Industry Committee on April 11th, 2017

[Jean-Guy Prévost]

In many countries, ultimately the chief statistician is named by the government, by the equivalent of the Governor in Council. In some countries he may be nominated by Parliament, a bit like the Auditor General is. There are countries with quite an elaborate set of procedures for the selection, with international experts brought in to assess the candidates. In other countries it's a more opaque procedure.

Here we are moving to new territory. We had no fixed mandates. We had a chief statistician from 1985 to 2008. That's 23 years, which is very long. I haven't seen this in any other country. Before that, in the 1970s, we had six chief statisticians during a very short period of time.

So we're moving to something new. The idea that there's a selection committee that advises the Governor in Council would give some confidence as to the impartiality of the process.

[Lloyd Longfield, Liberal, Guelph, ON]

Right. Thank you.

Mr. Allen.

[Dr. Brian Allen]

I would echo those comments.

I think one of the advantages of having a selection committee is not so much who is the final choice but to ensure that there is a very strong short-listed group, the very best in the world. You need a committee, a broadly based committee, to beat the bushes worldwide, to perhaps contact people they know but who others might not, in order to have the very best list to present to the Governor in Council.

[Lloyd Longfield, Liberal, Guelph, ON]

In terms of where we're heading with this, of having it open to all Canadians but also open to reflect diversity and merit-based, we're looking at that across all government departments right now. Trying to bring this into the fold of how we manage the appointments process is maybe.... I'm giving a speech, but it's something new for us, new territory.

We're in the Westminster system, so there will be some differences in how ministers involve themselves with the process. Regarding the selection process in the OECD, is this something that's consistent with the OECD? Is there a general statement that can be made, or is it varied across countries?

[Paul Schreyer]

It varies a lot across countries. There is no clear pattern about the nomination of the chief statistician. We recommend that whatever the process is, it should be transparently spelled out. It should be clear what the process is. I think that is the first requirement.

The other point, which I made earlier, is that we think it's good practice to somehow state that there should be some professional competence or that the chief statistician should be selected on the criterion of professional competence only. I would add that you do not consistently find this in the statistical laws of OECD countries. There is a lot of variation.

[Lloyd Longfield, Liberal, Guelph, ON]

Thanks.

While we have you on the line—thank you so much for dialling in from another time zone in another part of the world—I'd like to mention that in my previous role, I was on a board of a multinational, and we had great data coming from Europe. We had okay data coming from England. The data from the United States and Canada, or the North American data, wasn't quite up to the same standard in terms of market information on how our company was progressing in different markets and different regions.

To you and perhaps Mr. Allen, with regard to the OECD, how do the different statisticians in different countries interact with each other? We continue to see this global interaction between companies around the world. In terms of the standards used in different countries, how do we come to some agreement on how we manage that?

[Paul Schreyer]

There are a number of international fora that are specifically designed to deal with statistical standards. With the OECD, for example, there is a committee on statistics and statistical policies where things are discussed. On a more global level, it is the UN Statistical Commission that basically comes up with the worldwide standards for statistics.

I think that is handled relatively well internationally, especially in those areas that are part of the standard statistical system: trade statistics, national accounts, prices. We do have relatively good international standardization.

[Lloyd Longfield, Liberal, Guelph, ON]

Mr. Allen, in terms of this legislation, on the importance of independence so that this type of work can go on without government interference, could you highlight any of that?

[Dr. Brian Allen]

Actually, my colleague Jack Gambino would have a much better view of the details of the collection of data, for example, industry data, than I would on that matter.

[Lloyd Longfield, Liberal, Guelph, ON]

Okay.

Ms. Eichler, the points that you were making in terms of dependence or independence of the data, is there anything from this conversation that you would like to add that you didn't get a chance to add?

We only have a few seconds.

[Dr. Margrit Eichler]

Well, we asked the former chief statistician to give us details on that, and he outlined the type of information that was delayed, some of which was affected by Service Canada, because it did not have the new website ready for the release of the census. That's a problem for industry because you want to have those data, and they were not in the form that they would have been otherwise.

[The Chair, Liberal, Dan Ruimy]

Thank you very much. We'll have to move on.

Mr. Lobb, you have five minutes.

