Debt can be defined in more than one way

Each year, the federal and provincial governments issue audited financial statements prepared in accordance with accounting policies and guidelines established by the Public Sector Accounting Board (PSAB). Under the 2003 PSAB model, the financial statements are required to convey key information on debt, surpluses/deficits and changes in a government’s cash position. Government fiscal estimates under this framework are considered to be on a Public Accounts basis.

With respect to debt, two different measures are made available to readers of annual financial reports. The first is the accumulated deficit, which is calculated as the difference between the government’s total liabilities and its total assets. And, second, is the net debt position, which is computed as the difference between a government’s total liabilities and its financial (i.e., excludes non-financial) assets.

In budgets, governments usually emphasize one of the two measures, if only to spare the public from being overwhelmed by a plethora of accounting concepts on budget day. In recent budgets, the federal government has cast its focus on the accumulated deficit. In contrast, most provinces highlight net debt. The definition of accumulated deficit has some intuitive appeal, since most Canadians would associate a change in debt to total income (or revenues) less expenditures regardless of whether financial assets or non-financial assets are accumulated. At the federal level, where non-financial assets are not large relative to total liabilities, the difference between the two estimates is small. On the other hand, since non-financial assets are illiquid, there may be some merit in excluding them on the grounds of conservatism.

As shown in the table on page 2, the federal government reported debt on a Public Accounts basis of $481.5 billion (35.1% of GDP) in fiscal 2005-06. Combined, the federal-provincial debt burden was estimated at $761.2 billion (55.5% of GDP).

If those were the only concepts of debt, then there would be little reason for confusion. However, other definitions exist that are less commonly known than those contained in the Public Accounts. One alternative debt measure is released by Statistics Canada, which is based on an international standard referred to as the National Accounts. The statistics agency makes adjustments to the Public Accounts figures to bring them into line – as much as possible – with the standard. Notably, the combined asset holdings of the CPP/QPP, which currently amount to about $110 billion, are factored into the count. Another differentiation between the Public and National Accounts is that the latter also contains detail on the liabilities and assets of local governments. On a net basis, these adjustments act to lower the Canadian overall government debt ratio to 45% of GDP (see table on the previous page).

Given that the guidelines in the National Accounts leave room for interpretation, inconsistencies exist across countries. As a result, the OECD – an international body which publishes statistics on as many as 70 jurisdictions – makes further refinements to the National Accounts data in an attempt to enhance international comparability. These changes show Canada in a much more favourable light. Most importantly, the organization excludes the unfunded liabilities of government employee pension plans, which shaves a full 15 percentage points from the combined federal-provincial debt-to-GDP, taking the ratio down to 30.2%, (as shown in the federal Update).

The OECD figures show that in 2005 Canada recorded the lowest debt-to-GDP ratio among the G-7 countries, but a considerably higher ratio than Sweden, Australia, Denmark, New Zealand and Ireland. In fact, Sweden and Australia enjoy net asset positions.

What are others saying?

Government or quasi-public debt estimates are not the only ones garnering attention in the media. Other organizations – including bond rating agencies, financial institutions and think tanks – also release varying estimates often depending on the needs of their clients. For example, bond rating agencies tend to focus on the liabilities side of the ledger – and more specifically, on taxpayer-supported debt (i.e., total debt less that of self-supporting commercial crown corporations) or marketable debt outstanding. Both of these concepts are found to be more relevant to investors in terms of pricing risk. For example, based on the computation of the Dominion Bond Rating Service (DBRS), Canada’s overall debt-to-GDP ratio is estimated at 67%, which is above those of the Public Accounts and National Accounts.

Still, to provide an example of how definitions of debt can be in the eye of the beholder, consider an estimate released by the Fraser Institute in its study Canadian Government Debt 2006, which pegs Canada’s total debt at a stunning $2.7 trillion or 222% of GDP. This estimate, which is made on a Public Accounts basis, factors into the equation a myriad of additional liabilities that the Institute deems to be material, including government debt guarantees and contingent liabilities (i.e., potential claims such as those on Aboriginal lands). Above all, the 2006 study adds on a whopping $1.5 trillion alone for a category referred to as “program obligations”. These obligations largely include the unfunded liabilities of the CPP/QPP, OAS and Medicare.

With respect to CPP/QPP, unfunded liabilities can be defined as the difference between what the current cohort of workers will be entitled to in the future and the value of net assets in the fund as of a certain date. The Fraser Institute has cited an unfunded liability presented in the 2003 Actuarial Report of the Canada Pension Plan (page 116). The unfunded liability of the QPP was simply assumed to be one-third that of the CPP. More generally, unfunded liabilities refer to existing obligations or promises that, based on current actuarial assumptions, cannot be met without raising taxes. For Medicare and OAS – which are also programs that government has committed to providing but that are not deemed to be entitlements – the study’s authors have estimated liabilities through models of their own. It is noted that the calculations are highly sensitive to underlying assumptions.

