merely name, address and age? The United States contains a vast array of different people, and we are not all alike. Over the years it has been useful, not to mention fascinating, to learn about the variety of ways people live – the size of their households, whether they live with an extended family, how much they commute, and whether they enjoy modern conveniences such as indoor plumbing.

It is understandable if you do not remember the long form. Many households never got it. It took about a half hour to fill out. Because it was a burden, it arrived to a random set of households during the decennial census, as a way to collect deep information about the US population, but not to burden anybody too much.

The long form was replaced by the ACS, which is more periodic. Same idea. Only a small fraction of randomly chosen households get it, with the intent of trying to get deep information about the US population. And it leads to more timely information, since it is not done every ten years.

Most of the news reports framed the complaints about the ACS as a standard bit of Tea-Party symbolism about government intrusiveness.

If I might quote one article, one of the leaders of the complaints, Representative Daniel Webster (R-Fla.), calls the long form an invasion of “people’s privacy by requiring them to answer questions such as what time they leave for work, how long their commute is, and whether they need help going shopping.” There is also an objection to the fine levied against somebody who does not cooperative. It seems they would prefer it be voluntary, or “opt-in.”

Also, according to the news reports, there are other scores to settle with the Census Bureau, and that also plays a role. I am not a Washington insider, so I have no idea what this refers to, but I take that as a signal that there are other things – turf wars, personality conflicts, and other parochial DC stuff – going on in the background, adding a big of emotional edge to events.

Anyway, getting rid of funding seems pretty dumb to me. And that is a professional opinion.
As noted, the US has a very diverse population. It is useful and important to know about that. Moreover, we are talking about a very small burden, one that hits a household very rarely.

This opinion partly comes from observing the Census up close. I served on a Census Advisory panel as a professional economist, mostly in the first half of the last decade. It was me and seven other economists. We were appointed by the American Economics Association. There were 32 others too, appointed by associations representing demographers, marketers, and statisticians.

To be frank, it was not the most interesting government panel I have ever served on, but that did not matter. I was asked to do it, and I had a professional obligation to do it. So I did. That was how most of the others thought about it too. I do not regret doing it. I learned all sorts of stuff about how the Census worked. I hope I helped them think about their challenges.

It was not central to my reasons for being there, but the ACS was discussed extensively during that time, when I was on the oversight panel. It was being introduced at the time.

I just do not recall any major objection to it. It obviously improved thing, especially in comparison to the long form. This seemed like a much better way to do things; spreading out the work made a great deal more sense than the long form. It was a more efficient way to organize this activity, and it meant that the information was more timely, not a slave to the decadal timetable.

I cannot believe there is any desire to go back to the long form. Rather, the Tea Party would like to dispense with any of the extensive information altogether.

I will admit that this motive has me puzzled, and news reports apparently picked up on the same tension. Are these people Republicans? Business groups do not want to see the ACS discontinued. It feeds into a vast array of general information that business uses. That is why the marketing groups were on the same advisory panel, for example. They are a major user of the data. So are demographers, who do lots of research for business.

Moreover, getting rid of any information about the US population makes no sense for policy. Such flying blind would be unthinkable for policy in many settings. Every Congressional bill that shapes transportation infrastructure, for example, comes with a forecast for the size of growth it will produce and the number of jobs it will create and the pain it will alleviate, and so on. Nothing comparable can be done for many bills designed to shape public infrastructure without measuring the size of the household economy.

A personal story

I have a personal story to explain my reaction. Let me put it this way. Filling out the long form was one of the many odd jobs I had in college.

No, seriously. I was a temporary employee for the 1980 Census. It was me and several other students, many retirees looking to kill some time, and dozens of underemployed housewives. We all were looking for a little extra cash, doing government work at night.

