Public Accounts Committee on Oct. 30th, 2014

[The Chair, NDP, David Christopherson]

Thank you, Mr. Smith.

Mr. Falk, you now have the floor, sir.

[Ted Falk, Conservative, Provencher, MB]

Thank you, Mr. Chairman.

Thank you, witnesses, for coming and for your presentations.

Mr. Smith, the Auditor General's report indicates that Statistics Canada adheres to international standards, definitions, and classifications for its four main surveys: the consumer price index; the labour force survey; the national household survey; and the survey of employment, payrolls, and hours.

Can you explain how the agency has demonstrated its commitment to maintaining a high standard for all those surveys?

[Wayne Smith]

Statistics Canada is very actively engaged internationally in defining all those standards, and defining them in ways that ensure the resulting statistical programs respond to the needs of Canadians.

The industrial classification we use in the survey of employment, payrolls, and hours, for example, is one that's negotiated between us, the Americans, and the Mexicans, to ensure it is well adapted to the structure of Canadian industry and therefore responsive to the needs of the data users.

The classification standard we use for commodities in international trade is also a product that we negotiate jointly with the United States and Mexico, to ensure that it responds effectively to the needs of Canadians.

We're one of the leading countries in the development of international standards and guidelines around consumer price indexes, and we have a very large investment, a five-year investment in improving the consumer price index because of the importance it has in monetary policy and in transfer payments, the management of income tax thresholds, the Canada pension plan, pensions, and so on.

In the case of the labour force, we're very engaged internationally in the definition of standards, guidelines, and methods. Our Australian colleagues are having some difficulty with their labour force survey. They asked us to give them some advice and some assistance in dealing with the issues they're currently struggling with as, in their view, we're the leading country in this area.

The other thing is that in any budget reduction exercise, we have protected all of them. We understand the importance. we understand the impact on Canadians of getting this right. Therefore we have protected these programs from any impact of any budget reductions Statistics Canada needs to make.

[Ted Falk, Conservative, Provencher, MB]

Thank you.

Mr. Ferguson, in point 11 of your opening comments, the last comment you made was, “The Agency also informed users that they should use caution when making comparisons between data from the National Household Survey and data from previous censuses.”

Can you explain that comment or your observation a little more, and why that statement is there?

[Michael Ferguson]

That goes back to the earlier question about coherence. We explain that in paragraph 8.30 of the report. Again, the agency identified that the change to the national household survey was a significant change in method, thus there could be an impact on the comparability of the data over time.

One step that the agency took was to make sure that users understood that they have to use caution when making those types of comparisons. Again, that's explained under the section of the audit that deals with the coherence aspect of their quality framework.

[Ted Falk, Conservative, Provencher, MB]

So it's not a reflection of the survey.

[Michael Ferguson]

It's a reflection that moving from the mandatory questionnaire under the census to the national household survey was a change in method. All the discussion that we've had before about response rates and that sort of thing would mean that the end result would be something where you have to use caution when you are comparing to data produced under another method, the previous method. As I say, they identified that because there was a change in method. The impact was that change in method and, because there was that change in method, users would need to use caution when making comparisons.

[Ted Falk, Conservative, Provencher, MB]

Okay.

When you do your surveys and send them out from period to period, Mr. Smith, are you quite careful to ensure that the wording is the same on the questions?

[Wayne Smith]

We either use the same wording or we have tested to ensure that we're getting comparable results with the different wording. Otherwise, we're changing a question in a way that we intend to get a different result because we were modernizing it in some respect or adding a dimension to it. We're very deliberate in ensuring we understand what's going to happen when we change the survey.

[The Chair, NDP, David Christopherson]

I'm sorry. Your time has expired, Mr. Falk. Thank you.

Moving along, Mr. Giguère, you have the floor, sir.

[Alain Giguère, NDP, Marc-Aurèle-Fortin, QC]

Thank you, Mr. Chair.

Mr. Smith, I want to go back to the calculation of the unemployment rate in first nations communities, because the information you gave me somewhat contradicts other sources of information that come from your services.

I can understand that 400,000 people in Canada may not have a large impact on the national unemployment rate, but this is hugely important for those 400,000 people.

Of the 1,128 census subdivisions on whom data was not published, only 36 were aboriginal communities. Of the data on 608 census subdivisions that included aboriginal communities, only 36 were not viable and were not released.

You tell me that a financial problem is what prevents you from collecting that data, but how is it that all of the other information is good, except the data on the unemployment rates? I would not like us to wind up with a type of segregation through silence. Even if the process goes back to 1944, currently this information is urgently needed to solve social, economic and political problems those communities are experiencing.

