Beyond The Commons

Aaron Wherry covers all the goings-on in and around Parliament Hill. Follow Aaron on Twitter: @aaronwherry

Categories: Beyond The Commons, Capital Read

The libertarian cavalry

by Aaron Wherry on Saturday, July 17, 2010 1:21pm - 0 Comments

The economists, statisticians, city planners, social groups and religious leaders have had (and continue to have) their say and so now the folks at the Western Standard have helpfully compiled census opinions from their libertarian orbit, including takes from PM Jaworski, Walter Block, JJ McCullough, Terrence Watson, Martin Masse, Hugh MacIntyre and Paul McKeever.

Here’s another opinion sent directly to me from Matt Bufton, whose opinions, whatever his professional associations, are his and his alone.

It is likely that the government’s decision to discontinue the mandatory long-form census will have little to no effect on public policy making. Those who feel that the information provided by the census is vital to the government’s ability to make sound resource allocation decisions should consider the general quality of government decision-making with the vast amount of information currently available to policy makers. Governments are notoriously bad at allocating money efficiently – to take one example from thousands, how many Canadians support the recent decision to spend $16 billion on new fighter jets?

These types of decisions are not the result of corrupt or incompetent leaders. They stem from what economists of the Austrian School call the “knowledge problem.” It is simply not possible for any individual or organization to know enough to consistently make good decisions.

When those who make policy have access to vast amounts of data they may become convinced they know enough to intervene in the economy in just the right way. The results of these interventions can range from benign waste (fancy fighter jets) to tragic suffering (Mao’s attempt to restructure the Chinese economy lead to the starvation of over 20 million people).

Much of the knowledge our society finds useful or interesting is obtained without passing a law forcing randomly selected individuals to provide detailed personal information. Pollsters are able to provide data on topics ranging from how many Canadians would vote for a given political party to which world leader we’d rather have a beer with. Perhaps it’s time for the government to start getting its information from willing respondents rather than threatening us with fines and jail time for non-compliance.

Tags: census, libertarian, libertarianism, libertarians, Statistics Canada, Western Standards

More by Aaron Wherry

Emily

Well, that's certainly confused.

I was not only unaware Mao had a census to work with, I was unaware we've done any political polls on the fighter jets as yet.
http://intensedebate.com/people/psiclone psiclone

Excellent article by Matt and spot on – you see I am in the datamining biz!
Wascally Wabbit

Clearly – psiclone belongs to the Harper School – estblish a position and look for facts to support it! If you don't find them – plunge ahead anyway!
As I read Mr. Bufton's general thrust – he is absolutely supportive of the necessity of compiling facts – statistically supported – to – as the Austrian School suggests – to provide an individual or organization with as much information as possible as input to the policy making / decision making process!
I agree – how much information one can practically collect is constrained by time and cost & availability limits – that – for example – is why sampling techniques are used – to collect the information required from a subset of the total domain.

Wascally Wabbit

As one who has been in the Information gathering business for – I would hazard a guess – a lot longer than psiclone – and as one who has been designing databases for about 40 years – I know how impossible it is to anticipate all the possible combinations and permutations of information requirements ahead of time.
That having been said – nothing the information managers can do can stop political decisions makers making arbitrary decisions that do not make sense to someone looking at the facts. The resolution of that disconnect is to remove that Bad decisionmaker from the power that permits him or her from doing it again!

http://intensedebate.com/people/TJCook TJCook

Thoughts:

1) Government is hardly the only user of this data, as has been pointed out over and over of late. The magical free hand of the market makes extensive use of this data.

2) My gut says that lower levels of government probably make better use of the data than, say, Peter MacKay. This is solid, tangible data that's directly relevant to the management and growth of cities and municipalities.

3) The fact that government makes imperfect use of the data doesn't justify making the data worse! The census data is the gold standard and can't be replaced by polls, politicians' "guts" or the whims of the public. The data can also serve to evaluate policies once they've been in place a while.

4) The reference to Mao in a discussion of Canadian policy makes me think that Godwin's Law needs a corollary.

Olaf

1) True

2) If your gut has a moment, my brain has a question for it: how is my religion, whether I sometimes have trouble bending, or other similar curiosities "directly relevant to the management and growth of cities and municipalities"?

3) True. The point is, using statistics in decision making, especially in the political realm, is an art, not a science. Therefore scientific precision isn't necessarily required, nor will it necessarily improve decision making by any discernible degree. If it did, we might consider having severe and rigorously enforced penalties for not filling out a form, fact checkers to verify the information provided, etc. While imperfect use of the data doesn't justify making it worse, it is good reason to be skeptical as to whether or not the inevitable pan-Canadian public policy meltdown will ever come to pass as a result of the change.

