Abstract
This Selected Issues paper assesses the long-term fiscal position of Canada. Simulations based on current tax and spending policies suggest that the fiscal position will remain favorable until well into the middle of the century, and relatively modest adjustments would be required to make these policies sustainable in the long term. The analysis also illustrates that these conclusions could be easily overturned if pressures to spend the planning surpluses that are expected to emerge in coming years are not resisted, and if measures are not put in place to contain the cost of health care.
I. Assessing the Long-Term Fiscal Position of Canada1
1. The Canadian government has set fiscal sustainability and intergenerational equity as important policy objectives, and these goals have helped shape the broader fiscal strategy, as well as specific tax and expenditures policies. However, in Canada, as in most other industrial countries, demographic trends and pressures on health care systems are expected to make fiscal sustainability an increasingly challenging goal to achieve. This chapter assesses the extent to which Canadian fiscal policy of recent years has left the fiscal position well-placed to meet these challenges.
2. The results below are generally encouraging. Simulations based on current tax and spending policies suggest that the fiscal position will remain favorable until well into the middle of the century, and that relatively modest adjustments would be required to make these policies sustainable in the long run. At the same time, however, the analysis also illustrates that these conclusions could be easily overturned if pressures to spend the planning surpluses that are expected to emerge in coming years are not resisted and if measures are not put in place to contain the cost of health care.
A. Background and Assumptions
3. The analysis is based on a generational accounting framework, which first requires estimating the distribution of taxes and transfers across age cohorts in the current year.2 Long-term fiscal projections are then constructed by assuming that taxes and transfers paid by the average person in each age cohort, and the level of other government outlays received by the average person in each cohort, increase in line with productivity. In the case of some expenditure or transfer categories, however, different rates of growth may be assumed to take into account policy commitments or other aspects of these programs. Given projections for the size and composition of the population, as well as for future productivity growth, these assumptions determine the evolution of aggregate public sector revenues and expenditure.3
4. Previous studies on Canadian long-term fiscal stance have typically projected revenues and expenditure of Canada’s federal and provincial governments separately. However, the analysis below focuses on the consolidated government accounts, allowing a more comprehensive representation of the broader government sector, which is particularly important given the federal nature of Canada’s fiscal system (Table 1).4 In this chapter, the government sector is defined to include federal and provincial/territorial governments, the Canada and Quebec Pension Plans (CPP/QPP), non autonomous pension plans, health and social service institutions, and universities and colleges.
Canada: Revenues and Expenditures of the Consolidated Government, 2001–2002 1/
(In percent of GDP)
Includes Federal, Provincial, Territorial, and Local governments plus Canada Pension Plan/Quebec Pension Plan.
For year 2001/2002. Source: National Balance Sheet Accounts.
Canada: Revenues and Expenditures of the Consolidated Government, 2001–2002 1/
(In percent of GDP)
| Total revenue | 41.8 | Total expenditure | 41.0 | |||
| Income taxes | 16.7 | General government services | 1.3 | |||
| Personal | 12.8 | Protection of persons and property | 3.1 | |||
| Corporation income taxes | 3.5 | Transportation and communication | 1.6 | |||
| Other | 04 | Health | 7.0 | |||
| Consumption taxes | 8.0 | Social services | 12.7 | |||
| Property and related taxes | 3.7 | Social Assistance | 8.7 | |||
| Other taxes | 14 | Income maintenance | 0.6 | |||
| Payroll taxes | 0 7 | Family allowances and child tax benefits | 0.7 | |||
| Health and drug insurance premiums | 0.2 | Employment Insurance benefits | 1.2 | |||
| Contributions to social security plans | 5.4 | Old age security and other public pensions | 2.3 | |||
| Employment Insurance contributions | 1.7 | Canadian Pension Plan benefits | 1.9 | |||
| Contributions to the Canadian Pension Plan | 2.0 | Quebec Pension Plan benefits | 0.6 | |||
| Contributions to the Quebec Pension Plan | 0.7 | GST tax credits | 03 | |||
| Contributions to the employees pension plans | 0.4 | Other social assistance | 1.1 | |||
| Contributions to workers’ compensation boards | 0.6 | Workers’ compensation benefits | 0.5 | |||
| Sales of goods and services | 3.1 | Employee pension plan benefits | 1.2 | |||
| Investment income | 3.0 | Change in pension equity | 0.7 | |||
| Royalties and remitted trading profits | 1.3 | Veterans’ benefits | 0.2 | |||
| Interest income | 1.7 | Other social services | 1.4 | |||
| Other revenue from own sources | 0.3 | Education | 5.8 | |||
| Resource conservation and industrial development | 1.4 | |||||
| Environment | 0.8 | |||||
| Recreation and culture | 1.0 | |||||
| Labour, employment and immigration | 0.3 | |||||
| Housing | 0.4 | |||||
| Foreign affairs and international assistance | 0.4 | |||||
| Regional planning and development | 0.2 | |||||
| Research establishments | 0.2 | |||||
| Debt charges | 4.8 | |||||
| Other expenditures | 0.2 | |||||
| Surplus (deficit) | 0.8 | |||||
| Net Debt 2/ | 63.5 | |||||
Includes Federal, Provincial, Territorial, and Local governments plus Canada Pension Plan/Quebec Pension Plan.
