Government Operations Committee on May 28th, 2013

[Mike Wallace, Conservative, Burlington, ON]

Oh, the PA is there, too.

[Sally Thornton]

But recall that the reports on plans and priorities go into even further levels of granularity, and we didn't bring the lower levels into the main estimates.

[Mike Wallace, Conservative, Burlington, ON]

Those are my questions, Mr. Chair.

[The Chair, NDP, Pat Martin]

You were very concise, Mr. Wallace.

We'll move to the next questioner.

For the Liberals, we have John McCallum.

[John McCallum, Liberal, Markham—Unionville, ON]

Thank you.

It seems to me there's a strong connection between the proposal to go by programs and the $3.1 billion of missing money, because I think the secretariat said that one of the possible reasons for it was that money was transferred to other programs without being tracked properly. It seems to me that under a program-based model, any change from one program to another would have been known and would have helped prevent this outcome with the $3.1 billion, because there would have been a paper trail all the way to Parliament.

If we don't want problems of this nature to reoccur in the future, is this not an additional argument to go via the program-based route?

[Bill Matthews]

It's a question of two things. It's a question of information provided to Parliament versus the vote structure. You now have a world where planned spending by program in 2013-14 is available and departments must explain in a narrative any changes in plans versus actual. That exists regardless of what you base your vote structure on. So you have that in play.

If the question is whether a program-based vote would have solved this problem, the answer is no, it would not have, because in that particular case you are taking some existing programs and adding money to them.

The best example I can give you is if you had stand-alone programs for that, it would have stood out, but where you are adding money to existing programs, that money wouldn't necessarily be tracked. Moving to a program-based model would not have solved that problem. What you need is better tracking of horizontal spending.

If the question is whether the program-based model would have prevented the $3.1 billion, the answer is no, it wouldn't have.

[John McCallum, Liberal, Markham—Unionville, ON]

I thought that under the model of program-based activity, if you change from one program to another, it has to be tracked properly and reported to Parliament.

[Bill Matthews]

It would have to be tracked properly and reported to Parliament and approved—that is correct—but then you're in a situation whereby, if you have an existing program.... Think about the economic action plans. You had existing programs which you would augment with additional funding. Basically, you're taking a glass of water and filling it up. If it was half-full before, where is the new water?

If you have a stand-alone program that is new, it's very easy to track. If you're augmenting existing programs, it's a little more challenging to track where the new money goes. That's what you have.

[John McCallum, Liberal, Markham—Unionville, ON]

It seems to me that the choice of granularity has to do in part with the degree to which Parliament wants to control spending.

I think we would prefer the intermediate case of program activity. It seems to me in that case it would be much more difficult to transfer money from, for example, border infrastructure to G-8 funds, because you would need the approval of Parliament if the model were based on program activities. That would increase the degree of parliamentary control. Is that correct?

[Bill Matthews]

If you go down one level, certainly the degree of parliamentary control increases.

I believe I mentioned earlier that if you look at strategic outcome, you're looking at roughly 290 strategic outcomes, and if you go one level down, you're at about 550. You're quite right that from a parliamentary control perspective it puts greater control at that level.

From a departmental perspective, that would cause a fundamental shift in how they actually do their planning and spending. When departments got back to us about what the cost to implement such a structure might be, they said the cost would go up by about half.

There were some concerns about what it would do from a management perspective. To actually control at that next level down would cause them some issues. That's not to say it's not the right thing to do, but understand there is a significant impact on departmental operations the greater the degree of control by Parliament.

That's the trade-off. Departments right now plan at a very high level where their controls are. The further down you go, the more process you add at the departmental level.

[John McCallum, Liberal, Markham—Unionville, ON]

I think the whole premise of our study was that we wanted more parliamentary control. If the cost of that is a little bit of inconvenience in the managing of the department level, so be it.

[Bill Matthews]

I think that's the question. I distinguish between parliamentary control and information for parliament. We clearly have improved what Parliament gets for information in terms of detail. We are talking about a potential change to vote structure. It makes for a good discussion whether the improvements in the information alone are enough, or whether you still want a more detailed level of control.

[John McCallum, Liberal, Markham—Unionville, ON]

I think we want a situation where transfers cannot be made between two unrelated programs without the knowledge or consent of Parliament. Going to program activity would achieve that by and large, and that's the foundation of our recommendation.

