Evidence of meeting #25 for Government Operations and Estimates in the 39th Parliament, 1st Session. (The original version is on Parliament’s site, as are the minutes.) The winning word was model.
A recording is available from Parliament.
November 7th, 2006 / 11:05 a.m.
Liberal
The Chair Liberal Diane Marleau
I will call the meeting to order.
We're happy to have you back--both sides, all sides--in this wonderfully exciting study that my committee thinks is just terrific. I must say that I'm interested, and I think it's been good so far.
Specifically to Monsieur St-Jean and Monsieur Moloney, we're anxious to hear the answers to our questions.
By the way, we're pleased to have had a copy of the PricewaterhouseCoopers study. That was very good, thank you.
Ms. Fraser, do you want to go first?
Sheila Fraser Auditor General of Canada, Office of the Auditor General of Canada
Madam Chair, we hadn't planned to give an opening statement. We would just refer back to the statement we made when we appeared on September 26.
We are very pleased that the discussion is continuing.
Charles-Antoine St-Jean Comptroller General of Canada, Treasury Board of Canada Secretariat
If I may, Madam Chair, perhaps we could dive right into the subject of the meeting. My colleague will go through the presentation, and after that we can have a discussion.
David.
Liberal
David Moloney Senior Assistant Secretary, Expenditure Management Sector, Treasury Board of Canada Secretariat
Thank you very much.
Good morning. Thank you to the committee for inviting us back again.
We realize there are questions left from our last appearance. We expect there will be additional questions, based on what you heard from some of the witnesses, particularly the representatives from B.C. and Ontario. I certainly enjoyed reading the transcript of that meeting.
I'd like to take you through a brief presentation, which will hopefully move the yardstick forward in terms of our discussion.
Slide 2 outlines where we've been with the views of the financial plan. As I think I advertised last time, then we want to move to the questions around parliamentary control and what it specifically means.
I'm going to beg your indulgence and ask you to work through an example the Comptroller General put on the table last time. We used that example to work through some of the specific approaches and what, in practical terms, they would mean for Parliament, what you would vote, the means of control, and pros and cons. Then, we'll talk very briefly about next steps.
On slide 3, we have a graphical presentation. The committee I think is well briefed on this now, so I don't want to dwell on this. As we've discussed, the federal budget and the government's financial statements are prepared and shown on a full accrual basis. Assets are fully valued on an accrual basis, from heritage assets to military assets. So that issue is addressed.
The plans and actuals are addressed. I think other of your witnesses have stressed the importance of comparing apples to apples.
It is important to clarify that it is departments that make the accrual adjustments and do the accrual accounting, with the exception of certain items--for example, the public service pension costs and aboriginal claims. Most accounting is done in the departments. It is the central agencies that are collecting that information to prepare the government's summary financial information.
One way that Parliament will see that in fact is in the departmental performance reports, which are scheduled to be tabled in the House in a matter of days, I believe. You will see accrual-based financial statements in each of those departmental performance reports. These have been prepared in the departments, as I said.
Slide 4 takes that federal budget and shows graphically what the government plans to do. What Parliament votes maps into the main estimates. Planned expenses are now adjusted into planned expenditures, in accounting lingo, by department--those that are authorized under existing legislation or those that need to be authorized under annual appropriations through an appropriation act. It is the form of the appropriation act that we need to drill into here.
We'll move on to slide 5. Parliament votes on the overall fiscal plan, as embodied in the budget. Parliamentary control becomes the critical element that drives the design of the model for detailed departmental budgeting and appropriations. So the key questions are--and these have been in front of the committee for a while now--what control Parliament wants to exercise over the specific expenditures or revenues within the fiscal plan set by the budget, and what information is needed to support the controls.
The appropriations are Parliament's tool for exerting financial control over government. The standard we apply in terms of appropriations is what will satisfy parliamentarians. So among the possibilities, whether it's cash or accruals--and we're not talking about this directly here today--or annual or multi-year, gross versus net, ministry, department, or program activity level within departments, it is what Parliament will find useful and necessary.
As the slide shows, this obviously then influences the form and coverage of the appropriations, the information we put in it, and as we have discussed, how departments will actually conduct their financial management.