For year 2001/2002. Source: National Balance Sheet Accounts.
Canada: Revenues and Expenditures of the Consolidated Government, 2001–2002 1/
(In percent of GDP)
| Total revenue | 41.8 | Total expenditure | 41.0 | |||
| Income taxes | 16.7 | General government services | 1.3 | |||
| Personal | 12.8 | Protection of persons and property | 3.1 | |||
| Corporation income taxes | 3.5 | Transportation and communication | 1.6 | |||
| Other | 04 | Health | 7.0 | |||
| Consumption taxes | 8.0 | Social services | 12.7 | |||
| Property and related taxes | 3.7 | Social Assistance | 8.7 | |||
| Other taxes | 14 | Income maintenance | 0.6 | |||
| Payroll taxes | 0 7 | Family allowances and child tax benefits | 0.7 | |||
| Health and drug insurance premiums | 0.2 | Employment Insurance benefits | 1.2 | |||
| Contributions to social security plans | 5.4 | Old age security and other public pensions | 2.3 | |||
| Employment Insurance contributions | 1.7 | Canadian Pension Plan benefits | 1.9 | |||
| Contributions to the Canadian Pension Plan | 2.0 | Quebec Pension Plan benefits | 0.6 | |||
| Contributions to the Quebec Pension Plan | 0.7 | GST tax credits | 03 | |||
| Contributions to the employees pension plans | 0.4 | Other social assistance | 1.1 | |||
| Contributions to workers’ compensation boards | 0.6 | Workers’ compensation benefits | 0.5 | |||
| Sales of goods and services | 3.1 | Employee pension plan benefits | 1.2 | |||
| Investment income | 3.0 | Change in pension equity | 0.7 | |||
| Royalties and remitted trading profits | 1.3 | Veterans’ benefits | 0.2 | |||
| Interest income | 1.7 | Other social services | 1.4 | |||
| Other revenue from own sources | 0.3 | Education | 5.8 | |||
| Resource conservation and industrial development | 1.4 | |||||
| Environment | 0.8 | |||||
| Recreation and culture | 1.0 | |||||
| Labour, employment and immigration | 0.3 | |||||
| Housing | 0.4 | |||||
| Foreign affairs and international assistance | 0.4 | |||||
| Regional planning and development | 0.2 | |||||
| Research establishments | 0.2 | |||||
| Debt charges | 4.8 | |||||
| Other expenditures | 0.2 | |||||
| Surplus (deficit) | 0.8 | |||||
| Net Debt 2/ | 63.5 | |||||
Includes Federal, Provincial, Territorial, and Local governments plus Canada Pension Plan/Quebec Pension Plan.
For year 2001/2002. Source: National Balance Sheet Accounts.
5. The demographic projections adopt the “medium-growth” case prepared by Statistics Canada.5 Accordingly, during the next ten years, Canada’s demographic profile is expected to remain roughly unchanged. However, between 2010 and 2030, the old-age dependency ratio (the ratio of retirees to those in the labor force) begins to increase rapidly, from around 20 percent in 2010 to slightly above 42 percent in 2036, before stabilizing at around 44 percent after 2050.
6. Labor productivity in the simulations below is assumed to grow at a rate of 1½ percent per year. This is broadly in line with the assumption used in other long-term fiscal projections for Canada (Jackson and Matier, 2002), and is consistent with recent Statistics Canada data that show that business sector labor productivity grew at an average rate of 1.4 percent over the period 1981–2000.
