Evidence of meeting #45 for Government Operations and Estimates in the 41st Parliament, 1st Session. (The original version is on Parliament’s site, as are the minutes.) The winning word was estimates.

A recording is available from Parliament.

On the agenda

MPs speaking

Also speaking

  • Bill Matthews  Assistant Secretary, Expenditure Management Sector, Treasury Board Secretariat
  • Sally Thornton  Executive Director, Expenditure Operations and Estimates, Expenditure Management Sector, Treasury Board Secretariat
  • Douglas Nevison  General Director, Economic and Fiscal Policy Branch, Department of Finance

3:30 p.m.


The Vice-Chair Mike Wallace

Here we go. Welcome, everyone, to meeting 45 of the Standing Committee on Government Operations and Estimates. This is a televised meeting.

Under the orders of the day today, we are going to start with Standing Order 108(3)(c)(vii), the study of the process for considering the estimates and supply.

I want to welcome back our guests. We had to chase them away at the last meeting because of votes, unfortunately, but I hear we won't have that interruption today.

From the Treasury Board, we have Bill Matthews, Sally Thornton, and Kenneth Wheat. From the Department of Finance, we have Douglas Nevison.

We'll start with the Treasury Board presentation, if that's okay. Then we'll go to Finance, and then to questions.

The floor is yours, Bill.

May 14th, 2012 / 3:30 p.m.

Bill Matthews Assistant Secretary, Expenditure Management Sector, Treasury Board Secretariat

Thank you, Mr. Chair.

We are pleased to be here today. As you mentioned, I have some colleagues with me.

We have been following this study with great interest. The witnesses you have lined up have been quite impressive in terms of their backgrounds. You will have noted, I'm sure, that you have quite a variety of opinions in terms of how you should proceed. That likely confirms what you already knew when you started this study, which is that this is complex material. It's no great surprise that you're getting a range of views.

My read of the testimony so far is that the only issue you have not had conflicting testimony on is the issue of whether or not there should be studies of tax expenditures. Not all witnesses have weighed in on that front.

Our intent today is to take you through a fairly quick presentation that highlights some of the themes you have been hearing about so far with great regularity. I don't feel it's our place to advise you on what to do as parliamentarians. I do feel it's important that you understand the consequences of some of the choices you have and what they would mean in terms of the supply process. That's my intent.

I'll take you through some of the themes we would like to comment on today. The first is the alignment of budget and estimates. The second is the cash and accrual issue, which has been around for quite some time. The third issue is around what Parliament should vote on—program activity or the current structure of capital, operating, and Gs and Cs. I'll make a couple of quick comments on the deeming rule, and then I'll turn to my colleague, Sally Thornton, to speak to maybe some small things we could do to improve how the information is actually reported.

In our deck, Mr. Chair, we have a distinction between the budget and the estimates. The issue here is not whether they should be linked. It's clear the budget and the estimates have a relationship to each other. I think what may not be well understood about the relationship is that they both feed each other.

The estimates from the previous year are certainly used by my colleagues in the Department of Finance in formulating their next budget. That budget will then influence the main estimates for the following year. It is very much a circular relationship. You will never have complete alignment between the two. I think what we're talking about is whether there is a way to strengthen the links between the two documents. You will never see a complete alignment.

I know that some members or some witnesses have expressed views that they should be totally aligned. I think that's a little simplistic. The reason I say it's simplistic is that you do have the basis of accounting to consider—is it cash, is it accrual—but you also have to understand that to get into the estimates document, an expenditure item needs to be approved by Treasury Board.

So to get perfect alignment, you need time for items to get through the Treasury Board process and to make sure the due diligence and challenge function is done. If you respect that process, you will see things in the budget that do not make it into the estimates document for potentially several years. That's an important point to consider.

The budget document is a policy document. It does set out a broad plan. Estimates are all about expenditures for the current year. They are—I know you've heard me say this before—an “up to” amount, the basis to establish a ceiling for what departments can spend.

There's also a view that supplementary estimates are a bad thing. We will never get rid of supplementary estimates. Things will happen during the year that require expenditure authority. We've had three the last few years. Is the right number two or three? That's debatable. We will never have a system where we're going to Treasury Board for approval where we don't need supplementary estimates. So I just thought I'd make those points.

What can be done to strengthen the relationship? Certainly the greater the amount of time between the budget and the estimates, the stronger the link you'll have. Some years the budget is tabled before the main estimates, and some years the budget is tabled after.