Our job was repetitive but interesting. We phoned people who had inadvertently left some questions unanswered in their Census forms. For several hours, many days a week, we all would report to a temporary office with a large phone bank. Somebody would hand us a stack of Census forms, and we would call some stranger, ask them one or two questions they left blank, fill them in, and then move to the next questionnaire. The job only lasted a few months, since this was temporary work, by definition, but it was interesting and important work, and at that stage in my life, I was grateful for any job.

“Survey fatigue” accounted for the vast majority of unfilled forms. That is, somebody would leave a blank by accident. It happened with the short form and the long form. It was totally understandable. Nobody is perfect.

I can still recall how enjoyable it was to talk to so many people. Informed that they left something blank, most people politely and respectfully answered the question, and the phone conversation was done in sixty seconds or less. No big deal. The vast majority of people were very polite and cooperative. That respectable behavior cut across race, income, education gender, or location, whether inner city or suburb.

Now and again, one of us would get a “non-cooperative response.” Our instructions were clear. If we got that, then we were not supposed to press our luck. Those were put those in a separate pile for the professionals.

Usually the people who did were not the type of person who should have a major voice in government policy. Indeed, talking to any “non-cooperative respondent” left quite an impression on me at that age. It was the first time I had ever encountered somebody who was deeply mistrustful of government authority in any form. As soon as I identified myself as a Census employee, many of them would ask a few questions to understand the situation, then swear at me, or declare their unwillingness to talk, or just hang up.

Understand what this was. These phone calls touched a part of the country (e.g., Berkeley and Oakland, CA) where plenty of locals were distrustful of government. The attitude was partly due to hangover attitudes from the Vietnam War, and partly because people grew marijuana in their backyards. This was just the Census, to be sure, and not the police, DEA, or FBI, but it did not matter. Mistrust was mistrust.

I can recall getting a few of these, but not many. It was rare. Let me say that again: it was unusual, even in a part of the country where rebellion was in the air.

So when I heard about the complaints about government intrusiveness, I thought “really?” Has the world changed that much in thirty years that a large fraction of respondents are not polite? Has the country really become that impolite in thirty years, that they now find this survey to be excessively burdensome?

Call me an optimist, but I recall that the vast majority of my fellow citizens did not mind filling out a government form. I dunno’, but methinks the complaints have been exaggerated.


I sympathize with the desire to be left alone, but I guess I just do not understand what is so darned difficult about a survey. Is it too much to expect everyone to cooperate during the once-in-a-lifetime moment when the Census sends the long form — that is, the ACS — to a household? Isn’t that a fairly mild obligation for every citizen?

Like I said, removing the funding just looks stupid.

4 Replies to “Do not cut the American Community Survey: an editorial.”

  1. “Every Congressional bill that shapes transportation infrastructure, for example, comes with a forecast for the size of growth it will produce and the number of jobs it will create and the pain it will alleviate, and so on. Nothing comparable can be done for many bills designed to shape public infrastructure without measuring the size of the household economy.”

    Perhaps that is the point. It’s easier to garner support against government action when it’s poorly conceived & executed. Call it “Lobomizing the Beast”.

  2. I was sent the ACS a few months back. I respect your opinion on the survey and agree that the information no doubt has value. However, I feel that you too quickly gloss over the fact that the Census Bureau has made this random survey mandatory and repeatedly threatens up to a $1,000 fine for not completing it. I find it very easy to see how a recipient (and not just a “tea-partier”) would feel their privacy has been violated when being required to answer very personal questions like how long their commute is, what method of transportation they use, is there a special needs person living in your home, and what your utility bills are. Maybe this program should not have been cut outright, but making it a voluntary response seems like a much more reasonable approach to me.

  3. Give me an assurance* that the information will remain anonymous and coded to, say, a city block or school district, and I’d fill out the long form every year. .Gov has a legitimate interest in gathering aggregate statistics. If it wants specific information about me, there’s an app for that. I’ts called a warrant or subpoena.

    I don’t think this is a particularly unreasonable position.

    *I mean a technological limitation, not a pr statement.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s