7. Revenues: The simulations in this chapter take into account future tax reductions that have already been announced by both the federal and provincial governments. For the federal government, personal and corporate income taxes are projected to grow over the period 2003–2008 in line with the October 2002 Economic and Fiscal Update. For the provinces, projections provided in the Conference Board of Canada’s report Fiscal Prospect for the Federal and Provincial/Territorial Governments (2002) were used. Contributions to the Employment Insurance (EI) scheme are also assumed to move in line with the forecasts contained in the Economic and Fiscal Update until 2008.6 After 2008, all per-capita taxes and social security contributions are assumed to increase with labor productivity.
8. An exception to this assumption involves the CPP and QPP pension schemes. For the period 2002–2075, contributions to (and transfers from) the CPP are assumed to grow in line with the Office of the Chief Actuary’s projections (OSFI, 2002). These projections show that, given the economic and demographic assumptions used, the CPP is actuarially balanced, i.e., transfers net of contributions from these pension plans are broadly equal to current assets on a net present value basis. After 2075, CPP contributions are assumed to grow in line with benefits. Projections for the QPP were also based on the Office of the Chief Actuary’s report, and benefits and contributions from this scheme are assumed to grow in line with those for the CPP.7
9. In the projections, the ratio of overall tax revenues to GDP responds to changes in the composition of the population. For example, revenues from income taxes are projected to remain constant as a share of output, as both revenues and GDP are affected by the decline of the working-age population. Conversely, property tax revenues and consumption tax revenues increase as a share of GDP, owing to the increase in the elderly population relative to those in the working-age cohorts (Figure 1).
Canada: Tax Revenues
(In percent of GDP)
Citation: IMF Staff Country Reports 2003, 034; 10.5089/9781451806953.002.A001
10. Expenditure: Employment Insurance benefits and elderly benefits are assumed to grow in line with the Economic and Fiscal Update over the period 2003–2008. Non-contributory social benefits—including income maintenance, family allowances and child tax benefits, old age security, and other elderly benefits after 2008—are assumed to be partially indexed to real wages. This represents a deviation from a “current-policy” assumption, since these programs presently are indexed only to CPI inflation. However, assuming CPI indexation over the long run implies that these payments would decline steadily as a share of GDP per capita, which would seem unrealistic and inconsistent with past experience in which ad-hoc increases in the benefit rates have been introduced.8 Nonetheless, in the baseline scenario, these per capita benefits are assumed to rise by only half the rate of labor productivity growth, which still implies only partial indexation to wage growth and a declining ratio of these payments to GDP.9
11. Pensions from government pensions plans are also assumed to be partially indexed to wages, in order to approximate the fact that benefits under these schemes are based on wages prior to retirement and are indexed to inflation thereafter.10 Employment Insurance benefits and workers’ compensation benefits paid on average to each member of current and future cohorts are assumed to rise in line with contributions, which in turn grow in line with productivity.11
12. Based on these assumptions, spending on social services is projected to rise from 12¾ percent of GDP in 2002 to a peak of around 14½ percent in 2035 and to slowly decline thereafter, as the effect of the partial wage indexation of social transfers offsets the effect of population aging. Figure 2 also shows the paths of social services expenditure under two alternative indexation rules. In the first, non-contributory transfers increase only with prices on a per-beneficiary basis, and spending falls to below 10 percent of GDP by 2075. In the second, these transfers are fully indexed to wages and increase more rapidly, converging to just above 17 percent of GDP by 2075.
Canada: Social Transfers
(In percent of GDP)
Citation: IMF Staff Country Reports 2003, 034; 10.5089/9781451806953.002.A001
13. Health outlays per each age cohort are assumed to grow in real terms at a rate that is ¼ percentage point faster than productivity growth. The growth of real per capita health expenditure implied by this assumption is broadly in line with historical experience.12 As a result of this assumption and demographic projections, aggregate health expenditure doubles over the next 50 years as a share of GDP, from 7 percent to around 14½ percent (Figure 3). It is noteworthy, however, that this increase is principally the result of population aging rather than the higher price inflation assumed for health care. Simulations based on unchanged population structure would cause health care spending to rise only modestly to around 8 percent of GDP by the end of the projection period.13
Canada: Health Care Spending
(In percent of GDP)
Citation: IMF Staff Country Reports 2003, 034; 10.5089/9781451806953.002.A001
14. Given the uncertainty surrounding projections for health care outlays and their importance for the estimated debt dynamics of the next section, two alternative scenarios are considered. In the first, defined as the tighter health-care spending scenario, average real per capita health expenditure is assumed to grow at the same rate as productivity, so that health care outlays as a share of GDP move solely in response to demographics. Health care spending in this case would grow at a slower rate than in the past, and would stabilize at 12 percent by 2050. In the second scenario, defined as the looser health-care spending scenario, larger price/technological pressures are assumed to cause spending on health care for each to cohort grow at an annual rate of 2 percent, ½ percentage point faster than productivity growth. In this case, the impact of population aging would combine with the other pressures to take public health expenditure to around 17 percent of GDP by 2075.
