Public Accounts Committee on Oct. 29th, 2018

[The Chair, Conservative, Kevin Sorenson]

All right.

Mr. Ferguson, are you confident in some of the processes they've talked about today and that they're trying to implement?

[Michael Ferguson]

I think what you've just heard is that it is extremely difficult to get accurate information at the micro level about jobs, where jobs are available and what types of skills are needed.

They have consulted with the statistical experts in the country in Statistics Canada about how they would go about gathering that type of information. I think at this point what I am hearing is that they're saying they have a way they can try to move forward on this issue, but they are not yet sure it is going to succeed. If there are ways for them to try to do some of that work without trying to do it across the whole country, to try to understand whether they would be able to succeed at it or not, then it would be prudent to take some steps into it.

Maybe it's time for the department to say they know what they would have to do under the Statistics Canada approach, but I think that before they go too far down that road, they should go back to what they are trying to do here. The fundamental piece of their recommendation was to know what types of training these indigenous people need, what types of jobs there are, what skills are in demand, where those jobs are, what skill sets people have, and how we would train people. It's going back to those fundamentals.

Is there any other way to find out what jobs are available and what the skill gaps are? This may be the only approach to doing it. If so, they need to take some initial steps to see whether it can be successful. If those initial steps seem to indicate success, then carry on. If not, take a step back and ask if there's another way they can look at getting this data.

[The Chair, Conservative, Kevin Sorenson]

Thank you both.

Mr. Kelly, I stole some of your time, but go ahead.

[Pat Kelly]

That's all right. You've generously given it back to me.

[The Chair, Conservative, Kevin Sorenson]

There's been a little extra time for everyone here.

[Pat Kelly]

I want to return to the accountability aspect. This is a committee for delivering accountability to Canadians.

I appreciate some of the remarks we've heard. There are many other committees of Parliament that can examine different ways we can solve policy issues. There are committees on government operations, human resources, indigenous affairs. This is about accountability for what has happened.

I didn't really get much of an answer in my round. Mr. Christopherson very bluntly and succinctly pointed out that there is no answer to the question of the “why”. I think we probably would mostly agree that every government for the last 40 or 50 years has, in good faith and earnestly, examined the systemic barriers to employment and to better conditions, both on reserve and for indigenous Canadians who live in cities, yet, as has been pointed out, we struggle to answer basic questions about what happened to the money we spent on programs.

I'm going to ask Mr. Flack for another crack at the answer to the “why”, but before I do I want to also say, Mr. Chair, that I will add my voice to support Mr. Arya's point that it would seem we really do need to speak to Mr. Flack's predecessor.

[The Chair, Conservative, Kevin Sorenson]

Go ahead, Mr. Flack.

[Graham Flack]

Let me unbundle the “why” and give some different categories.

On the question I just answered on the unavailability of labour market data at the micro, granular level, I think the “why” is that no such instruments existed. That's why we didn't collect it. That's why we had embarked, starting in 2016, on a program to try to collect it, but that program is not yet advanced enough to have that data.

In sum, the answer to the “why” is that there were technical limitations on what we could do. We are working through those, but as the Auditor General indicated, we are doing that on a pilot basis to test it—

[Pat Kelly]

How could that have really come to be a surprise at this point, when we've been struggling for decades with the policy outcomes that aren't being met? How did it take until 2016 to have it come to light that data isn't collected, or that there are difficulties, indeed, such that you are pouring cold water on expectation going ahead and that there are maybe intractable problems of collecting data?

[Graham Flack]

It's not a surprise, but the centrality of that data in allowing programs to operate on the ground—because that labour market information is critical to how they operate—is the reason we're pushing further on it.

There are other areas where we had collected data, for example, as the Auditor General indicated, where we didn't measure whether it was long-term employment and we just measured whether there were forms of employment. I would characterize that as deficient collection, in that we should have worked to collect more granular data about the length of employment.

