“64 percent of all the world’s statistics are made up right there on the spot
82.4 percent of people believe ’em whether they’re accurate statistics or not”
— Todd Snider, Statistician’s Blues
Munir Sheikh, the former head of Statistics Canada who resigned last during l’Affair Long-form Census, has released his cri du coeur in defending his professional integrity, the necessity of quality statistical data, and willingness to accuse politicians of acting politically or something (h/t Aaron Wherry). While his undated paper, subtly entitled “Good Data and Intelligent Government”, is a reasonable though flawed argument which came too late to affect the government’s plans for the long-form census, it happily provides an deep insight into the mindset of the technocratic career public servant.
The paper is rife with inconsistencies. Sheikh says that “housing information collected in the census is the only source of data for the condition of housing for aboriginals living on reserves.” But he writes later that “data on aboriginal populations are particularly weak … especially true of populations on reserves where the only source of data for reserves which participate in the census.” However, Sheikh is also adamant that StatsCan produces “the highest quality data”. In fact, “on quality, we are not aware of another statistical organization that produces census data of a higher quality than Statistics Canada.”
So we are given world-class, high-quality census data, except for data about aboriginal reserves, the area arguably most in need of high-quality census data.
He emphasizes the importance of political neutrality and independence of the statistical agency, that “credibility takes long to establish but can be lost quickly if the Agency lets itself be politically manouvered [sic].” He recommends that “government should ensure that Statistics Canada is at arms length from the government” and that we should “achieve this objective in law” through revision of the Statistics Act.
Yet the beneficiaries of quality census data include “federal programs … for the promotion of bilingualism nationally”, hardly a non-partisan issue, while education census data “are fundamental for studies related to groups that draw policy attention such as immigrants, aboriginal peoples and official language minorities.” These groups would include, I presume, the Centre of for the Study of Living Standards, which posted Sheikh’s paper on their website and just happens to be an organization heavily invested in the role StatsCanada played in rooting out information necessary to promoting their cause.
Which brings us to the agency’s stakeholders, who are described as “data users, including federal departments, provincial and territorial governments, municipalities, researchers and the media.” Notice who’s not a stakeholder? If not, try looking in a mirror.
In the latest round of “public consultations” for StatsCan, the general public were generally not consulted:
From April to November 2007, Statistics Canada met with over 360 people during more than 50 in-person consultations. The majority of the participants were data users. Meetings were held with federal, provincial and territorial governments, municipalities and regions, non-profit organizations, community groups, academia, private industry (which includes licensed distributors and the media), advisory committees and the general public.
A total of 1,276 comments were made on a wide variety of census topics during this public consultation effort: 835 comments came from federal, regional, provincial, territorial and municipal departments and agencies; only 15 were made by private citizens.
Who’s driving the questions being asked in the census? A look at the current long-form census, now known as the National Household Survey, will tell you. In addition to personal info, the surveyed citizen is also asked series of questions on “activities of daily living” for data on physical or mental health, “sociocultural information” to better understand our cultural mosaic, “mobility” data on labour movement, “place of birth of parents” to, uh, better understand our cultural mosaic, “education” info, “labour market activities”, transportation between home and places of employment, “income in 2010” (one is given the option of providing access to tax files) which includes “income from employment” and “income from government”, and finally a section on housing and living expenses. Apparently, this is all information deemed vital to those all-important stakeholders, of which you and I are not a part.
His thought experiment gives us an example of how StatsCan information could be used to help develop a new neighbourhood in a community:
Just about everything it needed could have used information from the long-form census from building roads an sidewalks, to strip malls, to the establishment of schools, to the setting up of health care facilities, to the opening up of a community centre, to the provision of services for minority groups. And the list can go on and on.
Yeah that, or you could just have an experienced developer build a new neighbourhood based on market expectations.
The questions asked, however, seem so random, don’t they? Why ask about home renovations, but no questions as to whether you own, say, a cabin at the lake? Questions about your work abound but why aren’t you asked about your various volunteer activities? You’re asked about how much your utility bills are, but not how much money you spent in local restaurants or shops? And just who decides who is a stakeholder to begin with? Why are social justice units involved but, for example, those who produce such important products like gasoline or consumer goods are excluded?
Regardless, at this point, you might be getting the impression that Sheikh might be the type of bureaucrat who believes that if governments had enough information, they could solve everyone’s needs, individual rights be damned. If this is your impression, you’d be right. He certainly argues that StatsCan data is the backbone of good governance, hence the title of his paper, but it goes further than that.
Sheikh mocks the federal government’s call for a “balance” between the needs desires of stakeholders and the individual. For example, he places sneer quotes around the term “balance” whenever the term “balance” is used in this context. His concern is over the quality of data from the mandatory short-form census is if Canadians don’t decide to fill out the short form because of “the government’s argument that Canadians are right in not providing information they think the state does not need to know.” The audacity!
There is, of course, a system to deal with non-compliance. It is illegal under the Statistics Act of 1971 to not complete a mandatory census form in which offenders could be prosecuted and even sent to prison. This is the crux of the government’s position, but Sheikh doesn’t think that’s a big deal. “Nobody was ever sent to jail for not completing the census questionnaire.” Besides, only “a total of 64 Canadians were referred to the Public Prosecution Service of Canada in a population of over 31 million,” which is a statistically insignificant amount of persons subjected to the prosecution process. How significant their referral would be to one of the 64 Canadians would be another matter of course.
Sheikh clearly misunderstands the federal government’s role in this debate. He may be right that the data quality will be statistically affected by moving away from the mandatory census process (though the quality of data acquired without personal incentive and that which cannot be appreciably verified is another issue). But he thinks the big issue against the mandatory long form is privacy. It’s not — the issue is placing the needs of the state above the wishes of the individual.
The federal government’s role, then, is to take the place of the missing stakeholder — to speak on behalf of the citizens. Not all citizens, of course. Many people are more than willing to hand over everything they own (sorry, everything you own) in exchange for more government control. But despite his best efforts, the prime minister’s libertarian bent runs deep. This decision might be a mistake for him, but it’s comforting to have a government willing to err on the side of the individual, even if we lose a bit of statistical accuracy in the wake.
Sheikh and his like-minded defenders of the old system are adamant that changing long-form data collection from being mandatory to voluntary would create havoc in the little world of policy wonks and social justice advocates, yet somehow, as crazy as it sounds, life will go on.
I’ll leave the last word to F.A. Hayek:
Today, it is almost heresy to suggest that scientific knowledge is not the sum of all knowledge. But a little reflection will show that there is beyond question a body of very important but unorganized knowledge which cannot possibly be called scientific in the sense of knowledge of general rules: the knowledge of the particular circumstances of time and place. It is with respect to this that practically every individual has some advantage over all others because he possesses unique information of which beneficial use might be made, but of which use can be made only if the decisions depending on it are left to him or are made with his active cooperation.
Just because the state doesn’t have all the information doesn’t mean people can’t solve their own problems. It would do Sheikh some good to brush up on his Hayek once in a while.