Abstract
The issue of productivity growth in Canada has received considerable attention reflecting its marked slowdown since the early 1970s and concerns about its implications for Canadian competitiveness. To better understand productivity developments in Canada, it is useful to decompose total factor productivity (TFP) into investment-specific productivity change (ISP) and technologically neutral productivity change (TNP). The gap in manufacturing productivity growth between Canada and the United States originates mostly in the strong performance of specific industries, such as electrical products and commercial and industrial machinery.
V. Debt Reduction and Fiscal Policy Alternatives in Canada1
1. Having tackled its fiscal deficit and with substantial surpluses in prospect on the basis of unchanged tax and spending policies, the federal government in Canada faces a critical decision on how to set its long-term fiscal objectives. The ratio of government debt to GDP in Canada remains high, particularly in relation to most other industrial countries (Figure 1). Further debt reduction should clearly be an important objective, but how much and how rapidly debt should be reduced are important unanswered questions. The experiences of other countries with debt targets, as well as economic theory, do not provide much guidance on answering these questions. For those countries adopting targets for public debt, a great deal of judgement, reflecting the country’s circumstances, was used in setting these targets. In the economic literature, specific conclusions about optimal public debt levels have to be derived from hard-to-quantify economic tradeoffs and alternative criteria for evaluating social welfare, with results varying widely depending on the approach adopted and the parameters assumed in the models. More pragmatic approaches, based on model simulations of alternative debt paths and analysis of the economic tradeoffs between them, suggest for Canada that it would be useful to bring the debt-to-GDP ratio down to 20–30 percent in the early part of the twenty-first century, which is a target easily within the government’s grasp.
Selected Countries: Government Debt, 1998
(Percent of GDP)1
Citation: IMF Staff Country Reports 2000, 034; 10.5089/9781451806922.002.A005
Sources: WEO September 1999; and IMF Government Financial Statistics.1 Figures for both federal and general government are provided for Canada; general government debt figures are shown for all other countries.A. International Experience with Government Debt Targets
2. In recent years, several industrialized countries have started to include targets for public debt in their fiscal policy objectives. New Zealand and the countries of the European Union are the most prominent examples. The adoption of specific debt targets has been motivated by several considerations, including concerns about increasing public expenditures as a result of aging populations.
3. In New Zealand, one of the legislated principles of fiscal management under the Fiscal Responsibility Act of 1994 is that public debt should be reduced to “prudent levels” in order to provide “a buffer against future adverse events.” In principle, the government should achieve debt reduction by running operating surpluses and, once the debt has been reduced to prudent levels, it should maintain an operating balance on average over the business cycle.2 The Act provides the government with scope for flexibility in setting the debt target and in deciding how to achieve it. The Act leaves it to the government to interpret many of the key terms, including the level of debt that is “prudent.”3 The government is not legally bound to observe the debt target at all times, but it is required to explain any departures from the legislated principles (including debt reduction), as well as how and when it plans to reverse such departures. In practice, the debt target in New Zealand has been derived as a residual based on the target for projected expenditures (which in part reflects demographics) and the desired level of taxation. With rapid progress being achieved in debt reduction in recent years, the government has steadily reduced its debt target. The initial target, following passage of the Act, was to reduce net public debt to below 30 percent of GDP by 1996/97. With this target well in hand and continued rapid progress in debt reduction (debt was 24 percent of GDP in 1997/98), the target was revised down to below 15 percent of GDP in 1998.
4. The debt target for European Union countries is part of the convergence criteria in the Maastricht Treaty. As such, the debt target functions as a reference value: specifically, governments are required to aim at keeping the gross debt of the general government below 60 percent of GDP. The target is intended to be consistent with a sustainable fiscal position in the member states. The reference value of 60 percent happens to coincide with the average level of general government gross debt in the European Union countries at the time that the Maastricht Treaty was drafted in 1991. Although the actual debt target was chosen with a large measure of judgment, it has subsequently been pointed out that the target is consistent with a steady state for the level of debt when the general government deficit is 3 percent of GDP and nominal output growth is 5 percent (e.g., Buiter et al., 1993). In turn, such a level for the general government fiscal deficit would be consistent with the Maastricht Treaty requirement of deficits at no more than 3 percent of GDP, which is equal to the average rate of public sector investment for the EU countries in the period 1974 to 1991. Hence, it can be argued that the deficit target for the EU countries has been established on the basis of a golden-rule principle of public finance.4 This principle suggests that the accumulation of public sector debt should mirror the accumulation of physical capital by the public sector, which is assumed to increase the productive potential of the economy. However, the European Union has never officially referred to the golden rule when justifying the public sector deficit target.5
B. Economic Literature on Optimum Debt-to-GDP Ratios
5. Conclusions in the economic literature regarding the optimum quantity of government debt vary widely based on the kind of approach that is taken. At one extreme are the models characterized by Ricardian equivalence, in which the size of the debt is of no consequence for the allocation of resources, provided that the debt does not grow at an unsustainable pace. The theoretical assumptions to reach such a conclusion are, however, restrictive. Under more plausible assumptions—such as the presence of liquidity constraints and/or distortionary taxation—Ricardian equivalence breaks down and there is in fact a tradeoff between financial spending through debt or taxes. Considerations then arise regarding an optimum time-path for the debt. Empirically, Ricardian equivalence has rarely been observed in its strong form, and most empirical research has focussed on determining how large the departures from Ricardian equivalence actually are. It is thus appropriate to focus on approaches in which there is some meaningful tradeoff between debt and other forms of financing government expenditures.
