INDU Committee Meeting

     Welcome everybody. We're going to get started. This is meeting number 55. We're continuing our ongoing study of Bill C-36, an act to amend the Statistics Act.

    We have with us today, as individuals, Mel Cappe, professor, University of Toronto School of Public Policy and Governance; and Munir Sheikh, former chief statistician of Canada. Via video conference we also have Paul G. Thomas, professor emeritus of Political Studies, University of Manitoba. From the National Statistics Council we have Ian McKinnon, chair.

    Welcome, everybody.

    We're going to start with Mr. Cappe.

    You have up to 10 minutes. I wouldn't be hurt if you kept it under 10 minutes.

    I hope to make you happy today, sir.

    Mr. Chair, thank you for giving me this opportunity to share my thoughts on Bill C-36.

     The last time I appeared before this committee was August 27, 2010, on the issue of the long-form census. I went back and reviewed that testimony and found that what I said then I really want to say again, so I'm going to quote myself, I'm afraid, and perhaps bore you all.

    First of all, I start off by saying I've never been partisan. For five years, I was president of the Institute for Research and Public Policy, and then, I quote, “I spent over 30 years in the public service of Canada. I served seven prime ministers. I was the Clerk of the Privy Council and Secretary to the Cabinet to Mr. Chrétien, but lest you think that somehow taints me as being a partisan in anyway”—remember this was the last government I was talking to—“the first order in council naming me to the deputy minister ranks was by Brian Mulroney, and I ended my career in the public service loyally serving Prime Minister Harper”.

    In the spirit of that non-partisan public service, my objective today, again, is to try to help the committee deal with the government's objectives, as well as the opposition's objectives. I think both can be met. Although the government and the opposition have switched, my point is as applicable today as it was then.

    My first point is I'm not partisan, and the second point is that statistics are a public good. I quote myself, “That's a technical term, but it's a good one. [They're] used by a wide array of real people: banks, charities, and public health authorities.” The state can collect and analyze statistics at lower cost than requiring everyone to collect their own. One person using statistics does not impede others from using the same ones, and that's what makes statistics a public good.

    My next point is that the Statistics Act should minimize the use of coercion, which I think was an issue back in 2010. We should be minimizing intrusiveness and maximizing the privacy of the data as much as possible.

    What I meant then was that “you can remove jail terms”, but “you can review the questionnaire and minimize the intrusiveness of the questions” as well, “and I would add to what the National Statistics Council has said, you can increase the penalties for the divulgation of private data.” So this secrecy of data provided to Statistics Canada is fundamental and important. “I think anybody who releases census data inappropriately should be seriously fined.”

    My next point is that the governance of Statistics Canada can be improved. The higher principle is, quote, “...to ensure the integrity of the statistical agency. I think the events over the course of [2010]...raised questions about this larger significant issue. I think the committee should take its time...to consider the UN Fundamental Principles of Official Statistics.”

    “To my surprise, there are [such] principles”, but in 2010 many of those issues were raised “about who's responsible for methodology in collecting statistics. There are several principles in that UN charter that deal with independence, methodological integrity, and the role of politics”.

     I said in 2010 to this committee, “You might consider—and I'm not suggesting this is 'the' answer, but it's 'an' answer—amending the Statistics Act to make clear that the Chief Statistician, who is a statutory officer named in the Statistics Act and appointed by Governor in Council, has the sole responsibility for methodological and technical issues.”

    However, I also think—I thought then, and I think now—there is a legitimate role for politics in statistics—politics, but not partisanship. Statistics Canada is a:

    Those comments in 2010 are as apt today as then, and I stand by them. It's through that lens that I reviewed Bill C-36, and I want to make four points.

    One, statistics should not be a partisan issue. The efficacy of the agency and the integrity of our institutions should be important to all parties. The bill does not appear to me to be particularly partisan. It revalidates the independence of the agency.

    Two, statistics are indeed a public good, and it's highly appropriate for the state to collect and analyze statistics, so I'd support the objectives of this bill. It reconfirms the legitimacy of the collection and analysis functions of the agency.

    Three, in minimizing coercion and reducing penalties for violation of the act by removing jail terms, this bill meets my objectives and, I would suggest, the legitimate objectives of this committee.

    Finally, the changes to the governance structures of the agency strike me as appropriate. Creating the statutory council and legislating the independence and responsibilities of the chief statistician for methodology are apt.

    I would just note that there's a trade-off here, Mr. Chair, between the independence of the agency and the need for more direction from government, when it's a department of government that already is there. The more independence you give the agency, the more formally in statute the relationship has to be articulated. Thus, I think this is a very good piece of housekeeping to modernize the Statistics Act. While I have strong views on other parts of the bill dealing with independence and directives, I'll await your questions to deal with them.

    Thank you, and I'll be happy to answer your questions.

     Thank you very much for keeping that brief.

     Now we're going to move on to Mr. Sheikh.

     First of all, I would like to thank the committee for inviting me to come here today and share my thoughts with you on Bill C-36. It's good to be back at this committee after almost seven years, when I appeared following my resignation as chief statistician.

     I think it would be useful for me to set a context for my comments on Bill C-36 and proposing changes to the Statistics Act. An official statistical agency must have the utmost trust of the users of this data. To achieve this objective, both the government and the official statistical agency should play their parts.

     For its part, the agency must strive to achieve excellence in the production and dissemination of needed data. I can say, based on my experience, that the agency has always tried to do that. On the part of the government, it needs to find the appropriate mechanisms to make the official statistical agency accountable to the citizens of the country, while at the same time ensuring that the agency is appropriately independent and avoids political interference, both in fact and in appearance.

     Avoiding political interference is important because Statistics Canada is in the business of producing facts, not in the business of policy-making, where political decisions are normal. Political interference can damage the trust that citizens must have in the official agency that is producing the data, which could make all official data suspect for users.

    Turning to Bill C-36, I would like first of all to commend the government for setting the objective of increasing the independence of Statistics Canada and for introducing legislation to that effect. Let me now offer some comments. In doing so, I apologize to the committee, in that I have actually more questions than I have answers.

    The existing Statistics Act is flawed, in that it gives the authority to make technical statistical decisions to the minister responsible for Statistics Canada. Bill C-36 rightly shifts some of that responsibility to the chief statistician. I'm pleased that the government has proposed this important change in proposed subsection 4(5).

    However, the bill does not stop there. It allows the minister to send a public directive to the chief statistician in cases where the minister disagrees with the chief statistician on these matters. This is in proposed subsections 4.1(1) and 4.2(1). I understand fully that this is done to preserve accountability; however, it does raise a number of questions. Let me mention just two.

    If a chief statistician is perceived not to have made appropriate decisions on statistical matters, so that the minister needs to intervene, how can the government afford to have such a chief statistician stay in the job for five years? Next, as an example, given that Statistics Canada is a national statistical agency and not a federal agency, as I understand it, what happens if a minister orders the cancellation of a survey that is of critical importance, say, to a province?

     As I just mentioned, another proposed change in the bill is a fixed five-year term for the chief statistician. Presumably, the purpose of the five-year fixed term is that the chief statistician should be able to withstand political pressure.

     However, let me mention that I am not aware of any such problem ever happening in the long life of Statistics Canada, but it may happen in the future, of course. I am, however, aware that the Prime Minister makes changes to the ranks of the senior civil service to match the best people to the types of deputy minister jobs that exist. This prime ministerial prerogative includes the chief statistician at this time. This raises the question whether it is worth sacrificing a known benefit that is part of a Prime Minister's authority at the moment to achieve a potential benefit with an uncertain and very small likelihood.

    Another important proposed change is the establishment of the statistics advisory council. Its members would be appointed through an order in council and could be asked to provide advice to both the minister and the chief statistician on issues related to the "overall quality of the national statistical system". This also raises some questions. Why does the minister need advice from outside when they have the chief statistician for all such advice? What happens if the minister and the chief statistician have different views on the advice they get, particularly in the context of the chief statistician's five-year term? Is there a risk that a government would make politically motivated appointments to the council? Let me emphasize that, if a council is indeed established, it is of the utmost importance that it should be set on the right foot at the start of its life.

    Let me now turn to the question of how things could have unfolded if this law had been in place in 2010 at the time of the cancellation of the long-form census. There were two issues running at that time, if I could remind the committee members. First, the inappropriateness of cancelling the census, which seriously reduced data quality, as I warned in my resignation statement. The second issue was the nature of the statements made by the minister in response to the criticism the government rightly received for making a very bad decision. These statements led to the resignation of the chief statistician.