15. On the basis of the projected revenue and expenditure, it is possible to simulate a path for the fiscal primary balances. This then allows projections for net debt of the government sector (total liabilities less financial assets), assuming that all primary balances are applied to the previous year debt and debt servicing costs. For the purposes of the present exercise, the effective interest rate on net government debt, in real terms, is assumed to be constant at 5 percent.14
B. Debt Dynamics and Fiscal Sustainability
16. In this section, Canada’s long-term fiscal situation is assessed with regard to several criteria. First the debt dynamics are explored, to examine whether current tax and spending policies imply growth in the debt-to-GDP ratio to levels that would appear excessive in qualitative terms. Second, an assessment is made of the long-term fiscal imbalance, defined as the size of the immediate and permanent increase in income taxes needed to satisfy the intertemporal budget constraint—by ensuring that the present value of future primary surpluses matches the current level of net debt.15 Third, the question of generational equity is considered, including by examining the inequities that arise by targeting alternative debt-to-GDP ratios.
17. Debt dynamics; A baseline scenario was first simulated, in which it is assumed that the government projected surpluses—including the “planning surpluses’ projected in the federal government’s Fiscal Update—are not spent and are devoted to debt reduction. In this case, the overall surplus of the consolidated government rises from 0.8 percent of GDP in 2002 to 1.9 percent in 2008, while the primary balance rises from 4 percent to 4.2 percent of GDP during the same period. The baseline scenario implies a relatively favorable path for the debt ratio—the persistence of primary surpluses over next two decades allows net debt as a share of GDP to decline steadily from around 60 percent in 2002 to a low of slightly below 4 percent in 2027 (Figures 4 and 5). However, since then population aging begins to place greater pressures on expenditure, the debt ratio begins rising again and by 2050 the ratio is projected to return to its initial value (see Figure 5).
Canada: Primary Balances
(In percent of GDP)
Citation: IMF Staff Country Reports 2003, 034; 10.5089/9781451806953.002.A001
Canada: Net Debt
(In percent of GDP)
Citation: IMF Staff Country Reports 2003, 034; 10.5089/9781451806953.002.A001
18. These simulations, however, are sensitive to the underlying assumptions chosen. Even more benign scenarios can be constructed, including by assuming that transfers are indexed only to the price level (rather than partially indexed to wages), that growth of health care spending is constrained, or that productivity growth is stronger (maintaining the baseline path for health care spending). In each of these cases, net debt falls below zero and the government remains in a net creditor position in the long run.
19. However, considerably less benign scenarios also can be envisaged. For example, if the bulk of the surpluses projected over the next five years is devoted to increases in expenditure, government outlays are placed on a permanently higher level, with a permanently lower primary balance. As a result, much more modest debt reduction is achieved over the next 20 years, and the simulation illustrate that additional policy measures would be needed to avoid rapid debt growth thereafter.16 Alternatively, if social transfers are required to keep pace with productivity gains and are indexed to wages, the debt ratio falls much slower in the nearer term and rises well above 100 percent of GDP by mid-century. Similarly, looser constraints on public health care outlays would significantly worsen the debt profile.
20. Fiscal solvency: As illustrated in Table 2, the re-emergence of fiscal deficits in the longer term means that even the baseline scenario does not fully meet the standard fiscal solvency criterion.17 In other words, the net present value of future primary surpluses is not sufficient to cover the current outstanding net debt of the government sector, i.e., these surpluses do not satisfy the intertemporal budget constraint. The estimates suggest that closing this fiscal gap would require an immediate and permanent increase in taxes roughly equivalent to a 1.4 percent increase in the level of income taxes—this would be equivalent to C$2.5 billion or about 0.2 percent of GDP in 2002. Making this adjustment would allow a faster reduction of the net debt-to-GDP ratio than in the baseline and would facilitate the ability of the government sector to absorb the impact of demographic change on the budget through borrowing.18
Indicators of the Fiscal Gap
(In percent of income tax revenue)
Immediate and permanent change in income tax revenues (in percent) needed to satisfy the intertemporal budget constraint.