[Ben Lobb, Conservative, Huron—Bruce, ON]

I want to throw this out here, because I have brought it up at a number of meetings. My belief is that for Stats Canada to be truly independent—or at least to the level of independence that most Canadians would aspire it to have—in the performance of the backbone of the entire system, it should at least have the very basic ability to determine how and where it stores its data.

Currently, it's my belief that they are hamstrung by Shared Services Canada. If this were anything else but government, the business, which is Stats Canada, would have severed its relationship with Shared Services Canada many years ago, and probably sued them in court and received quite a settlement.

Is there anything that should be in this bill, or anything that we should try to amend in the bill, to make sure that the current chief has the ability to put its data, on behalf of Canadians, in a place where it is safe and secure and robust enough to withstand the spikes in user traffic?

[Dr. Margrit Eichler]

Is that addressed to me?

[Ben Lobb, Conservative, Huron—Bruce, ON]

Yes, sure.

[Dr. Margrit Eichler]

I think Statistics Canada used to have the capacity to run its own operations, and it should be part of the bill that this capacity is being restored to it. That includes all aspects of its work.

[Ben Lobb, Conservative, Huron—Bruce, ON]

Are there any other thoughts from the members here?

[Jean-Guy Prévost]

I was going to say the same thing.

[Dr. Brian Allen]

I would agree as well.

[Ben Lobb, Conservative, Huron—Bruce, ON]

I understand some of the arguments, that this is critical data and should be stored with government and on and on, yet life insurance companies have critical personal information, and there never seems to be a breach or a hack with them. Banks, credit card companies, have our most personal financial information. The e-health records are probably even more personal and vulnerable. All of those records seem to be stored somewhere on private servers, and never leaked or hacked, at least to date. I think the private sector has proven that it has ability to probably produce a better product at a much lower cost.

The other thing I want to ask you about is this. In tying in its independence with Stats Canada—but also what I would call the “scope creep” of Stats Canada and how it makes sure it's addressing its core mandate but not stretching itself too thin—is there a role for the minister to be able to tell the Stats Canada chief, “No, you're going too far here. This is the line we want you to look at.”

Is that the role of the minister? Are there any thoughts on that?

[Jean-Guy Prévost]

Well, if we are talking about programs, there needs to be input from the political side to statistics, and there are various forums for this, including various councils where ministries can express their wishes. So that is one thing: the ability to ask for information on this and that. I think the Minister is there to do that.

It's another thing when there's a technical dimension that has various consequences for the quality of the data. With regard to this aspect, the expertise lies with the chief statistician, and at least—as we find, more or less, in proposed section 4.2 of the present bill—if he does not agree with what is proposed, he should ask for it to be made public and could express himself publicly in a manner other than by resigning.

[The Chair, Liberal, Dan Ruimy]

You have about 30 seconds.

[Ben Lobb, Conservative, Huron—Bruce, ON]

I have one last quick thought. The example I would give is the current study, or beginning of the study, on housing data and the perils that may lie within that if it gets too big. You have the Canadian Real Estate Association, the banks, and CMHC tracking data, so I think Statistics Canada has to be very careful about how far in scope this is. It's obviously been given a pretty big budget to study the data. Many can argue that it's important. However, it doesn't seem logical to have all these other agencies recording this data and then Statistics Canada starting fresh or tying in in how it works. That's just one example I would give you.

[The Chair, Liberal, Dan Ruimy]

Thank you very much.

Now we're going to move on to Mr. Sheehan. You have five minutes.

[Terry Sheehan, Liberal, Sault Ste. Marie, ON]

Thank you very much to all our presenters. It's been very informative and a great discussion thus far.

My first question is for Brian Allen. I've asked the question before. I'm from Sault Ste. Marie, and when the long-form census was not there, we, as a community, didn't get very good data for planning. We also heard testimony that it was not only the communities like Sault Ste. Marie, but also chambers of commerce, businesses, and non-profit organizations that just didn't quite get that data that was available.

Bill C-36 is trying to ensure that Statistics Canada has a clear mandate and remains responsive to the needs of Canadians and their governments. How does it do this? Could you comment also on the importance of that statistical data being available for those organizations?