     On the first issue, regarding the cancellation of the long-form census, I have not found anything in Bill C-36 to suggest that things would have been different in 2010. The census would still be cancelled. Let me emphasize here that I am assuming that section 21 of the Statistics Act overrides section 22. If it does not, and my assumption is incorrect, I think the law should be clarified on that.

     This raises another question. Given what we went through in 2010, and perhaps one of the reasons we are all gathered here today, is there not a need to avoid repeating that problem? On the issue of the nature of the minister's statements that led to the resignation of the chief statistician, I fully understand it is not possible to legislate that a minister cannot say those types of things.

    Thank you.

    Thank you very much for your opening remarks. Now we're going to move to Mr. Paul Thomas, by video conference.

     Thank you very much, and thank you for the invitation to appear before the committee to offer some thoughts on Bill C-36.

    I've served on the National Statistics Council since 1996, but my remarks today are very much my own opinions.

    I prepared and submitted a brief to the committee, and I understand that it has been translated and circulated, so I'm going to use my brief opening time this morning to highlight some of the recommendations I made in that document. There won't be time to go into at length the rationale behind the recommendations I'm making. If there is time and interest in the question period, I'd be pleased to provide additional comments on the reasoning behind my recommendation.

    Let me say at the outset that I think Bill C-36 represents an important and valuable first step toward modernizing the Statistics Act. In particular, Bill C-36 provides for a less detailed and prescriptive governance framework for Statistics Canada. This more flexible governance framework means that the chief statistician and the agency at large will have more independence and autonomy to make informed, impartial professional judgments about statistical programming, the methodology to be used, and the interpretations to be applied to the data they collect.

    I think the bill also clarifies the accountability relationships that Statistics Canada must maintain with the government, with Parliament, and with the public. In the past, these matters were largely governed by long-standing practices and conventions. Some of those conventions and practices are now being codified in the form of this legislation. Therefore, my recommendations are not to approve the bill, but I think there are a number of areas covered by the bill where improvements could be made. I'll go through those very quickly.

    In both the backgrounder to the bill and the remarks by the minister, the government is saying that the contents of this bill align with the fundamental principles of official statistics of the United Nations. I think it would be better, in fact, if a preamble were added to the bill to indicate that this is the foundation for the contents of the bill. Such a preamble would serve both a contextual and a constructive role in the interpretation of the statute. It would confirm the spirit of the law and help with the interpretation of any ambiguity therein. It would provide a foundation for the development within Statistics Canada of a culture of independence, impartiality, and objectivity in the production and publication of official statistics. Such a preamble would also provide a basis for discussions and negotiation between the chief statistician and officials of the government when issues of independence arise.

    The second point I would make is that the bill presumes a policy operation split. In other words, the policy remains the prerogative of government and Parliament, whereas operational and technical matters are supposed to be the domain in which the chief statistician and other experts at Statistics Canada prevail.

    As already mentioned by the previous two speakers, the bill, I think, needs to create greater clarity regarding those instances in which the responsible minister believes a technical and operational matter is of such importance that it rises to the level of becoming a matter of national interest and the minister can issue directives to Statistics Canada. I think in that instance there should be a requirement in the law that such directives be tabled in Parliament and be subject to a 60-day notice and comment period so that there would be debate about the appropriateness of government involvement with an operational matter.

    There is also authority given to the minister to issue more general policy directives that are binding on the chief statistician and the agency. In that case, I think it would be better if those directives came not from the minister solely on his or her own behalf but were subject to prior approval by cabinet. I suggest a procedure for that.

     The further recommendation I have on this general policy directive is that no such policy directive should amount to an indirect amendment to the Statistics Act in any fundamental way. Amending the Statistics Act is a responsibility of Parliament.

    I then turn to the position of the chief statistician, which I think has a crucial role in all of this, the catalyst that makes for a high quality national statistics system. I think that the provisions proposed in Bill C-36 should be amended to provide for an advisory panel of three eminent or distinguished persons with appropriate background knowledge to conduct the recruitment activity and to review the applications, nominations, for the position of chief statistician. That panel would then recommend one name and place two alternative names before the Prime Minister for possible recommendations to the cabinet.

    If the Prime Minister found none of the nominees suitable, he or she could nominate their own choice, but would be required to give reasons for not accepting someone from the list provided by the panel.

    Turning now to the proposed Canadian statistics advisory council, I think there is a very real ambiguity here about the role of this new council that could lead to problems down the road.

    I think two questions of clarification need to be asked. The first question is, does the government understand whether the council is to serve primarily a representational role or is the new council presumed to play a governance role, serving as the eyes and ears of the minister by overseeing the performance of the agency?

    The second question is, as order in council appointees, will council members see themselves more as agents of the government than as trustees of the long-term interest of the national statistical system?

    What happens if there is a disagreement between the council and the chief statistician over what advice should go to the minister? I have not read or heard a clear statement from the minister on these points. A greater clarity would be advisable.

    On the basis of that clarification, I would suggest that proposed subsection 8.1(2) of the bill regarding the new Canadian statistics advisory council should be amended to provide for a greater number of members, possibly in the range of 20 to 25 members, including a chairperson appointed by the Governor in Council to reflect a wider range of interests served by the statistical programming of Statistics Canada. Recruitment should be done on the basis of an open process of application and nomination.

    In conclusion, I would make the following point. Legislation that distributes authority and creates structures and procedures is the starting point for achieving an appropriate balance between independence and accountability for Statistics Canada. Even more important, however, is the appointment of a chief statistician and other leaders of integrity within the agency who are committed to strengthening an already strong shared culture within Statistics Canada that's based on the principles and values of a high quality national statistical system.

    Thank you very much. I look forward to any questions.

    Thank you very much.

    We're going to move to Mr. Ian McKinnon, chair of the National Statistics Council. You have 10 minutes, sir.

     Thank you, Mr. Chairman.

    I appreciate the opportunity to appear before this committee. Let me begin with a brief introduction to the National Statistics Council.

    The NSC was created 30 years ago to advise the chief statistician on matters affecting the Canadian statistical system. It complements the more focused work of the subject area advisory committees and the federal, provincial, and territorial working groups.

    It is my strong view that we've been very useful in pressuring StatsCan to look at new areas of data collection, for example, finding ways to reduce respondent burden and to make StatsCan's data more widely available to users.

    With the controversy that attended the 2011 census and the national household survey, the council also began to consider more explicitly and even speak out on issues that could affect the continued quality of the overall Canadian statistical system.

    The importance of this bill, Bill C-36, lies precisely in addressing some of the issues that were raised at that time.

    We believe it is fundamentally important for our statistical system that we get this right. For the information produced by a statistical agency to be fully used, the quality of the information must be trusted. That information must be relevant, accurate, timely, and available widely.

    Similarly, for Canadians to provide information—this is the other side—freely to a statistical agency, that agency must itself be trusted. From looking at both Canadian experience and that of similar countries, trust is dependent on several aspects of an agency's operation and its mandate.

    Specifically, the statistical agency must operate transparently, with strongly protected guarantees of independence in terms of exercising its professional judgment.

    There must be a relationship of trust with the individual providers of the information to the statistical agency. They need to understand why the information is collected, how decisions about what is collected are made, and above all, there needs to be a commitment to reducing the burden on respondents and an assurance that the information they provide will be held in confidence.

    A statistical agency earns both its credibility and its social licence, if you will, by its success at embodying those attributes.

    I'd like now to turn to the bill and the council's response to the direction of the bill.

    The first is simply removing imprisonment as a possible penalty for individuals refusing to respond to mandatory surveys. The council has publicly endorsed removing the threat of imprisonment from the initial debates over the census in 2010 when that issue was raised, and we continue to do so.

    Secondly, moving to the second topic, confidentiality, this is a central issue for the council because it is essential to holding public trust. If public trust is undermined, the provision of information is undermined. In contrast with our view that the potential punishment for not responding to mandatory surveys is too severe, the council has also suggested that penalties for unauthorized disclosure of data by employees or designated research should be monitored to make sure that they constitute a significant deterrent. In this I echo fully Mr. Cappe's earlier points.

    Maintaining confidentiality is far more than simply having appropriate penalties, however. There are also matters like the security of computer systems and data protection procedures. This committee has already heard about further securing computer systems and the changes at Shared Services Canada.

    The council does not possess the expertise to make a judgment on secure computing environments. However, we do believe that the core practices that ensure the protection of personal information flow in part from corporate culture. We can attest, from our experience, that confidentiality is a deeply rooted value in the culture of Statistics Canada.

    On the aspect of confidentiality related to making census returns available after 92 years, the council agrees that the benefits to historians and genealogists outweigh concerns that this change might affect people's willingness to respond to the census. We simply haven't observed significant public concern over this.

    Turning to strengthening the independence of the chief statistician and Statistics Canada, many of the changes in the act are consistent with advice that the National Statistics Council has given.