6. The literature in this area has focused mainly on optimum fiscal positions, and thus indirectly shed light on the appropriate size of government debt (since in an accounting sense the steady-state level of the debt is determined by the budget balance and the growth rate of the economy). There are two main approaches to assessing optimum fiscal balances. In one approach, it is argued that the government should seek to minimize tax distortions over time by keeping tax rates relatively constant (e.g., Kydland and Prescott, 1980). Government debt then functions as a means for minimizing tax distortions over time (“tax smoothing”). For example, if large unfunded government liabilities are envisaged in the future, then debt reduction in the near term would be appropriate. By reducing debt-servicing costs, the government would free up resources with which to meet its liabilities without resorting to sharply higher tax rates in the future. In the other approach, the government seeks to maximize intergenerational fairness. When Ricardian equivalence does not hold, debt places a burden on the capital stock of future generations, and the level of the debt should be carefully chosen in such a way as to distribute the burden evenly across generations. A key parameter in such an optimization is the rate of time preference, which represents the importance placed by the current generation on its lifetime consumption and welfare relative to the consumption and welfare of future generations.
7. Another part of the literature has focused directly on the optimum quantity of government debt, and tried to determine the optimum quantity based on the tradeoff between the benefits and costs of government debt (e.g., Aiyagari and McGrattan, 1998). The benefits of government debt include the role that it plays in enhancing the liquidity of households by providing an additional means of smoothing consumption and by loosening borrowing constraints. The costs include the adverse wealth distribution and incentive effects of the taxes needed to repay the debt, as well as the crowding out of capital through higher interest rates and the lowering of private consumption. The optimum quantity of debt depends positively on the effectiveness of debt in smoothing out private consumption, negatively on the extent to which debt crowds out private capital, and negatively on the extent of the disincentive effects of distortionary taxes.
8. None of the above theoretical approaches lead to strong conclusions about how large the government debt should be. The government is assumed to try and maximize social welfare, with the social welfare function usually involving a tradeoff between equity and efficiency and both an intra-and an intertemporal dimension. Optimization involves several unobservable parameters—such as the weights that are placed on competing objectives, the rate of time preference, and the elasticity of labor supply with respect to tax rates—and the results can vary significantly depending on the assumptions about these parameters. In addition, in the optimum debt calculations derived by Aiyagari and McGrattan (1998), the welfare function calibrated for the U.S. economy is found to be relatively flat. Thus, the welfare cost of having a government debt equivalent to 0 percent of GDP instead of 60 percent of GDP (which is found to be the optimal level) is estimated to be less than 0.1 percent of total consumption.6 If government debt were to increase to 100 percent of GDP, only a marginal (0.02 percent of consumption) additional loss in welfare would result. The small welfare losses incurred by deviations from the optimal debt level suggest that undue importance should not be attached to the specific value of the debt ratio that emerges from this analysis.
C. Alternative Debt-Reduction Paths and Tradeoffs
9. While acknowledging the lack of clear conclusions from economic theory, a recent empirical analysis of Canadian federal government finances, which takes a relatively pragmatic approach, provides some target ranges for the debt ratio over the longer term. Scarth and Jackson (1997) note that considerations of economic efficiency do not lead to any specific conclusion about the size of the debt. However, considerations of intergenerational equity would call for reducing federal government debt to 20–25 percent of GDP. In their analysis, debt reduction raises living standards of future generations by allowing for lower future taxes and by reducing foreign debt-service obligations.7 The positive effect of debt reduction on living standards are judged to offset what otherwise would be a negative effect arising from the projected aging of the Canadian population.