We have been happy to be here a number of times. With respect to the committee's discussions, I think there is pretty broad agreement on the goal we want to get to in terms of departments' abilities to make the right financial decisions to properly manage their resources. If that is the goal we all agree to, and I think it is important to underline that, how can we best increase the amount of accrual information to support that further step of improving departmental financial management?
Slide 6 then gets us to the example. This was in the previous deck that the Comptroller General presented.
I'm not an accountant, but I and some of my colleagues in TBS found this to be a useful example for us to be able to walk through. What would it mean?
As a reminder, this is an example of a department or an agency that is spending $100 million to produce a piece of equipment. It will take over two years to put that in place, and so in the example, $25 million is spent in the first year, $75 million is spent in the second year. The equipment is put into use in three years, and will then last for ten years. So the acquisition cost, clearly summarized in the little grey bubble on the left, totals $100 million over two years.
On a cash basis, these expenditures, as they are recorded in the fiscal accounts and as a department needs to budget for them today, as Parliament votes for them today, total $100 million over those two years.
On the right side in the accrual column, you see that an asset is created, and then that asset is expensed. It is used up at a pace of $10 million a year, from years three through 13, again for a total of $100 million. By construction, the expenditure and the expense totals are the same, but the annual amounts differ, and they occur on the government's books in different years.
Immediately one of the things we see from this is, moving from an expenditure or cash base, or near cash base, to accrual, we could potentially see different amounts being charged to the fiscal framework. As it so happens, as we look across the government today, in the last two years, the pattern of capital assets that exists, the pattern of capital spending that's taking place roughly balances out so that in the last fiscal year, 2005-06, capital acquisitions totalled $4 billion. Amortization, coincidentally, was $3.9 billion. The year before the difference was slightly bigger, but not that material. Acquisitions in 2004-05 were $4.6 billion, amortization was $3.7 billion. So the aggregate fiscal impact on the fiscal framework is not overwhelming at this point from moving from one side or the other.
Slide 7 is a bit complicated. We'll walk through it.
Looking again at tangible capital assets, and as we've discussed before—we'll come back a little later—tangible capital assets are only one of the areas where accruals and cash differ, but certainly for non-accountants it's maybe the easiest to understand. It's also one of the most important ones, just in terms of dollars and cents.
So there are four basic models, two of which you heard about from your B.C. and Ontario interlocutors.
Number 1, you see the box on the left with a little arrow down to the left bubble, the current model, vote acquisitions. That's the status quo. Model or option number 2, vote amortization expense, so that box is showing with an arrow down to the bubbles with the tens. This is a logical construction, it's really a heuristic, as I'll come to.
Options 3 and 4 are real live options that governments are using today, and you heard about two of them. Model 3 is to vote both the acquisition and the amortization. That's why there is a line down to the $25 million and $75 million. There is also a line down to the bubble with the tens. Model 4 is to vote the acquisition again, but to have amortization treated as an information or statutory basis only.
Each of these models has certain benefits. Each of these models has certain drawbacks. As the slide notes, we need collectively to come to a balance that gives optimal departmental asset management, transparency for Parliament and Canadians, and parliamentary control and oversight. All of this we want to do in as efficient a manner as possible. So I'm going to walk through each of these four models.
Model 1 is the status quo. We've been through the positive points and the many issues identified in great detail.
As for what we're going to follow on through here, there is a benefit in clear accountability for the expenditure decision, the actual cash outlay when you buy an asset.
The downsides, of course, we've been through in some detail. The Auditor General has pointed out the problems created in departmental lease-buy decisions: the difficulties of a bias against capital purchases, and rust-out problems; the annual availability of funds leading to a short-term focus. This is a model that I think we have a general agreement about. It suffers quite a few weaknesses, and we need to get past this model.
Model 2 is really just heuristic. Nobody actually does this, but it's the opposite side of the spectrum, if you like. It's worth looking at, because it's a pure accrual model. In this model, one would say, if the fiscal framework is going to essentially be charged $10 million, why doesn't Parliament vote that, and only that?
There are some benefits. Clearly it would remove the constraint of the availability of current year funding from lease-buy decisions, for example. It would remove the constraint, in fact, of available funds completely, perhaps so much so that acquisitions would be too easy. Certainly it would deal with the problem of the complete cost of programs, because you would be voting the amortization.