I'll give you a sense of the process behind the two documents. Expenditure authorities and documentation for expenditure approvals are largely cut off around December to make the main estimates that are tabled in March. If you were looking for alignment between the budget and the main estimates, just remember that in terms of cut-off for expenditure items from our internal executive approvals, we're looking at December.

If you were to establish a greater amount of time between the budget and the main estimates, that would allow for greater alignment, but it also might delay some of the things that actually have to get parliamentary approval. I'll speak to that later on.

You could have a process where main estimates are presented in the fall. That would give you greater alignment between the mains and the budget, absolutely. It would also mean that departments are operating on interim supply for six or seven months. I don't think parliamentarians would enjoy a system where it was six or seven months into the year before they were dealing with main estimates, so do understand that distinction. But the greater the amount of time between the two documents, the greater the potential for alignment. There are certainly some things we can do to maybe strengthen the links between the two.

I've already touched on changing the actual timing. It would allow for some additional items to be reflected, but not all, as I mentioned.

If you think about events like H1N1 or the earthquake in Haiti, where there are actually urgencies where we have expenditure approvals required, you're always going to see the need for supplementary estimates. Do remember that the budget is an accrual document and the main estimates and the estimates documents are cash basis, so you will always see a difference between the two numbers. What might be helpful is a way to do a crosswalk between those two numbers so people at least understand the differences.

I'll just touch on cash and accrual for a few minutes. It is overly simplistic to say we have a purely cash system. Our appropriations are modified cash basis, but you do need to understand the context and the other documents that are used in our system. The budget is full accrual. The financial statements of the Government of Canada, which link back to the budget, are full accrual. Estimates and appropriations we have on a modified cash basis, and there is reporting in the public accounts, volume two, on the same basis of accounting.

I'm not aware of any system in the private sector or in the public sector that doesn't need both types of information. The question for us today really is what is the right basis for appropriations? No one here is saying accrual accounting is a bad idea. I think it's actually critical for the budget. Independent accounting standard setters have said accrual accounting is absolutely critical for financial statements. There is no such pronouncement on appropriations. So I think we have some flexibility there.

Understand that if we were to change the basis of appropriations from cash to accrual, it would represent a significant change. It would be a change in legislation. It would be a change in systems. It would be a major change in how departments work. Two points impact there. Such a change would require several years to implement. This is not a change that could be made overnight. And there would also be a cost to this. Do understand that the systems and our controls are currently built around ensuring that departments do not exceed their cash appropriations. If we were to redo the system to make the control on a different basis, there would be time required and there would also be some dollars required.

I would like to highlight for you the experiences of some other countries. Some of this I've spoken on before, but some is new. The Australian government, as you know, were the leaders to go to accrual appropriations, and they have since switched back to go to a net cash basis. You have heard information on this in the past.

It's the Netherlands' experience that I would like to highlight for you. In the Netherlands' experience, they did a study of other countries, but they are the only country I know of that took a focus around what appropriations mean for parliamentarians. It's a bit of a unique decision or assessment they put on things.

There are a couple of things—complexity and transparency. Their conclusions are from a parliamentary perspective. Cash results in more transparency and less complexity, and the two of those are clearly linked. From an accountability perspective, when they looked at other countries there was definitely acknowledgement that cash works in terms of monitoring expenditures. If you are dealing with something more complex than that, which would encompass the assets of the government and its liabilities, clearly accrual is a more appropriate model. That's why we have accrual accounting for both the budget and the government's financial statements. I just thought I would highlight the work of the Netherlands in that case, because they did look at the experience of other countries, but very much with the view of parliamentarians.

We should spend a few minutes on the basis for appropriations. This really comes down to how Parliament controls votes. When you see the word “vote”, that means a department needs Parliament's authority to actually move money between votes. If you think about the current structure we have, most departments have operating and maintenance votes, they have a capital vote, and they have a G and C vote—grants and contributions. Some of our smaller departments have a program vote, which is all of those things rolled into one vote.

Currently we have 135 organizations that get appropriations. There are 191 votes for those 135 organizations. If you think about how you might change that structure, the number of votes is important. The more votes you have, the more cumbersome the system becomes in terms of letting departments actually manage. So there's a balance in there somewhere.

There has been some discussion at this committee about changing to a program basis for votes. I'd just like to maybe spend a few minutes on what that might mean.

