What happens if I don't fill out my census form? (2019)
people.howstuffworks.comThe US census was infamously used to assist in rounding up Americans of specific ethnic backgrounds for concentration camps.
It would go a long way to easing people’s minds if they had some way to guarantee that this could not happen again. I realize there is value in gathering attributes beyond just “one person”, but storing these attributes in a way that can be connected to addresses seems seriously problematic.
I agree that the data should be treated anonymously and confidentially, with severe punishment to abuses.
Would the lack of census data prevented the concentration camps from operating? It may have instead taken longer and required more people, but I would expect that at the time anyone with a Japanese/Chinese look or name to have been detained anyway. Side note: It seems even the 2nd amendment didn’t help citizens resist agains oppression.
I suspect these days the government has enough information already on its citizens through other means (intelligence agencies etc) than the census, to still target based on race if they wanted to.
Immediately after 9-11 the discrimination against Muslim, Arabic and other minorities was rather pronounced. No census data needed to that either. People just judged by visual appearance or name.
> I agree that the data should be treated anonymously and confidentially, with severe punishment to abuses.
What constitutes "abuse"? Differential privacy is hard. Once the data has been collected, simply not being an expert in statistics can be enough to accidentally leak sensitive data. The government is simultaneously the data collector, data user, and prosecutor of abuses, so I'm skeptical that much will come of any such promise. Maybe a letter of apology in 50 years.
> I suspect these days the government has enough information already on its citizens through other means (intelligence agencies etc) than the census, to still target based on race if they wanted to.
Pointing out that ethnic minorities are likely to be screwed anyway in such a situation, even without census data, is not going to ease anyone’s mind.
"the data should be treated anonymously"
The problem is "anonymous" data is often really easy to de-anonymize, especially when there's so much unique, identifying information there, as there is with the census.
Even if it did somehow manage to remain anonymous, they could still tell that a person with some undesirable characteristics lived at such and such a location. That would be enough to round people up. They don't necessarily have to know your name.
France had the same issue with Jews being rounded up during Nazi occupation with help from government data, their answer was to no longer collect data on ethnicity.
This is worth quoting here IMHO:
"However, even if you don't get fined for not filling out the census form, there are some good reasons you should do it anyway. Seats in the House of Representatives seats are apportioned by population, with the most populous states receiving the most seats. Federal and state governments rely on census data to budget for social welfare programs that assist the poor, elderly, disabled and veterans. Cities and private industry use demographic figures to plan new hospitals and housing developments, and to assess the need for new schools or new strip malls. So, not filling out the census form may cost you something in the long run."
> not filling out the census form may cost you something in the long run.
On the average each uncounted person loses their district $2,000 per year in federal payments (i.e. tax money returning to the district). Losing $20,000 over the ten years between censuses is a big deal for school and other program funding.
PLEASE BE COUNTED!
Got a citation for that? You're suggesting $600B+ is allocated by the federal government based on census numbers?
https://gwipp.gwu.edu/counting-dollars-2020-role-decennial-c...
They calculate that in 2017 there were $1,504,191,364,000 in Federal spending guided by the census numbers, which is about $4,600 per capita.
In this paper [1] they look at some specific programs:
> Five grant programs administered by the U.S. Department of Health and Human Services (HHS) use the Federal Medical Assistance Percentage (FMAP), based on the 2010 Decennial Census population count, to determine reimbursements to and payments from each state government (totaling $286.1 billion in Fiscal Year 2015). The five FMAP-guided programs are Medicaid, the Children’s Health Insurance Program, Title IV-E Foster Care, Title IV-E Adoption Assistance, and the Child Care and Development Fund. In Fiscal Year (FY) 2015, reimbursements to and payments from state governments under these five programs totaled $286.1 billion (48.1% of all federal grants to states and 13.0% of all state budgets).In FY2015, 37 states forfeited a measurable amount of funds for each person missed in the 2010 Census.
> Among these 37 states, the median FY2015 loss per person missed in the 2010 Census was $1,091. FY2015 loss per person missed ranged from $533 for Utah to $2,309 for Vermont. The median state is Tennessee.