The large unfunded liability of the CPP is a good example of how different debt concepts can lead to markedly different results. The federal government’s Public Accounts estimate does not factor in an unfunded CPP liability. What’s more, as we noted earlier, the OECD actually lowers the debt calculation by including the public pension plan’s current asset holdings. No future liabilities are considered.

In any event, there is nothing particularly binding about these definitions of unfunded liabilities, since designs of the programs can always be altered and/or taxes be raised. In the case of CPP, returns on investment from the current excess of premiums over payouts can make the plan sustainable. Indeed, under the changes introduced in 1998 – including a doubling in CPP premiums – the pension plan has shifted to a more fully-funded status. And with the fund assets expected to grow sharply on the back of investment earnings in the coming decades, the Auditor has concluded that the fund is on a solid footing.

In sum, there is no “single” measure of the federal or all-government debt. Figures are highly sensitive to the basis chosen and the assumptions made.

A closer look at the federal government’s debt target

In the November 23rd federal Update, Minister Flaherty allocated considerable time to the issue of debt reduction. His speech touched on the significant progress the federal government has made in reducing debt, highlighting the fact that a string of budget surpluses had cut the federal debt-to-GDP ratio from a lofty peak of 68.4% in fiscal 1995-96 to 35.1% in fiscal 2006-07 (Public Accounts basis). Indeed, the federal government has been a leader in debt reduction both within Canada and on the international landscape over the past decade. Since the mid-1990s, the combined provincial debt-to-GDP ratio has also declined, albeit to a much lesser degree.

In the Update, Minister Flaherty also reminded Canadians about the government’s previously-stated longer-term target to lower its debt-to-GDP ratio further. In the 2006 budget, the government committed to a 25% target by fiscal 2013-14. This was based on an annual commitment to pay down the debt by $3 billion per year and nominal growth in the economy of about 4.5%. However, with a higher-than-expected $13 billion payment against the debt in the last fiscal year, Minister Flaherty announced that the 25% target was being brought forward one year, to fiscal 2012-13.

However, few Canadians would have noted the announcement, since it was lost in the media frenzy that centred on a newly-announced debt target – that being a goal to eliminate Canada’s net debt by 2021. This debt measure is is very different than the standard one underpinning the government’s 25% debt-to-GDP goal. Rather than the Public Accounts definition of federal debt, the new target is based on the OECD debt concept. Recall from page 3 that the OECD concept is on a National Accounts basis, includes the debt of provinces and municipalities, excludes the unfunded liabilities of public employee pension plans and includes the value of CPP/QPP assets.

A few points we would like to highlight regarding this new target:

• While some may see the zero net debt target as unrealistic or heroic, it can be viewed as an extension of the federal government’s 2012-13 objective, since the government is assuming $3 billion per year in annual reduction in the federal net debt (see table below). In other words, the zero net debt goal essentially represents the status-quo.

• The new target relies on the provinces/municipalities keeping their net debt steady and actuarial assumptions for CPP/QPP assets and unfunded pension plan (UPP) liabilities to materialize as projected. Accordingly, compared to the 25% debt-to-GDP target announced in the 2006 budget, the number of variables outside of the control of the federal government under the new target has increased from one (i.e., GDP growth) to four (i.e., GDP growth, unfunded pension liabilities, CPP/QPP assets, and provincial/local government balances).

• There is a difference between the language used in the November 23, 2006 speech by the Minister and that in the background document. In the speech, the zero Canadian net debt appears to be put forward as a target, whereas the wording in the document suggests that zero net debt would simply be the end result if the federal government pays down debt by $3 billion per year, the provincial and local governments balance their books and the CPP/QPP actuarial projections are realized.

• The zero net-debt objective implies that the federal government would alter its fiscal plans if these “external” variables went off track. For example, lower-than- expected CPP assets would require increased annual debt reduction to meet the target. However, not only would this not make sense, but we doubt that the federal government had this in mind when it decided to establish the new goal.

• The choice of the zero net debt sounds like an attractive goal. But as we have discussed, the OECD adjustments are carried out for convenience of comparability across countries. It doesn’t mean that the concept is the most relevant to Canada.

• Perhaps most importantly, this broadly-defined target deflects attention away from the federal-only debt. Yet the implications on the federal debt burden beyond fiscal 2012-13 are indeed significant. More specifically, continuing annual debt reduction of $3 billion per year, combined with a growing economy, would push down the debt (on a Public Accounts basis) to 16% of GDP by fiscal 2021-22. This would represent the lowest reading in the Post-WWII period and eclipse the previous trough of 18.4% set in fiscal 1974-75.

What should the debt target be?

An even bigger question is whether the government should be targeting zero net debt in the first place. In other words, what is an optimal level of debt-to-GDP ratio that Canada’s federal (and provincial) governments should be striving for?