4) I don't know if you're suggesting he was drawing an equivalence, but it seemed to me he was using an example of extreme government intervention in the economy (in that the "range" does not go beyond Mao, he's the worst example), which would seem to be accurate.

http://intensedebate.com/people/amherstvw amherstvw

Having graduated university at age 65, my mother last time lamented she did NOT receive the LONG form and therefore "they won't know about 'our' religion" (United Church of Canada). BOOKMARK THIS SPOT !! A CITIZEN WANTED THE LONG FORM !! She had been in class with many Chinese students who couldn't understand WHY people believe 'these things' … my mother said THEY "just wanted to get ahead". So there are actually people who want the long form … I think she, like Dean, took a few statistics courses too …
Once working with ESL newcomers in a conversation group I interpreted aspects of Christmas as a non-religious person and they asked me the same question. So maybe ask not why there are Jedi knights … ask how our rapidly changing religious profile will affect Canada's outlook toward "Christian charity" here and abroad.
People who have trouble bending sometimes see physiotherapists at clinics a lot and as this service is de-listed in Ontario they claim the medical receipts on their taxes.
It would be fine to have only historians, sociologists, health care planners, etc setting up the questions. As I suggested previously it was perhaps the cost-recovery and service-fee mentality of Mulroney et al which undermined the better angels of our census a few decades ago. Sure, make the census questions less 'commercial' if it would make feel better about its appropriateness.

http://intensedebate.com/people/doug_rogers doug_rogers

http://dougsamu.wordpress.com/2010/07/17/from-the…
Alsadius

If she wants to give out large amounts of personal information publicly, she doesn't need a census form to do so. She should just do it the old-fashioned way, and start a Facebook account.

http://intensedebate.com/people/Thwim Thwim

Yes, because that's terribly useful for the purpose..

..oh wait.. it isn't.
http://intensedebate.com/people/amherstvw amherstvw

1. United Church of Canada has recently declared "The Facebook" haraam.
2. Religious trivia/bet you didn't know … drawing a stick figure resembling Dr Robert McClure will meet with dire retribution.

http://intensedebate.com/people/Stewart_Smith Stewart_Smith

I want to respond to your 3rd point because it is central to your quest to understand why some of us are convinced of the negative impact of the degradation of the census data.

When you describe the decision making in the political realm as art, you don't really mean art in the Miles Davis/Lawren Harris sense. Others might use the term inexact science. The point is that there are so many uncontrolled and in many respects unknown variables that impact the decision making process that there is no tractable approach to even begin to write an deterministic equation that links decision making to the statistical value of the data used.

Yet, you readily acknowledge that information is important… i.e. completely false information would be very bad, no information would be bad (unless you are a libertarian sorted into the crackpot bin). However, your main point is that perfect information is unattainable, so as long as your information is pretty good, don't sweat it.

In essence, you are stating there is a statistical relationship between the use of statistics in decision making and the quality of the decision made. Moreover, you are asserting that there is a positive correlation between the quality of the data and the quality of the decision.

Given the above, the rest is not science but mathematics. Certainly anyone would agree the scatter in the quality of decisions plotted against quality of data would be large. So no one decision will ever be decisively impacted by a modest degradation of the data. However, the census data is used in a very large number of decisions. Just like in Vegas, eventually the odds always win out.

Olaf

That's an interesting theory Stewart, I can't say it's without merit. I'm not sure that I'm saying there is always a statistical relationship between the use of statistics and the quality of the decision being made. It really depends on how one uses those statistics, or in what circumstance it is used. But I do think certain core statistics are necessary and useful in the provision of services, like population numbers, age, gender and income. So if your municipality doubles in population between censuses, and if somehow this escaped your notice, you might think about adjusting policies and programs to meet this need. Likewise, if the average age in a city is steadily increasing, and there's reason to think that trend will continue, there may be a need to adjust services prospectively, or at least prepare for the needs of an older population. Great, fine, use the short census data.
Olaf

Do I think that when people are using numbers such as "how many hours do you spend doing housework per week?" and "what is your religion?", the people using those numbers are probably involved in a certain degree of social engineering and have their positions rather firmly established before the census data comes out. So I don't think that information improves decision making quality in the public sector. Or rather, I should say that absolute precision in these matters is not required to make a decision as to whether to fund this or that religious festival to curry favour with this or that increasingly populous religious group or have the government pay to do housework as a measure towards gender equality.

http://intensedebate.com/people/Stewart_Smith Stewart_Smith

What you say is no doubt true… i.e. that the decision makers and wantabe decision makers are biased. Nevertheless for most rational people there is a tipping point at which facts overwhelm inate bias. Poorer data means fewer tipping points are triggered more, bad decisions made.
Olaf

Nevertheless for most rational people there is a tipping point at which facts overwhelm inate bias.

Well if people are making decisions based on arbitrary "tipping points" instead of a rational analysis of the data anyway, what does it matter whether the numbers are slightly more inaccurate than they used to be? Sometimes the inaccuracy may lead to a good decision "The earths temperature has increased 10 degrees over the past decade? That's it, I'm biking to work" or a bad thing "25% of the Canadian population is Muslim? That's it, I'm starting up a xenophobic far-right European style protest party". But in both cases they're arbitrary and the statistical inaccuracy was incidental to whether or not we'd describe the decisions as "good" or "bad", which is generally determined on the basis of opinion anyway (e.g., a few, like Mark Steyn, may have inverted my examples, switching the 'good arbitrary decision based on inaccurate data' with the bad).

The notion that "slightly less accurate data" is linked to "making specific decisions wrongly" seems like blind faith to me.
http://intensedebate.com/people/Stewart_Smith Stewart_Smith

I did not say they were arbitrary… they just are not known apriori. In any case, you already knew this, basically you make a decision (route to work, favorite lunch at restaurant, people you like etc). We would be paralyzed with if continually re-evaluated every decision/opinion we had. We may do it periodically because we are feeling introspective (if Gaulion reads this… that is what jazz is all about), however sometimes we simply run into contrary evidence that forces use to rethink.