[1] https://gwipp.gwu.edu/sites/g/files/zaxdzs2181/f/downloads/G...
The article links to this other article, https://people.howstuffworks.com/census.htm, which states:
> Every year, federal and state governments use census data to allocate more than $675 billion toward public services and infrastructure [source: U.S. Census Bureau].
The link is to https://www.census.gov/content/dam/Census/library/visualizat....
It certainly seems possible. That’s not even 7% of the annual budget. We spend a lot of money. Can’t find a citation on my phone though.
Are the apportionments done based on the number of census forms returned, though, or based on the Census Bureau's best estimate of population? Presumably the Bureau have a better guess than that, using an adjustment factor that could be different for each state.
Apportionment is based on direct census count only. Indeed there is better science available for population estimation (and IIRC some of it indeed gets done by the Census Bureau's reporting), but that's not how the law was written.
It's called the "American Community Survey" (https://en.wikipedia.org/wiki/American_Community_Survey). It's a great piece of high-quality, modern, scientifically valid demographic work that the C.B. does between its enforced stints with 18th c. `demographicology' that the constitution requires.
Apportionments (and redistricting) are done every 10 years based on the actual census enumeration.
This is required by the constitution: Article 1, Section 2, Clause 3 - as amended by the 14th Amendment.
Some of us live where we oppose all of the current elected officials. In that case, it sounds like opting out is the only effective "vote" we can make in a FPTP system
That's like cutting of the nose to spite the face.
You can reasonably oppose the officials, sure, but the schools and other services in your area still need funding. Avoiding the census is a vote not to fund them, and unlike a normal vote every single return or lack of it has a direct impact on the level of that funding.
Also the census results are applied for a whole decade during which officials can change and many elections will take place.
The effective vote is the one with your feet.
>Seats in the House of Representatives seats are apportioned by population, with the most populous states receiving the most seats.
People in the minority party in solidly red/blue states would want their state to lose votes in order to increase the representation of their party.
Not everything is about national politics. I live in a Republican controlled state, and even though I’m not a Republican, I want us to be represented and properly funded. So I will fill out my census form. Just because I don’t have the same political views as my neighbors doesn’t mean I want to disenfranchise them (and myself while I’m at it).
In 2010 I just kept forgetting to fill out the census. My census guy left notes on my apartment door begging me to call him. Finally, one day, he literally popped out of the fucking bushes with the census form.
I kept getting some mid-census domestic abuse questionnaire a couple years ago. I had zero interest in filling it out but they were persistent about that too.
They missed me in 1990 and 2000, when I was unconventionally housed (living in the office).
Hopefully you are in a better situation now. For anyone who is currently homeless, but wants to be counted, the census bureau has the following plans:
March 30, 2020: Counting people who are in shelters.
March 31, 2020: Counting people at soup kitchens and mobile food vans.
April 1, 2020: Counting people in non-sheltered, outdoor locations, such as tent encampments and on the streets.
See this for more information: https://2020census.gov/en/what-is-2020-census/focus/people-e...
>Note that the Sentencing Reform Act of 1984 has increased the fine for any criminal misdemeanor to as much as $5,000. In practice, though, no one has been prosecuted for not filling out the census since 1970, according to a 2014 Politifact article.
That article is wrong, and the fine is absolutely not $5,000. If that article were correct, then there would be no federal fines under $5,000 anywhere. This is misinformation that the census bureau likes to spread around to scare people. In face, the Sentencing Reform Act of 1984 applies only to criminal cases.
This means that the fine is actually $100.
I went through this rigmarole when I got one of those off-year census forms during the Obama administration, asking about what my family eats, how many toilets we have, etc. I refused to fill it out. A guy showed up at my door. He showed me the brochure that said it was mandatory, and what the potential fine was. The same $5,000 number was quoted. I told him I would not be participating in this program, and that his brochure was lying because this was not a criminal offense. He agreed it was not a criminal matter. And that was that.
One thing I’m curious about is, why does a country need to do this in the first place?
