Notices of Meeting include information about the subject matter to be examined by the committee and date, time and place of the meeting, as well as a list of any witnesses scheduled to appear. The Evidence is the edited and revised transcript of what is said before a committee. The Minutes of Proceedings are the official record of the business conducted by the committee at a sitting.
Welcome to the 42nd meeting of the Standing Committee on Industry, Science and Technology. We're here today pursuant to order of reference on Bill C-625.
We have some esteemed witnesses with us, but before I introduce them, Jean-François is here and he's the expert on iPads and on our move toward a paperless committee, so if anybody needs any help, Jean-François will be here for the first bit to coach and massage your technological expertise in order for you to have dominance of your iPad.
I don't have anyone here in that regard, but I'm certain we could work through that.
Thanks, Jean-François, I appreciate it.
We have before us Joe Preston, member of Parliament for Elgin—Middlesex—London, as well as Wayne Smith, who is Canada's chief statistician. With Mr. Smith is Mitch Davies. He's the assistant deputy minister, strategic policy sector, Industry Canada.
We'll begin with opening remarks. I think there have been discussions among everyone at the table and because of the nature of the bill, when we move to questions it will be whoever has questions, and we'll be able to have a very orderly and efficient meeting.
Mr. Preston, if you want to go ahead with your remarks first, we'll go to Mr. Smith after that.
The principles of this bill are simple, Mr. Chair. It addresses two very important issues. First, the bill seeks to eliminate the threat of jail time for Canadians who refuse to complete mandatory surveys, and second, it will ensure historians have access to related census records, where Canadians have given their permission, 92 years after the information is collected.
With these changes to the Statistics Act, we are again delivering on promises made. Our government committed to removing the penalty of jail time for anyone who refuses to complete any mandatory survey administered by Statistics Canada. The members of my constituency asked me to do the same. The bill delivers on this commitment by eliminating the threat of jail time under sections 31 and 32 of the act. It removes this threat for those who refuse to complete the mandatory surveys and for those who deny access to administrative records. It also proposes to eliminate the threat of jail time for failing to pay a fine under those two sections.
Canadians understand the importance of the census and other surveys in which they participate. I believe that people provide their information to Statistics Canada surveys because they know that it is by far the most reliable and accurate source of Canadian socio-economic information at the national, provincial, territorial, and community levels.
Jail time is a punishment that should be reserved for the most severe crimes. Canadians should not be threatened with jail time for not filling out a survey. We have an obligation to eliminate the penalty of jail time from this section of the act and replace it with a more reasonable penalty.
This brings me to the second part of the bill, which addresses the release of historical household survey records related to the census of population 92 years after its collection. This bill delivers on that commitment by adding a provision to the Statistics Act that allows access to these records. This change echoes the decision that was made in 2005 to amend the Statistics Act and allow for the release of census records after 92 years.
The adjustments in this bill will allow for the wealth of information collected through the 2011 national household survey to be released in the year of 2103 for the Canadians who have provided their consent. I'm sure we can all agree that it is important to leave a record of present-day Canada for future generations, and with this amendment to the Statistics Act, we're giving Canadians that choice.
I would encourage my colleagues to support this important piece of legislation.
Mr. Chair, members of the committee, thank you and good morning.
I welcome the opportunity to speak to you about Bill C-625. I'd like to begin by saying that Statistics Canada fully supports the proposed amendment to remove the penalty of imprisonment from the Statistics Act for Canadians who refuse to comply with mandatory data collections.
Under the Statistics Act, all data collections are mandatory by default and refusal to participate in a mandatory collection is subject to the penalty clauses that are addressed by this bill. Since 1982 Statistics Canada has had the authority to declare participation in individual data collections, with the exception of census of population, voluntary for the purposes of the act and refusal to participate gives rise to no penalty at all.
Today, virtually all business surveys, including agricultural surveys, are still collected on a mandatory basis. Household surveys are generally conducted on a voluntary basis and there are two current exceptions. One is the census itself and the other is the labour force survey. While business surveys and the labour force survey are mandatory, and while refusal to participate could result in prosecution, and therefore, penalties, Statistics Canada in my 34 years at the agency has never referred either a business data nor a labour force survey refusal case to the Public Prosecution Service.