Several perspectives have been put forward to evaluate what an optimal debt burden might be. Economic efficiency and inter-generational equity are typically featured prominently on this front. With regard to efficiency, there is widespread recognition that high debt keeps taxes propped up higher than they would otherwise be. And since taxes can be very economically inefficient – notably those on income and capital – there are costs that result in terms of foregone economic activity. The intergenerational argument is based on the notion that some debt can be acceptable if it is being used to finance capital that will be passed on to future generations. At the same time, however, debt accumulated to finance consumption by the current generation is generally considered to be “unfair”.

Unfortunately, there appears to be even less consensus on what level of debt would be optimal than actual estimates of Canadian debt. This point is well illustrated in the following quotation from the book “Is the Debt War Over?”, released by the Institute for Research on Public Policy (IRPP) in 2004:

“(Lars) Osberg (of Dalhousie University) observes that among economists there is quite a wide range of opinion about what level of debt would be optimal. An informal poll of attendees at this conference (IRPP) found that opinion ranged between 20 and 50% of GDP (federally), which will strike most readers as a substantial variation – except that, as we have seen, within the literature on this question, opinions range from -300 per cent (a net asset position) to +70 per cent. A “fiscal anchor” whose size is so hard to pin down may be of dubious value.”

Ragan and Watson, “Is the Debt War Over?”

Not much economic bang from debt reduction

Part of the challenge in nailing down an optimal debt ratio is that the perceived economic benefits of a declining debt ratio have not been clear in the research. For example, Myatt and Rugerri (2004) of the University of New Brunswick assessed the efficiency gains from an accelerated pace of reduction in the debt burden. They concluded that the growth impacts from the lower debt profile were trivial. Later research by Ruggeri, Zou and Garrett (2005) found that “given the very small reductions in the aggregate tax rate generated by (accelerating the pace of reducing the debt burden and hence allowing a lower tax burden which in turn reduced economic inefficiency), we suggest it would be preferable to let future tax rates fall even further, not through debt repayment, but by increasing economic growth through selective tax cuts and public investment today that would increase the productive capacity of both current and future generations.”

Passing on assets (and debt) to the next generation

There is fairly broad support within the economics community of establishing a debt target based in part on the principle of intergenerational equity:

“Studies focusing on inefficiency indicate that virtually any debt ratio is as good, or not, as any other … the chosen target for any debt ratio must be based primarily on its consistency with society’s equity objectives, and particularly its objectives for intergenerational equity.”

William Scarth, “Is the Debt War Over?”

Based on the intergenerational argument, assets being accumulated today will be enjoyed by future generations. Hence, it is not unreasonable to ask that future generations provide a helping hand in financing them. According to Scarth, this principle involves letting the debt ratio rise or fall when the nation faces major events, such as a war, that can reasonably be expected to lower living standards (so-called “consumption smoothing” policy). In his view, the looming aging of the population warrants “working the federal debt ratio down to the 20-25% range (and the consolidated federal-provincial debt ratio down to the 45-50% range) within the next 10 years.”

At the extreme, one could boil down the intergenerational principle to a simple rule of thumb – that being that the debt-to-GDP ratio be set equal to the level of the capital stock-to-GDP ratio. In 2005, the assets of the federal government totaled 12% of GDP, which is lower than the Scarth objective but in the same ballpark as the status-quo federal forecast for 2021 (16% on a Public Accounts basis) implied under its new target. Still, the consideration of such a rule would also need to take into account future capital spending plans, such as any push by governments to ramp up outlays to address past shortfalls in investment. In any event, if assets are going to start getting counted in the process of designing a long-term debt target, then there needs to be an appropriate view of the liabilities. And the concept used in the Update is probably too limited for that purpose.

More research and debate required to support target

While the federal government’s newly-unveiled net debt target for 2021 has been widely misunderstood and criticized, it has served the useful purpose of sparking debate about where the debt should head over the long run. There appears to be widespread support of the federal government’s plan to cut the debt-to-GDP ratio to 25% by fiscal 2012-13, as the Minister announced in the 2006 Economic and Fiscal Update. At the same time, however, there is not a strong consensus on what an appropriate target should be for the federal debt burden – never mind the total Canadian public sector debt burden – beyond that point. There is not even agreement on how best to measure total government debt. The OECD convention of aggregating federal, provincial, local and CPP/QPP positions, but excluding federal and provincial employee pension plan liabilities is not a terribly useful concept.

Canadians should be pleased and proud of how their governments have reduced their debt burdens over the past decade. One of the benefits is choice, and in particular, the options now open with respect to allocating public resources. It is now time for economists and other researchers to hone and sharpen their analytical tools to bring new insight into the best course of action once the federal government’s 25% debt-to-GDP ratio is realized. Should government debt continue to be reduced? Or would there be a better bang for the economic buck from cutting distortionary taxes and/or investing in key areas of public concern? Minister Flaherty has put the spotlight on this issue. Now let the debate begin!

Don Drummond, SVP & Chief Economist
416-982-2556

Derek Burleton, AVP & Senior Economist
416-982-2514