All departments have programs and activities. Those are rolled up into a high level called strategic outcomes. If we were to go to a strategic outcome basis for votes, you would be dealing with just under 300 votes. We currently have 298 strategic outcomes. That's an increase in the number of votes—again, more complexity. That's an option. One thing you could do to sort of lessen that number is, again, take your smaller organizations and move them to one vote as a way to reduce it.

If you went to program activities as the basis for voting, it's 593. So you're dealing with a substantial change in the number of votes, which would actually become quite cumbersome.

I've heard some witnesses say, “Move to programs”. Over 2,000 programs—so if you can imagine combing through 2,000 different votes, it becomes, I would say, overly burdensome. So do keep in mind that if you're contemplating a change to a program structure vote, the number of votes becomes important, because it does become cumbersome to manage if you go over those.

Of those 135 organizations, just to give you a sense of their size, only four of them have voted expenditures over $5 billion. If you go between the $1 billion and $5 billion mark, you have 21 organizations, then eight between $500 million and $1 billion, and then 102 of less than $500 million. So if you were to actually conceive of a structure where the smaller organizations only had one vote, that is a possible way of actually implementing a program-based vote without creating so many votes that it becomes cumbersome. Do keep that in mind.

I do want to mention that there is significant information already available on program activity. The main estimates currently are based on votes. Operating and maintenance, G and C, and capital are your vote structure. That is supported by information around expenditures by program activity and type of expenditure.

So there is a way to provide information to parliamentarians about which programs the money relates to without actually changing the vote structure. That is also an option: to strengthen the information around programs.

Part III of the estimates—our RPPs and departmental performance reports—have additional information as well. So one thing the committee might want to consider is whether there is a way to strengthen the provision of program-based information underneath the current structure.

What I will say about the current structure of having votes that are operating, capital, and G and C is that it's easy to actually understand them. Everyone has the same vote structure. If you move to program-based votes, you have to understand the programs of each department to actually understand the votes. One benefit of the current system is the comparability between departments. You can look at National Defence and understand why they have a bigger capital vote than, maybe, HRSDC. So that is one thing to consider: programs are very much unique by department.

On program activity, I think I've already touched on most of these, Chair, but the level of detail in the vote is something that's important for this committee to consider. There are resource implications. Again, it's the same as if you went through a change from cash to accrual. If you move from the current structure to a program activity basis, it is complex and would take some time—that's not to say it's not doable—and there are some policy issues that we would have to get our mind around.

Specifically, the one I will highlight today is that if you went to a program-based vote, there would be expenditures that would have to be allocated across different programs. You're likely dealing with some sort of formula there. It's not a bad thing, but it just means that it wouldn't have the exact nature, whereas in the current structure you have capital and operating, and people know the difference. You're into a case with a program-based structure where certain people may work part time on one program and part time on another, and we would have to allocate their time.

I don't plan on saying much on deeming, except that I do want to mention that the nice thing about the deeming rule is that it forces a study to come to an end—or no study at all, but it forces an end to the process. The longer that process goes on, the implication is that the longer departments have to operate on interim supply. So just be aware of that constraint.

Currently, departments operate on interim supply for April, May, and June. For most departments, that's three-twelfths. If you were to extend beyond that, you're into the summer, when parliamentarians are typically not here, so you could be extending into the fall. So do understand that if you are considering changing the deeming rule, there is a direct link to how long departments would operate under interim supply if that gets changed.

I think I will now turn to my colleague, Sally Thornton, who will talk briefly about what improvements we might make on the reporting of information to parliamentarians.

3:45 p.m.

Sally Thornton Executive Director, Expenditure Operations and Estimates, Expenditure Management Sector, Treasury Board Secretariat

Thank you.

One of the common themes that has come up in the various meetings has been that of increasing parliamentarians' understanding, but also better connecting the dots. I'd like to talk about some smaller things that could be done that might help you in that area.

First of all, in terms of increasing understanding, we have had in camera sessions with parliamentarians before your consideration of estimates, just to walk you through documents and understanding.... We are happy to continue. That might be something you'd like to take up on a regular basis.

I'd really like to stress that if you really want to follow the dollar, give questions in advance to departments. It helps a great deal to know what the question is and to have the opportunity to respond before sitting here.

Connecting the dots, though, is a frequent issue, and there is a lot that is out there. Mr. Matthews talked about a number of the pieces of information out there. It is possible, with direction, to have it brought together, whether it's us or another organization. There are some very specific ones, though, that I'd like to talk to you about, and some very modest changes, which could be either in material or online—we can take advantage of some of the open data capabilities as well.