From the example I’m familiar with: the last ever census in Turkey was held in 2007. There on, there is now a live, real-time count of everyone in the country. Why doesn’t that work in the US?
Surely US is capable of it, but I wonder if the blockage is inertial, political or just not enough of a priority.
In short? Because the Constitution specifies how it must be done, any ambiguity is enshrined in tradition, and neither is realistically changeable.
Can you provide any technical detail as to how a real-time count of every person is possible? What databases are used to track births and deaths and emigration in real time? How about income/housing/education characteristics? Are you a Turkish citizen? If so, are you able to query the government for the exact population count right now?
The population counts are released yearly and I believe you can do what is the Turkish version of FOIA request to get the live count, though don’t quote me on the latter part.
Turkey has a central database, called TCKN, that everyone is assigned by birth or by getting an immigration visa. The births and deaths are tracked by hospitals and the Revenue Service. Income / education / housing - all of these are tracked by the central registers and licensing boards. Every Turkish citizen and resident has an online account at https://www.turkiye.gov.tr/?lang=en_US (for English version) which you can use to view taxes paid, your previous health records (MRI scans and all should be downloadable) see scripts fillable, paperwork for legal action by you or against you, census records, even mobile and landlines registered on your name, and basically pretty much everything you’d expect a state can do. All of the outputs from this site are printable PDFs and they have a barcode on them which I think the government cryptographic signature. So you can get some doc from here, print it out and bring to someone else and they’d be able to verify it by scanning the barcode.
In other words, they have total knowledge. It’s not just the number of people in the country. What’s interesting is that it doesn’t rely on self-reporting, but on ambient data providers. It’s kind of dystopian, but technically fairly impressive and so far (ominously) relatively benign. Failing all else, it’s very useful.
Thanks for the thoughtful reply. I remain skeptical that an accurate detailed count can be automated (at this point in history) but I won’t claim to have any insight about contemporary life and society in Turkey.
I think you are right to be skeptical and I don't claim to have an insight into how accurate it is. However, mind that while Turkey is culturally about 80% western and 20% non-western, it's really unpredictable where that 20% shows up. For example, the reason the census count can be reasonably accurate even in the adversity of undocumented immigrants is that Turkish police can walk up to anybody they see as suspicious on the street and ask for their TCKN / ID to do a spot check for arrest warrants — decidedly free of the need for the western concept of a 'probable cause' or anything like that, it's a member of your community asking to validate your ID. Since they have the readers attached to their mobile phones, the check takes a couple seconds.
This also, by design, makes it really hard to be undocumented in Turkey. There's a significant skin tone difference between people from the Middle East and those in Istanbul, and the police does absolutely use this to be more scrutinous if you're even just walking by them. From an US perspective this is discrimination, and it definitely is, but this is how it works — and they're pretty good at spotting non-Turkish people. As a real example in action, there was a recent directive that required refugees to keep their residences within the borders of the first municipality they registered in (many of the Syrian refugees had moved to Istanbul from where they first registered for benefits), and the police managed to significantly clamp down on unregistered immigrants fairly quickly and send them back to the cities they registered to.
In this specific case this was the right move since the resources are allocated to states (ils) based on where refugees are registered and Istanbul alone does not have nearly the capacity to house that many refugees. My point is that Turkish police does have a lot more leeway than the US police before it becomes socially unacceptable — and while this is overall not a great thing (loose oversight), it also makes them much more effective at making a census in the right ballpark.
Turkey has at least two ... porous ... borders that I've seen in action, so I am doubtful that counting births and immigration visas can provide an accurate real-time count of every person in the country.
Which borders?
I think the border to Syria was at one point fairly porous, for humanitarian reasons, but as of now even that is sealed shut pretty tightly. It's actually third largest border wall in the world after Great Wall of China and US - Mexico one. [0]
How do they count undocumented immigrants? If you don’t go to school or have an “on the books” job, are you going to show up in the count?
