Public Accounts Committee on Oct. 17th, 2017

[Bill Matthews]

The default is, if they're grants and contributions dollars, or if it's above 5% of operating or 20% of capital, it goes back into the general coffers. The responsible department can then make a case for saying they'd like to re-profile that money—that means spending it in a future year—and then the government has a decision to make, but there's nothing automatic.

The default is, if nobody does anything, it goes back into general coffers. If there's a decision made to delay the spending to future years, that's more a fiscal discussion that has to be had.

[The Chair, Conservative, Kevin Sorenson]

Thank you, Mr. Matthews and Mr. Nuttall. We'll come back to you soon.

We're in the second round now.

Mr. Chen, please. You have five minutes.

[Shaun Chen, Liberal, Scarborough North, ON]

Thank you, Mr. Chair.

Mr. Ferguson, as Auditor General, you have identified three key observations, one of which is the government's selection of discount rates used to estimate long-term liabilities. To illustrate the material impact of this, you give the example on page 2.43 that a 1% reduction in the discount rate for estimating the accrued benefit obligation for unfunded pension benefits results in an increase of this liability to the tune of $7.7 billion.

This would clearly be significant, and it would overhaul significantly the financial picture. By using a higher discount rate, one could effectively lower the estimate for long-term liabilities.

On page 2.43, it states, “While we have concluded that the assumptions underlying the Government's significant estimates are within a reasonable range, historically, certain discount rates have been at the high end of the acceptable range when compared with market trends.”

Is there a difference between what you deem a reasonable range vis-à-vis an acceptable range? If there is, I'm assuming that a reasonable range would be better than an acceptable range. If that's the case, how do you determine what is reasonable and what is acceptable?

[Michael Ferguson]

I think that's perhaps unfortunate wording on my part in the audit observations.

When we refer to reasonable range and acceptable range, we're referring to the same thing, within a range that is what you would call either reasonable or acceptable. We use them as synonyms, but I understand why the fact that we use different words could seem like we mean two different things. In fact, it means the same thing.

[Shaun Chen, Liberal, Scarborough North, ON]

With respect to the government's method of selecting discount rates, is there a difference compared to industry practices, whether in the public or private sectors?

[Michael Ferguson]

I think when it comes to selecting discount rates for things like the unfunded portion of the pension plan, that's where there may be some differences between what the federal government is using to select a discount rate and what other governments would use.

One of the parts of the analysis that needs to be done on the discount rates is looking at that comparison to others and also looking to make sure there's consistency between the different liabilities that the federal government has and how they're selecting discount rates for those various liabilities.

There are some places where the way the federal government is selecting the discount rates may be unique to the federal government, but all of that needs to be part of the analysis that we've recommended they do to rationalize the discount rates they're using.

[Shaun Chen, Liberal, Scarborough North, ON]

Thank you.

One of the financial highlights in volume I is with respect to Canada's debt-to-GDP ratio. Much discussion happens around debt-to-GDP ratio and comparisons between Canada and other G7 countries. I just want it to be clear on the record. When you look at different countries, they are set up differently. Some might have provincial or state levels of government as well as municipal levels of government.

When the IMF compares the net debt-to-GDP ratio for the G7 countries, do they take the approach of looking at the figure in a total government lens? Are they accounting for all levels of government and their respective total net debts when making those comparisons?

[Michael Ferguson]

Thank you, Mr. Chair.

I'll start by explaining how I understand the IMF calculation. Perhaps Finance may want to jump in on this as well.

To me, the IMF calculation of net debt to GDP is a good measure from a relative point of view, but not from an absolute point of view. Relatively, the measure is very good because it shows how Canada compares on a relative basis, in apples-to-apples comparisons to other countries. It essentially shows that the next best country in the G7 has a debt burden that is roughly twice Canada's—and I believe that's Germany—and then after that I believe it's the U.K. and the U.S. that are four times that of Canada. On a relative basis, it's a good measure.

I'm not so fond of the percentage comparison, though, because it takes the federal government's net debt and, as I understand it, deducts the federal government's pension liabilities. The $240 billion that the federal government owes in pension liabilities to its employees, I believe, is deducted. It adds in the net debt of the provinces and of the local governments, so it expands net debt to not just a Government of Canada definition but to a definition of all net debt of all governments in Canada. Then it deducts the assets of the Canada pension plan and the Quebec pension plan. Again, it goes through all of that to try to put things on a comparative basis to other countries, because not all countries record their pension liability. One reason Canada can get 19 years' worth of clean audit opinions is that Canada records its pension liabilities, while other countries don't.

In the IMF's calculation, they have to balance all those things off to try to get an apples-to-apples comparison, and that's why I say that on a relative basis I agree with the results of it. When it comes down and says that Canada's net debt to GDP is roughly 27.6% or something like that, that's a number that doesn't include all the debt of the federal government, because it excludes the pension liability and reduces the debt by the assets in the Canada pension plan and the Quebec pension plan, which are not assets that can be used to pay down any government debt. I don't like the 27%, and I don't think people should focus on the 27%, but relatively, it's a good measure.