What are some of the downsides of just doing this? There is a potential loss of control by virtue of losing sight of the cash information, and I believe that in the study we, your interlocutors, have been clear: cash continues to matter for a government as for any other entity.
There are particular issues around charging the amortization, taking it as the fiscal consequence of a decision. Amortization of assets is generally set by accounting standards. Managers and parliamentarians should not be arbitrarily changing the amounts that would be charged. Doing so would either incur a qualified opinion from the AG on the financial statements or might even lead to wasteful asset disposition and re-acquisition.
There's also the question of the consequence for a vote that might be exceeded solely because of an unexpected non-cash expense; for example, the need that if a building is destroyed by fire, for instance, one would have to write it off. “Blowing a vote” in this sense, which has a very strong normative impact today, would be something that could take place essentially through no fault of the manager. How would we interpret that ahead of time?
The model also potentially weakens accountability. Under such a model, current governments and managers would not be held accountable for the ongoing fiscal consequences, or the consequences in terms of a parliamentary vote, for their expenditure decisions. In contrast, current managers and governments would be held accountable for what the fiscal consequences are of past expenditure asset acquisition decisions.
Models 3 and 4 take a couple of different approaches to try to address those weaknesses.
Model 3 is what you heard the Province of British Columbia has now adopted in recent years. In this case, parliamentarians are voting on both the current year cost of acquisition and separately on the amortization of the existing stock of assets, which would be included in an operating vote.
There are significant benefits to this model. Departments and parliamentarians would both be considering the two things that really matter here: the cash requirements and the long-term expenses of amortization. There's no loss of information associated with cash management; it would be much clearer. Should a government be in a world of needing to borrow in a period of surplus because of a cash requirement, that would be quite visible through the detailed accounts, and Parliament would be aware of it. It fosters consideration of both short-term and long-term impacts of decisions.
There's still a constraint of available current year funding, but the presence and greater familiarity with accrual information should make it easier, both at the government-wide level and in departments, to obtain necessary, more lumpy funding streams for acquisitions, because the fiscal consequences are smoother at amortization.
Amortization would be voted, so there are still the same questions, however, about the meaning of control of amortization as in the previous model. Would a department be, for example, forced to dispose of an asset to avoid the associated amortization expenses if a vote was reduced? We would need to work very carefully on the wording of votes or on the consequences of an amount being reduced.
The votes on acquisitions partially solve the accountability issue. Amortization would continue on, but since we have the double vote, the accountabilities would still be clearer.
Parliamentarians would have to accept voting twice on the same asset: once at purchase, and once over the entire life of the asset. That's more complicated, but if it's clear and clearly understood—and I think you've heard this—it will need to be very transparent so that it is not confusing, for the public as well as for Parliament.
Model 4 has a different take. As I mentioned earlier, model 4 is the model that Ontario has adopted in recent years. It's quite similar in that expenditures there are voted. However, amortization is not included in an operating vote, it's included in a statutory appropriation. These items are presented, as all other statutory items today are presented, in our estimates for information. The legal authority is not provided in an appropriation act.
In our case, we would likely have to amend the Financial Administration Act to provide the statutory authority for these votes. I would note, however, that for reasons that don't deal as directly with the tangible capital assets, we would probably have to amend the FAA, in any case, for model 3 as well. So that's not really a distinction; it's a direct consequence of this aspect.
This model again provides the expenditure and expense information. Cash gets managed, and short- and long-term impacts do appear. One issue that has been identified around this model is that since amortization is for information, not voted directly, will there be any de facto loss of focus by Parliament, by managers, since this is for information as opposed to an actual vote? The upside is that we do not find ourselves in the potential box of a vote being reduced, and our having the lack of clarity about what to do with the asset if the amortization was not going to be voted.
Generally speaking, models 3 and 4 are real live alternatives. Those are the ones that you will see in the study come closest to some of where the study tends to look. Those are the ones that we are looking at more closely, and as I said, other governments inside Canada and abroad are using versions of this, but there is no one clear model, to our eye at this point. There are pros and cons, as we're trying to bring out here, between models 3 and 4.
Slide 12, then, just reviews the boarder scope of the departmental model. Going forward, as I mentioned, you will be seeing departmental balance sheets: the opening balance sheet at the start of a year and the closing balance sheet at the start of a year. These are relatively standard. As we go through a year, we have statements of operations and statements of cashflows.