"Counting everyone in the country" is one aspect of the census effort. Looking at where they live, what they do for a living, and if they live alone or with others are other aspects. In the US, when someone moves from one state to another, they're not required to notify the US government in any way. Most states have rules about getting your vehicles licensed within a certain time frame after settling in a state, or rules about when you must acquire a driver license and/or state ID. The rules are wildly different across all fifty states, and it's not feasible to "just track" everyone since the national government has no explicit authority to. The census, however, is explicitly assigned to the federal government in the constitution.
The idea that the US federal government doesn’t know where people are and who they are talking to is hilarious. We have the biggest spying apparatus in human history.
This is 100% political.
25% of Americans do not own a computer. 19% don't own a smartphone. 4% don't own a cellular phone at all. [1]
22% of Americans don't own a credit card -- although neither the Bankrate or Statistic Brain studies counted debit cards accurately, which are equally useful for tracking. [2]
All Americans do not even have Social Security numbers, as there are several moderately-sized religions who are excused from participating in the process. [3]
Finally, not everyone in America is here legally, and many of those people go to great lengths to avoid using technology or participating in programs which would lead to their identification and tracking.
All of these are good reasons to conduct a census now and then. But even if you find none of these convincing, the iron truth of the current situation is that none of these alleged panopticon efforts you're alluding to are allowed to release any of that information -- not to benefit allotment committees, not to infrastructure planners, not even to local law enforcement, so in practice it doesn't do anyone in the government any good at all.
1 - https://www.pewresearch.org/internet/fact-sheet/mobile/
2 - https://www.crnrstone.com/insightvault/2018/03/04/many-ameri...
The census predates mass surveillance. The bureaucracy changes slowly.
I doubt the agencies with the surveillance data want to share it.
The data isn’t a count of people. It’s one record per conversation, or whatever. It’s probably riddled with errors and duplicates, incorrect names, dates of birth etc. It probably misses a lot of people (like children). Getting a unique count of individuals, with accurate addresses and demographic information would be a Sisyphean task.
I wouldn't say political per se, at least not in the way most people would mean it. Taking the census every 10 years, in this particular manner, is required by the Constitution.
The short and slightly oversimplified answer is that one level or another of our government has all the same info on us most other countries do—if they don't, they can sure as hell ask any of several companies for it—but various interest groups demand, for a few different reasons, that they pretend like they don't, mostly in the name of some notion of freedom or advocacy of same (for the latter, see: making income taxes annoying for hundreds of millions of people only because some folks want them to be annoying)
> Why doesn’t that work in the US?
Because the US does not have border controls between states.
The US federal government has an exact, real-time count and location of every single person with a phone in the country due to universal bulk surveillance.
Cross-referenced with the subscriber records from wireless, cable, and utility companies, databases from data brokers that are fed by loyalty programs and supermarkets, and every bank or credit card swipe in real-time, they already have 100% of this data in far higher resolution than will ever be collected by the census. They know about your address updates even before you tell the DMV.
I will be ignoring the effort.
> I will be ignoring the effort.
Then you do so to the detriment of your state. Regardless of whether or not the government has the information via the means you describe, House and budget apportionment is determined by the census numbers, and not this "shadow data".
I think my cell phone setup throws a bit of a wrench in the tracking glue. I use a sim card from China with an account that's not in my name. While I've got a China bank account with Wechat pay connected to it, so I can just hit a button and reload it with funds each month. I can make all the phone calls I need while roaming in the USA and also use a reasonable amount of data, it still only costs me about $60 bucks per month max. I don't lose any sleep over anyone tracking my cell phone. For a USA number, I use skype with VPN.
I have 6 phones. Am I 6 people?
You registered them under 1 owner cluster that links up to you, so no :D
Citation needed
https://www.theverge.com/2019/11/14/20965354/nsa-intelligenc...
https://techcrunch.com/2019/06/26/nsa-improper-phone-records...
https://www.wired.com/2010/12/realtime/
https://en.wikipedia.org/wiki/Room_641A
https://www.aclu.org/blog/national-security/not-just-nsa-dat...
>https://www.theverge.com/2019/11/14/20965354/nsa-intelligenc....