    The council has agreed that giving the chief statistician a fixed, potentially renewable term during good behaviour increases independence.

     As with some of the statements made in earlier hearings by former chief statisticians, we agree that there should be a wide and aggressive search conducted when this position is to be filled. Again, this is consistent with the statements today from Professor Thomas and Mr. Cappe. As well, we suggest, as does Professor Thomas, the use of a senior panel in making the selection.

    The balance between independence and accountability is critical. The council believes that on questions such as appropriate methodology and other issues of professional judgment, Statistics Canada and the chief statistician should be responsible. Conversely, Statistics Canada, in particular the chief statistician, should have operational control of the agency, subject to the financial, personnel, and administrative disciplines governing federal organizations generally. The incumbent should be responsible to propose the statistical program of the agency, subject to written direction by the minister on topics and priorities.

    The importance of transparency and written directions I think is part of the pivotal mix, because it means that there is not necessarily a cost, but it makes clear where the responsibility lies. Transparency can help ensure this balance is sustained. The chief statistician's annual report through the minister is one element, as is the chief statistician's ability to make public the directives that are received from the minister.

    The final element in balancing independence and accountability is the creation of the Canadian statistics advisory council. The function of that body is significantly different from that of the current National Statistics Council, reflecting the changed position of the chief statistician and of Statistics Canada.

     Through their annual public report, this council can offer a more independent view of issues and challenges facing the Canadian statistical system. While it's not a board to oversee StatsCan, it can increase the transparency and general understanding of the competing pressures facing the statistical system. It can also provide the government with an external view of operational and professional issues facing Stats Canada.

    As you also heard earlier today, it means that appointment to that body is a critical issue because, given that new role, one that is different from that of the National Statistics Council, it's important that it is well understood how people are appointed and that people feel confident they can act in an independent manner. While we are intensely proud of the work of the National Statistics Council and what it has done over the years in terms of stakeholder engagement, ongoing professional consultation, and outreach to current and potential data users, all of those elements have become increasingly part of StatsCan's operating values.

     The advisory council fills a new role, one which would be difficult for the current statistics council to perform, frankly, and the creation of that new entity is I think an essential and pivotal part of the promise that this bill holds in transforming it from just letters into a well-operating and successful change to the Canadian statistical system.

    I await your questions. Thank you.

    Thank you very much, everybody, for your opening comments.

    We're going to move right into questions.

     Mr. Arya, you have seven minutes.

    Thank you, Mr. Chair.

    Thanks and welcome to all the distinguished guests who are here before the committee.

    Mr. Sheikh, a lot of Canadians applauded your decision when you resigned on the basis of principle. Now that the new census has come out, what do you think of it, and not just the end result, but the process by which it was implemented and delivered?

    I think it's great to have the census back, and I am quite pleased with the quality of the data it produces. I hope it stays around for a long time to come.

    Thank you.

    Mr. McKinnon, you mentioned that how people are appointed is important. Currently, how is your statistical council appointed? I don't know who the members are. How many do you have?

    We have 40 members, of whom I would say about 35 are very active. Sometimes they are selected by suggestions of the members of the council, or they are appointments by the chief statistician. Also, they are effectively a cross-section of users of the data.

    You say they're a cross-section. Are all provinces and territories represented?

     Yes, we try to have all regions. Currently, we don't have every province represented. We try hard to have regional balance. We have a mix of academics, NGOs, heavy users, and commercial unions, for example. That's the—

     What if I say that it is dominated more by academics than by the end-users?

    The academics who are on the council tend to be very heavy users, but in fact, I would say over half of the members of the council are not current academics. We have retired senior bureaucrats on there as well. We're very conscious that this not simply become a channel for, to put it pejoratively, academics to push for their favourite research project. There's a very strong internal ethic that that not take place.

    How is the corporate sector represented there?

    We have tended often to have a senior economist from one of the banks. Certain industries tend to be heavy users of data. We also have had people from groups like C.D. Howe Institute, that would be viewed not as part of the corporate sector directly but as interested in issues it finds important.

    Is advice from this council solicited by the chief statistician?

    I believe it has been very useful. Chief statisticians have often come to us saying, “There is an issue we are examining and we would like your feedback on it.”

    It also, in fact, tends to make us go back to our constituent groups and consult in the area and find out, so we find over several meetings that we'll get ongoing discussion with additional information about areas.

    We also tend not to hold votes. It is an opportunity for the chief statistician to see an array of discussion that allows him direct access to those external views.

    You mentioned that you don't hold votes. How do you arrive at consensus?

    We often go by consensus, or because our role is to provide a range of views and to make sure that StatsCan understands the range of user demands and concerns out there, often that is the end in itself. We rarely will then say to the chief statistician, “We formally believe that you should do X or undertake this survey.” We can push.

    I'll give you an example. For years and years, the statistics council, particularly the late Paul Bernard of Université de Montréal, pushed very heavily to increase the accessibility of data, both physically through the research data centres and also by reducing the costs and making the access to many well-established series free, to broaden use. That's an example of how we just pushed over and over again, not making a formal resolution, but making sure it was brought back to the—

    You mentioned that a range of views is submitted. Does that means that a dissenting report is also allowed?

    Absolutely. Occasionally, the chief statistician will ask for a committee to be formed on a particular topic. That committee will consult with the entire council and often produce a written piece of advice to the chief statistician, but in cases where there are mixed views, we make sure they're reflected.

    Once again, you just mentioned that the current members select the other members. How is that?

    No, they do not select. They are all appointed by the chief statistician. StatsCan, current members, and other people who have long-standing relationships with Statistics Canada tend to provide a pool that will make suggestions.

    For example, we have almost invariably had at least one member of the media, the press, there. There are a handful of people who are well known across the country, who analyze data and are reporters. That gives you your universe when you try to get some of that.

    Thank you, Mr. McKinnon.

    Mr. Sheikh, what do you think of the process proposed in the bill to appoint the new statistician? What are your views on that?

     What I see mostly, in the bill, is a fixed five-year term for the chief statistician. There's really not much detail as to how the chief statistician is selected.

    The point I was trying to make was that the new bill, by changing the balance of accountability, political interference, professionalism, raises a lot of questions. There are some questions that I think are truly important. Someone, hopefully, will give those questions some thought. I don't have the answers.

    One of the questions I raised is, if you trust in the Clerk of the Privy Council as the adviser to the PM for putting the right person in deputy ministers' jobs, and if you do trust that person to make an effort to avoid political interference, given that trust, do you really need a fixed term? When I look at the long history of Statistics Canada, my impression is that the existing system of appointments by the PM, with recommendations from the clerk, has worked really well. In circumstances where you have a mismatch between a job and a person, there's flexibility to change that.

    As an example, think about some really bright star in the ranks of the civil service who would become a great chief statistician, compared to the one who is in the job, who is doing a reasonable job, but he's there for five years.

    Thank you.

    Should you give up on that opportunity? The question I'm raising is you—

    Sir, I'm going to have to cut you off because we are way over time.

    Mr. Dreeshen, you have seven minutes.

    My apologies.

    Thank you very much, Mr. Chair. Thank you to our guests. Certainly, as a former math teacher, some of my former students have worked with StatsCan, so I'm extremely pleased that you do the great work you do. Again, we've asked many folks who have come here if they have any suggestions. Of course, I believe you've fleshed out a lot of potential things that can be dealt with.

    One of the things we looked at was, of course, the appointment of the chief statistician.

     Paul, I believe that in your discussions you had talked about the advisory panel, and that three eminent persons should provide the new chief statistician with names and so on. If the Prime Minister found that he didn't accept any of these, he would just have to explain why it is that he would choose somebody else.

    Could you flesh that out a little bit? Who exactly were you speaking of in your description?

    We go back to the fact that Statistics Canada is what's called a statutory agency; it's different from a regular department of government. The head of Statistics Canada is expected to embody both the expertise and the values systems that are needed in a national statistics system. Therefore, I don't think this should be seen as just another routine appointment made by the Prime Minister as part of a roster of deputy ministers. There is a specialized background that needs to be there in the person of the chief statistician, so therefore I'm suggesting that there be a customized appointment process. The new government, the Trudeau government, has an appointment process up online regarding order in council appointments, but it's not very informative in terms of how the process will be handled or the criteria.

    In this instance, I think we should have a specialized process of appointment, where three prominent individuals with appropriate background knowledge and expertise would both do the recruitment and the initial nomination of someone to the Prime Minister, who would then put the name forward. Then, as Mel Cappe said, if the political considerations or other considerations make the Prime Minister decide he can accept none of the recommendations from the panel, he or she has the right to do that and they can answer for that. You're locating ultimate responsibility and accountability where it belongs, with the Prime Minister.