The change in income tax revenues (in percent) required for every period from 2003 to 2010 to bring the net debt/GDP ratio to the target level at the terminal date and, thereafter, the immediate and permanent change in income tax revenues (in percent) needed to satisfy the intertemporal budget constraint.
Indicators of the Fiscal Gap
(In percent of income tax revenue)
| Income Tax Gap 1/ | |||
| Baseline scenario | 1.4 | ||
| Higher productivity growth (1.75 percent) | -4.8 | ||
| Lower interest rate (4 percent) | 2.3 | ||
| Higher interest rate (6 percent) | 1.2 | ||
| Spending projected surpluses over 2003-2008 | 10.4 | ||
| Looser health care spending | 7.5 | ||
| Tighter health care spending | -3.8 | ||
| Non-contributory benefits indexed to prices | -5.3 | ||
| All benefits fully indexed to wages | 11.6 | ||
| Income Tax Gap with an Explicit Debt/GDP Target 2/ | |||
| Net debt/GDP at 30 percent in 2010 | |||
| Tax change required during 2003-2010 | 5.6 | ||
| Tax change required after 2010 | -5.2 | ||
| Net debt/GDP at 50 percent in 2010 | |||
| Tax change required during 2003-2010 | -7.2 | ||
| Tax change required after 2010 | 12.0 | ||
Immediate and permanent change in income tax revenues (in percent) needed to satisfy the intertemporal budget constraint.
The change in income tax revenues (in percent) required for every period from 2003 to 2010 to bring the net debt/GDP ratio to the target level at the terminal date and, thereafter, the immediate and permanent change in income tax revenues (in percent) needed to satisfy the intertemporal budget constraint.
Indicators of the Fiscal Gap
(In percent of income tax revenue)
| Income Tax Gap 1/ | |||
| Baseline scenario | 1.4 | ||
| Higher productivity growth (1.75 percent) | -4.8 | ||
| Lower interest rate (4 percent) | 2.3 | ||
| Higher interest rate (6 percent) | 1.2 | ||
| Spending projected surpluses over 2003-2008 | 10.4 | ||
| Looser health care spending | 7.5 | ||
| Tighter health care spending | -3.8 | ||
| Non-contributory benefits indexed to prices | -5.3 | ||
| All benefits fully indexed to wages | 11.6 | ||
| Income Tax Gap with an Explicit Debt/GDP Target 2/ | |||
| Net debt/GDP at 30 percent in 2010 | |||
| Tax change required during 2003-2010 | 5.6 | ||
| Tax change required after 2010 | -5.2 | ||
| Net debt/GDP at 50 percent in 2010 | |||
| Tax change required during 2003-2010 | -7.2 | ||
| Tax change required after 2010 | 12.0 | ||
Immediate and permanent change in income tax revenues (in percent) needed to satisfy the intertemporal budget constraint.
The change in income tax revenues (in percent) required for every period from 2003 to 2010 to bring the net debt/GDP ratio to the target level at the terminal date and, thereafter, the immediate and permanent change in income tax revenues (in percent) needed to satisfy the intertemporal budget constraint.
20. Table 2 also shows the sensitivity of the fiscal gap calculations to macroeconomic and fiscal assumptions. Higher productivity growth—by ¼ percentage point—(without a commensurate growth in health spending) would eliminate the gap completely and would allow immediate and permanent tax cuts and/or an increase in expenditure consistent with fiscal solvency. Tighter limits on health care spending and indexation of non-contributory benefits solely to prices would have the same effect. Notably, assuming that the federal government spends the bulk of its planning surpluses over 2003–2008, inducing a permanent increase in spending, causes the fiscal gap to increase significantly, to the equivalent of 10 percent of revenues.
21. Intergenerational equity: Closing the fiscal gap through an immediate and permanent increase in taxes would be consistent with a policy of long-term tax smoothing, as it ensures that no further tax increases (including on future generations) are required to satisfy the government intertemporal budget constraint.19 In addition to sharing the burden of closing the fiscal gap equally across generations—at least those born after the start date of the simulations—tax smoothing also minimizes distortions to labor-leisure decisions (Barro, 1979).
22. The intergenerational redistribution implied by alternative fiscal paths can be illustrated by considering the impact of alternative debt targets. For example, aiming at a 30 percent net debt-to-GDP ratio by 2010 would require a 5½ percent increase in income taxes starting from 2002, since this would involve a more ambitious fiscal path than the baseline, which only brings the net debt ratio to around 35 percent of GDP by 2010. This adjustment would have a significant impact on the fiscal gap calculated in 2010—with a lower level of debt, fiscal solvency could be achieved with a permanent 5¼ percent cut in taxes after 2010.