>The NSA has stopped collecting location data from US cellphones without a warrant
uh, isn't that a counter-proof?
edit:
>https://techcrunch.com/2019/06/26/nsa-improper-phone-records....
>https://en.wikipedia.org/wiki/Room_641A
not exactly location data
>https://www.wired.com/2010/12/realtime/
sounds like a great way to under-count people with poor credit or access to banking
>https://www.aclu.org/blog/national-security/not-just-nsa-dat....
Sure, it'd probably be cheaper to buy the data off some data broker rather than doing a census yourself, but how clean is their data? Are they also going to disclose their methodologies? Is giving a few private companies the power to control government funding allocation and distribution of congressional power really a good idea?
Look at the second link for why it isn’t. Don’t believe spies when they claim to have stopped spying (after repeatedly demonstrating that they are willing to break the law to spy).
IP addresses are absolutely location data. The subscriber records of major network providers, both fixed and wireless, are known to the IC. An IP address is a street address and subscriber name.
Most states have programs to ensure that even the poorest have wireless phones, if for no other reason than to enable job application callbacks. Very, very few people do not carry any sort of phone, and all phone metadata in the US, including and especially location, is under continuous logging and surveillance.
I didn't fill out the last census forms in protest of the longform census that year(in Canada) that asked for what I considered to be unnecessary information I didn't wish to divulge. Nobody that lived in the house filled one out. Nothing came of it. Census people showed up at the door every day for a while, but would leave after a bunch of knocking with no answer. I do normally fill them out though. For reference Canada has similar census laws to America.
Next national census in Poland happens in 2021 and also here, you can be fined or sent to prison, get non-custodial sentence by not participating or providing false answers; persons covered by census are obliged to participate. Those who by gathering answers for census surveys and processing data will gain any profits can be imprisoned for up to 5 years.
So far no one from my family was chosen; since fall of the Eastern bloc there were 2 censuses, in 2002 and 2011
tldr: nothing
(Can confirm, was Census enumerator. You probably will get an in-person attempt from a very-low-paid enumerator, though.)
The pay is actually pretty decent, and is cost-of-living adjusted. In Portland where I live, for example, it's $18/hour, which is way better than some temp jobs I've had.
Interesting. Back when I did it, it was piecework. Maybe one to three dollars per household, depending on whether it was a short or long form, and whether you managed to get someone to fill it out. That was around minimum wage, though if you hustled (or pencil-whipped things), you could probably make more.
Also, I think I heard that these days you have to produce your draft registration number in order to apply, which sucks.
> you have to produce your draft registration number in order to apply
Not true. What you actually have to do is sign a form declaring that you are or aren't registered with Selective Service, and if not a brief reason why, which is the same information-gathering requirement for most or all federal jobs.
You can actually see the full set of paperwork with PDF forms here: https://www.census.gov/about/census-careers/new_emp.html The main one is the Declaration for Federal Employment, which is the generic form that everybody who ever wants to work for the government in any capacity has to fill out.
Glad to hear it.
So in theory, I should hold off on answering so someone gets paid to come to my house?
Do they have a quota to hit, like if I take my time with them do they get punished?
If you believe in the broken window fallacy and have time to kill, yes.
There's no quota, but enumerator hours-per-week are based on estimates of work and there's no particular effort to claw back unused time if the actual work is done early, so intentional foot-dragging just takes up de facto paid time the enumerator could spend on other things like school or job-hunting.
Or maybe they would prefer to spend their time finding people who are actually hard to enunerate, rather than wasting on someone who is just looking to consume their time.
The idea is to get "contacts" and tell them to be "hard to enumerate" and then collect a paycheck.
I was putting a hypothetical on helping them make more money, you shouldn't take it seriously.
Just don't set your dog or the cops on them, and they'll be happy enough.
watch "The Census: Last Week Tonight with John Oliver" on this topic.
Deleted: Made a mistake.
Your link does not support your claim, and this link gives more detail not supporting your claim.
https://www.census.gov/acs/www/about/why-we-ask-each-questio...
That's information which is 20 years out of date. Oliver is right, Trump is wrong.