    The suggestion has been that some of these three persons who would be on that panel or could be chosen would come from among the Clerk of the Privy Council, chief statistician of Canada, Governor of the Bank of Canada, chairman of the National Statistics Council, president of the Statistical Society of Canada, president of the Canadian Institutes of Health Research, president of the Natural Sciences and Engineering Research Council, or the president of the Social Sciences and Humanities Research Council.

    Are there other agencies and councils that you think could be added to that group of eminent persons who would be able to help in the decision-making?

     That's a pretty good list. I think that's Ivan Fellegi's list. I think that's a pretty good list. I'm not sure, but perhaps a former chief statistician from a provincial level might be an idea for an addition to the list, or someone with quite specialized knowledge. Statistics have become highly technical, and we want the best-quality statistics system we can have. It's not just about people having distinguished careers in other domains. We need to have some expertise on this body, which will do the initial appraisal of nominees or applicants for the position.

    Thank you very much.

    Mr. Sheikh, if a person were to add this proposal to it—that the Governor in Council should be consulting with leaders of every recognized party in the House of Commons on the appointment of a chief statistician from the list of candidates that has been described—do you feel that it would help in ensuring that the chief statistician has the support of Parliament?

    I think it would, but let me just mention another dimension that is important to any deputy minister or chief statistician of a department. Running an organization of 5,000 or 6,000 people is not a small task, so in addition to a person having statistical expertise, they should also have the ability to run a very large organization with very complex issues.

    There's another point that perhaps, Mr. Sheikh, you could flesh out as well. There are some thoughts on what the definitions would be of the census of population and the census of agriculture. There are amendments that have been floated around that would help prevent the courts from arguing, as they've done, that unless otherwise specified in the act, the census, by common definition, has to be addressed to all households. I believe that was part of the issue, that it had then gone to the courts, and that was the reason why things had progressed in the way they had.

    Could you describe any wording that you might have for the census of population or the census of agriculture that would allow for a smooth transition and without the concern about the courts looking at it in a different light?

    I don't think I'll be able to provide a definition right now. I haven't given it much thought.

    I think the system we have has worked really well. You have a past census that you can search very widely. You can make changes to what the census should look like, and then the cabinet approves it. I think that system really has worked.

    To my mind, that really isn't the issue, how you define a census. The issue for me is that in the Statistics Act right now, if you look at sections 21 and 22, there's a conflict. You need to avoid that conflict. Section 22 has a list of things that, under Bill C-36, the chief statistician would be able to make most, if not all, of the decisions on. Section 21 says quite explicitly that, by the way, the census questions are the responsibility of the government.

    To me, which one of those would override the other in Bill C-36 is a big, big issue, and the bill is totally silent on that.

    Thank you very much.

    We'll move to Brian Masse for seven minutes.

    Thank you to our witnesses for being here today.

    One of the more interesting aspects raised about the current situation we're in right now is trust. Trust is necessary, not only for those who want to compile the information in accordance with the questions we ask but also among the users. It should be noted that many of the users pay for this information. It's not just an expense with regard to the census. For those who are not familiar with that, it's a product that actually brings in substantial revenue. It's unique to government to be able to quantify, and hence the censuses going back in time to the Roman Empire.

    Our current system right now is looking for restoration of that trust. Will that be achieved as the bill is right now, without amendments? That's what I'm wondering, and I'd like everyone's opinion here. If this bill is not amended, then in its current form will it fully restore the trust necessary to provide the best data collection opportunity that we have in the clearest sense of understanding amongst those who are collecting information and those who are using the information for products and services that really, at the end of the day, have consequential expenditures in our civil society, both private and public sector?

    Mr. McKinnon, maybe you can start with that.

     There are at least two aspects to it. A change in legislation cannot, by itself alone, restore confidence that has been damaged or solve that problem. On the other hand, having appropriate legislation in place and ensuring that it creates greater transparency, that people are more aware of things, of what has happened, and that there is a structure that ensures confidentiality, competence, professional integrity, and so on, are part of the building blocks.

    For many Canadians, the trust is something that is built over time; it can be damaged quickly. If there is a silver lining in the cloud of the 2011 arguments over the census and the NHS, it is that the end-users, the people who build transit systems, locate schools, and help people figure out where new markets are so they can put up shops, all spoke out about its importance. Part of the product of that is the rapid, and actually better, response rates in 2016 than to earlier ones.

    I'm backing away from saying that this solves the problem. Rather, this is one brick on a path that requires action and broader understanding.

    Before the 2010 census problem, as chief statistician, whenever I met with chief statisticians of the world, what I kept hearing was that Statistics Canada was the best statistical agency in the world. That reputation, of course, suffered as a result of what happened in 2010.

    Therefore, the question I give to you, sir, is if Bill C-36 had been the law at that time, would that situation have been avoided? My answer is no, but you might come to a different answer. I think the answer is no because of this problem that I mentioned between sections 21 and 22 of the Statistics Act.

    So you believe that the bill does need amendment. I think everybody has come here with amendment suggestions.

    The bill does need some pretty important amendments. It's not because somebody disagrees with those amendments, but my view is that probably someone hasn't thought about the challenges—

    Of course. It's the same with me. My motivation is about how to improve the product at the end of the day, and we won't get many chances to improve this product. It will be several years, if not a decade, before this comes around again.

    Mr. Cappe.

    I think you're asking the right question. The question is, though, what's the role of the statute in this? You can't fix this with the statute. It's a crucial instrument, and this bill is an improvement.

    Can the bill be improved? Probably, but I don't think I've heard any improvements, so far, around the table. As Mr. Thomas suggested in terms of having a different appointment process, I think that can be done, but it doesn't have to be done in the statute. I don't think the law has to oblige the consultation the government does on the appointment of a new chief statistician.

    On the competence of the chief statistician, with all due respect to Munir—we were in graduate school together—Mr. Sheikh was an unlikely appointment as chief statistician but did a great job.

    I was looking more for your opinion on the bill and whether in its current context the trust was there.

    I know.

    I'm going to move on to Mr. Thomas and get his opinion before I run out of time.

    Right, because I disagree.

    No, you're talking about other personal stuff that I'm not really interested in and I would prefer to spend my limited time with Mr. Thomas.

    I make a distinction between external trust in the agency and its products, among the principal users, the organizations that rely upon it, but also Canadians in general. There was a bit of damage to the image and reputation of Statistics Canada in the outside world as a result of the events of 2011, and that will take time to fully restore, but most Canadians aren't that well informed about the role of Statistics Canada.

    However, there's another level of trust, and that's internally. We've relied in the past on these unwritten conventions and informal understandings on how the relationship between the government and the agency should operate, between the minister and the chief statistician. Trust was harmed by the events of 2011. That will be more codified and formalized in this bill.

    Trust is an elusive phenomenon. It's difficult to develop, it takes time, it can be quickly lost, and it's difficult to restore. Now we'll have a new set of more formal working relationships, and I think that's where we should be at this time in the 21st century.

     Thank you very much.

    We're going to move on to Mr. Baylis. You have seven minutes.

    I'll start off with a set of questions to do with the advisory council. First of all, Mr. Sheikh, was it of value to the advisory council in its present form when you were chief statistician?

    I found the council to be extremely helpful. In fact, whatever path we were going to be on, to make substantial changes to something, my approach was to take it to the council, hear out what they had to say, make my decisions, but then get back to the council to explain why we had done what we had done.

    So they were a sounding board for you.

    They were a sounding board, and I must say that I did get good advice.

    I've been led to believe that it was very narrow, with a particular interest. Is it a proper reflection that your advisory council was very narrow in terms of its scope and understanding? It was mostly made up of academics and people with very specific personal interests.

    No. We wanted to get the membership of the NSC to ensure that in every subject matter area, we had an advisory committee on their subject matter. The NSC would have to fill the gaps that were not filled at that level, and the appointment of the members was based on that. By the way, it was the minister who would send these letters of appointment.

    Okay.

    There is a question about moving that from 40 down to 10. Are you in agreement with that change? What are your thoughts on that?

    I think 40 is too much; 10 is probably too little. So it's somewhere in between. It's hard to pick out a number. A bit more than 10 would be helpful.

    Mr. Thomas, you had actually thrown some numbers out and had some thoughts on the advisory council. Perhaps you could expand on that, please.

    Yes, I'll reflect briefly on my time on the National Statistics Council. The first time I was appointed, the official letter to join the council came from the minister. On subsequent renewals, it came from the chief statistician. So it seemed to shift back and forth between the minister's office and the chief statistician. It would be good to have one consistent approach to appointment.