In my opinion, the best measure of debt to GDP—in terms of the federal government itself—is the net debt-to-GDP measure, which is the $714 billion of net debt in relation to the GDP of Canada, which is probably just slightly over $2 trillion. I believe that comes out to somewhere around 35%, and I think that is the best measure.

If you go through the financial statement discussion and analysis, you will see at least three other comparisons of debt to GDP. You'll see the accumulated deficit to GDP, the interest-bearing debt to GDP, and the net debt to GDP. You could also do a calculation of total liabilities to GDP, so there are many different ways of comparing debt to GDP, depending on your definition of debt. I think the best one is to do a comparison of net debt to GDP on a financial statement basis.

I like the IMF one from a relative point of view, but I don't like it from an absolute point of view.

[The Chair, Conservative, Kevin Sorenson]

Thank you. We're three or four minutes over, but it is interesting.

I want to go to Mr. Rochon—he was indicating he wanted to answer a bit there—and maybe even Mr. Matthews.

[Paul Rochon]

It depends on why you're looking at the debt, and that really goes to what measure you want to use. Absolutely, if you're comparing across countries, the IMF calculation is the more appropriate one. It allows us to compare Canada to the United Kingdom, for example, which is a unitary state. We are not, so you should add the provincial levels of government in.

You should also recognize that we include a number of liabilities related to pension obligations and benefit obligations that other countries do not, as well as the fact that we are saving quite a large amount, by international standards, to pay for future pension obligations that other countries are not, for the most part, accumulating.

If you were an international investor, or if you were asking how we stack up vis-à-vis other countries, the IMF measure is absolutely the right measure. As you know, we very much agree with the Auditor General. If you are looking at the federal government's debt situation from a sustainability or a purely financial perspective, we would agree that the net debt is the right number to be using.

[The Chair, Conservative, Kevin Sorenson]

Thank you, Mr. Chen.

Thank you, Mr. Rochon.

We'll now move to Mr. Nuttall, please, for five minutes.

[Alex Nuttall, Conservative, Barrie—Springwater—Oro-Medonte, ON]

Thank you for that answer from the Auditor General. I thought that question from Mr. Chen and the detailed answer were great. It provided a lot of clarification, certainly, in my own mind.

I want to follow up on a question that Mr. Liepert put forward earlier. I know that some of that information is going to be forthcoming and perhaps we could add this to that information.

It says the overall interest paid was lower than what would have been projected. Do we know what vehicles or tools allowed us to come in much lower than we had estimated at the beginning of the year?

[Bill Matthews]

I'll go ahead, and then I'll let Mr. Rochon give real content. There's a document that is not part of the public accounts, the debt management strategy document, a Department of Finance publication. It is an extremely useful document in terms of understanding the debt management strategy of the federal government. That is their publication.

We do have an answer to the member's previous question on small business, so if there's time at the end of this answer, maybe we'll have Diane answer your last question.

[Alex Nuttall, Conservative, Barrie—Springwater—Oro-Medonte, ON]

Sure. Why don't we try to answer both if we can?

[Paul Rochon]

Yes, I can make it short.

The answer to your question is that the 10-year government bond rate came in about 30 basis points lower than we had projected at the time of the budget. That is on page 1.3. There is the projection versus the actual for those interest rates.

[The Chair, Conservative, Kevin Sorenson]

Thank you, Mr. Rochon and Mr. Matthews.

Ms. Peressini.

[Bill Matthews]

I'm going to have Diane get on the record here.

[Diane Peressini, Executive Director, Government Accounting Policy and Reporting, Office of the Comptroller General of Canada, Treasury Board Secretariat]

Thank you.

The Canada Small Business Financing Act writeoffs are on page 2.9 in English.

In French, it is on page 2.10.

There you will see that there were 907 writeoffs totalling $60 million last year.

[Alex Nuttall, Conservative, Barrie—Springwater—Oro-Medonte, ON]

Totalling $16 million?

[Diane Peressini]

No, it is $60 million—six zero.

[Alex Nuttall, Conservative, Barrie—Springwater—Oro-Medonte, ON]

It's $60 million. Sorry, I just don't have it open in front of me anymore.

[The Chair, Conservative, Kevin Sorenson]

It's on page 2.9 of volume...?

[Diane Peressini]

Oh, sorry, it's volume III.

[Alex Nuttall, Conservative, Barrie—Springwater—Oro-Medonte, ON]

So, it was $60 million in writeoffs. What was the total of that program?

[Bill Matthews]

The only thing I would highlight for that question, Mr. Chair, is that those writeoffs are not representative of one year. You can write off six or seven years' worth of program data. We will get you the program spend, but do understand that when writeoffs happen, they can be for over 10 to 12 years.