I got randomly selected for the long one in 2010. There were several questions I was not able to answer but providing incorrect information is a crime and not answering is a crime. So I did nothing. They sent one or two threatening follow ups. Then an in-person enumerator showed up and ran me through the short-form questions only, which was fine.
It may be technically illegal, but when I worked for the 2010 Census we were told that it's fine if someone doesn't want to answer a question, and that's what I told people when I interviewed them.
According to this article, no census failures have been prosecuted since 1970.
https://thehill.com/opinion/white-house/381254-answering-the...
> providing incorrect information is a crime and not answering is a crime. So I did nothing.
The correct answer in this case is to say "I don't know". You can write in stuff on census forms and it's up to the local office to process those as special cases.
That's not something they say you can do.
In my case the correct answer was to do nothing, then cooperate with the enumerator on the short form questions, which as I said was fine.
I had the same thing happen. I sat down to do the long-form and it asked fairly invasive questions like what chronic diseases I had, and IIRC even how many sexual partners I'd had.
I pitched the thing, but they came to the door several times until I did the short-form.
Oof, yeah, I'm not filling out something that invasive. They can serve a warrant to Google if they need info like that about me.
You might be thinking of 2000, there was no long form in 2010. Could you not answer because you didn't know, or you didn't want to answer those questions?
OK, thanks, I managed to find the form they sent me. It was the Census' American Community Survey, which claims "Your Response is Required by Law". I agree that cooperation with the decennial census is required by law to the extent it collects ages and citizenship status of household individuals, but I do not agree that the Census clause of the Constitution authorizes the collection of the American Community Survey information, which is vastly beyond the scope of Congressional districting needs. The claim that my response is required by law is fallacious and is worthy of being challenged in court.
> Could you not answer because you didn't know, or you didn't want to answer those questions?
Didn't know many. Didn't see how the Census clause authorized the collection of personal and private information they were demanding I turn over accurately under criminal penalty otherwise. Not knowing was sufficient to not be able to respond.
> Didn't see how the Census clause authorized the collection of personal and private information they were demanding I turn over accurately under criminal penalty otherwise.
It's a law, not part of the Constitution.
https://www.law.cornell.edu/uscode/text/13/221
(Section 3571 and Section 3559 amend the penalty from $100 to $5000.)
> demanding I turn over accurately
Note the "to the best of his knowledge" in the law above.
I think the point was that the law, as applied to the ACS, does not fall under enumerated powers of the federal government, nor could it be reasonably seen to fall under the necessary and proper clause. The census? Sure. But not the ACS.
As someone who was sent the ACS and started to fill it out before abandoning it because the government has no business demanding answers to some of those questions, I felt violated. Then the Census Bureau harassed me. And then nothing came of it. Shame on them.
Politifact couldn't find evidence of prosecution for lying or refusing since 1970.
https://www.politifact.com/factchecks/2014/jan/09/us-census-...
From what I understand, the illegality is mostly used as (a) a tool to increase the baseline number of responses, and (b) to by consequence make it illegal for anyone else to force someone to not complete the survey for some reason.
Nothing. Nothing happens. You may get an in-person visit, and if you blow them off, nothing also happens.
As with voting, it only encourages them. They are passionate about getting you to participate because they want the appearance of legitimacy, so you get obvious fear-and-guilt-prop like this.
> Nothing. Nothing happens.
Except for your area getting federally undercounted, which loses representation and tax dollars for you and all your neighbors.
Is there any reason to believe that more people are failing to fill out the census in my area than in other areas? Because if the failure rate is about the same everywhere it probably does make little difference.
How do you know the failure rate is the same everywhere? It is very possible that the failure rate correlates with the region because it is not a purely stochastic process. Some users here think the government has no business in demanding census information and simply ignore it.
I don't have any data either way, do you? HN users don't represent a geographical region, per se, but I suppose you could argue that they're mostly in SF? Regardless, I imagine there are other communities that might also have a higher-than-baseline failure rate, but absent any evidence I have no reason to assume they're not fairly randomly distributed geographically. Especially at the granularity of US states.