It's a pleasure to be a good neighbour and to discuss the Canadian practice, even though I'm not an expert on it.
I want to begin by indicating that good practice is not necessarily a precondition for good budget outcomes. My own country, the United States, is an example of this. We have perhaps the most widely respected legislative budget organization in the Congressional Budget Office. We also have the largest deficits in the world.
I don't want to correlate the two, but I do want to caution that often budget reform is a substitute for budget policy, and it should never be a substitute. Near the end of my remarks, I'll get back to the issue of parliamentary assistance or staff assistance for parliamentary budget work.
I do want to mention that anything that deals with a budget touches the constitutional framework of a country—the relationship between government and Parliament, between the parties and government, the electoral system. Therefore, one has to be guarded in importing from another country practices that may not be well suited in your own country. This is particularly the case because the Westminster system, of which Canada is a member, is at polar ends from the American congressional system. One has to be wary about exchanging techniques when they may ill fit a particular country.
Having said this, I do want to mention that from a distance, my own distance, there are a few matters of Canadian budgeting that I think deserve some attention. Perhaps the most obvious is the disconnect between the budget and the main estimates.
As a matter of fact, I'm kind of puzzled that what you call the main estimates are not what political leaders would regard as the main estimates. The main estimates should be a statement of policy, if they're truly the main ones, and yet here they deal more with the ongoing work of government, that which is being continued, rather than with revenue changes and with policy changes.
The timing issue is well known in Canada, and that is that the main estimates, I believe, precede the tabling of the budget, and consequently the additional estimates, or supplementary estimates, have to be tabled later in the year to incorporate the policy changes recommended by government.
It would be a reasonable step, not a difficult step, for government to change the timing to coordinate, indeed to consolidate, the estimates and the budget. At one time, in fact, the United Kingdom had a distinction between the estimates process, which they called the “spending process”, and the budget we talk of, particularly with revenues in the U.K. case. Now they have been brought together on the same page, as it were. I think that is something that your own country should well consider.
In doing so, I would urge that the fact that you have a divided process, with estimates coming at one time and the budget at another time, allows you to restructure your entire budget process—not only that they are timed together; that indeed the two different sets of actions occur at different times. This is something that I would urge you to consider. Sweden is among the countries that have successfully introduced the procedure: that is, dividing parliamentary budgetary work into two discrete stages, a framework stage followed much later by an estimates stage.
The framework stage is a matter of policy, of strategy, of changes to revenues and to programs, of major changes to the estimates, and, most importantly, looking ahead to the macroeconomic environment, not only for the year for which estimates are being voted but also three to five years or more ahead.
If you combine strategy and estimates, the big picture and the detail, the likelihood is that one or both of them will be neglected. More often than not, it's the strategy, the policy, that is subordinated to the specifics of the budget.
So what a number of countries have introduced is a divided process in which at the first stage, which I label the framework stage, they do not delve into the details of the estimates. Instead, what they do is look at the economic environment, projections over the medium term, major policy changes by government, particularly with respect to the deficit, the debt, and other key fiscal variables.
While spending details are not tabled during this first framework stage, the government does provide, to use a famous or infamous Canadian term, the “spending envelope” that would be available during the estimates stage—in other words, what will be total expenditure, and that's divided by key sectors or policy arenas.
That's the first stage, and in some countries it's actually voted by parliament; in other countries it's merely discussed by parliament. It's presented by government, and depending on the role of the legislative branch, it either would be recognized by parliament or agreed by parliament. After this occurs, government ministries prepare their budgets consistent with the voted or tabled framework.
This leads to the second stage, which deals with the estimates and appropriation of authorized expenditure. The obvious rule in this case is that the estimates have to be consistent in two ways with the framework that was previously established. The first is with respect to the totals—the details cannot exceed the amount of money voted in a framework. Secondly, with respect to policy initiatives or changes taken by government, those should also be reflected in the estimates.
This would be a significant departure from the situation that currently prevails in Ottawa, but it would be consistent, however, with the fact that you have a divided system, and it would engage Parliament at two different times in the year on budgetary matters, one dealing with strategy and the big picture, the other with the specifics of expenditure.
