Conservatives sound hysterical

Ifind the comments coming out of the Conservative Party regarding Statistics Canada’s requests for individuals’ banking records very troubling. The decision by the Conservative caucus to use StatsCan’s proposed program to attack the Liberal government is unwise and unwarranted. When they label this initiative as something out of Orwell’s “1984,” they also sound hysterical.

The Tories’ record on StatCan is, to put it kindly, mixed; they happily place greater reliance on received “knowledge” than on empirical information in making policy.

It was a Conservative government that cancelled the long form Census of 2011 replacing it with a voluntary survey. But, the resulting data’s quality was so bad that it was essentially junked - at a cost of more than $60 million. It appears the party’s present antediluvian attitude towards StatCan is a holdover from the days of Stephen Harper.

In addition to the quinquennial enumeration of the entire population, StatCan has a mandate to provide frequent, accurate and timely information on a host of economic and social issues. That sounds easy, but it is not. Getting people and organizations to provide government with the raw data to produce these reports is increasingly difficult.

Up to the present, success in obtaining good data by voluntary compliance has been, in large measure, a function of the trust individuals and organizations placed in StatCan; people believed that StatCan would not allow an individual’s personal information to be made available to the public at large.

(While StatCan has the legislated authority to compel the provision of data, they have always preferred to obtain it by voluntary compliance. The use of the government’s power to force compliance is very expensive and, in any case, fails to win support from the citizenry.)

In my early life, I was a senior official in the Census of Canada and learned just how zealously that operation ensured confidentiality of the data. We decided in 1973 to produce a one per cent sample tape of the 1971 Census data. The sample had to a true representation of the national Census data base. That meant all the variables including ethnicity, age, language, family structure, housing type, location, rural or urban by province, education attainment, etc.

It took a team of more-than-10 working for five months to produce a version that met the confidentiality standards set by the chief statistician. Even then, we hired a group of academics to see if they could identify any respondent and after four months they admitted defeat. Only then was the tape released.

But now, StatCan is facing growing difficulties in meeting its mandate. Increasing numbers of transactions are digital and automatic such as subscription renewals. And, the payments system is undergoing massive restructuring that may make the gathering of basic financial information exceedingly difficult unless the mechanism is built into the system itself; this means individuals’ bank accounts.

There is no question that StatCan flubbed the launch of this program to improve the quality of the data it produces. The Office of the Commissioner of Privacy should have been involved from Day 1.

Focus group consultation would have helped identify potential concerns of individuals and the financial institutions. A comprehensive educational campaign should have been developed explaining why this new methodology is necessary and how the safety and confidentiality of data is to be ensured.

Canada is fortunate to have an excellent statistical institution with a worldwide reputation for quality and attention to detail. Its products are outstanding and it is constantly looking for ways to improve both the accuracy and timelines of its products while becoming ever more cost efficient.

A test pilot of this project with the major banks will take substantial time and funding, but it is necessary. I am confident that the agency will master this complex challenge and the quality of data produced will continue to be outstanding.

David Bond is an author and retired bank economist from Kelowna. Email: curmudgeon@harumpf.com.