    As I suggested, there is some uncertainty about whether this new body is to perform a representational role. In other words, would it capture the regional, linguistic, occupational diversity of Canada, and all sorts of other considerations, or is it more like a semi-governance body that is there to oversee the operations of Statistics Canada?

    Which role do you think it should play, with your experience?

    It would be a great loss if there wasn't a larger body with some representational role. I think it's a false dichotomy to say it's absolutely either-or. It could be a kind of hybrid model. But that takes you into the range of 20 to 25 people, as I suggest. As Munir Sheikh and previous chief statisticians have pointed out, there's a large array of committees that give advice to Statistics Canada, including provincial-territorial statistical advisory committees.

    So this body, though, didn't bring an institutional perspective. It wasn't a particular organization. We were encouraged as individual members to leave our institutional identity at the door and come into the room and ask the question: what's in the best interest of the national statistical system?

    It would be helpful, for example, to have somebody from the north, from one of the three territories—or maybe more than one.

    So it wasn't necessarily a fixed regional...but regional would have been interesting. But it didn't have to say, I've covered everyone. You mentioned that there are other advisory councils that are regionally based. Is that correct?

    Yes, and then there are specialized ones on a sort of functional basis. What does the mining industry need, or resource industries, more broadly? There's a wide array of those. It's not like the chief statistician and his executive team is without advice in terms of segments of Canadian society.

    This could actually just play a catch-all global role. As Mr. Sheikh said, he could come and bounce around ideas and then still have ideas coming from different sectors, different areas of the country.

    Some members of the current council, the National Statistics Council, are not so much extensive users of statistical data. They're people like me. They are more process and administrative people. Those kinds of people are useful as well, as the chief statistician deals with the Prime Minister's Office, the Treasury Board, committees of Parliament, and so on.

     Thank you for that.

    I'll turn to you, Mr. McKinnon, now and hear your reflections on both the composition of the board and the changes that are being proposed. There have been some arguments that there's been a lack of transparency. Maybe you could address that.

    First, I think the optimal composition of the proposed group flows very much from what it is you want it to do. I was pleased to hear what Munir Sheikh said, that we were a consultative group and tried very hard, as Paul has said, to leave our individual occupational hats at the door and to reflect regional topic areas in the whole country.

    For me, the critical issue is not what an ideal advisory group looks like until you've defined it. Therefore, what you really need to consider is what function that group should perform and where it fits into increased independence, transparency, and all of those, and this reporting function that it's given, and then flow out. Insofar as it acts as part of a series of balances and checks to assure independence and professionalism, it shouldn't have the same composition as we currently have. There should be significant overlap in terms of the composition.

    You need the specific areas of expertise, which we already have on the council, but because it will have a changed mandate, it should also reflect that changed mandate. I also believe, therefore, that the recruitment and appointment process should be more public, transparent, and meet the typical tests for that kind of an agency.

    Thank you very much.

    We're going to move to Mr. Lobb. You have five minutes.

    Thank you all for attending. We've certainly had some great panellists at our last two meetings, for sure. It is great to have feedback like this.

    My first question is for Mr. McKinnon. You're the chair. I apologize that I was late, but I was here for your words. Can you tell this committee how or if you were consulted by either the minister or the department on this bill?

    Yes. There were several periods. As soon as the government announced that it intended to increase the independence of Statistics Canada, the chief statistician asked us to begin to give him advice that he could convey to the minister about our thoughts. There was a working group formed. Professor Thomas was one of its members, along with other people who had greater experience with organizations such as this, or with changes.

    We prepared papers, conveyed them to the parliamentary secretary of the minister, had meetings with him, consulted back with the larger council, and have continued to iterate as the process has unfolded. We have met subsequent to the tabling of this legislation. We've had conversations in that same working group to provide advice to the chief statistician.

    Is it your position and the council's that most of your input and comments made it into the amendments, or are there others that still need to make their way there?

    In fact, before the details of the bill were out, we had already submitted, and we see some of the provisions of the bill reflect advice we gave. Our expertise is not in legislative drafting, as you can appreciate, so we focused on the principles and objectives, and we see many of them reflected in the bill and the amendments. There are areas of specific interest, such as Professor Thomas's, which some of these people have made suggestions about and also have resulted in our memos.

    Everything that you said I take with a great deal of respect, no doubt about it. The one point, though, that I continue to come back to is that the 2011 situation happened and we've moved on, and I think we're probably, overall, better for the changes. I truly believe that one of the key pillars of the independence is the actual behind-the-scenes performance of the computing and the data centres to truly make Statistics Canada a worldwide entity. It doesn't make any sense to collect all this great data and then have your servers crash when you go to publish it.

    What role does the council or what role do you or any of the other guests feel we need to play? Do we need to give the chief more latitude to be able to move where he or she feels the need to move to get better and more robust performance from our data centres? What is it, because I truly don't believe Statistics Canada to be independent if it's still got this cement block tied around its foot, which is Shared Services Canada.

     The National Statistics Council has not involved itself in what we would view as operational issues within the agency. If we see problems on things like delivery or process, we'll speak up. Frankly, as I said in my remarks, we don't have the expertise to make a sophisticated judgment about the appropriate case.

    However, I will say that as soon as the government announced its intention to make changes to enhance the independence of StatCan, one of the things the working group of which I was a member did was to canvass the former heads of other similar statistical agencies in Britain, Australia, and New Zealand. They said they felt a significant element of their independence flowed from their control of their informatic structure, and we passed that forward.

    Again, we don't have the capacity or experience to make judgements, but we convey that.

    That's fair enough.

    Professor Cappe, you may answer very briefly.

    I'm not going to give you a simple answer. I was deputy minister of three departments. When I was deputy minister of the environment, we had the supercomputer in Montreal doing the weather forecasting. We had our own people there. They're now part of this infrastructure, Shared Services Canada.

     It can work. There are advantages to having that kind of overall government purchasing and control.

    There are disadvantages. We know the RCMP and Statistics Canada have said they want their control back. I have to say that I think it is too early to tell, and there are potential advantages from having that.

    Thank you.

    Mr. Sheehan, you have five minutes.

    Thank you very much. This question, first of all, is for Mr. Sheikh.

    It was my understanding that an important factor in your resignation was the government interference in statistical matters in a way that made it seem like you as the chief statistician at the time had agreed with the decision. I just want to know if that is correct.

    Further to that, I'll get to Bill C-36. Had it been in place at the time, how might the transparency around the direct power provision have altered the events? Would it have prevented the perception that you were on board with the government's decision? Finally, would you have felt more comfortable keeping your position as chief statistician knowing that the accountability for the decision clearly and transparently rested with the government and not you?

    In my testimony to the committee seven years ago, I said that I had two jobs, one was to give my best advice to the government, and the second was to implement whatever the government decided.

    I think it's pretty obvious by now that the advice that we gave the government was to continue with the long-form census. The government said no. They wanted the national household survey. StatCan's response was, “You're the boss. We'll do whatever you want.” We tried to do our best with the survey.

    It was when the minister started to blame Statistics Canada for recommending the survey and saying that the chief statistician personally gave him that advice that it became very difficult for me to continue in that job, especially since he had done it a number of times.

    I did my very best to encourage him to stop, but he did not stop. I think the point came when I said, “I can't work in this situation because it leaves the impression that we gave bad advice to the government.” That was my only way to send a message that we did not give that advice.

    If Bill C-36 had been part of the Statistics Act at the time—I have already dealt with that in my comments, and I have said that I don't think anything would have been different. I would still be gone, and the minister would still have said the same things. We probably would be sitting here today talking about Bill C-36, but it would not have changed anything.

     Thank you very much.

    I know that my community of Sault Ste. Marie is very glad that the long-form census is back. The last data that were released are really important for a community of our size to have because they weren't really captured previously. That allows our economic development agency in our city to really advance some thinking and some planning for the future. I think your advice was well warranted at the particular time because it's being well received now. Thank you for your work in the past.

    My next question is for our friends from the Universities of Toronto and Manitoba. It's my understanding that many countries that have independent statistical systems continue to treat the decisions to make a survey mandatory as a regulation due to the political implications involved. In other words, they continue to ensure the government has some powers to decide what's mandatory and what's not. However, the statisticians who were here previously, Mr. Smith and Mr. Fellegi, seemed to think the government should not have any say whatsoever, no matter the circumstance, because the decision to make a survey mandatory or voluntary is strictly a methodological matter.

    As experts in your own right in the field of politics and public administration, do you think it's reasonable that the government, in a democratic, Westminster parliamentary system, keep some form of directive power when it comes to decisions that involve state, power, or coercion?

    I can go first, if you like.