First of all, within the document, within a main estimate or within a supplementary estimate, it's quite straightforward to do a high-level analysis of changes. It wouldn't tell you what the issues are, but it could focus your attention very quickly on those areas where you might wish to delve further. Specifically, you'll notice that in the past we have recently introduced major changes. You'll see that in our supplementary estimates. We have a section that highlights the big changes in each supplementary estimate.

We also have a section on horizontal items, where we talk about items that have implications for more than one organization. Also, as of last year, we've actually started tracking those horizontal items throughout a full year, so you see them in supplementary estimates (A), (B), and (C).

Another area that we could do, and that in fact you'll see in supplementary estimates (A), is our top 10 changes to votes—so basically just the 10 biggest vote changes in supplementary estimates (A). They don't answer the question as to why the change, but they quickly focus your attention as to “there is a change here and it's somewhat significant”, and then delve or not....

Among other things we can do, though, is improving the searchability of the document. One of the issues is about who the witness is that you call when you want to talk about something in particular. Also a common issue is this one: where on earth do I find this organization? Something as simple as going to alphabetical order for organizations would help that. I appreciate that we would have to provide a portfolio map so that you could also understand the ministerial accountabilities, but that's fairly straightforward, it's easily adapted by systems, and it's something that we would probably like to promote anyway.

Then there are things we can do between documents. So far I've been talking about what we can do within a supplementary estimate or a main estimate or between the estimates in one cycle. But there are also crosswalks between documents. You've talked a lot about actuals in public accounts; that could be presented. If you're in year X, we don't have X-minus-one actuals yet, but we have X-minus-two actuals, and to the extent that sort of thing is useful, that can be presented. We could also go forward a few years and do X-plus-one and X-plus-two, so you could actually see a five-year profile of actuals...and forward looking.

Mr. Matthews talked about the possibility of a crosswalk of some sort between the budget and the estimates. Obviously the documents are different, and you don't have a full reconciliation, but there could be a better understanding of the linkage.

The third area that we could work on is perhaps an area where we tighten up the linkages between main estimates and the RPPs and departmental performance reports.

Those are some very simple things that could be done to help connect the dots and improve the information within the current structure.

3:45 p.m.

Assistant Secretary, Expenditure Management Sector, Treasury Board Secretariat

Bill Matthews

To conclude, I think there are a couple of points I'd like to highlight.

I've already mentioned this, but I will again. Some of these changes that are being contemplated are of significance, so do keep that in mind when you're forming your recommendations.

The other one I would like to touch on is the reporting burden. When I speak to departmental chief financial officers and their staff, one of the things that frequently gets raised is the reporting burden they face in terms of producing information. If they are producing or we are having them produce information that is not useful to this committee, please tell us, as part of your reports. We are happy to look at stopping some reporting as well.

The one I would highlight in particular is that departments have been producing future-oriented financial statements or forecasted financial statements on an accrual basis for about two years now. They do one for the statement of revenues and expenses, and they also do one for the balance sheets.

I have yet to see a question asked on any of those documents. They are produced online. They are an awful lot of work for departments. It does improve the link between accrual and cash.

I would highlight in particular the balance sheet. In my opinion, it is of no utility whatsoever to forecast a balance sheet on an accrual basis.

So if, as part of your study, there are things that you do come across that you're finding are not useful, please let us know. If it's only being produced for Parliament, we'll look at stopping that production.

With that, we're happy to take your questions.

Thank you, Chair.

3:50 p.m.


The Vice-Chair Mike Wallace

Is there any overview from the finance department?

3:50 p.m.

Douglas Nevison General Director, Economic and Fiscal Policy Branch, Department of Finance

Thank you, Mr. Chair.

I think Bill basically covered the main points that we would raise.

3:50 p.m.


The Vice-Chair Mike Wallace

Thank you. We'll go to questioners now.

Just before we get started, we're going to have a discussion this Wednesday coming about what we want in the report. I know you prefaced at the beginning that you don't want to tell us what to do, but if you do have suggestions based on the report that you get, including which reports you're not sure are useful.... Because to be frank with you, Bill, I haven't even looked at one of those reports in two years, and I'm one of the ones who tends to look at that kind of stuff. So it would be interesting to see.