Now, there's another aspect of Canadian practice that to my distant eye was once quite common around the world. It's still common in many developing countries, but it has virtually disappeared from advanced countries such as Canada. That is the distinction between operating and capital expenditure. At one time it was common around the world that there were actually two separate budgets. We often called them divided budgets: one dealing with the investments of government and the other with current or operating recurring expenditure.
The historical basis for this distinction is that the two sets of expenditure had different sources of funding. One was out of current revenue, and the other was often out of borrowed funds—a kind of golden rule that government can borrow only for finance investment. And to ensure compliance with this rule, they divided the budget and the expenditure into these two categories. For two main reasons, this practice has disappeared from developed countries.
The first reason is that to the extent that government is concerned about its fiscal position, the key aggregates—total revenue, total expenditure, total debt and deficit—you have to have a consolidated picture that does not separate between capital expenditure and operating expenditure.
The second reason is that capital and operating expenditures often are not distinct but are interchangeable with one another. In so many areas of government policy they're actually substitutes. One can proceed down a policy course of action through investment—for example, building clinics in rural areas. That would be in the capital budget. On the other hand, government can seek the same aim to improve health services in rural areas by subventing physicians to practise in rural areas. One is in the capital budget, one is in the operating budget.
The more you do one, for example, building clinics, the less you may need of the other, or vice versa. Consequently, if you want to have a robust analysis of the policy options of government and the connections among them, it makes a lot of sense to merge the two types of budgets, keeping in mind that you still must have sufficient data on the investment position of government.
This doesn't mean that you remove investment and capital information from the budget. It means, however, that they are within one umbrella of the budget. The question then becomes what should that umbrella be. What should be the classification or the framework within which you see both investment and operating expenditure?
There are two main possibilities. One is widely practised. The other is widely recommended. The widely practised one is by organizational units. To the extent that an organization bears both operating and capital costs, they should be combined in that entity's budget.
The alternative is what we call a program budget, or program structure. To the extent that operating and capital expenditure contribute to the same objective, they should be located within the same budgetary program, regardless of organizational location. In other words, a program budget, in some cases, will ignore organizational or ministerial boundaries. This is precisely why the program budget approach is highly recommended but is rarely practised, because to the extent that government, in addition to wanting to make robust policy, which would require that you see capital and operating expenditure contributing to the same objective...government has another purpose in managing its finances, and that is maintaining accountability.
Accountability in almost every case requires that the organization responsible for the expenditure and the activity financed by the expenditure should be in the dock, so to speak. It should be the accountable party. This is deeply embedded in Westminster tradition, and it's something that may be very difficult to surrender.
In fact, many countries call it a program budget, but the program is simply a veneer for organizational entities. To give an example that comes immediately to mind, you may have a bureau of water resources, which is an organizational unit, but instead you label it as a water quality program. The boundaries of the program and the boundaries of the bureau are identical, so in effect you've labelled it a program budget, but it really is an administrative, organizational budget.
Regardless of the way you go, I would think it is worth reconsidering the connection between the operating and the capital budget.
That leads to the third issue that I would like to discuss, which is what should parliament's role be, and more important, how should parliament be assisted to carry out that role responsibly and in an informed manner?
There is a mock trend around the world, not in every country, and certainly less so in Westminster countries than elsewhere, to enlarge the capacity of parliament to review and even to amend the government's budget. Keep in mind what I said at the outset, that the extent to which you want parliament to be able to amend the budget rises to a constitutional issue.
The practice in a growing number of non-Westminster countries is for a vast increase in the volume of amendments tabled in parliament and some subset of them being adopted, but most of the amendments are specific, detailed. They are within the government's fiscal envelope. This is very important.
To the extent that a country enlarges parliamentary discretion with respect to the budget, it is urgent that Parliament be subject to some fiscal constraint in terms of what it does. The combination of an open-ended parliamentary work on the budget without a constraint is something that can lead to fiscal damage to the country.