10. Using a simple calibrated model for the Canadian economy, Robson and Scarth (1999) demonstrate the importance of how a particular debt target is achieved, given uncertainties about future economic outcomes and about the structure of the economy. A large number of simulations were run based on two broad budget approaches that have the objective of reducing the debt-to-GDP ratio to 30 percent over the next 15–20 years and maintaining it at that level, which the authors suggest would aim to accommodate foreseeable fiscal pressures associated with the aging of the population.8 The first approach is described as “drifting,” under which the government is assumed to target budget balance when the economy is strong and a budget deficit when the economy is weak. The second approach involves the government following an explicit debt-reduction objective. Within the latter approach, the government is given the choice between paths for the annual budget that emphasize debt reduction in the early years (i.e., target budget surpluses) or aim for budget balance. Within each of the two debt-reduction approaches the government faces two further choices: whether to maintain the budget target unchanged in the face of the business cycle (“rigid”) or to alter it (“flexible”).9
11. To capture the potential effects of uncertainties, Robson and Scarth incorporate in each simulation its own set of random economic disturbances, designed to mimic economic cycles and temporary shocks, and its own set of values for key parameters in the model. The estimates reported in Table 1 are the median values derived from the multiple scenarios. The results suggest that the debt-reduction approaches deliver substantially greater benefits—i.e., lower levels of government debt, higher budgetary payoffs to be used for tax cuts or increased government spending, and a greater increase in per capita consumption—than the “drifting” approach. Within the debt-reduction approaches, the approach that targets surpluses delivers larger benefits than the one that targets budget balance. In addition, the debt-reduction approaches on average result in debt-to-GDP ratios lower than 30 percent, which would leave more room in latter years for the budget deficit to expand to meet unforeseen costs associated with population aging.
Canada: Alternative Approaches to Debt Reduction
(In percent, unless otherwise noted)
When output is equal to or above potential, the government is assumed to target a balanced budget; when output is below potential, the government is assumed to target a deficit of 0.4 percentage point of GDP for each percentage point of the output gap. After 15 years, demographic pressures are assumed to move the budget in five equal annual steps to a deficit of 1¼ percent of GDP, which would be consistent with a debt-to-GDP ratio of 30 percent, provided annual nominal GDP growth over the period averages a bit over 4 percent.
The government targets annual budget surpluses of 1 percent of GDP over the period 2000/01 to 2005/06, with the surpluses then gradually declining and the budget being in balance by 2015. The budget would move in equal steps over the subsequent five years to a deficit of 1¼percent, of GDP, which would be consistent with maintaining a debt-to-GDP ratio of 30 percent indefinitely.
The government targets budget balance in 2000/01 to 2015/16, then shifts (like in the other alternative budget paths) over a five-year period into a deficit 1¼ percent of GDP, which would be consistent with maintaining a debt-to-GDP ratio of 30 percent.
Canada: Alternative Approaches to Debt Reduction
(In percent, unless otherwise noted)
| Debt Reduction Approaches | ||||||
|---|---|---|---|---|---|---|
| Target Budget | Target Budget | |||||
| “Drifting” 1/ | Surplus 2/ | Balance 3/ | ||||
| Approach | Rigid | Flexible | Rigid | Flexible | ||
| After 15 years | ||||||
| Debt/GDP ratio | 32.9 | 24.1 | 23.0 | 31.3 | 30.4 | |
| Budget payoff (percent of GDP) | 2.4 | 2.8 | 2.9 | 2.3 | 2.4 | |
| Rise in real per-capita consumption | 12.3 | 13.3 | 13.2 | 12.6 | 12.6 | |
| After 20 years | ||||||
| Debt/GDP ratio | 30.9 | 23.0 | 22.0 | 28.8 | 28.0 | |
| Budget payoff (percent of GDP) | 3.8 | 4.2 | 4.2 | 3.8 | 3.8 | |
| Rise in real per-capita consumption | 16.8 | 17.7 | 17.7 | 17.2 | 17.2 | |
When output is equal to or above potential, the government is assumed to target a balanced budget; when output is below potential, the government is assumed to target a deficit of 0.4 percentage point of GDP for each percentage point of the output gap. After 15 years, demographic pressures are assumed to move the budget in five equal annual steps to a deficit of 1¼ percent of GDP, which would be consistent with a debt-to-GDP ratio of 30 percent, provided annual nominal GDP growth over the period averages a bit over 4 percent.
The government targets annual budget surpluses of 1 percent of GDP over the period 2000/01 to 2005/06, with the surpluses then gradually declining and the budget being in balance by 2015. The budget would move in equal steps over the subsequent five years to a deficit of 1¼percent, of GDP, which would be consistent with maintaining a debt-to-GDP ratio of 30 percent indefinitely.
