Western Standard

The Shotgun Blog

« Stelmach: In defense of oilsands | Main | WS on the Census: Walter Block »

Friday, July 16, 2010

WS on the census: Why dumping the census might be a good (or bad) idea

Aaron Wherry, over on Maclean's, has sent out a call to all "conservative-minded" organizations to try and defend the Conservative government's move to scrap the mandatory long-form census.

We're "libertarian-minded" over here at the WS, but we thought we'd oblige Wherry a little, and offer some thoughts on the census.

But let's be clear. It's weird that the Conservatives, whose leader has given the signal to purge libertarians from Conservative Party ranks, have been sounding a libertarian tone on this issue. That tone rings entirely hollow to our ears.

It was Harper, after all, who gave libertarians the elbow during the Manning Centre get-together a few years back. And it was under the Tories that Marc Emery was unceremoniously shipped off to a federal prison in the U.S. to face five years for a "crime" he would have gotten a slap on the wrist for here in Canada. Also, mandatory minimums? And never mind the budget.

This? Scrapping the mandatory long-form census is supposed to be the fop that brings libertarians within the folds of Harper's skirt? If you want us back, dearest Conservative Party, then legalize pot, bring Emery back to Canada to serve his time in a Canadian prison, cut the income tax, reduce spending, and balance the freakin' budget.

Scrapping the long-form census won't lead to droves of libertarians lining up to join the Conservative Party. But, and this is an important but, we're happy to see movement towards greater individual liberty, regardless of motivation, and regardless of party label, and regardless of past attempts to give libertarians the cold shoulder (or the elbow). We're reasonable people over here. If you want to talk increasing individual liberty, we'll talk.

We've sent out requests to our little constellation of the Western-Standard-verse to offer reasons why we might not be so fond of the long-form census (or why some of us might like the census, the request was not restricted to only slagging the census). This weekend, we'll be posting responses as we get them in our inbox.

But I'll begin with myself:

Eliminating the mandatory long-form census strikes me as a good idea.

We're all familiar with polling companies, is there a reason why use of polling companies is not a good idea? Why this alternative is somehow unreasonable or ridiculous? (I really don't know, you tell me).

Come to think of it, why wouldn't compensating people for filling out the long-form census be a good idea? Forget spending money on advertising and census enforcement, spend it on compensating participants for crossing ts and dotting is. Would that be a wholly outrageous suggestion? Would that, dear statisticians, motivate enough people to fill out this thing so that the government can centrally plan much better? Would it yield sufficiently robust findings?

Okay, those are my constructive suggestions for the folks who think that, in principle, it's just fine to make people fill out forms under threat of sanctions. But why should we think that it's okay to make people fill out forms without paying them for it and without them having a choice about it?

I really don't know how much uncompensated time I'm expected to spend filling out forms for the government, but I really don't think it should be more than a minute or two. It's insulting enough for a taxpayer to work for the government half the year, but to have to spend several hours agonizing over hundreds of exceptions, exemptions, special tax breaks, brackets, and so on, just adds to the frustration. And never mind starting your own company and keeping up-to-speed with the latest wisdom in the forms of new regulations coming out of Ottawa or more local governments. I suppose I may be a special case, since filling out paperwork is an especially tortuous form of hell for me.

We might also wonder just how well the census helps the government make really good decisions. I'm not sure how familiar you are with economics, but a general conclusion is that governments are real bad at allocating resources efficiently. Do we have studies to demonstrate that the census has improved allocative inefficiencies? Census-supporters: Can you show me empirical studies demonstrating that the thousands of hours Canadians collectively spend filling out the long form of the census has actually made the Canadian government (or any government) more efficient?

Of course the census might be useful at doing just that. And I bet that, in several cases, it actually does help. But do you really think that government spending decisions are made on the basis of census data, rather than on the basis of possible electoral gain? Do you think it's a really surprising coincidence that the Conservatives have dumped stimulus money on Conservative ridings, or on swing ridings? Really?

But we're getting off track. I'm trying now to persuade you. Instead, I should really be focusing on explaining why I, personally, dislike the census. We can rephrase all of the above into this simple little explanation:

1) I think it's unseemly to expect Canadians to fill out forms without compensation.

2) I think it's extra unseemly to threaten people with sanctions for not filling out forms.

3) Electoral considerations trump other considerations when it comes to dispensing tax money.

4) Opinion polls might be just as good as the census in terms of getting the info, without it being unseemly.

5) A voluntary census with compensation can be not unseemly, and might be sufficient.

So there are my (constructive!) suggestions, and my explanation for why I don't like the census. Soon, Aaron, you'll get some more responses. Some, I suspect, will be fairly radical. So be on the lookout for them.

Posted by P.M. Jaworski on July 16, 2010 in Census, Libertarianism | Permalink


The trouble, once again, with this is the erroneous assumption that people can be pigeon-holed; that they are one hundred per cent this or that. The only "consistent" people are some of those in mental institutions. This is what makes us living, breathing human beings and not robots.

The scrapping of the mandatory census is long overdue regardless of what label you try to attach to it.

Posted by: Alain | 2010-07-16 3:36:55 PM

I used to do surveys for a living. We would call people back until they answered the survey. Even if they said no. People had the freedom not to answer the survey, we had the freedom to call them over and over until they did.

Posted by: Altavistagoogle | 2010-07-16 4:36:42 PM


I did that back in high school. Cheers to being a pain in somebody's neck :-)

Posted by: Terrence Watson | 2010-07-17 12:31:14 AM

Don't dump the census, but abandon as relics the questions which allow government slap labels on the population beyond:
1. Adult or minor
In theory government does not differentiate between citizens, guaranteeing equal treatment from the government to all. This is attenuated only by the age of majority (won't someone think of the children?); ideally the government can only be allowed to ascertain numbers about the two categories of people they govern, child or adult. Male, female, white, brown, wealthy, devout, pick a label; none of these are the governments legitimate concern unless the target is social engineering and demanding the data reveals the hypocrisy of 'equal treatment' by government. As far as the usefulness of the statistical data to rent seekers and for profit companies go, they can pay for the data instead of relying on extorted demographic data. Pay for your own polls, in other words.
2.Citizen or resident alien
I don't expect non-citizens in the country illegally to fill out the census, but I do expect the government to know the statistics regarding people that would continue to live here without becoming citizens.

I've taken to asking pollsters that phone me at home asking for free data to pay me. They have all declined so far.

Posted by: Dana | 2010-07-17 10:56:38 AM

The comments to this entry are closed.