The only instance in which Statistics Canada refers cases to the Public Prosecution Service for possible prosecution, therefore bringing individuals into the scope of the penalty sections of the act, is for refusal to participate in the census of population itself. The decision of whether to prosecute is ultimately taken by the Public Prosecution Service.
A long-standing practice of seeking prosecution for census refusal is in recognition of the constitutional importance of the census and its foundational role in the national statistical system. In the last six census cycles, between 18 and 74 refusals have been prosecuted per cycle. The very small number of cases reflects the difficulty of establishing an unambiguous case of refusal. For successful prosecution an individual's name must be determined. They must have refused multiple times. They must be the same person in the household refusing on each occasion. The field staff involved must be able to definitively physically identify the person who refused. The field staff involved must also be able to testify years after the event and the person concerned must have received and failed to respond to a registered letter sent by me advising them of their responsibilities and the risk of prosecution.
Our objective in proposing cases for prosecution is to establish the fact and seriousness of the legal obligation. The agency does not emphasize penalties in its communications or on the census, although the news media reliably report on this aspect of the census during the collection period.
While nothing prevents a judge from applying the jail term provision of the act's penalty clauses on a guilty verdict, this has only occurred once in my 34 years of career at Statistics Canada and in very unusual circumstances. Most judges will typically invite the accused to complete the questionnaire at the beginning of proceedings. If they comply, and two-thirds of people do comply, the charges are normally stayed or withdrawn. If they decline and are found guilty, which occurs about 90% of the cases that go to trial, community service or fines are typical of penalties assigned.
Essentially, the courts have also clearly viewed the imprisonment penalty as unreasonably harsh. Statistics Canada agrees that the jail term penalty is inappropriate and should be removed. I'm aware of no one external to Statistics Canada who is arguing for its retention and we anticipate absolutely no adverse impact on our operations as a result of removing the jail term penalty.
The bill also proposes to amend the Statistics Act to include census-related surveys, such as the national household survey and the provision to release individual census records to Library and Archives Canada 92 years after the information is collected, provided personal consent is obtained. Statistics Canada also supports this amendment.
Census records for individuals are important to genealogical and historical research in Canada. In 2005 it was noted that historical census legislation did not allow for release of individual records for censuses subsequent to the 1901 census because of confidentiality provisions. The Statistics Act was therefore modified to allow for full release to Library and Archives Canada after 92 years for every census taken between 1910 and 2005.
For the 2006 census and subsequent censuses, release after 92 years was only to be allowed with the consent of the respondent. This applied to records from both the short and the long census forms.
In 2011 when the long-form census became voluntary under the name of the national household survey, this change was approved under different dispositions of the Statistics Act and the census, and as a result the individual records could no longer be released, even with consent, after 92 years. Genealogists and historians therefore were at risk of losing an important future resource. Statistics Canada, anticipating that the act would be changed to allow for release, asked respondents in 2011 for consent to provide their information to Library and Archives Canada after 92 years, but we did not have the legislative authority to actually do so.
The proposed amendment will allow Statistics Canada now to provide these 2011 national household survey records, and any similar records from future surveys conducted jointly with the census, to Library and Archives Canada. This recreates Parliament's original intention from 2005. The 2005 amendment to the Statistics Act also provides for an eventual parliamentary review of the outcomes from that amendment to determine if the intended objectives of supporting genealogical and historical research were met.
In summary, Mr. Chair, Statistics Canada fully supports the amendments proposed in Bill C-625.
Now as discussed, colleagues, if you have a question, just identify yourself and I will recognize you. We'll just continue in a casual fashion in that way until the answers are given, and then we'll move to clause-by-clause.
Are there any questions for Mr. Smith and Mr. Preston?
Is there any information on a questionnaire, say, for a small business or agriculture, which is a small business in some cases, that would be on the census questionnaire but not be contained in an income tax return or some other federal mandatory return?
Yes, there is a significant amount of information that isn't necessarily available from other administrative data. Statistics Canada does in fact use administrative data to the greatest extent possible to avoid imposing a burden on businesses and small businesses, but nonetheless there are a number of cases where the data's not available either on a sufficiently timely basis, or the data itself is not available from administrative records.
I'm just curious as to what the intent of jail time was to begin with. What was the reason and logic to include that from the start? There has to be some history behind it because if we need a legislative club as opposed to regulation change....