We looked just briefly here at just some of the issues associated with capital assets. There are other items around capital assets themselves, such as writedowns or writeoffs, not catastrophic loss, other sorts of issues that we will need to look at. The annex takes you further into how these models would treat environmental liabilities or financial assets. They are analogous, but we will need to look at them.
Overall, there is quite a scope of accrual information that will be potentially there on the books. How much we bring into estimates and how much is controlled versus information are important parameters in terms of the decisions for how we can go forward.
Slide 13 just sums up. We obviously believe that these discussions are complicated enough. The consequences are material for departments, for Canadians, and certainly for Parliament. We have gone back and looked at how such decisions have been taken in recent decades.
In the 1980s, there was a case where the public accounts committee of the day recommended that the OAG and TBS, as I understand it--I wasn't doing this then--survey individual members of Parliament.
In the mid-1990s the Standing Committee on Procedure and House Affairs, in its 110th report, noted that “the primary focus of any revisions to the Estimates and other related material must be on the needs of Parliament.”
Certainly that has been our guiding approach. We are trying very much here to present estimates that Parliament will find useful and that meet its needs. We recognize the need for us to carry our side of this partnership, but we do view it in that sense, as a partnership.
It's for the committee to decide how it would like to move forward in this particular case, of course. In terms of the model that seemed to work well at that time, a parliamentary working group, as I understand it, was struck. Each party identified an individual with an interest. This working group met in camera with TBS officials, worked through some really detailed design issues, and mocked up what each of these approaches would look like. That culminated in some agreed pilots with individual departments, and this was the reform that actually broke apart what used to be called the part IIIs and created departmental performance reports and reports on plans and priorities. It culminated in a motion in the House to permanently split the part IIIs into these two documents that we now see today.
That's the end of the presentation. As we said, we believe this is a very important and, we hope, ongoing dialogue.
Liberal
Navdeep Bains Liberal Mississauga—Brampton South, ON
Thank you very much, Madam Chair.
Thank you very much for coming this morning to give us the presentation. I'm glad we were able to review the various options. That was very helpful, actually.
We're in the final stages of this, the final phases, and I'm hoping there's going to be consensus on this in terms of accrual-based budgeting, specifically for appropriations. I think we've examined this issue fairly extensively. We've met with various witnesses in the past, and you've alluded to the various committees that were struck. This issue has been examined every which way possible over the past few years, and I think there's a deep-rooted desire now to come up with some recommendations and move forward.
I think your remarks also alluded, Mr. Moloney, to the idea that you had read the transcripts from when we met with representatives from British Columbia and Ontario with respect to their feedback and views of the changes that they made when they went to accrual-based budgeting. Some of the issues that emerged from our discussions with them, at least in terms of the areas that I targeted, were system changes, costs, and implementation. Those are really the three areas that I focused in on in terms of my questioning, and my colleagues did as well.
I'm a bit confused about the PWC report, because I thought the mandate was a bit different. When I actually realized what the mandate was, it was interesting, but we'll talk about that later. But based on the models that you've highlighted, and particularly model 4, which I believe you just reviewed momentarily, I believe it's probably the best, with the least amount of cons. That's just my viewpoint, and I therefore wanted your feedback.
It was brought to our attention that seven major systems are used by various departments for budgeting purposes. If we were to implement model 4, what would be the impact to the system changes, the costs associated, and some sort of implementation timeframe with those?
Comptroller General of Canada, Treasury Board of Canada Secretariat
Thank you very much for that question. Maybe I can take the first stab at it.
You say model 4 is probably the one you would favour. I would say maybe model 3 and model 4 would be the two models that would be roughly in the zone of where it would probably be the most useful for members.
If I recall the presentation from our colleagues from B.C., they we are making reference to a cost of $5 million for training, but that was exclusive of the system costs.
You also mentioned that we have seven systems to deal with. What would be the costs of moving to this? It's very difficult to give you a hard number. PWC gave a range of between $10 million and $150 million—
Comptroller General of Canada, Treasury Board of Canada Secretariat
—depending on whether it's done with other activities or other changes that we're talking about. So it would take significant dollars, but that is not a bad ballpark figure in terms of how much it would cost. However, it could also be done over a number of years.