Our first questioner is from the NDP, Monsieur Blanchette.

3:50 p.m.


Denis Blanchette Louis-Hébert, QC

Thank you, Mr. Chair.

My thanks to our guests for joining us today.

As I listened to you, I got the impression that we are going to run into difficulty. The picture you paint seems difficult to change. The problem is that we need to change it. The material produced at the moment does not meet our needs, by which I mean that members of Parliament find it increasingly difficult to follow what is going on. But that is one of the things that the documents are supposed to do.

How is it that the estimates and the budget are so difficult to match in a short time, but it would be easier to do with more time? Is it because corners are cut? Why is it not possible to produce both in a reasonably short time?

3:50 p.m.

Assistant Secretary, Expenditure Management Sector, Treasury Board Secretariat

Bill Matthews

Thank you.

There are a couple of things. The one I would highlight is the role of the Treasury Board itself, which actually has responsibility for expenditure approvals. If you think about a new initiative that gets included in the budget, that's done at a certain level. Before that spending can actually be achieved by a department, it needs to actually go to a cabinet committee, Treasury Board, and get approved with details on the expenditure plan.

That kind of work would include setting forth—if you're thinking about a new program—the performance metrics, the evaluation plan, and some detail on how they might spend the money. There is a challenge function there.

If it's a brand new program it can often be a year or two from the time it's in the budget before it's actually ready for money to be spent. What you actually see is a timing difference between the budget and the estimates.

3:50 p.m.


Denis Blanchette Louis-Hébert, QC

I understand that some programs are designed very quickly, on the back of an envelope, as it were, that officials are not ready and that, after an announcement, they have to hustle to make up for lost time. I understand that. But for most routine expenses, the vast majority of government expenses, how is it that the two cannot be reconciled quickly? Everything is known; only the amounts may change, depending on what the government of the day wants to do.

3:50 p.m.

Assistant Secretary, Expenditure Management Sector, Treasury Board Secretariat

Bill Matthews

No, I would say that's an area where there could be some improvements made in terms of building a crosswalk between the main estimates and the budget, because there is clearly a link between the numbers. There is no such crosswalk at the moment to actually show that the numbers in the budget are reconcilable with what's in the main estimates.

If that's something that the committee feels would be helpful, we would be happy to do it. The key thing for your consideration today is that the budget is a full accrual-based set of numbers; the main estimates are cash-based. So you will never see a complete match between the two, but you can build a crosswalk to show the links, absolutely.

3:50 p.m.


Denis Blanchette Louis-Hébert, QC

At the beginning, you said that you wanted to warn us about the consequences of the choices we might make. If we did manage to bridge the gap between the two and to produce the estimates and the budget sequentially and in short order, what would those consequences be?

3:55 p.m.

Assistant Secretary, Expenditure Management Sector, Treasury Board Secretariat

Bill Matthews

“Considerations” is probably a better word than “consequences”, but I did say “consequences”.

The consequences are around the time—number one—and the cost attached to making some of these changes. Just to make sure, if you were to move to a new vote structure, for instance, there would be time and effort and cost required to make that change. As long as that's understood, that's okay. But I just don't want to leave the impression that changing a vote structure is something that can be done overnight. So that was really the intent there.

I would make the same comment on cash versus accrual. I think you've heard a lot of evidence about one versus the other. Generally speaking, cash is viewed as being more transparent for parliamentarians, and more easily understood. Accrual is definitely viewed as being the better way to build a budget and do the financial statements.

I'm quite comfortable having them on a different basis of accounting, but I do take the point that the two documents should be reconcilable.

3:55 p.m.


Denis Blanchette Louis-Hébert, QC

What would be the cost of building that crosswalk so that we could compare the budget estimates from year to year, so that we know where we are going and so that parliamentarians are given figures that they can understand? That is where the problem lies. Members of Parliament are finding it increasingly difficult to understand the figures they are being given.

3:55 p.m.

Assistant Secretary, Expenditure Management Sector, Treasury Board Secretariat

Bill Matthews

There the considerations are more around the time between the two documents. The main estimates need to be tabled by a certain date. There is flexibility in terms of when the budget can be produced. It can be any time. As long as there is flexibility on when and where this crosswalk could be tabled—it could be part of the estimates document or it could be part of a budget document—I don't think there's that much effort involved to actually do that.

That's something we should, if it's the will of the committee, take under consideration.

I'm not sure if Finance wants to add to that.