The government targets budget balance in 2000/01 to 2015/16, then shifts (like in the other alternative budget paths) over a five-year period into a deficit 1¼ percent of GDP, which would be consistent with maintaining a debt-to-GDP ratio of 30 percent.
Canada: Alternative Approaches to Debt Reduction
(In percent, unless otherwise noted)
| Debt Reduction Approaches | ||||||
|---|---|---|---|---|---|---|
| Target Budget | Target Budget | |||||
| “Drifting” 1/ | Surplus 2/ | Balance 3/ | ||||
| Approach | Rigid | Flexible | Rigid | Flexible | ||
| After 15 years | ||||||
| Debt/GDP ratio | 32.9 | 24.1 | 23.0 | 31.3 | 30.4 | |
| Budget payoff (percent of GDP) | 2.4 | 2.8 | 2.9 | 2.3 | 2.4 | |
| Rise in real per-capita consumption | 12.3 | 13.3 | 13.2 | 12.6 | 12.6 | |
| After 20 years | ||||||
| Debt/GDP ratio | 30.9 | 23.0 | 22.0 | 28.8 | 28.0 | |
| Budget payoff (percent of GDP) | 3.8 | 4.2 | 4.2 | 3.8 | 3.8 | |
| Rise in real per-capita consumption | 16.8 | 17.7 | 17.7 | 17.2 | 17.2 | |
When output is equal to or above potential, the government is assumed to target a balanced budget; when output is below potential, the government is assumed to target a deficit of 0.4 percentage point of GDP for each percentage point of the output gap. After 15 years, demographic pressures are assumed to move the budget in five equal annual steps to a deficit of 1¼ percent of GDP, which would be consistent with a debt-to-GDP ratio of 30 percent, provided annual nominal GDP growth over the period averages a bit over 4 percent.
The government targets annual budget surpluses of 1 percent of GDP over the period 2000/01 to 2005/06, with the surpluses then gradually declining and the budget being in balance by 2015. The budget would move in equal steps over the subsequent five years to a deficit of 1¼percent, of GDP, which would be consistent with maintaining a debt-to-GDP ratio of 30 percent indefinitely.
The government targets budget balance in 2000/01 to 2015/16, then shifts (like in the other alternative budget paths) over a five-year period into a deficit 1¼ percent of GDP, which would be consistent with maintaining a debt-to-GDP ratio of 30 percent.
12. The Government’s Fall 1999 Economic and Fiscal Update provides a framework for analyzing alternative debt-reduction paths.10 The five-year projections in the Update are status quo estimates based on current tax policies and an assumption that program spending will rise in line with inflation and population growth. Built into the spending projections is a $3 billion contingency reserve and an explicit allowance for economic prudence. To look at the longer-term consequences of alternative debt paths, these status quo projections were extended through 2014/15.11 If all of the status quo surpluses after 1999/2000 were used for debt reduction, net federal government debt would be eliminated by 2011/12, and the government would accumulate net assets equivalent to 20 percent of GDP by 2014/15 (Figure 2). If instead the government continued to target ex ante budget balance (including a $3 billion contingency reserve) and, ex post, used the contingency reserve to pay down debt, in the absence of significant adverse economic shocks, the budget would tend to be in surplus each year by $3 billion (roughly ¼ percent of GDP). On this basis, the debt-to-GDP ratio would fall to around 29 percent by 2014/15. Alternatively, if more ambitious surpluses of ½ percent or 1 percent of GDP were pursued, then the debt ratio would drop to around 24 and 20 percent of GDP, respectively, by 2014/15.
Canada: Federal Government Net Debt Under Alternative Fiscal Scenarios, 1998–2015
(In percent of GDP)
Citation: IMF Staff Country Reports 2000, 034; 10.5089/9781451806922.002.A005
List of References
Aiyagari, S. Rao and Ellen McGrattan, 1998, “The Optimum Quantity of Debt,” in Journal of Monetary Economics, Vol. 42, pp. 447–469 (Amsterdam: Elsevier Science Publishers).
Auditor General of Canada, 1998, Report of the Auditor General of Canada—April 1998, Chapter 6, “Population Aging and Information for Parliament: Understanding the Choices,” (Ottawa: Office of the Auditor General of Canada).
Barro, Robert, 1979, “On the Determination of the Public Debt,” in Journal of Political Economy, Vol. 87, No. 5, Part 1, pp. 940–971, (Chicago: University of Chicago Press).