My own sense is that this is not where Canada is right now. Canada is not going to break away in a fundamental way from its Westminster legacy. Consequently, the issue becomes one of informing parliament rather than empowering parliament. Empowering parliament would mean that parliament can make significant changes to the government's budget. Informing parliament means that what parliament does is to hold government to account by having a robust debate on the options in the budget, the estimates tabled, the economic and programmatic assumptions that underlie those estimates, and the longer-term sustainability of the government's position. This clearly is consistent with the role that your committee has.
In fact, in your opening remarks, Mr. Chairman, when you indicated that you are a member of the opposition, it brought to mind the historical role of public accounts committees, in which that was the basic mechanism for well over a century, perhaps two centuries, for holding government to account: that opposition would chair the public accounts committee, the committee would operate on a non-partisan basis, take evidence from government, and thereby hold government to account.
Perhaps this system is sufficient. It's certainly survived a long time. But the fact that Canada a number of years ago established a Parliamentary Budget Officer tells me, hundreds of kilometres away from Ottawa, that at least there was unease in Ottawa about whether simply having a committee chaired by the opposition was sufficient to hold government to account and to allow for informed debate. After all, if it was sufficient, there would be no need to establish a Parliamentary Budget Officer.
Canada, in so doing, was following a trend that is quite widespread around the world, and that is staffing up parliament to be able to better perform its budget-related responsibilities. In most countries, however, I should note that the staffing occurs at the committee level, so that the additional staffing that is available to parliament to review the estimates, to offer options, to challenge the assumptions when it's appropriate—these are committee staffings, and therefore it has a low profile and is subordinate to the committee process in parliament.
In a small number of countries, including the United States, Mexico, and Korea, rather than relying on the committee structure as the main means for improving parliamentary budgetary work, a separate and independent office has been established. This has been done in Britain. In Britain it's not formally attached to Parliament, but it advises Parliament.
The role often is to review the estimates to see whether they are reliable. The key budget work today around the world is not simply whether the money should be spent, but are the assumptions underlying the estimates robust? Are they reliable?
Now keep in mind what assumptions are. If the table over here is the waterline, everything above the table is the budget and the estimates. They are open and transparent. They can be reviewed and published. Everything below the table line is assumptions. The assumptions are not transparent. They are not visible. But the numbers above the table are dependent on the assumptions below the table, and there's very little sunlight in government around those assumptions. This becomes the difficult task—and perhaps the most important modern task for Parliament—in dealing with a budget.
On the revenue side, clearly, the revenues are driven. They are a function of the economic performance of government. The future economic performance of government can never be known; it can only be assumed. That is the assumption; that is something parliament has to invest in.
What about the sustainability of the budget over a long term? That's critical to the future course and the future fiscal and economic health of your country. That rests on a bed of assumptions.
What about a government introducing a policy change? You want to know over the medium term what would be the budget implications of that policy change. After all, the first-year cost of a policy change usually is quite minimal, quite modest, but it cascades and enlarges in the future. Has the government been forthcoming? Is it using reliable estimates?
One of the reasons why assumptions lie below the waterline, below the table, is that they do not do well in sunlight. Very often, assumptions.... How do we describe them? The back of an envelope, right? Guesswork, okay? Sometimes they're politically massaged, but even when they're not politically massaged, the best thing you can do about assumptions is just to say this: let's assume a different set of assumptions, let's do a critical sensitivity analysis, and let's—to use a modern term—stress-test the assumptions to see whether they can stand the light of day.
I think this is something that parliament can benefit from. I think this is likely to be something that I would urge your committee to consider, whether in the context of the PBO, the Parliamentary Budget Officer, or in a larger framework—I don't know enough about the situation.
But I do think that you have an organization already, the PBO, which in international quarters is widely regarded, and building on it I think would be helpful. My understanding is that while it's a parliamentary budget office, it works closely with committees so that it's not a completely adrift entity. I would think that's something your committee might wish to consider.
These are some of the thoughts I have. I would invite questions or venturing into other areas where you think I might be of assistance.