[Bill Matthews]

You used two words, “knowledge” and “consent”. They're two different things. You said “knowledge and”, and I think that's the question. We're saying that with this model the knowledge is there. Whether you need Parliament's consent to move money between programs or program activities or strategic outcomes is the question.

[John McCallum, Liberal, Markham—Unionville, ON]

Okay, thank you.

[The Chair, NDP, Pat Martin]

You're out of time, John, but that was an excellent exchange.

Next, for the Conservatives, we have Jacques Gourde.

[Jacques Gourde, Conservative, Lotbinière—Chutes-de-la-Chaudière, QC]

Thank you, Mr. Chair.

I thank the witnesses for being here with us this morning.

I would also like to thank you for following up on your commitment of October 30, which was to do a study on a vote structure using accrual accounting.

Can you tell us whether you will also take into account the effectiveness of parliamentary oversight in the context of preparing the basic model for programs and activities?

[Bill Matthews]

That's the key question. If you go to too low a level, or at the program level, for instance, where you're dealing with close to 2,000 programs, supplementary estimates would become largely a vote transfer exercise. To move money between programs, you would need parliamentary control and you would actually be increasing greatly the number of transfers Parliament would have to approve.

Whether you can achieve the same level of information for Parliament by providing spending plans on programs while leaving a higher level of control is really a question of what the right balance is. We have factored in our views in terms of providing a model based on strategic outcomes for discussion. I think some of the key points have been raised here today about the desired level of control by Parliament.

What was clear to us was that Parliament needed more information on program spending. We think you now have it with the latest changes, the supplementary estimates, and what will come with the database. We think that will complete that story. What's left is a discussion about the level of control.

[Jacques Gourde, Conservative, Lotbinière—Chutes-de-la-Chaudière, QC]

Can you tell us what legislative changes would be required to do that?

[Bill Matthews]

As Sally mentioned earlier, if we left the model at the strategic outcome, we're not aware of any legislative changes that would need to be made. If we went down further levels, there potentially would be some changes needed.

I'll turn to Sally to see if there is an example she can provide.

[Sally Thornton]

In terms of changing legislation if we went to a lower level of granularity, it may be that we would have to modify some departmental legislation where they have specific authorities.

The other side, though, is there are a number of government-wide policies having to do with expenditure management that would have to be revisited. Right now we have some policies based on having votes on operating, votes on capital, and we'd want to modify those and see how to implement them. They're in place for a good reason: to help good management and to promote desired behaviours. We'd have to reassess and redo those in a different context. We would still need to maintain some way of managing that information to promote those policies, but they would have to be amended.

At the strategic outcome level, there is probably no need to change departmental legislation. At the program level it's possible that certain departmental legislation would need to be changed, but our policies would have to be adjusted regardless.

[Jacques Gourde, Conservative, Lotbinière—Chutes-de-la-Chaudière, QC]

I would like to know when Parliament will be able to see the changes made to the vote structure and whether there will be a pilot projet in that regard.

[Bill Matthews]

With regard to the changes that were proposed in the government response, because there were more than just recommendations 1 and 2, you will have seen in the latest version of the RPPs and in upcoming versions of supplementary estimates....

In terms of whether there is a pilot project, no, there is not. When we're dealing with parliamentary control, we thought it would be important to actually keep all departments on the same basis.

If there is a decision taken to change the structure, we're suggesting that it would be roughly a five-year process. It would be a number of years before parliamentarians would actually see something. There is no intent at this stage to pilot a few departments. It would be the whole government moving to the new model at the same time.

[Jacques Gourde, Conservative, Lotbinière—Chutes-de-la-Chaudière, QC]

Will changes be made to supplementary estimates?

[Bill Matthews]

There have been changes proposed to supplementary estimates in the government response. There are two key ones that I'll highlight.

One is a better format of information, where members could see what each department had in previous estimates, what they actually spent, and what their current year authorities are. The other change I should highlight is that in future supplementary estimates, to help parliamentarians better see the link between the budget and something in the supplementary estimates, when the government commits for the first time that new spending will appear in supplementary estimates or in estimates, it would reference back to the appropriate budget. Members would actually see the first time there's a new program that it came from budget 2013, for instance, so that would help at least to improve the link.