Buiter, Willem, Giancarlo Corsetti, and Nouriel Roubini, 1993, “Maastricht’s Fiscal Rules,” in Economic Policy, April, pp. 56–100, (Cambridge: Cambridge University Press).
Kydland, Finn and Edward Prescott, 1980, “A Competitive Theory of Fluctuations and the Feasibility and Desirability of Stabilization Policy,” in Stanley Fischer (editor), Rational Expectations and Economic Policy, (Chicago: University of Chicago Press).
New Zealand Treasury, 1999, “New Zealand Economic and Financial Overview 1999,” (New Zealand Treasury).
New Zealand Treasury, 1995a “Fiscal Responsibility Act 1994: An Explanation,” (New Zealand Treasury).
New Zealand Treasury, 1995b, “New Zealand Economic and Financial Overview 1995,” (New Zealand Treasury).
Organisation for Economic Co-operation and Development, 1999, “Canada: Annual Review, 1998–1999,”, (Paris: OECD).
Robson, William and William Scarth, 1999, “Accident Proof Budgeting: Debt-Reduction Payoffs, Fiscal Credibility, and Economic Stabilization,” (Toronto: C.D. Howe Institute).
Scarth, William and Harriet Jackson, 1997, “The Target Debt-to-GDP Ratio: How Big Should It Be? And How Quickly Should We Approach It?” in Thomas J. Courchene and Thomas A. Wilson (editors), Fiscal Targets and Economic Growth, (Kingston, Ontario: John Deutsch Institute for the Study of Economic Policy).
Swagel, Phillip, Brenda Gonzalez-Hermosillo, and Yutong Li, 1998, “Macroeconomic Effects of Government Debt in Canada,” Chapter V, Canada: Selected Issues, IMF Staff Country Report 98/55.
Wyplosz, Charles, 1991, “Monetary union and fiscal policy discipline” in European Economy – special edition No. 1, (Brussels, Belgium: Commission of the European Communities).
Prepared by Vivek Arora and Anders Matzen.
By focusing on operating surpluses, debt reduction could not be achieved simply through the sale of government assets. By looking at the average budget balance over the business cycle, there would be some flexibility to operate counter-cyclical fiscal policy while still ensuring that the government does not borrow on an extended basis (see New Zealand Treasury, 1995a).
In fact, it is recognized that no single level of the debt is likely to be considered prudent at all times, since such a determination would depend on factors that change over time, including the structure of the economy, its vulnerability to shocks, demographic pressures, and the costs of debt servicing.
Such a “golden rule” is explicitly a part of the fiscal strategy in the United Kingdom. The main purpose of the rule is to ensure that public sector investment does not get squeezed out in situations where the budget is in overall deficit. Containing the level of government debt is ensured by the second element of the U.K.’s fiscal strategy, the debt sustainability rule, which requires public sector net debt to remain below 40 percent of GDP.
The golden rule principle of public sector finance is mentioned in Wyplosz (1991). Two qualifications would apply to the golden rule argument however. First, the public sector deficit is compiled on the basis of nominal interest expenditures. Hence an inflation rate higher than zero would imply that the “real” deficit was lower than public sector investment. Second, consideration needs to be given to the depreciation of public capital, as well as the quality of public sector investment.
In 1998, the ratio of U.S. government debt to GDP was 44 percent, compared to the estimated “optimum” level of 60 percent.
The model treats labor supply as exogenous and does not allow for the possibility that lower taxes may increase labor supply and national income, nor does it allow for the gains (such as the positive effects on output and consumption) arising from the lower interest rates associated with debt reduction. While endogenizing the labor supply response would raise living standards, the authors argue that lower interest rates would spur output but may discourage private saving, so that the long-run benefits for living standards may be questionable.
The authors suggest that achievement of such a debt target over this timeframe would be sufficient to meet the budgetary implications of population aging as estimated in a recent report by the Canadian Auditor General.
Specifically, the government is assumed in the “flexible” case to increase or decrease the budget target by 0.4 percentage points of GDP for each percentage point by which output is above or below potential, respectively.
The alternative debt paths discussed here are static projections in that they are all based on the same set of economic assumptions and do not factor in the effects of debt reduction on the economy. The debt ratios are, however, of the same order of magnitude as suggested by the staff’s earlier work, which examined a similar set of scenarios and was based on a dynamic analysis which took into account the positive effects of debt reduction on output (see Swagel et. al. (1998)).
In extending the projections, it is assumed that the output gap (potential output minus actual) will rise from around zero at the end of 2005 to 1 percent in 2009 and remain at that level thereafter. This assumption is made to effectively simulate the effects on the budget estimates of “normal” business cycles.