I think it's an anachronism. It's been embedded in the act from the 19th century and it's simply been carried forward. In some ways it's surprising that it was carried forward in the 1970 revision to the Statistics Act. That was the moment when I would have thought that the world had evolved and people would no longer think that jail was an appropriate penalty. Possibly it's simply that the penalty has never or rarely been applied so people just never focused on it.
I can't satisfactorily answer the question, but I really think it's something that's been simply carried forward thoughtlessly through time.
You mentioned the word “constitutional”. Obviously there must have been some thought at that time that data was so valuable that it related to the country.
In terms of our survey now and the long-form census change, what's our response rate right now? What types of results are we getting? Because that would be the only thing. This seems like a very crude element to try to keep the numbers up for reporting, but now we've moved to a voluntary process and we're eliminating—it shouldn't be used anyway—another motivating factor for people to reply to the census. We do have some people, as you've noted, who are actually going to the courts over this. It's hard to believe that's the best use of your time, but I guess some civil libertarians, perhaps, and some others might have some objection to this.
What's our return rate? Have we looked at any analysis as to whether or not this will diminish the return rate, or are there other strategies to include that?
The census portion of the 2011 census program was mandatory, and the response rates held up very well. They were in the high nineties, around 98%, for the 2011 census population per se.
For the national household survey, which was conducted on a voluntary basis, the response rate was around 69%. Strictly speaking, the way it was designed means the response rate was 77%, but without getting into the technical details of why the two rates are different, I'll just stick with the 69%.
Going to a voluntary survey did reduce the response rates, and the reduction in response rates has consequences in terms of data quality and the ability to publish data. Nonetheless, for survey research in Canada, that's a very respectable response rate. I have great confidence in the data from the 2011 census and the national household survey.
We don't really emphasize jail time. Penalties remain under the act; people can still be fined. In the wake of the 2016 census, a similar number of cases will probably be prosecuted. Of those cases, as in the past, probably two-thirds of the people will choose to fill out the questionnaire at the moment they go before the court. The others will ultimately be found guilty, and now they will potentially be subject to community service or a fine, not jail time.
We've never really focused on jail time as an argument to convince Canadians to participate, so I'm confident there will be no operational consequences for the mandatory 2016 census of population as a result of the removal of the jail time penalty.
I'll conclude with this. I was part of the complete count. We did door-to-door at that time because it was important.
I'm a little disturbed by a 69% response rate. That's 30% of our population that's not communicating data to us to be used for science research, decisions with regard to how we spend our resources in this country, and social planning. That's a significant rate, so I'm hoping that we'll see some other improvements.
I thank Mr. Preston for bringing this bill forward and agree that jail time is inappropriate. At the same time, what's clearly inappropriate is when we move down from a higher rate to 69% on the first throw. I think this shows that the government is going in the wrong direction with regard to our census data collection. I'm hoping there are going to be proper resources for Statistics Canada to improve that voluntary number, because that data is critical for our social planning.
Thank you very much, Mr. Chair. It's great to be here with the industry committee today.
It's great to have my friend, the honourable member, here today. As a matter of fact, I appreciate the fact that he was able to solve this problem, because I've spent a number of late evenings with him and we've solved quite a few of the world's problems. It's nice to see one of the resolutions brought forward in legislation.
That being said, more out of curiosity than anything, Mr. Smith, I have a question about the fact that Statistics Canada supports this legislation. Is it the norm, on anything that impacts Statistics Canada, for the organization to take a position on legislation that comes forward; that is, whether you support it or don't support it? I'm thinking of Ted Hsu's bill about resorting back to mandatory census forms. Is it common for you guys to take a position?
The occasions on which the opportunity presents itself have been...so I can't really say there's a track record. But I take your point. It wouldn't be the norm, and perhaps I shouldn't have indicated our support.
Before becoming an MP in 2011, I spent more than two years working at the Institut de la statistique du Québec. I find statistics fascinating.
While you may find it acceptable, I agree with my colleague that a 69% household response rate falls well below the significance threshold in terms of providing critical data. Having spent more than two years at the Institut de la statistique du Québec, I saw the difference in response rates between mandatory and voluntary surveys. It was day and night. The responses provided by people who are busy and have little time to spend on the questionnaire represent vital information that has to be part of the whole in order to reduce the margin of error as much as possible.