Then there's stuff that's in between. The stuff that's in between is, as in my example, this longitudinal tracking of an individual and how the individual did. It's perfectly rational to ask if we can know, for individual A, after five years, what the effect of the intervention was and what the outcomes were. If those individuals are moving outside the zone of the community, that is a very difficult thing to track, which is why I think the department focused on these short-term measures. It's because they were trackable and you could deal with them with the account holder.

There, as I indicated, we've tried to develop a new methodology that maintains the anonymity of the data and will allow us to track that by category.

[Pat Kelly]

Here's a quick question just for the record.

You said you're three weeks into it, as the current deputy. Where did you come from? Are you from the department?

[Graham Flack]

I've been at the Department of Finance, the Privy Council Office, National Resources Canada, and most recently the Department of Canadian Heritage, where we did extensive programming regarding indigenous languages, which is I think the closest proximity to this one.

[The Chair, Conservative, Kevin Sorenson]

Thank you, Mr. Kelly.

We will now move to Madam Mendès.

[Alexandra Mendes, Liberal, Brossard—Saint-Lambert, QC]

Thank you, Mr. Chair, and I will cede the rest of my time to Mr. Sarai.

Thank you all for being here.

Actually, I don't really have a question. I'm going to make a statement on Mr. Ferguson's message—not these last reports, but the ones from fall 2017, and on the fact that we tend, as governments, to focus a lot on process and not necessarily on outcomes in how we deliver for people, for Canadians.

I think that when we ask ourselves the “why” of these failures, we have to look very much at the relationship we had with first nations, Métis and Inuit, and how we saw ourselves as the service providers and almost as the supervisors of how it was being done. It's reasonably recently that we've changed this approach and changed the relationship we have with indigenous peoples, and I think that's part of the “why” of the failure. For decades we saw this very much as an “I pay, you deliver” kind of service, and now it's a partnership. It will take some time to correct the way this is done, and I do appreciate the efforts you're making to develop these programs with indigenous peoples by getting them to tell you what they need and what they actually will be able to work with. Perhaps that will be the change that we need to see for the future.

I just wanted to make that statement, and I would like to pass the remainder of my time to Mr. Sarai.

[The Chair, Conservative, Kevin Sorenson]

Thank you, Madam Mendès.

Now we have Mr. Sarai.

[Randeep Sarai, Surrey Centre, Lib.]

Thank you.

My first question is to you, Mr. Flack.

You were saying that now you're going to use SIN numbers as data but be anonymous in doing that. Was that not available before? That seems like an old way to track where somebody works and to see the success of it. I'm a rookie at this, but there are people who have been here for a while, and this is an epidemic problem. Would that not be a metric that was used before?

[Graham Flack]

The trick on SIN, the real trick on outcomes, is not just measuring employment but income levels as well, and for that we have to link it to the CRA database.

Now we're not able, obviously, to do that in a direct way, by tracking a specific individual, so the work that had to be done was on linking the information we have on EI and SIN and the CRA database on an anonymized basis that met all the privacy standards. This was not just to give us a global result that would allow us to say of the people in the program that their average income was x; the tricky part was working through how the model could kick out for us what the results were after two years, three years, four years, on the income for individuals who had these types of interventions—literacy training, etc.

That was extremely tricky to work through, and that's what has been, in fact, worked through. It isn't just an awareness of employment. That's the other thing. For the self-employed, knowing the SIN number doesn't tell you if they're actually with an employer who's issuing slips to them at work. If they're self-employed it's different. Because of the diverse clientele, we think this linked database is the best way to get at the data that the Auditor General has rightly said we need in order to track not just how we're doing now but how those individuals do over time, but it is tricky.

[Randeep Sarai]

To the Auditor General, how is this program different from other accountability mechanisms, say, in immigration, with the settlement agencies that do employment training? There is also ESDC, which does it for other non-indigenous actors. Are they tracking in a better way than this program is tracked? Are there best practices they can utilize for them, or is this a problem with all employment training programs?