    I agree with you that our system of responsible government is based on the prime minister and ministers of the crown being required to explain and defend what they've done in public and to be held accountable ultimately through elections, before the House of Commons, and even, dare I say, the Senate. You can't expect ministers to be completely indifferent to the scope of questioning in a census, the types of questions they're asked, how intrusive they are, and the range of things that are inquired about. I think there has to be some final role for the cabinet.

    What I was suggesting in my brief was, on that relevant section, proposed subsection 4.1(1), where it authorizes the minister to intervene on a technical operational matter, they made clear that the chief statistician be allowed to voice objections in a public manner to that, and that it's not fatal to their reappointment.

    Undoubtedly, when you go to work the next Monday morning, it's a little tense between you and the minister and the government. There is no doubt about that. This is a last resort. It comes after long negotiations. It's to recognize the fact that there's no clear dividing line between policy and operations. It's a blurred line. In the mind of the minister, this may look like technical matters to the chief statistician, because in mind of the minister this is a sensitive policy matter that he heard a lot of conversation about from the people he interacts with. I think there's a way to strengthen the protection against a repeat of 2011.

    Thank you very much.

    We're out of time.

    I, I—

    I'll use my time.

    It's up to you.

    You have five minutes, Mr. Nuttall.

    You can jump in there.

    I was just going to answer Mr. Sheehan that the Minister of Finance has just asked StatsCan to do some survey work on housing. He didn't say to make it a mandatory, compulsory piece.

    It's conceivable that the government wants a universal, census-like survey and wants to make it mandatory. I can't imagine what that might be, but I wouldn't think it's a....

    I understand Mr. Fellegi's point. Yes, it is methodological. What's a right survey? But we shouldn't constrain the government to say that you can't make it mandatory.

    Thank you.

    Thank you, Mr. Chair.

    If I may continue, I want to jump back to some questions that I had asked of the minister and members of this committee previously with regard to the appointment process for the new advisory council. I think there has been some talk about the right number of individuals on the council as well as how that happens.

    Is 10 the right number? Does it make a difference?

    I think the interesting question is, what do you want it to do?

    If it's a methodological advisory committee, it's different than if it's a representational committee. If it's representational, then yes, you want to make it a broader group, as Mr. Baylis raised earlier. I'm not sure what you want it to be.

    The other thing is, it isn't the only advisory committee. You could have a complementary methodological based committee. You already have all these other groups advising on specific surveys. There is not a right answer.

     So the outcome you're looking for will determine the number of individuals in the community.

    Yes.

    The follow up to that is this: with it being transitioned back into an appointment process through the cabinet itself, is there a fear that this may become a partisan dumping ground? It almost goes back to the original question, which is, what type of committee it is, and so on.

    I have more confidence because, frankly, there's a political consequence to making it a political dumping ground. So you're not going to let them get away with that. That's the confidence that I have in the process.

    Is that shared?

    Maybe I could offer a perspective. I was twice chair of boards of crown corporations in Manitoba, appointed by different governments and advertised to be a non-partisan board chair. I had colleagues on those boards of directors who came from political backgrounds, but they had appropriate expertise. Just because someone at some time in the distant past, perhaps, had been a Liberal, a New Democrat, or a Conservative doesn't mean they should automatically be disqualified. Again, it goes back to the purpose of this body. A very small body of 10 people I think would send the message we're all about governing. That we're there to oversee the chief statistician, we're invited to step over the line between the policy-making inside Statistics Canada and become interested in the budget process, in staffing the agency, in operational matters. I think that would be a danger. It's not like a board of directors of a crown corporation, or a private corporation, a commercial firm. It's a different entity. The body would be there to serve the public interest. There, I think you need broader representation than just 10.

    That's perfect. Thank you, Mr. Chair.

    We're going to move to Mr. Longfield. You have five minutes.

    Thank you for the questions from across the way.

    I agree with Mr. Lobb that these last two meetings have been exceptionally good for presentations. Thank you for all your non-partisan presentations on a difficult subject.

    In 2010-11, I was president of the Chamber of Commerce in Guelph, and tracking the labour statistics. I was really disappointed, to say the least, that the long form was taken away. I found I wasn't the only one. I was looking at it from business, but the not-for-profits really surprised me; how much they rely on census information. When we restored the long form, they all wanted to come and hug me, which is okay.

    They were looking at homelessness and at issues around poverty reduction. When those issues from our communities are coming forth through the government, could you talk about how governance picks up on that? What leads the charge in what surveys are done, what's regulatory, and as Mr. Dreeshen said, we have the census of population, the census of agriculture, but what other mandatory way could we have to address homelessness and poverty reduction in our communities using StatsCan?

     I'll go to the front. You've raised an enormous issue, and I tried to emphasize in my testimony earlier that what matters is the Canadian statistical system. The census is critical because it's the foundation and the benchmark against which we track everything else. The reason that the changes, the independence, and the trust are so important, is that it's not just a question of whether we should do a single survey, change a question in the census, but also, where do we get administrative records? Will provinces...? When you're into issues like homelessness, you need social data, education data, and health data to become the trusted repository of information that may be collected, particularly through administrative records. That's what I think is really at stake here.

    Is that addressed in this legislation?

    I do not believe you can compel any of that by legislative fiat. You have to reinforce the climate that said this is a trusted system and agency that we will entrust with our data.

    When it isn't trusted.... I have two elements in terms of customer service, and we had a lot of testimony around customer service at the last meeting. In March 2014, Guelph's labour force went from 82,000 to 92,000 in one month. It jumped by 10,000. It's never been corrected. I've always been suspicious about that number. The process to correct it is one issue.

    Along the same line, we have an employer that has 26 locations and has to submit census information for each of the 26 locations, instead of consolidating and doing the job once. It takes them a day and a half to prepare the information. They would like to get that corrected.

    In either, how would corrections be taken forward to Stats Canada under the current system, and would it be improved in the new legislation?

    Mr. Cappe, I'm a University of Manitoba grad, so maybe I have to direct it towards you to start off.

    You mean Mr. Thomas.

    I'm sorry.

    Mr. Thomas.

    Here in the middle of the country, the real centre of Canada—

    Absolutely.

    —bone-cracking winters and killer mosquitoes...we've got it all.

    They'll smack you back.

     It's an important point that goes back to the role of the current National Statistics Council. I'm one of two members on it from Manitoba. We had a few fights break out between Manitoba's statistical agency and Statistics Canada over whether there was an undercount of population. That matters. It means millions of dollars in equalization payments, for example.

    Manitoba officials came to me and said, “Let us hear your side. We want to tell you our side of the story.” It turns out that there were methodological disagreements, but I'm saying that's an example of how it helps to have someone in a particular province who's at least somewhat identifiable as a figure of this and who has connections with Statistics Canada but is not on the payroll of Statistics Canada.

    That's the kind of thing I would offer as a suggestion for why 10 people is not enough.

    Thank you very much.

     We're going to move to Ms. Blaney. You have two minutes.

    Thank you so much.

    Just to reiterate what the former member said, I ran a non-profit before this, and the value of the information was tremendous. When we lost it, it was a huge detriment.

    Could you tell us, Mr. Sheikh and Mr. McKinnon, why you think the government has resisted the idea of adding a preamble consistent with the United Nations Statistical Commission declaration?

    I don't know why the government has resisted the idea, but as a former chief statistician, I find the principles to be extremely valuable. For example, Bill C-36 talks about the chief statistician doing his or her job. It makes it very clear, and the language in that is what I would say is a one-sentence summary of the principles. That's one reason.

    I'm not quite sure what extra value you would get by adding those principles. It really is right there, in my view.

    But they would change, too.

    They would change as well. The principles change, so putting them in the legislation creates the challenge that you have to amend that law.

    The other thing I want to mention here is that Canada plays a huge role in the UN. I, as the representative of StatsCan, was put in that position to play that role. On many occasions, we actually lead the UN in many things. I'm somewhat concerned about putting the UN in the law, when Statistics Canada may be doing a better job than anybody else. To me, it's a bit of a challenge.

    That actually gives me a flexibility, so I think what we have here is just fine.

    It's not the Statistics Council's business to endorse, or not, the UN Charter. We've been well aware of it, and we think it is a very terrific series of guidelines. In our discussions, we have always said that whatever Canada does should be consistent with those guidelines.

    Frankly, on the issue of legislative drafting, I can't opine as to why they did or did not include it.

     Thank you.

    We have to move on. We have enough for half a round so we're going to go back to Mr. Baylis.

    Thank you, Mr. Chair.

    There has been a consistent message from all four of you including Mr. Fellegi and Mr. Smith, people who testified about the need to make sure we have a differentiation between the methodological statistical issues and political governance.

    It reminded me of the old saying, “There are lies, damn lies, and statistics.” I figure you want to leave the lies and damn lies to us, and you will take care of the statistics. I would like to explore that.