Yes, the concern over global warming is a contributing factor to people's decision to bike to work.

Good or bad in my usage means consistent with the intent of the decision maker, not the Final Decision Maker (for Gaulion). Yes, within that definition bad data could encourage a "good" decision.
I suspect that there is a solid mathematical argument that statistics dictate that this is alway outweighted but I am not that good a mathematician. However, what is certainly true is that if decision makers lose faith in the data, they will discount any that it inconsistent with their apriori beliefs. (Check these boards anytime a political poll is discussed.)

Your last point… blind faith. Faith in the absense of any knowledge. Absolutely not, mathematics is the one area in which faith is not required. We are talking about a mathematical proof. Can I recreate the proof for you… no. But I can't recreate the proof of Fermat's Last Theorem either and yet I have "faith" that it is correct.

http://notquiteunhinged.blogspot.com Catelli

2) Are there no questions on the long form that could be considered relevant by municipalities? Or are you just cherry-picking?

3) If there is no-one making good use of this data in a true scientific manner, why not advocate for dumping the census altogether and save some money? OK Yes, this won't matter in the long run because everyone is partisan. Doesn't justify making it worse (I know you said it didn't) but that then just means we have another case of a government being inefficient and wasting money. Oh yay!

http://notquiteunhinged.blogspot.com Catelli

Further to 3). Statistics Canada has many customers that are willing to pay for the compiled results derived from the census data. Now that these results will be less than perfect, is it not reasonable to concluded that this source of revenue will fall off, decreasing the revenue that offsets the costs of running Stats Canada?

Olaf

The results are less than perfect now. But yea, I have no idea how much they sell to the private sector. It does seem, if you browse through their stuff, that pretty much everything is available for free online, but I have no clue.
http://notquiteunhinged.blogspot.com Catelli

If you dig deep enough, there are certain data that are only available for a fee, along the lines of raw data tables (with personal information scrubbed) for your own analysis.
Jan

Only the overview of the raw data is available – free – online.

Olaf

2) It's not that none of the information is "relevant", in some sense or another. It's that policy making inside government isn't the type of thing where you punch numbers into a formula and out comes your accurate policy. If it were, the hemming and hawing that less accurate data will somehow lead to substantively worse outcomes would be justified. So, would it be good to know that the french speaking population in your municipality has increased by between 40,000 and 50,000 over the past decade? Yea, although it won't necessarily change anything based on a million other factors, but it would be good to know. But would it be necessary to know that your municipality saw a net 43,567 increase in french-speaking people? No.

http://notquiteunhinged.blogspot.com Catelli

It might make a difference if there were a funding decision based on numbers of people of X descent. Given that many decisions are based on a line (if over 40,000 then you get X if under you get Y). Doesn't have to be direct subsidy for having French speakers, but it may determine whether you get a language services centre OR just a regional representative.
Olaf

Well if it's based on a line it's pretty arbitrary anyway so what's the difference. Some reason why 39000 french speakers only require a regional representative but 40500 need a languages centre?
http://notquiteunhinged.blogspot.com Catelli

All lines are arbitrary, but they do exist. Everyone accepts that a town with pop 250 doesn't justify a full centre, but a city with 55,000 probably does. It's trying to find that line in the middle to cut off funding that's always the problem. Always will be. However, these distinctions are more palatable if everyone can agree on the accuracy of the data. (Those that don't get the funding will still grumble, but if they feel they were slighted due to inaccurate numbers then they really revolt. There will always be complaining, but you want to minimize it, or alternatively don't give them another reason to bitch.)

Olaf

3) Like I said, I think general trends are useful to know. To the extent that these can be collected from other sources (e.g. Revenue Canada, Voting registries, immigration Canada, other more or less exhaustive government databases) they probably should be, but to the extent that they can't I would guess that a short form+voluntary form+private sector surveys would be relatively useful in determining broader demographic trends.
Alsadius

2) So you defend mandated violations of privacy, because some of the questions asked might give the government information that might be useful to somebody. I don't know about you, but I prefer to hold egregious violations of privacy to a higher standard than that.

3) The census has one fundamental and necessary purpose – the allocation of electoral districts. If we want to have Parliamentary ridings, we need to know how to divvy them up to ensure proper representation by population(or at least, as proper as our Constitution allows for). Everything else the census does is a tacked-on afterthought, put in place because someone thought it'd be an easy way to collect data they wanted. It is an easy way to collect data, of course, but mandating that people answer questions about their income, relationships, race, religion, and the like needs a better reason than "because I want to know".

Dave

There you go — the Cook Law. Thanks!
Terrence Watson

Some questions on the "statistics enable government planning" argument:

(1) If we deprive the government of data, does that mean we'll get less planning? Or worse planning?
(2) If government is to be scaled back (in the real world) this will only occur through projections and planning; does depriving government of data actually make it less likely we're going to get a smaller government?
(3) Do we make the lives of special interest groups or rent-seekers more difficult if we deprive them of data? Could that contribute to smaller government, by making it harder for special interest groups to coordinate their activities and exercise influence over the government?

My answers:
(1) is "we'll get worse planning."
(2) is impossible to know.
(3) is a tentative yes: but don't ignore legitimate social science research that might be negatively affected if the long form is scrapped.

http://bigcitylib.blogspot.com bigcitylib

3) Yes, but depriving these people of data also involves depriving hard working marketers and other entrepeneurs of information. Do you hate Capitalism?

