Abstract
1. In anticipation that the budget of the federal government will be balanced within the next year, the focus of the fiscal policy debate in Canada has shifted toward the extent to which the expected “fiscal dividend”—the surplus that would accrue under current policies—should be used to pay down the public debt, to lower taxes, or to increase program spending. Economic theory does not provide a clear guide on the “optimal” level of public debt. Paying down government debt in itself produces interest savings for the government. It also reduces interest rates as the government’s demand on savings declines and the risk premium on its debt falls. Lower interest rates, in turn, crowd in investment and raise national income. However, the pace at which debt reduction takes place can have important transitional effects on output and employment. Moreover, there is an opportunity cost involved in running budget surpluses to reduce debt that has to be addressed, reflecting a question of whether resources used for debt reduction might produce higher returns to the economy if used instead to lower taxes or increase spending on public goods.
V. Macroeconomic Effects of Government Debt in Canada1
1. In anticipation that the budget of the federal government will be balanced within the next year, the focus of the fiscal policy debate in Canada has shifted toward the extent to which the expected “fiscal dividend”—the surplus that would accrue under current policies—should be used to pay down the public debt, to lower taxes, or to increase program spending. Economic theory does not provide a clear guide on the “optimal” level of public debt. Paying down government debt in itself produces interest savings for the government. It also reduces interest rates as the government’s demand on savings declines and the risk premium on its debt falls. Lower interest rates, in turn, crowd in investment and raise national income. However, the pace at which debt reduction takes place can have important transitional effects on output and employment. Moreover, there is an opportunity cost involved in running budget surpluses to reduce debt that has to be addressed, reflecting a question of whether resources used for debt reduction might produce higher returns to the economy if used instead to lower taxes or increase spending on public goods.
2. To illustrate potential trade-offs, an analytical framework drawing on MULTIMOD is used to assess the effects of alternative paths for the fiscal balance and associated debt-to-GDP ratios. The framework captures the effects of debt reduction on investment and income. Assessing the potential benefits of alternative uses of the fiscal dividend would require a detailed analysis of specific possibilities, and is not attempted here.
3. The analytical framework focuses on supply-side relationships between savings, investment, and output. It does not directly examine the effects of fiscal policy on short-term demand because in the kind of longer-term analysis presented here, the properties of the model are such that the economy would tend to move back to its potential level of output over time. It is worth noting, however, that the magnitudes of the fiscal surpluses that are simulated in this paper would not be expected to have a significant effect on output growth in the short run. Since economies do not typically operate at potential output for extended periods of time, the framework was adjusted so that the level of GDP was maintained over the long term at 1 percent below potential, which is slightly larger than the historical average of the output gap in Canada.
4. The framework is based on the block of equations for Canada in MULTIMOD. Output is produced using a Cobb-Douglas production function with capital and labor as inputs. The labor supply is taken from long-term population projections that show a significant aging of the population in the first half of the twenty-first century.2 Hence, over time, labor force growth is expected to slow, and growth in potential output would decline as a result. The capital stock evolves according to Tobin’s Q theory, in which new investment depends on the relationship between the book value of capital and its replacement cost, with the additional assumption that it takes time to build new productive capacity. This assumption slows adjustment in the capital stock, which smooths the response of output to changes in interest rates.
5. Since Canada is a small, open economy with respect to global financial markets, the long-term interest rate in Canada is modeled as being equal to the long-term interest rate in the United States plus a premium that reflects risks specific to Canada. This premium vis-à-vis the United States is specified as reflecting differences in output growth, fiscal deficits, and the magnitude of the two countries’ respective government debt-to-GDP ratios. Econometric analysis using data for the period 1965 to 1996 suggests that, ceteris paribus, a 1 percentage point fall in the ratio of government debt to GDP in Canada leads to a decline in long-term interest rates in Canada of between 5 and 10 basis points, with the smaller figures occurring for the average of the entire 32-year period, and larger responses found using data for the period after 1973. A response of 5 basis points is used in the simulations, since it is expected that the risk premium on Canadian debt will respond by less in the current and future environment of low inflation than in the more recent past when the specter of higher inflation made market participants particularly wary of, and quick to react to, adverse fiscal developments. Use of this relatively small response of interest rates may be considered as providing a lower bound estimate of the gains from debt reduction.
6. Four alternative scenarios were developed for the federal government budget over the long term. In the current policy scenario, the federal government is assumed to broadly continue with current tax. and expenditure policies, with the additional assumptions that expenditures affected by demographic shifts reflect the effects of these changes, that other program spending remains constant in real terms, and that the employment insurance premium is lowered in steps to a level where receipts are roughly in balance with benefit payments. On this basis, the budget surplus would grow sharply over time, and the federal government’s debt would be eliminated by 2011/12 (Charts 1 and 2). The balanced budget scenario assumes that the federal budget is balanced in 1998/99 and that balance is maintained thereafter. Even on this basis, the government debt-to-GDP ratio would decline from 71 percent in 1996/97 to slightly less than 40 percent 2012/13.
CANADA: FEDERAL GOVERNMENT NET DEBT
(Public accounts basis, in percent of GDP)
Citation: IMF Staff Country Reports 1998, 055; 10.5089/9781451806915.002.A005
Source: Fund staff estimates.CANADA: FEDERAL GOVERNMENT FISCAL BALANCES
(Public accounts basis, in percent of GDP)
Citation: IMF Staff Country Reports 1998, 055; 10.5089/9781451806915.002.A005
Source: Fund staff estimates.7. Two additional scenarios were developed based on the assumption that current policies would be generally maintained over the next three years, yielding budget surpluses of around 1 per-cent of GDP a year, and thereafter surpluses would be run that were sufficient to reduce the debt-to-GDP ratio to 30 percent and to 20 percent by 2012/13, respectively. A budget surplus of around ¾ percent of GDP a year over the period 2001–2013 would be required to bring the federal debt-to-GDP ratio down to 30 percent by 2013 and a surplus of 1½ percent of GDP a year over this period would be needed to reduce the debt-to-GDP to 20 percent by 2013. After 2012/2013, the budget is assumed to be maintained in balance in both of these scenarios.