In Quebec, some people are wondering whether the province shouldn't collect all its own statistical data, questioning whether data obtained by Statistics Canada on a voluntary basis can really be trusted. They are wondering where the response threshold should be in order to provide meaningful data.
The line of questioning has absolutely nothing to do with the legislation that's before us right now. Prior to the member's time in office, we've had countless hours of conversation about the census changes prior to the last election.
Maybe we could keep to the subject matter. We could be here all day if we decide that we're going to go into every aspect of the census, and what could or could not be there, but certainly the line of questioning has nothing to do with the question before us today.
It's true that the response rate has dropped, going from 93.5% to 69%. And the problem that gives rise to is this. If we do nothing and keep the sample size as is, we will lose statistical reliability. But we have offset that effect. Instead of having a sample size of one in five households, we established a sample size of one in three households. In the end, more people responded in 2011 than in 2006, with the mandatory long-form census. So we rectified that problem.
A lower response rate can skew the data. Some segments of society happily respond, whereas others refuse altogether. We invested a considerable number of resources and made sure we had a tremendous amount of information at our disposal to make every possible effort to correct that kind of bias. We'll never be able to fix it 100%, but our analysis shows that we've been able to largely offset that response bias. There has always been a certain measure of response bias, even in the census data.
We had all the resources necessary, but we can only do so much. There's no way to offset the bias 100%. When we identified the problem, we made information to that effect available. We were transparent with Canadians, telling them where exactly we had found problems.
Furthermore, because of the methods used to collect the data, we observed greater volatility in the data on small regions with very small populations. So, owing to quality concerns, we didn't release a portion of the data that we normally would have.
We couldn't provide Canadians with data that was, in our view, inappropriate and quite possibly misleading if used as the basis for decision-making. Nevertheless, the quality and reliability of the national household survey data are very good, overall. Although we did experience some loss and there were consequences, it's important not to blow them out of proportion, either. They weren't as significant as some claimed.
It's just that there are some blanks, if you will. Certain pieces of data, as far as certain regions and certain groups are concerned, are missing. These gaps are the reason that you weren't able to complete the work. As for the rest, I understand.
Thank you, Mr. Sullivan, for your kind comments about my piece of legislation but it is only about that, the pieces that I've talked about. There are a number of mandatory forms that are still out there, not counting the household survey, and my bill doesn't really talk about the household survey except for the release of information. The answer here is removing jail time from all of those mandatory surveys whether they're agriculture-related, small business-related, or censuses in themselves.
Maybe you didn't understand my question. I understand that would remove it from a bunch of mandatory surveys. But the reason the government gave for eliminating one of those mandatory surveys way back in 2010 was because of the jail term. Now that's gone.... That was one of the reasons, that it's coercive and people shouldn't have to go to jail for refusing to do this.
Having taken that out, can we expect that there will only be one asterisked year in Canada's history books, one dark period where we didn't have a national household survey?
I leave it to greater minds than mine to decide that. The answer here is.... In talking to constituents in the riding in that period of time I found it was about the threat. No one for the sake of not filling out a form for the Government of Canada should ever be threatened with jail time or receive jail time, so it truly is both.
I tried to make this as simple as I possibly could. I'm not saying I'm simple, the bill is. It has two things. First is to remove that threat and remove the possibility, even though it's not been used. Mr. Smith has shared that with us. It's not used, but there is still the opportunity to threaten its use so it's just to remove that. In a country as great as ours to talk about incarcerating people for something as small as not filling out a form.... Canadians were appalled by that and told us so, so let's fix that.
The other piece is simply the release of historical data, and there are many citizens and eco-social groups that need that information, so let's keep doing that too.
My final question is for Mr. Smith and it's on the release of data with consent. It may be my own ignorance of the process but is that the consent of any person named on the form or is it the consent of the person who filled out the form?
In principle, as I recall the instructions, it is supposed to be the desire of the person who is named on the form but it obviously poses problems for children. Clearly parents have opted to answer on behalf of the child in some cases, so the result, the effect, has been probably a mixture of people who have chosen not to answer for their children and others who have.
Therefore, it's a mixture of the two scenarios. In some cases it really reflects the views of the person completing the form and in other cases it reflects the opinion of the individual who is directly concerned.
In 2103 when the 2011 census is able to be able to be revealed, if a parent of a two-year-old didn't specifically authorize the release of that two-year-old's existence we won't know about that two-year-old in the census, or...?