Mike T.

marketers and entrepeneurs ARE special interest groups.
hosertohoosier

One of the arguments for free enterprise is that it is better at operating in a world of limited information. This is because you have many decentralized decision-makers. Some may make good choices, some bad, but over time the ones making bad choices tend to either change or fall out of the system. Governments have far less of a basis for comparison, and inherently different selection mechanisms.

Moreover, advertising is not necessarily a perk of capitalism. It provides limited public benefits, rather, it involves people spending money to improve their market share – a zero-sum game. The important decisions of capitalism involve research and development and the innovation of new products. In a world of high information this may orient innovation towards niche products (which often have fewer spillover effects), because you can tell there will be a market for it. In contrast, in a low information world, it makes more sense to invest in productivity improvements (produce what already sells more efficiently) and broader improvements (eg. it was relatively obvious that trains were better than wagons, even without detailed reports).
Terrence Watson

Aw, come on. Wal-Mart will have to muddle through with data they gather on their own. Less reliable data, perhaps, but Wal-Mart and its corporate friends will make do.

http://bigcitylib.blogspot.com bigcitylib

Walmart will have no standard against which to measure their results. You some kind of communist, Terrence?

And H2H is blithering. He's either a commie or a teenage libertarian that got hold of an economics text, and thinks he knows better what Capitalism is than the Capitalists do.

Terrence Watson

Wal-Mart doesn't need maximally reliable and accurate information to do what it does. I'm sure it would like information of that quality, but I'm not aware of any plausible or even coherent notion of political morality that demands we organize society to maximally benefit Wal-Mart.

And, of course, we know corporations love free market capitalism — almost as much as they love getting subsidies and other perks from the government. And I love the irritation of the business class almost as much as I love the annoyance of the political one, which is less communism and more anti-everythingism. :-)
hosertohoosier

Well I may not be a big city liberal, but…

You are making a reductionist fallacy that because individual firms are doing something it is something that they want to do, or something that they are doing consciously. This is false. Imagine I chose to open a shoe store that gave away shoes for free. I would soon run out of money, and my business would fail. Even if people starting shoe stores didn't realize this as an inherent law of business, however, I posit that you would see few shoe stores giving away things for free. Why? Because any store doing so would go out of business (and to some extent because other firms would adopt the strategy of successful firms, even if they did not understand the mechanisms behind that success). What we see at any given time is a snapshot of the firms that survived.

Even if businesses operated in a complete vacuum of demographic information, they would soon resemble something approximating a world of high information as poor strategies failed and good strategies succeeded. You look at businesses in terms of firms, I see firms as being involved in a collective evolving ecology.

Single decision-makers are not subject to the same pressures. The primary selection mechanism operating on government is the fact that politicians must be elected. What you will see across time are policies that get governments elected and re-elected. Parties that ignore the polls are like shoe stores giving away shoes for free. Bad public policy could continue indefinitely if it got people re-elected. Worse, there are fewer relevant comparisons from which governments can adopt leading strategies than exist for firms. Moreover, the pressures by which badly run countries get selected out of the system (historically, war has been the main one) appear to have ceased operation.
Anon 001

Ah yes, efficient market hypothesis as applied to real life! Perhaps you should issue public-policy default swaps and then, one day, you too can apply to the government for a bailout.
hosertohoosier

How does the subprime mortgage crisis vindicate the idea that more information benefits corporate decision-making? Quite obviously it doesn't, and in many ways the subprime mortgage crisis would have been impossible in say, the 1960's because firms would never be able to get enough information on lower income investors to even think of participating in the subprime mortgage market, or engaging in risky derivative trading.

Also, the efficient market hypothesis doesn't hold that markets will always instantly converge on efficient outcomes. It is a long-term argument, and one that surely involves many bumps on the road.

Aongasha

Memo to Wherr et al
We don't care. Next scandal please!
regards
average Canadian (voters and non-voters alike)
Aongasha

Sorry for the mis-spelling Mr Wherry
http://bigcitylib.blogspot.com bigcitylib

"Pollsters are able to provide data on topics ranging from how many Canadians would vote for a given political party to which world leader we’d rather have a beer with…"

Pollsters direct calls into those areas that they consider to be representative of certain significant segments of the population. Representativeness is determined by refering to census data. Bonehead.

http://intensedebate.com/people/LynnTO LynnTO

….that's only if there's a quota structure set up. Basically, in true random sampling, all the possible telephone numbers are put into a sequence and a randomizing equation pulls the numbers that are to be called. Even in quota structures, all the households that fit a certain profile (households in Ontario with a man age 35-44, for example) have their numbers put in the same process.

Once the survey is complete (IE, it reaches the target number of completed surveys) the data goes through a process to weight it, so that those random responses are then controlled so that overall, they are representative of the population overall.

http://intensedebate.com/people/Jenn_ Jenn_

I wonder if it would be worthwhile to discuss a small honorarium for filling out the long-form, similar to the type I get when participating in one of those focus groups I so love to go to. I mean, instead of advertising or 'enforcement', how about keep it mandatory but give, say, $25 when you've returned the long-form you were lucky enough to get?

http://intensedebate.com/people/PhilCP Phil

Or how about a tax rebate/credit (fully refundable) if you do fill in the long form.