23. Alternatively, a looser fiscal path, one that aimed at a 50 percent net debt-to-GDP ratio by 2010, would be consistent with a cut in income tax revenues of 7¼ percent for every year to 2010. However, the burden on future generations to close the fiscal gap thereafter would be the equivalent of a 12¼ percent permanent increase in income tax revenues. Similarly, policy that aimed at maintaining the debt/ratio constant at its present value would place a greater burden on future generations.
C. Conclusions
24. The analysis above illustrated the possible effects of population aging on the consolidated fiscal accounts of the federal and provincial governments under alternative economic and policy assumptions. Three main conclusions are suggested:
The substantial improvement in the fiscal position achieved during the past decade—including the achievement of surpluses at both the federal and provincial levels, and the reform of the public pension systems—has left the Canadian government sector relatively well placed to withstand demographic pressures, especially compared to many other industrialized countries.20 Baseline scenarios illustrate that, even on the assumption of more rapid increases in health care as the population ages, the government sector debt/GDP ratio would continue to decline and remain at a low level for the next 20-30 years.
At the same time, however, the simulations illustrate that important fiscal risks and policy challenges remain. Despite present policies imply only a modest intertemporal fiscal imbalance, the relatively benign scenarios described above are vulnerable to the assumptions regarding productivity growth, the indexation of social programs, and the degree to which real per capita health care expenditure grows in excess of real income.
In addition, the simulations also illustrate the risks associated with policies that allow planning surpluses that are projected for the next five years to be used to boost spending rather than contribute to debt reduction. Increased outlays on health care and other programs in coming years, especially if not easily reversed in subsequent years, could compromise the achievement of fiscal sustainability and intergenerational equity.
These points underscore the importance of ensuring that the recent debate over health care reform results in substantive improvements in the capacity of the system to contain costs, and of resisting pressures to erode the fiscal surpluses that have been achieved in recent years and slow the pace of debt reduction.
Appendix: Methodology and Data
Assume that Xt denotes the amount of revenues from personal income taxes in the base year t. This aggregate amount is allocated to age and gender groups using age-gender profiles derived from the taxfiler information gathered by Canada Custom and Revenue Agency (in particular, from Table 4 of the income statistics, reporting taxable employment income by age and gender). Ignoring the gender subscripts for simplicity, Xi,t denotes the personal income tax paid by members of the age group i at time t;
where Ri,t is the relative amount of the revenues that should be allocated to the age group i at time t. Denoting with Ni,t the number of individuals belonging to group i at time t, the personal income tax paid by a member of the age group i is:
This amounts to defining the aggregate Xt as:
To project Xt forward demographic projections are needed that show how the number of persons in each age group evolves over time. Keeping the relative age and gender profiles Ri constant for the whole projection period, and assuming that real taxes and expenditures for the average person in age group i increase in line with productivity at rate g future values of X (at time t prices) are obtained as:
which can be also put as:
The latter expression shows that aggregate taxes and expenditures will evolve in the future depending only on changes in the demographic structure of the population (reflecting both size and compositional effects) and on the productivity adjustment. While this is true in general, one also needs to take into account that several transfer payments in Canada have been de-indexed to wages and are currently only linked to inflation. In this case, there will be no adjustment for productivity growth (g = 0).
The age profiles for income tax revenue, payroll taxes, and health insurance premium were derived from the age and gender distribution of taxable employment income reported by the Income Statistics of the Canada Custom and Revenue Agency (2002). Contributions to social security plans were allocated using the age and gender distribution from the same source. The age and gender profiles for consumption taxes, property taxes, and other taxes were derived from Statistics Canada’s Social Policy Simulation Database and Model (SPSDM), as reported in “The Age Distribution of the Tax Transfer System in Canada,” Hicks (1998). Revenues from the sales of goods and services and from natural resources and remitted trading profits were allocated equally across age groups.
Several social service benefits (including old age security pensions, CPP and QPP benefits, and the employment insurance benefits) were allocated based on the age and gender distributions reported by the Income Statistics of the Canada Customs and Revenue Agency (2002). The age and gender profiles for child tax benefits and family allowances, other social assistance, and education were obtained from Hicks (1998). The profile for health spending was obtained from Health Canada (“Health Expenditure in Canada by age and sex, 1980-81 to 2000-2001”, 2001). All other expenditures were allocated equally across age groups (the age and gender profiles were flat).
