During Wednesday CBC’s town hall,Terrence Watson asked:
…if the census is only used to _justify_ government programs, but never to shrink or eliminate them, then can anyone see why somebody interested in smaller government — like me — would be suspicious of it?
It is a bit of an irrelevant point, because changing the format of the long-form census from a voluntary to a mandatory one isn’t going to get you less government, it will only give you worse government.
But if Terrence wanted to see census data used to argue against a government program, all I he needs to do is open an issue of Canadian Public Policy. I picked an issue at random…
March 1, 1995. Sure enough, an article: Canadian child benefits: behavioural consequences and income adequacy. The conclusion of that article, in part:
This paper has two objectives: (1) to evaluate the EIS component of the new Canadian child-benefit system; (2) to assess the adequacy of the new benefit levels.
An evaluation of the earned-income supplement component (EIS) of the new Canadian child-benefit system leads to the conclusion that the IS will not be an effective means of increasing the labour supply of parents with low earnings… Thus it is recommended that the EIS component of the new child benefit system be dropped.
I’m sure our readers (who aren’t rushing off to have dinner) can find many other studies that reach similar conclusions using long-form census data.
If you’re one that believes that government programs are likely to be highly ineffecient, then shouldn’t you want as accurate data as possible in order to prove this to others?
Worthwhile Canadian Initiative 
Thanks – that sort of thing was on my to-do list.
No worries. It seemed like a really obvious point to me, because like anyone who has studied economic policy in Canada I’ve run across many studies that use census data to suggest that some policy is ineffective and should be altered/dropped. I believe I’ve even read one or two by Kevin Milligan – hope to hear from him!
Some of these guys have a very odd mental model of what academic and government economists do.
It reminds me of the Trotskyite sociologists I’d run into at TA union meetings at Wisconsin. They were absolutely convinced we just sat around proving the first welfare theorem to each other. At least they gave us credit for doing something hard.
You are wrong about this.
Law of demand…make government more inefficient, and people demand less government.
Libertarians should strive to make every dollar transferred incredibly inefficient. They should strive for distortionary taxes and foolish social programs. People will vote for less taxes and less government programs if that’s their only choice.
Fortunately, they have liberals to help them in their quest!
David, that’s fairly disingenuous. You’re calling for libertarians to become Vandals. Nothing brings on the dark ages like a good sacking.
Hi Brian, the 1st theorem proof is easy–four or five lines. Now, the 2nd theorem proof still gives me shudders . . .
Mike–thanks for the shout out. I will see if I can scare up some links of my research. One example of the top of my head. I find the RESP-CESG grants go (went?) way disproportionately to high income families, but they were meant to improve education access. That was a research piece. In a policy piece for the CDHowe Institute, I argued they be scrapped as wasteful. These grants are incredibly popular among those who receive them, so this isn’t exactly my most popular line of research. But it was factual. And the government subsequently reformed the CESG to target it more at low income families. So the research mattered. I didn’t use the census, but the survey I used relied on the long form for the sampling frame and the survey weights.
So, libertarians, what do you think? Should the government just spend $600million on CESGs without knowing the impact? More importantly, given the government WAS spending that kind of money, would the world have been a better place if I had not been able to write that paper?
There is a huge conflict of interest for economists to argue for more government data collection. Economists are one of the main beneficiaries of government data and I don’t think Mike Moffatt has made a full disclosure of the benefits he derives from the use of government data. Economists asking for less data would be like a heroin addict asking to be given less drugs.
Kevin Mulligan has given an example of government data being used to reduce the size of government. However, there is probably a high correlation between the growth in the amount of government data collected and the size of government. Correlation does not imply causation, but my gut feeling tells me without government data the size of government would be smaller (perhaps some economist can run a study using government data to prove or disprove my gut feeling). A government that is blind might be more cautious in expanding. It is probably less controversial to propose that without government data the government would be less efficient.
Libertarians are probably right that without data government would be smaller but certainly less efficient. The genie is out of the bottle, however, and there is no going back to a time where the government collects little data. So it is too late to argue against collecting long form Census data in order to reduce the size of government.
“I don’t think Mike Moffatt has made a full disclosure of the benefits he derives from the use of government data. ”
I did at Macleans.ca. Answer is – I’ve never used nor purchased long-form census data. We all derive benefits from its existence, however.
I miss the Ayn Rand libertarians who were all about the “facts of reality.” A is A!
” but my gut feeling tells me”
You go with that. I’ll stick to the data.
“You go with that. I’ll stick to the data.”
Well, then prove it, big man.
Tom, I don’t understand you. I’m not terribly attached to a larger or smaller government, I want a BETTER government. What you’re proposing would seem to be very clearly not better. I can understand why you’d want to constrain government in certain ways if you thought it would improve our standard of life, but why on earth would you want to do it if you’re fairly certain it would make us worse off? I assume you’ll concede that a less efficient government is not a good thing.
Plus, prove what? That census data is a valuable public good? I think Stephen has made that point half a dozen different ways.