That rebate/credit could be on the line immediately following the rebate/credit for voting.

Alsadius

A refundable tax credit, a honorarium, and a subsidy differ only in name.

Phil

Understood.

I was speaking to a possible mechanism of transferring the funds, and also drawing a parallel to another task often associated with citizenship, a task that also seems to have participation "problems".

Thanks!!

http://intensedebate.com/people/Jenn_ Jenn_

Actually, I greatly prefer the refundable tax credit. It ensures you get people filing their tax return as an added bonus (either they do or they don't get it) and it is cheaper than cutting a separate cheque.

hosertohoosier

The problem is that might worsen sampling bias. A $25 honorarium is a relatively larger incentive for a poor person than a rich person. We need for the survey-takers to be a random sample, which is why coercion is necessary.

http://intensedebate.com/people/Crit_Reasoning Crit_Reasoning

Exactly. I'll add that there's no need for incentives; it's enough that census compliance is mandated by law.

http://Www.westernstandard.blogs.com P.M. Jaworski

“Mandatory” means threat of fine/jail which *is* a kind of incentive.

And people will respond differently to incentives like this, probably based on their income and/or libertarian attitude toward the census. We don’t have 100% compliance with the law. For example, war on drugs.

http://intensedebate.com/people/Jenn_ Jenn_

I understand wealthier people won't care about the $25 or whatever. But it is still mandatory and whether they like the compensation or not, they're still obliged to fill it out. So, I can't see how it makes it worse. Of course as previously mentioned, we haven't seen the jail terms for non-compliance.

http://intensedebate.com/people/Crit_Reasoning Crit_Reasoning

Why should we give away $25 to millions of households who were randomly selected to fill out a long form? If compliance is mandatory, there's no need for an incentive at all. So how could you justify it? Compensation for the time to fill out the form? Meanwhile, the "small honorarium" scheme would cost taxpayers more than $100,000,000.

http://intensedebate.com/people/Jenn_ Jenn_

Well, mostly because that seems to be the complaint–you are being asked to do something with no return for your time. And $100,000,000 sounds awfully high–let's change the honorarium to $10, although, that's still $62,000,000. We'd have to find out how much it costs to enforce and advertise the thing. I suspect I'd lose money, in which case, never mind. Bad idea. Let's not do that!
http://intensedebate.com/people/JustinWordswrth JustinWordswrth

"If compliance is mandatory, there's no need for an incentive at all. So how could you justify it?"

The incentive might be difficult to justify. I wish the same could be said for mandatory compliance.

Alsadius

You use this word "need". I do not think it means what you think it means.
http://Www.westernstandard.blogs.com P.M. Jaworski

I don’t see how this differs from enforcement. Threats of fines/jail is a bigger or lesser incentive for people of different income as well.

http://intensedebate.com/people/E_B_ E_B_

Make the incentive a chance of winning $1,000,000? The census lottery?

Canadians love lotteries…

http://intensedebate.com/people/Jenn_ Jenn_

Hey! That's a good idea!

http://intensedebate.com/people/ChrisWPG ChrisWPG

Jenn, as soon as you change the method of obtaining the data by adding ANY form of incentive you risk corrupting the data, and at the very least make the data incomparable to years where the data was collected with non.

http://intensedebate.com/people/Jenn_ Jenn_

I get that the previous data won't be quite as accurate as the new lottery census, but I don't get how the new lottery census is corrupting the information. I mean, it is still mandatory, it is still one-in-five, etc. Could you expand on that part?
ChrisWPG

I can try, but first I disagree with your premise that the lottery census data will be more accurate. I actually believe that as soon as you attach any reward to the respondents you will have a greater number of returns from the poorer end of the spectrum, which is why and how the data becomes corrupt.

But forget that for a second. If 95% of the households already submit the form without a reward, how much more accurate would our data be if we got the last 5%? Why are we looking for a solution when there isn't a problem?
http://intensedebate.com/people/Jenn_ Jenn_

Why are we looking for a solution when there isn't a problem?

Good question! I'm going on the not-very-plausible concept that there is something behind Clement's move. Other than Harper, I mean.

I was taking it as a given that 100% return of the questionnaires gives the non-corrupt data. If it is already weighted assuming that a certain percentage will not be returned, and that certain percentage will be 95% from the poorer end of the spectrum or whatever, then okay. But I was assuming it wasn't weighted at all.
http://intensedebate.com/people/PhilCP Phil

I gather (without being sure) that there are about 50 questions on the long form. Those 50 quesitons could be arranged in a list from most intruxive to least intrusive..

Then we could look at the list and evaluate the loss in effectiveness of planning that would occur if we ditched the top two or the top three or even the top five "Most Intrusive Questions" or replace them with better questions.