    We will begin with Mr. Cappe.

    In your testimony, you mentioned that it was necessary to clarify that point of view.

    Personally, I don't think it is possible to clarify that.

    However, it would be possible to face that tension and try to find a balance between the two. In my opinion, that is one step, and as we were saying, the law isn't going to be changed.

    Mr. Fellegi had asked that it be written down that the methodological decisions are to be made by the Chief Statistician.

    I think that is clear enough now. However, if it needs to be added, I have no objection.

    However, I think the government should have the power to make the really political decisions.

    Mr. Sheikh, you mentioned that you're not sure this legislation would have addressed the issue that forced you to resign. Mr. Fellegi had mentioned the need to put in the wording that methodological decisions should rest strictly with the chief statistician. I believe your words were to the effect that technical statistical decisions should rest with the chief statistician. Is this something you would like to see inserted in the legislation?

    As Mel said, I find it really difficult to define those terms in legislation.

    Mr. Fellegi, for example, used the words “methodological decisions”.

    I wouldn't worry about putting in those words. I don't think it's helpful in solving the problem you're talking about. The problem for me is this: I think the bill quite correctly tells the chief statistician, “You will do this. However, the government can send you a directive”. My focus was when the government sends a directive—and it should be able to send a directive—it would be under exceptional circumstances.

    You had said earlier in your testimony that it would not be appropriate to have decisions on statistical matters come from the minister.

    It is not appropriate for the government to do that.

    I understand that, but how do we address it?

    At the same time, I understand the difficulty of having something in law to do this. My indirect way to get to that is to strengthen the clause where the government can send a directive by saying “under exceptional circumstances”. Professor Thomas suggested you table it in Parliament. I think it's their discussion, depending on the nature of the issue, that will bring out whether or not it is a statistical issue. It's really hard to define.

    If it was a statistical issue, the first thing we should do is make sure that it sits with the chief statistician.

    Yes.

    Some way or another, we need to find this balance, as Mr. Cappe has said, to make sure these statistical decisions are left with the chief statistician.

    I'm trying to solve this problem indirectly. The biggest issue right now for me in the law is that section 22 lists the things that the chief statistician can make decisions on. Section 21 says that the census questions are done by the cabinet.

    You mentioned the—

    If you make the simple change that says section 21 is subservient to section 22, to me that would solve a huge problem.

    We should make subsection 21 subservient to section 22?

    Yes.

    Okay, we'll come back to that.

    I'll move on to Professor Thomas. I believe you had specific ideas on how to address this concern about carving out the difference between what would be a political decision and what would be a statistical decision. Maybe you could elaborate on that.

     Yes, I said that there's no clear dividing line between policy, operations, methodologies, and so on. What you don't want is to give the ministers the best of both worlds, what I would call “discretionary accountability”, where, when it pleases them, they can intervene and get their way through dinnertime directives, as I used to call them when I chaired the boards of crown corporations.

    These instructions to the chief statistician at some point should come out in the open, but that's after a long process of back and forth. Then when the minister decides that he wants to overrule the chief statistician, or the government does—I think it should be a cabinet decision—then that becomes public, and the minister can boast and confess why he or she thought this was the right thing to do in the national interest. Presumably, there are skeptical, or if not, hostile, people in the opposition who will ask challenging questions for why this was the case.

    I would even go so far as to say that the chief statistician should be allowed the opportunity to state publicly the reasons for his or her objections to a directive that related to operational matters.

    In your way of handling it, say, compared to Mr. Fellegi's, who wanted something in the bill very clearly delineating...perhaps, as Mr. Sheikh says, making section 21 subservient to 22, you're saying put molasses into the system, so if it's attempted it has to go through this process. We're going to be able to shine a light on it. That might act as a deterrent for the minister trying to get the best of both worlds, where he's changing the statistical methods to get a political gain.

    Yes. There's no foolproof way of stopping a minister from trying to do something either for political reasons or because they have a different substantive judgment from that of the chief statistician.

    Let's be clear. As you said, if we've hired a chief statistician, we're not looking for a minister to have statistical expertise over and above what a chief statistician and his advisory council have.

    Absolutely not, but you could say that the chief statistician and the advisory networks they have in place create a narrow perspective on some of these issues. The minister and the government bring a broader perspective: what does this mean nationally or regionally for different segments of Canadian society? They apply a much wider matrix of considerations than does the chief statistician perhaps, and they should have the right to do that at the end of the day. At some point, we have to trust the politicians to do the right thing.

    Thank you.

    I see, Mr. McKinnon, you're eager to jump in. Be very brief, please.

    I have two quick notes.

    First, it is very difficult to draw a very firm line. To use the example of response burden in the technical sense, with 15 different operations a minister might say to a chief statistician, that's a huge burden. That is also a core methodological issue about how you collect the data, and a good example of where those intermesh.

    Thank you.

    Mr. Dreeshen, you have seven minutes.

    Thank you very much.

    I had the pleasure of going to Belgium for the Blue Sky 3 forum and so you get a chance there just to see the value of data analysis and innovation, and how important it is throughout the world. Of course, there are different ways that the international community assesses information. Of course, our universities are paramount in being able to use that as training ground and then as an expansion. That's a critical part of it.

    One of the things I'd like to ask—and perhaps, Mr. Sheikh, you could reflect on this—is what statistical tools were used to correlate the long-form census data with the results that were obtained from the long-form survey? The data came in two different formats, but it's similar to 60,000 Canadians being of the Jedi religion. There has to be some analysis and some way in which you relook at the data.

    I can give you an example of a comparison of the survey with the census. In 2011, there were some areas where questions were asked in the short-form census that were the same as in the long-form census, the national long-form survey. If you look at that small number of questions and compare the responses, you have a fairly good idea as to how good or bad the survey is. The census is the census, and everybody had to answer it.

    I'll give you an example of one area. In the Toronto metropolitan area, there were hundreds of units that responded. I looked at many indicators, but let's say apartments, so how many apartment buildings existed in various areas of Toronto. The ratio of the response in the short-form census to the survey was 0.25 to 3, and that tells you the error in the survey.

    There are ways of comparing those, and in fact StatsCan does it all the time. We have a lot of voluntary surveys, and we test them against the census and we adjust them.

     That's one of the points I've tried to make. Statistics is a tool. You work with it; you take the best information that you have, and you go forward. I know, for example, with the homelessness surveys and so on, people were saying they may not have had that data, but people have found their ways around it and there has been success that has come through that.

    With the long-form census, as far as farmers are concerned, if you get that it's as though you've drawn the black straw. They're after you constantly about why you haven't got this in yet or why you haven't done that. Prior to that it was, “Well, you're going to go to jail,” if you were hauling your own grain across the border.

    If these are the situations you have, it's about the timing of them. Lots of times when these things are sent out, you're on your tractor. You may have your phones and you're all linked up, but I know this is one of the reasons why in the agricultural communities you say to give you a break. Of course, in your case, you say, “I have all this information they're asking for anyway. Now they want me to go back and drag all of this out and complete all these forms.” Yes, it's important, but again, there are a lot of different ways of analyzing that.

    The other thing I want to talk about is the census questions that are there. When they come from government, from any government, they can be couched in the type of commentary that you might want. There may not be any questions on how much carbon tax you are paying right now, because they might want to talk about something different from that. They may want to talk about green tech and where that is going, but they may not want to talk about clean coal technology and all of these sorts of things because those are outside of a policy position that they have.

    How do you on your council look at these sorts of things and say that you know which is the political side of this and which is the natural resource side and how best that should be presented? How do you then tell the ministers or whoever is making those decisions to back off a bit here, that you know what the difference is, and you know what the situation is?

    First, I'd like to distinguish between the census, for which there is a very long history.... Frankly, statistical agencies are very conservative organizations. Given the choice between changing a question over time to reflect changing circumstances and asking the same question twice so you have the same comparability over time, they always go for the comparability, or almost always. With the census, there is a very well-structured, extensive consultation with stakeholders, both with the people who provide the information and with the users.

    With regard to individual surveys, there they are fairly extensive. If we were to do an energy-use survey, a lot of it would be from administrative data just explaining what it is. In that case, there are many....

    Here's where policy—and not methodology—intervenes. If the government of the day says it has an important issue area that it needs to work on, be it housing prices, the debt of Canadian households, homelessness, or the role of NGOs, it can—and I come to Statistics Canada—say, “This is our policy interest. Design, and you will be funded for, a survey.” This would usually come through a department, through extensive consultations, and on it would go. There is, I think, a good example of where the broad policy interests of a government are distinct. The chief statistician can say, “Here is the way to gather information on that topic”, and they would control....The government would say, “Here's our policy interest area.”