[Wayne Smith]

I think we are talking about two different things.

For the National Household Survey, which did cost some $650 million, we went systematically to visit all of the reserves in Canada that allowed us to do so, because not all of them do. In some cases there were forest fires or flooding that prevented us from going onto the reserves.

On the reserves where we were granted access, for the National Household Survey, we collected data not on a population sample, but on 100% of the population. Those are good data that also include figures on the working population and the May 2011 unemployment rate.

As for the Labour Force Survey, it was carried out on 56,000 households throughout the country. So this represents only a small fraction of all households. That survey does not include all the reserves. If we went to the reserves for that survey, the data would not include all of the reserve households, but only some of them. Consequently, we would never publish data on those reserves. The same thing applies when we go to Calgary, for instance. For the Labour Force Survey, we do not go to every household in town, buy only to some of them to represent all of the households.

So, even if we went to reserves for the Labour Force Survey, we would not have data on every reserve. The impact of that data on the national, provincial or economic regions unemployment rate would be minor and almost imperceptible.

Be that as it may, I quite agree with you. If we could produce monthly data on reserves, that would be extremely useful and desirable, but we would have to allocate $650 million to that every year, which is not realistic.

[Alain Giguère, NDP, Marc-Aurèle-Fortin, QC]

Mr. Chair, there is a problem. The witness is talking about an amount of $650 million, but that is for all of Canada.

You say that you do not go onto the reserves, but there are only 36 reserves for which the information was not sufficient, whereas there are 608 reserves. So, you had information on the other 572.

[Wayne Smith]

The National Household Survey provides such information.

[Alain Giguère, NDP, Marc-Aurèle-Fortin, QC]

There was no problem and those communities were not refusing to cooperate.

[Wayne Smith]

I think approximately 36 reserves refused to give us access.

[Alain Giguère, NDP, Marc-Aurèle-Fortin, QC]

Yes, that is the figure I have.

If we ever want to find solutions for these problems, we will need exact figures. Your participation would be helpful in that regard.

As for the constitution of your committees, your support committees more particularly, the Auditor General indicated that they are too often made up of academics, or of federal, provincial and municipal public servants, exclusively.

Auditor General, would it be pertinent to ensure not only that they come from various fields, but also, as regards the consumer price index, that these people not be economically orientated to satisfying a particular market, that is to say employers rather than unions, workers.

[Michael Ferguson]

I think what we were identifying, in terms of the makeup of the committee, was that having broad representation from the various stakeholder and user groups of Statistics Canada would give a broader range of opinions, and we felt that would be desirable.

[The Chair, NDP, David Christopherson]

That's very good, thank you.

The time has expired. I'm sorry.

We are now returning to Mr. Woodworth, who now has the floor.

[Stephen Woodworth, Conservative, Kitchener Centre, ON]

Thank you, Mr. Chair.

I'd like to ask Mr. Smith a question about the census subdivisions. In fact, I'm looking in the Auditor General's report at paragraph 8.53 in which, I believe, there is a quote from the “Statistics Canada Census Dictionary” to define “census subdivision”. Among other things, it says:

The population contained in a census subdivision can range from zero (uninhabited)....

I wonder, in fact, if there are census subdivisions that range that low—zero or maybe just a few hundred people—in the course of your survey?

[Wayne Smith]

There is actually quite a fair number. The zeros tend to be cottage country municipalities, where even though there is a significant population in the summer, nobody's principal residence is there. When we do the enumeration, we count them where they normally live, not where their cottage is. That's how you can get the zero.

[Stephen Woodworth, Conservative, Kitchener Centre, ON]

In quite a number of census subdivisions where the population is close to zero or virtually uninhabited, are these likely the kinds of census subdivisions where you would withhold data as a result of quality concerns?

[Wayne Smith]

The quality concerns really occur in provinces and in areas where there are very small municipalities of this nature, or of even a few hundred people. Certainly all of it, or almost all of it, is in municipalities below 5,000, and most of it is below 1,000.

In provinces where they've done amalgamations of their municipalities into larger municipalities, we didn't really have a significant problem. In provinces where there are still large numbers of very small municipalities, that's where we encountered the difficulty. Saskatchewan, for example, was very badly affected by this.

[Stephen Woodworth, Conservative, Kitchener Centre, ON]

That's sort of what I was expecting. This number of census subdivisions where quality concerns led data to be withheld seems to centre on those census subdivisions where the population is, if not zero, at least in the very low numbers of inhabitants. I think that's what you've told me. Have I summed that up correctly?

[Wayne Smith]

It's very much concentrated there. Yes.