Volia, problem solved.
http://intensedebate.com/people/ChrisWPG ChrisWPG

Volia, problem created. Much like offering a reward, changing the order of the questions from one census to the next has the potential to corrupt the data. People may answer differently depending on the preceding questions.
http://intensedebate.com/people/PhilCP Phil

It seems that I wasn't as clear as I could have been…

The purpose of placing the 50(?) questions in order was only to allow the worst two or three or five to be either deleted completely or replaced with "better" questions, not to actually change the order on the form

I understand that shrinking the existing list of questions could create the same type of data corruption that you mention – even so eliminating the most offensive questions is still an option that I would at least consider.
http://intensedebate.com/people/ChrisWPG ChrisWPG

Agreed.

http://Www.westernstandard.blogs.com P.M. Jaworski

That’s a great idea, Jenn. At least, I thought so when I wrote my post (it’s linked by Wherry). I haven’t yet heard why paying people (or a tax credit) is a really bad idea.

http://intensedebate.com/people/Jenn_ Jenn_

Yes, well, I read it there and thought I'd move it down to the thread. So, we discussed it, and it turns out it is not a great idea after all. We can't afford it is why its a really bad idea.

brooster

Given libertarians' proclivity for rejecting government intrusion into their lives, and their consequent refusal to volunteer information, they are ironically proposing a world in which their viewpoint/needs/preferences will be the information missing (i.e., built-in selection bias) from government databases on which public policy is built.

Perhaps not a bad idea after all!

http://intensedebate.com/people/BGLong BGLong

But you have to be happy that the folks have place to hang out.
Olaf

You make it sound like the government is asking for our opinions. Libertarians aren't defined by their skin colour, income, religion or ability to bend, so there's no reason why their viewpoints/needs/preferences will not be taken into account on that basis.

brooster

You're quite right, but I also mentioned their needs, based on their demographic characteristics.

Alsadius

Precisely which of my needs are demographic in nature?

brooster

For a simple example: number of times requiring visit to emergency in last year (among a myriad of other similarly useful questions). You will note that the answer is not a matter of preference or opinion.

My point (made with tongue in cheek) was that libertarians, by refusing to complete census data, would be self-selecting out of what is generally regarded as one of the best information databases on which to develop social policy.

If you counter that libertarian values are independent of demographic variables, I would suggest that they may well correlate with age, level of education, and income (demographic factors).

So, as I originally suggested (tongue in cheek…I forgot that libertarians are evidently humorless and incapable of appreciating irony), go ahead…self-select out.
Alsadius

Oh, I don't deny that politics can correlate with demographics. Obviously, it does. But that doesn't really apply to the question at hand. How exactly does identifying my race or religion help the government figure out what sort of services it needs to provide?

Also, for what it's worth, if I got a long-form census in the mail, I'd fill it out. Most libertarians I know would do the same(though not all, admittedly). I just don't think it should be mandatory.
http://intensedebate.com/people/TJCook TJCook

Health care, pension, social housing, ESL, schools, transit, off the top of my head.

Jan

If you can't bend you are no doubt going to need help with daily living, if not now, in the future, Provisions for homecare, care homes etc. have to be planned for. Just because you don't understand why a certain question is being asked, doesn't mean it does not have significant meaning.

Olaf

Are you being serious? Do I really have to explain again and again why this makes absolutely no sense?

hosertohoosier

While I don't think the census decision is a good one, I also don't think critics of the census have used the best possible argument. Generally, they have focused on the idea that the census is either intrusive or unnecessary because other sources of information exist. The argument in the post above is essentially that government is imperfect, therefore we should do away with a tool that may make it less imperfect.

I pose another argument. Why is it that we often wax enthusiastic for the days of the 60's when public policy broad-based and bipartisan? While some of this is pure nostalgia, I think part of it has to do with information. The more information politicians have, the more they can do to target narrower and narrower constituencies. Whereas economists tend to imagine bureaucrats using census data impartially, they ignore the role of elected officials in public policy-making. Politicians are concerned with re-election and rent-seeking, neither of which are necessarily in the public interest. With good data, they can ensure that public spending is directed efficiently towards the few narrow constituencies needed for re-election (eg. Harper's micro-targeting strategy). This both increases the scope for rent-seeking (since it leaves more resources for such a purpose), and deprives many people of their share of public spending.

Absent good data, governments would not be able to efficiently target core constituencies. They would ultimately be forced to do what our leaders did back in the 60's – craft broad-based public policy that serves the national interest because it is the best way to get re-elected absent more efficient targeting techniques.

Jan

The answer can't be eliminate data. The answer has to be to demand better governing. Surely.

http://intensedebate.com/people/Halo_Override Halo_Override

It is. And don't call me Shirley.
hosertohoosier

That is a hope not an answer. The mechanism that drives microtargeting and narrowcasting is public demand for better governance. The problem is that voters want better governance for themselves, rather than for country. Politicians do what they do because it makes them more, not less successful. It might be possible to develop appropriate institutional constraints, but they are unlikely to work because by definition, those in power have been well-served by existing structures.

Anon 001

Given that the only people against the mandatory census are cranks like Kinsella and a few no-name libertarians, who seem to be shilling for the private polling industry, I'd say that this decision was yet another Harperian blunder on a grand scale.

http://Www.westernstandard.blogs.com P.M. Jaworski

I don’t see how my being a “no-name” libertarian detracts from my argument.

Can you please tell me why an argument is better or worse when made by someone with a “name”?

Also, shilling for polling companies? Utter nonsense.

Anon 001

When I said "no-name," I meant nobodies i.e. people of no stature, sort of like anonymous commenters like me. And no, arguments don't necessary become invalid if put forward anonymously.

However, given that all the libertarian arguments put forward to-date has the same tired old "government should not be interfering in the lives of private citizens" ring to it, and has nothing to do with the important role of statistically relevant data for important policy-making, I would tend to disregard them.