[Michael Ferguson]

I can't tell you. I'm not quite sure of the last time we did an audit on this type of program, which says to me that maybe it's time for us to go back and look at some of the others. That would be something to come. I don't have the answer right now.

[Randeep Sarai]

My last question is this: Is this just an issue of denial, of not wanting to state that the effectiveness percentage of those programs is low, and therefore program directors are trying to find a number that shows a higher success rate, or is it specifically a problem with these types of programs?

I would be more satisfied knowing that an employment program had a 5% or a 10% success rate and knowing it's effectively helping 5% or 10% and is better than the last one, which had a 3% success rate.

Is this just a cultural problem of nobody wanting to admit that it's not a successful program?

[Michael Ferguson]

I wouldn't put it that way.

The way I would put it is that it's a common problem that we have seen in many government programs. Sometimes the way performance is measured is based on the information that is easy to capture and easy to report, rather than putting together a performance measurement regime that really looks at whether the program has achieved its intended outcome.

We don't make up the objective of the program; we look at what the departments say the objective of the program is. They said that the objective of this program was “sustainable and meaningful employment”. If the objective of this program was simply to know how many people manage to get at least part-time employment, then our report would be very different. However, because they said it was about sustainable and meaningful employment, then we wanted to see a performance measurement strategy that would let us know whether it was being successful or not. What we ended up seeing was that there were some things that were being tracked, but, again, they didn't differentiate between somebody getting full-time employment versus part-time employment, so is that really sustainable?

I don't think this is an issue that is unique to this program or this department. I think it's a problem that I've talked about many times in front of this committee, in terms of the way that the different programs measure their performance.

[The Chair, Conservative, Kevin Sorenson]

Thank you, Mr. Sarai.

We're coming up close to 5:30. However, because we missed so much earlier and we have bells coming again in another half hour, would it be all right with the committee and with our guests to extend this a little longer for some of the people who haven't had the opportunity to ask questions? Are we good with that?

The consensus is yes.

All right. Now I have Mr. Kelly, and then we'll go back to Mr. Arya.

[Pat Kelly]

Thank you.

Mr. Chair, perhaps in the interest of time, I won't take a full five-minute round.

Mr. Ferguson, being able to accurately judge the success of a program is extremely important. It was mentioned in part of the testimony we heard that there are barriers to employment for indigenous Canadians that are far beyond skills training. At the individual level, people need to have skills to obtain and keep a good full-time job.

I've travelled to the north with the finance committee, where we heard about other significant barriers: absence of roads, absence of all kinds of other infrastructure that impedes economic development in remote communities.

There may be some critics who would say that if programs can't be judged to be successful, then the funds ought to be allocated to other things, such as roads, water treatment systems or airports in remote areas.

What can we really tell Canadians about success that has been achieved in this program, so that Canadians might not demand of their parliamentarians that the money be spent in other ways?

[Michael Ferguson]

I think that's part of the fundamental question we were looking at when we decided to undertake this audit.

Again, as you said, among the indigenous population there are higher unemployment rates and lower participation in the workforce. There are all of those indications that indigenous people need help getting into the workforce. That's something, as I think the department has rightly said today, that they need to work on, and not on their own but with the indigenous groups and communities as well.

Fundamentally, if there is a program that is intended to help indigenous people get sustainable and meaningful employment—and, at this time, the program was spending about $300 million a year, and I think we heard today that there will be another $100 million a year, taking it up to $400 million a year—we would hope that we would finally see a change in some of the outcomes.

I think the deputy minister did mention some very specific performance measurements that they are going to track now. I think one was reducing the skills gap by a certain percentage, by some very specific measurement.

I think these are the types of measurements that would give us all a better sense of whether this money is achieving what it was intended to achieve, which is sustainable and meaningful employment for these populations. That's what we're looking for.

I can't answer the question about whether there is another approach. What I can say is that we need to have some way of knowing whether the current approach is working or not.

[The Chair, Conservative, Kevin Sorenson]

Thank you, Mr. Kelly.

We'll now go to Mr. Arya.