8. The path of federal government spending and debt from the balanced budget scenario was used to establish a baseline projection in the analytical framework.3 This deficit and debt profile, together with assumptions for the government deficit and debt in the United States, determines the differential between long-term interest rates in the United States and Canada.4 This interest rate feeds into the equations for investment and the production function to determine the capital stock and output and, in turn, future interest obligations and resources available for discretionary fiscal actions.
9. Alternative paths for the deficit and the debt were then derived such that the ratio of federal government debt to GDP would decline by 2013 to 30 percent and 20 percent, respectively. As would be expected under both of these scenarios, long-term interest rates fall more rapidly and to a lower level than in the balanced budget scenario (Chart 3). Consequently, by 2013, the level of GDP is permanently higher by around ½ percent in the 30 percent debt-to-GDP ratio scenario and by 1¼ percent in the 20 percent debt-to-GDP ratio scenario (Chart 4).5 As a result of the influences of the reduction in government debt, the decline in interest rates, and the rise in national income, the federal government’s interest payments over time take up a substantially smaller portion of its total revenues in the scenarios where the debt-to-GDP ratio declines to 30 and 20 percent than in the balanced budget scenario (Chart 5).
CANADA: LONG-TERM INTEREST RATES
(In percent)
Citation: IMF Staff Country Reports 1998, 055; 10.5089/9781451806915.002.A005
Source: Fund staff estimates.CANADA: INCREASE IN GDP 1/
(In billions of 1997 Canadian dollars)
Citation: IMF Staff Country Reports 1998, 055; 10.5089/9781451806915.002.A005
Source: Fund staff estimates.1/ Relative to the balanced budget baseline.CANADA: FEDERAL GOVERNMENT INTEREST PAYMENTS
(Public accounts basis, in percent of total revenues)
Citation: IMF Staff Country Reports 1998, 055; 10.5089/9781451806915.002.A005
Source: Fund staff estimates.10. While the scenarios presented here do not provide a definitive answer regarding the optimal fiscal policy path for Canada, they do shed some light on possible trade-offs. With the decline in interest rates and the higher level of GDP, the primary budget surpluses (which provide an indication of fiscal effort) needed to achieve the debt-to-GDP ratios of 30 and 20 percent diminish rapidly over time (see Chart 2) suggesting that ample scope would be provided in time to use fiscal dividends for purposes other than debt reduction. Thus, a policy of running small budget surpluses over the medium term would entail a relatively small sacrifice relative to pursuing a balanced budget, but it would provide a lasting benefit to the economy in terms of the economy’s ability to reach a higher level of GDP. Moreover, increased resources would be made available to the public sector over time, which could be used to reduce taxes or increase program spending.
List of References
Macklem, Tiff, David Rose and Robert Tetlow, 1995, “Government Debt and Deficits in Canada: A Macro Simulation Analysis,” Bank of Canada, Working Paper 95-4 (May).
McCallum, John, 1997, “Fiscal Targets and Economic Growth,” paper presented at Department of Finance conference, Queens University, Kingston, (August).
Office of the Superintendent of Financial Institutions, Canada Pension Plan Sixteenth Actuarial Report (September).
Robson, William and William Scarth (eds.), 1994, Deficit Reduction, What Pain, What Gain?, CD. Howe Institute, Policy Study 23, (Toronto, Canada).
Scarth, William, 1997, “Debt/GDP Dynamics and Optimal Policies: How Quickly Should the Debt/GDP Ratio be Reduced? What Should be the Ultimate Target?” paper presented at Department of Finance conference, Queens University, Kingston (August).
Statistics Canada, 1994, Population Projections for Canada, Provinces and Territories 1993-2016, Catalogue 91-520 Occasional (Ottawa, Canada).
Uctum, Merih and Michael Wickens, 1996, “Debt and Deficit Ceilings, and Sustainability of Fiscal Policies: An Intertemporal Analysis,” Federal Reserve Bank of New York Research Paper, No. 9615 (June).
Prepared by Phillip Swagel, Brenda González-Hermosillo, and Yutong Li.
Demographic changes in Canada are discussed in Section VI.
Long-term projections for the rest of the general government sector were taken in the case of provincial governments from Scenario 1 in Section VI, and in the case of the Canada Pension Plan/Quebec Pension Plan from estimates prepared by the Office of the Superintendent of Financial Institutions (1997).
Inflation is assumed to fall to 2 percent in the United States by 2005 (compared with 1 percent in Canada); the U.S. federal government budget is assumed to be balanced in 2002 and to remain in balance thereafter; and real GDP growth in the United States is assumed to slow from 2¼ percent in 1998 to about 214 percent in 2015. These assumptions lead to a gradual decline in the U.S. debt to GDP, from nearly 52 percent in 1997 to around 25 percent in 2014.
The effects on output arising from these more ambitious debt-reduction paths are probably understated. As noted above, the framework incorporates a conservative estimate of the effect of debt reduction on interest rate premia. It also does not include the possible effect of improved confidence on investment and saving and any positive response of labor supply or productivity growth to increased investment and saving.