Besides, so many of these famous libertarians I keep hearing from, Tom Flanagan, Ted Morton, etc etc, have done nothing in their lives except feed off of the public trough. So it's kind of hard to take these people seriously when they go off on their less-government-intrusion rants.

Alsadius

There are valid reasons for the government to intrude into the lives of private citizens. I don't think that "Because we want to know" is one of them.
http://Www.westernstandard.blogs.com P.M. Jaworski

Neither Ted Morton nor Tom Flanagan are libertarians. Whatever gave you that idea?

Did you read the arguments that Wherry linked to? Some of them do have that, as you put it, tired, old ‘mind your own business’ tone to them, but others do not. And J.J. McCullough’s is pro-long form (although he isn’t a libertarian, so he doesn’t count as a member of the cavalry).

It’s fine if you disagree with libertarians, most everyone does, but do please address the arguments and not the individuals or their motivations (motives are relevant only to test the sincerity of those who claim to believe something, but the belief is true or false independently of the sincerity of any of the advocates or detractors of a position).

Terrence Watson

By the way, can we be known as the libertarian posse from now on? "Cavalry" sounds way too organized!

Jan

You do know that if you don't have uniforms you will not be covered by the protections of the Geneva Convention. Pick a t-shirt colour. Black is already taken.
http://intensedebate.com/profiles/peterjaworski peterjaworski

No. Cavalry is perfect.
http://intensedebate.com/people/Halo_Override Halo_Override

I generally use "cohort", although after a couple of drinks I can occasionally be caught replacing that with "stain".

http://intensedebate.com/people/A_logician A_logician

For demographic statistics, "cohort" is already taken.

http://intensedebate.com/people/Halo_Override Halo_Override

Yes, I was being A_humorist.

Ceeger

There's a lot of rhetoric floating around here. Let's get down to brass tacks on this issue:
The long-form, census questionnaire is still being sent out as usual, and those who want to fill it out can do so. The only change is that it is not mandatory to do so.
Basic information that is vital to government planning is still part of the normal census questionnaire; it is only the intrusive private life, work life, ethnicity questions that are used for social engineering that have become voluntary.
I'm not seeing a huge problem with that, especially in light of the news this year that Francophones were fibbing on the last census about their language proclivities, in order to ensure more federal money for federal French language programs.
You nanny-state lovin', Government interventionist appreciatin' lefties can still fill out the long-form questionnaire to your hearts' content – the rest of us who think the government doesn't need to know what time we leave for work each morning, or the age of the wiring in our houses, will forego the opportunity, thank you very much.

Steve

1) The issue is that making the long form census voluntary will probably allow for greater selection bias that skews results. It removes the element of randomness that comes from a mandatory survey of 20% of Canadians. Surely you can see the danger of only having 'lefties' fill out the form – it invites gross misrepresentation. Because census data is used to normalize results from other surveys that do not have the resources to be as accurate (be they government, corporate, non-profit or academic), making census data less reliable impacts everybody engaged in social research.

2) "Basic information that is vital to government planning is still part of the normal census questionnaire".

Define 'vital'. You already give far more detail about 'work life' dealing with taxation. UBC economist professor Kevin Milligan has written about the importance of data from the long form in shaping pension reform. Are these 'social engineering' schemes? I don't know. Personally I find these libertarian arguments a red herring. We live in a free society and there are extremely few responsibilities pressed on Canadian citizens, surely taking a few minutes to fill out a form accurately is not too much of a civic duty to ask for?
brooster

And you gummint-hatin' paranoid libertarians can sit in your darkened homes and lock yer doors agin' them evil agents of progress and mutual support. We'll be lookin' after each other long after you've run outta' cat food.

http://intensedebate.com/people/lgarvin lgarvin

Maxime Bernier is belatedly joining the cavalry with a pretty stunning revelation of his own: But Bernier said that when he was industry minister during the last census period, he received about 1,000 complaints per day about the long census being too nosy..

1000 complaints a day! How did the poor fellow find time to pursue a love life?

http://intensedebate.com/people/Thwim Thwim

More importantly, why did he not bring these forward to enact changes when he was the minister? Surely he wasn't simply ignoring people who complained, was he? I feel bad for the people in his riding.

ChrisWPG

Hint: Get an account, then you can edit your posts, and perhaps reflect on how poorly you represent the "average Canadian" with posts that garnish -19.
wilson

From Igarvin's link above:

"If some special interest group wants data on Canadians, they can do that, they can pay for that and they can do it," Bernier said.

"But we're not there to please special interest groups.
We're there for the silent majority of Canadians. And I'm sure that the big majority of Canadians understand that and they will agree with our decision."

Culture war!!!
Silent majority vs. elite special interest groups
http://intensedebate.com/people/RobinBC RobinBC

The bizarre and hysterical outcry about the disasters which will occur with the long form census questions being made voluntary beggars belief. Of course these are the same people screaming disaster because of "human caused" global warming and every other manufactured "crisis" (the latest being the Gulf oil spill). Save the forests, save the whales, save every threatened microbe we can find.

Meanwhile more and more of us live longer, have more wealth, educational and other opportunities than ever before.
Get a life people.
brooster

"these are the same people screaming disaster because of "human caused" global warming and every other manufactured "crisis" (the latest being the Gulf oil spill). Save the forests, save the whales, save every threatened microbe we can find."

Really?