    Do you have any other comment here?

    A voice: The same.

    Mr. Earl Dreeshen: I have one last question. Is it important to reverse the mandatory nature of StatsCan's participation in the Shared Services Canada initiative, and would it maintain or restore management control over IT to Stats Canada including IT's hardware budget? Is it important to allow them to have that flexibility?

    As I mentioned to Mr. Lobb, I think it is too early to answer the question. There are advantages to having Shared Services Canada. There are disadvantages. I know that when I was managing departments, and when I was deputy minister of Human Resources Development as it then was, which is now ESDC, all of the informatics for the Canada Pension Plan and the Employment Insurance Commission were done in-house. Frankly, we would have benefited from having an outside manager who was handling this for us, because we ended up cancelling a contract for $750 million because the contractor wasn't delivering. We could have had a better result if it had been done by a centralized government organization. So I've seen both—where it's better to have it under the control of the department and where it's better if the government, the whole of government, is trying to manage it.

     Thank you very much.

    Ms. Blaney, you have seven minutes.

    Thank you. Again, thank you so much for your time.

    One of the questions that I have is around the nature of collecting information through the two different methodologies.

    To use a concrete example, in the context of Bill C-36, will comparisons between data collected from the mandatory census versus data from a voluntary long-form census be statistically reliable?

    I think StatsCan made it pretty clear in one of the releases, which is on the website, that the two are totally different and you cannot compare the results.

    I would just note that my students have been writing papers, and 2011 is an anomaly.

    Furthermore, and to go back to something Mr. Sheehan talked about, the voluntary nature of the NHS meant that for small areas and specialized subpopulations, no data were released, because they didn't have the confidence they required.

    Thank you.

    Dr. Thomas, can you elaborate on your proposal for the appointment process for the chief statistician? Why is it important that the government engage in this process in a collaborative and transparent manner?

    I've tried to make the case that the chief statistician position is different from that of the regular roster of deputy ministers, in my opinion. That individual needs to be part of the deputy ministerial community, meet regularly with the clerk of the privy council, and be involved in the discussions of the statistical needs of the various departments and agencies of government. In other jurisdictions—as Ian McKinnon mentioned, we talked to people in the U.K., Australia, and New Zealand—there is a separate process of appointment.

     I think that we should have a more customized, tailored appointment process for the chief statistician. There was one private member's bill that even suggested that individual should become another officer of parliament or agent of parliament. That's a terrible idea, a very bad idea, but—in recognition of the status of Statistics Canada as a “statutory agency”, which implies more independence, more impartiality, and more autonomy from the central decision-making apparatus of government—I think we should change the appointment process.

    It will be hard to get any prime minister, regardless of which party is in power, to give up the prerogative to make these appointments, so I'm just saying we need a different advisory mechanism while retaining the right of the prime minister to select the final name.

    Someone mentioned earlier the possibility that there might be advance consultations with leaders of all recognized parties in the House of Commons. That could be done, but it will slow down the process and we already have a significant backlog of order in council appointments. These sensitive positions like chief statistician should not be held on a probationary basis for very long.

    I think there's a way here to give some opportunity to get better background knowledge about someone and make a qualified appointment. Getting the right person is critical, as I said earlier.

    Thank you so much.

    Mr. McKinnon, I'm going to come back to you.

    Does the outsourcing of data collection responsibilities to an agency outside of Stats Canada—in this case, Shared Services Canada—introduce another area of risk to data privacy? I'm not talking about firewalls and hacking. What I'm really wondering about is the data collected by Shared Services Canada on behalf of Stats Canada being used by other agencies or departments for purposes unrelated to the original data collection purpose.

    First, in many cases, the collection itself, as I understand it, will continue to be done by Statistics Canada, so we're talking about the data processing operations. I want to demur on the question because I really do not have the technical expertise, nor does the council view itself as having the kind of expertise to make a sophisticated judgment.

    As I said earlier, in addition to the technology involved—the machinery, literally—corporate culture is critical. We do have confidence in how deeply ingrained confidentiality is in the corporate culture in Stats Canada. I cannot speak for Shared Services. That's not to say they do not have it or will not develop it. I simply don't know.

     Thank you so much.

     Those are all the questions I have, so if there is any time left, I'm happy to share it.

    Thank you.

    For the final five minutes, we'll go to Mr. Jowhari.

    Thank you, Mr. Chair.

    I welcome you all.

    I want to go back to the long-form census, and I want to talk about two things: number one, whether it should be made mandatory or not; and number two, whether the scope of the long-form census should be legislated or not. Prior to you gentlemen coming in, witnesses talked about the fact that they all welcome the long-form census and the fact that it's giving us more data, but they identify as a gap the fact that it hasn't been made mandatory in this bill. Also, a few of the witnesses said it is an issue that the scope is not legislated.

    I'd like to actually go around and hear from all of you for about 45 seconds each.

    I'm going to start with Mr. Sheikh to get your input vis-à-vis what the issues are if it's not made mandatory, because right now in the bill it is not mandatory, and I want to understand what the issues are.

    The Statistics Act is totally unclear on the issue and Bill C-36 is silent on the issue.

    What I am suggesting, making section 21 subservient to 22, would—in my view—solve the problem. The way it would solve the problem is that section 4 gives the issue of methodological decisions to the chief statistician, covering all areas in section 22. So as soon as 21 is made subservient to 22, the chief statistician would then be able to prepare a long-form census, and he is the one who makes the decision as to whether something is voluntary or mandatory. Of course it would be mandatory.

     However, for accountability, you want to make sure the Government of Canada has the final say on it, and with section 21 around, the Government of Canada can approve it, reject the chief statistician's advice, and send him a directive at that time and say, “For the following reasons, we are not going to accept what you are recommending.” To me, that simple change would solve the problem.

    Perfect.

    Should the scope be legislated?

    No, I think you need to have the flexibility to ask the best questions.

    Mr. Thomas, can we go to you next?

    Well, on whether the census should be legislated, I think we're trying to get away from an overly detailed prescriptive act, to give the chief statistician more autonomy, while at the same time resetting the relationship between the agency and the government. It is interesting that the decision in 2011 led to such a wide outcry across the country. It brought protests from both public and private organizations, which suggests that there is political content to these decisions and that therefore it's appropriate at the end of the day to have responsibility reside with elected and accountable ministers and the government.

    Great.

    Go ahead, Mr. McKinnon.

    I agree strongly with Professor Thomas, that trying to legislate the scope is a mistake. On the other hand, I'm sensitive to the concerns that were raised in the earlier committee hearings that the government might instruct that the census be extremely brief and that everything else be done by surveys.

    A voice: [Inaudible—Editor]

    Mr. Ian McKinnon: Centrally.

    My first point is that they would now be much more public in owning the responsibility. I'm not sure about the legislative impact, not being a lawyer. I would say it would be similar in scope, but tying the scope is a mistake, and I will use the example of the agricultural survey. StatsCan is well aware that it's burdensome, and it is trying every which way. We have on the horizon technologies like remote sensing, which might allow the census to collect that kind of material directly.

    And on the mandatory side?

    On the mandatory side, I believe with the census that we must have our benchmark data mandatory for sound methodological reasons and because we base all of our voluntary surveys on benchmarks back—

    Do you agree, Mr. Sheikh, that rather than making it mandatory as part of the bill, just making section 21 subservient to section 22 will do the job? I'm just asking—

    Let me just clarify. The reason, in my mind, it would work is that the chief statistician—if I understand the bill correctly—has the option of making something voluntary or mandatory but informing the minister. So I cannot imagine that a future chief statistician would want to make the existing long-form census voluntary. The minister may, but not the—

     Mr. Cappe.

    Mandatory, yes; scope in the legislation, no.

    Thank you.

    A hearty thank you to our panellists; you've been very informative and given us lots to think about.

    Before we break I just want to do a quick housekeeping. On the 11th of next week we have three more witnesses, perhaps four, we're still working on a fourth. The 13th—as you know, it's the day before Good Friday—is going to be a regular committee day for us. We are pushing the minister from 8:45 to 9:45. He's going to go from 9:45 to 10:45. In the first hour we're going to talk about our Washington, D.C., trip because we won't see each other before then. We really need to button down our witnesses and so on. Beyond that we're in Washington, and when we get back on the 4th we will be doing clause-by-clause. I did forget to say that on April 20 your amendments are due for Bill C-36; that's very important. We're leaving the 9th open for potentially more clause-by-clause if we need it. If not we can move other things forward to that day.

    Let's say goodbye to our guest, Mr. Thomas. Thank you very much.

    Thank you very much, Professor Thomas.

    Thank you.

    The meeting is adjourned.