What might also be important to look into is the fact that PricewaterhouseCoopers also gave us an idea in terms of a timeline having to do with the first three years, and then doing it over five years. That's probably the way it should be done. Phase it in so that people can get used to the new way of doing business, the new information.
That number, as I say, is not a bad number to—
Liberal
Comptroller General of Canada, Treasury Board of Canada Secretariat
It was over five years. It was a PricewaterhouseCoopers study recommending some specific timelines in terms of what you do after year one, year three, year four, and year five. That would not be a bad model.
Liberal
Navdeep Bains Liberal Mississauga—Brampton South, ON
In your opinion, you said the best one might be a mesh somewhere between models 3 and 4, or something along that line. Could you elaborate on that?
Comptroller General of Canada, Treasury Board of Canada Secretariat
I like model 3 because of the transparency and visibility, but also the controls it gives to parliamentarians.
Also, one of the concerns that we all have is that somebody can play with the amortization, depending on whether you have room or not. But given that we're having our statements audited annually and one of the key accounting principles is consistency, I would say that if there were some inappropriate changes to the amortization or the depreciation policy to suit a certain need, that would be picked up. There would be a strong incentive not to go there.
So model 3 is not bad at all, because it—
Liberal
Navdeep Bains Liberal Mississauga—Brampton South, ON
And that's the one British Columbia has adopted, correct?
Comptroller General of Canada, Treasury Board of Canada Secretariat
Liberal
Navdeep Bains Liberal Mississauga—Brampton South, ON
I just have a question for the Auditor General.
Based on the models that I'm assuming, I'd just like your feedback on which model you think would be best. Should it be a matrix or a combination of models?
Auditor General of Canada, Office of the Auditor General of Canada
Madam Chair, we would obviously prefer either model 3 or model 4, because those are the ones that really do deal with the accrual-based information. Models 1 and 2 are essentially—
Auditor General of Canada, Office of the Auditor General of Canada
—cash, as it is now.
Either one of those models would work, but it really would be in consultation with parliamentarians to see which one they would prefer. We don't have a particular preference between models 3 and 4.
Liberal
Navdeep Bains Liberal Mississauga—Brampton South, ON
In terms of timeline, according to the PWC timeline that was indicated based on their report and what the Comptroller General has said as well, do you think five years is a fair timeline for implementation? What's your opinion?
I know we've talked about this in the past, but I just want to make sure we have clarification at the present time. Do you think that's an appropriate timeline for implementation?
Auditor General of Canada, Office of the Auditor General of Canada
I don't think five years is inappropriate. I've learned that it does take time to do these things in government, even though we would always like things to happen perhaps faster than it would seem to take. I think it's important to give a reasonable timeline, but with clear milestones along the way, in order to make sure the project doesn't slip and then become a seven-year or ten-year one.
And there will be changes. Just because of the cycle of budgeting and the presentation of votes, there is a certain delay. Even if you started today, it would probably take two years to put it in. So five years would seem to be reasonable.
Liberal
Navdeep Bains Liberal Mississauga—Brampton South, ON
In your opinion, with respect to the system changes, the systems that existed in British Columbia were legacy systems, so they were really outdated. There was an issue there. I believe it was Oracle that they went to as a system in terms of just managing departmental budgeting.
In your opinion, do we need to have a major system overhaul, or can we work with the current systems that we have in place?
Auditor General of Canada, Office of the Auditor General of Canada
When we refer to the seven systems, those are really the systems for what we call the general ledger. Those are the ones that can produce, if you will, the financial statements. In government, though, those systems are supported by a multitude of legacy systems, some of which are very old.
We have actually started to do a lot of work on assessing those legacy systems. We probably don't want to get into all that today, but I think the government does have a major investment that it has to make in updating or replacing some of those legacy systems. Many of the challenges that government faces right now, even in its accounting, come about because the systems were never designed to give the kind of accrual information that you need.
In this committee, I think we reviewed, for example, the collection of income tax receivables. The systems in the Canada Revenue Agency were never designed to manage receivables, and we see that in other departments too. So there is a major investment and change that will be necessary over several years—much more than five years—to adapt those systems even to meet the current accounting requirements.
This has to be integrated into that and the change has to be thought out. How are they going to do this overall not just in terms of the accrual part of the appropriations, but in terms of the other management information that's going to be needed going forward?