References
Auerbach, Alan J., 1994, “The US Fiscal Problem: Where We Are, How We Got Here and Where We’re Going”, NBER Working Paper No. 4709
Auerbach Alan J., Laurence J. Kotlikoff and Willi Leibfritz, 1999, “Generational Accounting Around the World.” The University of Chicago Press.
Barro Robert 1979, “On the determination of Public Debt”, Journal of Political Economy, Vol.87.
Canadian Institute for Health Information, 2002, “Preliminary Provincial and Territorial Government Health Expenditure Estimates, 1974/75 to 2002/03”.
Congressional Budget Office, 2000, “The Long Term Budget Outlook”, Washington DC.
Hicks, Chantal 1998, “The Age Distribution of the Tax Transfer System in Canada,” in Government Finances and Generational Equity, Miles Corak, Statistics Canada.
Jackson Harriet and Chris Matier, 2002, “Public Finance Implications of Population Ageing: An Update”, Finance Canada.
Kennedy Suzanne and Chris Matier, 2002, “Comparing the Long term Fiscal Outlook for Canada and the United States Using Fiscal Gaps”, Paper presented at the Canadian Association for Business Economics & Moneco-Econtro combined conference, Kingston Ontario, August 29th.
King Philip and Harriet Jackson, 2000, “Public Finance Implications of Population Ageing”, Finance Canada Working Paper, 2000-08.
Matier Chris, Lisa Wu and Harriet Jackson, 2001, “Analyzing Vertical Fiscal Imbalance in a Framework of Fiscal Sustainability”, Finance Canada Working Paper, 2001-23.
OECD, 2001, “Health at a glance,” OECD, Paris.
OSFI, 2001, “Actuarial Report (18th) on the Canada Pension Plan as at 31 December 2000”.
Oreopoulos Philip, 1999, “Canada: On the Road to Fiscal Balance”, in “Generational Accounting Around the World”. The University of Chicago Press.
Prepared by Roberto Cardarelli.
For examples of this approach’s use for the United States, see the Congressional Budget Office (CBO, 2000). For Canada, see by King and Jackson (2000); Matier, Hu, and Jackson (2001); Matier and Jackson (2002) and Kennedy and Matier (2002). The Appendix describes the methodology and the data in more detail.
These projections do not incorporate any feedback between the fiscal position and macroeconomic variables, such as productivity growth or interest rates. They are best interpreted as an indication of the magnitude of long-term pressures for fiscal balances if current tax and spending policies remained unchanged, given long-term economic and demographic assumptions.
These data are produced by Statistics Canada under the Financial Management System (FMS). The FMS uses the public accounts of the Federal, Provincial, and Territorial governments and transforms them into a common format. It also defines government more broadly than the public accounts, as it includes agencies, boards, commissions, and funds which are used for government policy even if these entities are excluded from government budgets.
In this scenario, by 2026 fertility rate is 1.48, life expectancy is 80 for male and 84 for females; and immigration is 225,000. Population projections for the years beyond 2026 were obtained by assuming that the above parameters remain constant at their 2026 levels.
As these forecasts assume that the economy grows at potential by 2008, starting to project from this year helps “sterilize” the effect of the economic cycle on the results of the simulations.
The long-term real rate of return on the CPP assets used by the Chief Actuary’s report is around 4 percent. However, the baseline scenario for the debt dynamics and debt sustainability analysis below is based on an effective net interest rate of 5 percent. For consistency with this assumption, both CPP and QPP projected contributions have been adjusted to make sure that the present value of the net transfers at a 5 percent discount rate is equal to the stock of CPP and QPP assets at 2002.
A similar strategy has been followed by the CBO (2000). Its baseline long-term projections are based on official taxes and transfers forecasts for the next ten years, after which all taxes and transfers are assumed to be indexed to real wages, despite the fact that only few of these transfers are formally indexed to wages under current legislation. For this reason, CBO refers to the policy implied in its baseline projections as “prevailing,” rather than “current,” policy.
Oreopoulos (1999) adopts a similar assumption for Canada. In contrast with King and Jackson (2000), who assume zero wage indexation of social transfers, Jackson and Matier (2002) assume that the average transfers received by members of each age cohorts will grow in line with wages, as the long-term nature of the projections make inflation-indexing appear “umeasonable.”