Jeez, I don't remember the Marketing Research and Intelligence Association, the Quebec Community Groups Network, the president of the CD Howe Institute, the Canadian Council on Social Development, the United Way in Toronto, or the Association of Municipalities of Ontario ever speaking out regarding global warming, whales, trees or microbes.

Show me where they did.
Jayce

I understand the theory. Unfortunately, currently we get the most incompetent people are the ones that run for office. The most incompetent of those are the winners. And they are the ones who chose what goes where. Now we're shocked ("shocked, I tell you!") that the more money they spend, the more gets wasted.

The most shaming of all, is the electorate keeps putting these idiots into office.
http://intensedebate.com/people/Stewart_Smith Stewart_Smith

I really liked the Star Trek photo.
Mike T.

With a strong stand like this, Mr. McKeever looks ready to bring his party up from 117th place in the polls.
http://intensedebate.com/people/Crit_Reasoning Crit_Reasoning

Perhaps it’s time for the government to start getting its information from willing respondents rather than threatening us with fines and jail time for non-compliance.

First, the threats of fines and jail time for non-compliance are mostly empty threats. When was the last time that someone was jailed or fined for not complying with the census?

Second, if only willing respondents provide long-form data, the government will only have detailed information about willing respondents. But the government needs data about everyone, not just the subset of our population that will provide such data willingly.
Olaf

When was the last time that someone was jailed or fined for not complying with the census?

We might just not hear about it. "Breaking: Man forced to pay government total of $500 in census-refusal, smoking-less-than-five-meters-from-a-door crime spree". But anyways, if they're empty threats then aren't the forms voluntary already? And then doesn't the whole argument collapse into itself?
http://intensedebate.com/people/amherstvw amherstvw

Shannon Proudfoot had a good article on this today. No one gets hauled off in chains. Some fines. Probably like cops … press their buttons, then don't expect favours.

If the data is only seen in aggregate and StatsCan people swear an oath, I don't see what the objection is. Time? We've (looking around me) got all KINDS of time for computers … fill it out online if we get our average one long form during the next 30 years.

There is no compulsory military service in Canada. There are generous social programs. Health Care is cheaper than any alternatives people can suggest. It is peaceful. It's pretty good to be Canadian. This is one of the few "onerous" things our society asks of us.
Kat

During the last census…I was almost late getting my form in and called because I had a question. I told the guy who answered that I didn't want to go to jail and he told me that a loooooong time would have to pass before they would consider throwing me in jail.

I was slightly disappointed…a short period in jail where someone else cooked for me sounded like a vacation of sorts. And if my crime was being late on my census, well, that would be something I would laugh over with the grandchildren…should I ever have grandchildren.
Crit_Reasoning

They're mostly empty threats, but the simple fact that non-compliance with the mandatory long-form census is illegal is enough to ensure that most people comply. That, and the persistent nagging by census workers if you don't return it on time: "I'm sorry, Mr. Olaf, but you have to provide this information – it's the law! Here, let's take care of this right now. It should only take us 15 minutes…"
Olaf

This is from the McKeever piece, apparently quoting Marc Garneau:

"…no one has gone to jail over the census, at least as far back as 1981. Only about 50-60 people are charged over each census, with about six having to pay fines".

So yes, mostly an empty threat, depending on how many people refused and were given exemptions because they're a bigger pain in the ass than the government. Like WK.
Alsadius

So your justification here is "It's okay if we threaten them with jail time for not giving us personal information, because we're only using the law of the land as a gigantic fraud to make people give us what we want"?
Alsadius

"It's good to be Canadian, because the government doesn't demand much of us. Because of that, it's okay when the government starts demanding things from us, and we shouldn't complain, because it's good to be Canadian". Have I got that about right?

It's not enough to say that something doesn't make us into a bad country. Of course it doesn't – canada was a good country with a mandatory long-form, and it'll be a good country with a voluntary long-form. The question you need to ask is whether it's a good policy or a bad one.
http://intensedebate.com/people/amherstvw amherstvw

Well almost since its inception, Canada has had a "long form". It was administered by humans at your door. It wasn't much of an issue until the passive aggressive wedge Mr Harper just applied. It has been a good policy if 100 years later you are investigating literacy levels, where settlers came from and how this affected politics, conjugal status and rates of divorce … etc. … less than 1% nation-wide compared to today's about 50%. Misery or happiness? … I don't know.

It is unfortunate that Canadians copy Americans' crankiness about the same issues. American citizens have lost control of their political process … we are just in the early stages. Americans have a moral right to be cranky about most things.

I just wish Canadians would learn a bit about our history and stop aping Americans. For the greatest benefit, we could start with the Prime Minister. One hundred years ago 22% in his province were American born … and they weren't UEL !
Alsadius

And we'll still have a long form after this change, and it'll still provide a lot of valuable data. Historians will still use it a century from now. We just won't be threatening to throw 20% of the population into jail if they don't give out every imaginable piece of statistical personal information to a bunch of government bureaucrats.
http://intensedebate.com/people/amherstvw amherstvw

It has to be properly randomized to be worth anything. People living in democracies should actually spend some time learning about their government – rather than taking their ideas from random rogue cabinet ministers who have no background in the national institution they are knee-capping.

Since widespread use of the internet, all sorts of organizations have Alsadius-specific information – e.g. Apple, Google, Microsoft, your ISP. The stuff going to the government is looked at in aggregate.

If you don't trust the government, then don't vote for the Steve's Wrecking Crew next time.