The government non-autonomous pension plans are fully-funded schemes, as the government contributions are deemed to cover the cost (net of employee contributions) of the accrued benefits of the schemes. However, it is worth noting that, despite indexing benefit rates more slowly than the rate of growth of wages (and thus contributions), these schemes do not appear to be actuarially balanced, as the difference in present value terms (using a 5 percent real discount rate) between future pension benefits and future contributions exceed the reserves of the funds. The simulations presented in this chapter do not incorporate recent decisions to increase contribution rates and to establish an investment board that will invest contributions in financial markets, both of which are likely to reduce the magnitude of the estimated actuarial imbalance.
Pension benefits from government pension plans do not incorporate changes in pension equity. These are the surpluses of the government pension plans, that is, the difference between the receipts of the funds (contributions plus interest income on accumulated reserves) and pension benefits. These surpluses are considered as an expense in the FMS, as they represent a household claim on the government. Once future values of contributions and benefits are obtained as explained in the text, these surpluses (in real terms) are projected by applying an interest rate of 5 percent on the reserves of the funds. Given this rate, and based on the fact that pension benefits are much larger than contributions, the surpluses are projected to disappear in about 15 years.
Real per-capita spending on health is projected to grow at an average yearly rate of 2.4 percent over 2003-2075, slightly above the 2.2 percent increase over the 1994-2002 period (obtained by deflating nominal spending on health using the implicit price index for government consumption of goods and services, as in Canadian institute for Health Information, 2002). However, considering a more extended period of time, this rate of growth does not seem unreasonable, and it is actually below the 2.9 increase over the period 1970-1998 (OECD, 2001). King and Jackson (2000) also show that real per capita growth in health expenditure (adjusted for population compositional changes) has been 1.8 on average over 1975-2000.
Consistent with the “current” policy approach adopted in this chapter, future values of health spending are obtained by keeping the age profile for health spending constant in the future. However, the projected increase in life expectancy may cause this profile to shift, given that a large share of lifetime health costs takes place in the final years of life. All other things being equal, incorporating this shift into the simulations would lower the projected growth in overall health spending.
On a consolidated basis, the net effective interest rate for the year 2002 was slightly below 7 percent, in nominal terms. The rate of 5 percent is obtained based on the assumptions that the rate will remain constant in the future and that the long-term inflation rate is 2 percent. The initial stock of net debt is derived from the national accounts as the net financial debt of total government, which corresponds to the government sector considered in this chapter. The March 2001 figure was estimated by interpolating end-calendar year national accounts data. For 2000/2001, total government net financial debt was estimated at C$ 698 billion, or 63.5 percent of GDP, which includes the assets of the Canadian and Quebec pension plans (at around 5 percent of GDP), and the unfunded liabilities of government employee pension plans (around 16.5 percent of GDP).
Satisfying this constraint typically only requires that the stock of debt grows no faster than the interest rate, or that the debt-to-GDP ratio will grow no faster than the difference between the interest rate and the GOP growth rate. A narrower concept of fiscal sustainability that is sometimes adopted would be the immediate and permanent increase in taxes needed to bring the net-debt to-GDP ratio to its initial level (or an alternative debt target) at some time T in the future, but as Auerbach (1994) points out, this will understate the fiscal gap when primary deficits are projected in the years following T.
In this scenario, in every year between 2003 and 2008, only C$3 billion (0.2 percent of GDP in 2008) is devoted to debt reduction (this is the federal government’s contingency reserve forecast by the Economic and Fiscal Update for the next five years). The remaining surpluses are assumed to be spent proportionally on all expenditure categories reported in Table 1.
The stock of net debt used in the sustainability analysis does not include government pension liabilities. Further, changes in pension equity are not treated as an expenditure. The unfunded liability associated with the government pension plan is captured by the present value of the difference between future pension benefits and contributions.
Adopting a similar methodology, Kennedy and Matier (2002) find that Canada’s flscal gap is negative. indicating that current fiscal policy is sustainable. However, their analysis focuses on the long-term fiscal stance of the federal government plus the CPP/QPP, while the simulations in this chapter include provinces and nonautonomous pension plans as well. Further, Kennedy and Matier adopt a narrower concept of the fiscal gap, defined as the immediate and permanent increase in taxes required to bring the net-debt to-GDP ratio to its initial level at some time T in the future (i.e., 2075).
Clearly, the fiscal gap could be closed by an equivalent immediate and permanent change in expenditure, or some combinations oftax and expenditure changes.
For an international comparison of the degree of fiscal long-term intertemporal imbalances measured in a generational accounting framework see A. Auerbach, L. Kotlikoff and W. Leibfritz (1999).
