Box 1. Policy Assumptions Underlying the Projections Fiscal policy assumptions for the short term are based on official budgets adjusted for any deviations in outturn as estimated by IMF staff and also for differences in economic assumptions between IMF staff and national authorities. The assumptions for the medium term take into account announced future policy measures that are judged likely to be implemented. In cases where future budget intentions have not been announced with sufficient specificity to permit a judgment about the feasibility of their implementation, an unchanged structural primary balance is assumed. For selected industrial countries, the specific assumptions adopted are as follows. United States: For the period through FY 1999, fiscal revenues and outlays at the federal level are based on the administration's March 1996 budget proposal (as updated in the July 1996 Mid-Session Review of the 1997 Budget), after adjusting for differences between the IMF staff's macroeconomic assumptions and those of the administration. From FY 2000 onward, the federal government primary balance as a proportion of GDP is assumed to remain unchanged from its projected FY 1999 level. Japan: Measures that have already been announced are assumed to be implemented over the medium term. These measures include an increase in the consumption tax rate from 3 percent to 5 percent in 1997 and a simultaneous end to the temporary income tax cut, implementation of the 1994 pension reform plan, and the medium-term public investment plan. Germany: The 1996 projection for the general government is based on the 1996 federal budget and recent official estimates for the other levels of government, adjusted for differences in macroeconomic projections. The 1997 projections are based on official tax estimates and envisage implementation of most of the government's proposed fiscal package. Elements of the package that have gained the requisite agreement (e.g., from the unions and the Bundestag) have been incorporated, while measures that still need Bundesrat approval or lack specificity at the Länder level have not been fully taken into account. Projections for 1998 and beyond are predicated on an unchanged structural primary balance. France: The projections for both 1996 and 1997 take into account policy measures that have already been implemented. In addition, it is assumed that the government's plans for the 1997 budget of the state (a freeze in nominal expenditure yielding an adjustment of about 1/2 of 1 percent of GDP) will be fully achieved, given the strong commitment to meeting the criteria for European Economic and Monetary Unit (EMU). As regards social security spending in 1997, the projections assume that Parliament (using the new powers conferred by the constitutional amendment adopted earlier this year) will set a ceiling consistent with the government's overall objectives. Italy: The projections take into account policy measures that have already been implemented, including the June 1996 supplementary package. In addition, it is assumed that the measures announced in the 1997–99 plan are fully implemented and yield the officially estimated amounts. Projections beyond 1999 assume an unchanged structural balance. United Kingdom: The budgeted three-year spending ceilings are assumed to be observed. Thereafter, noncyclical spending is assumed to grow in line with potential GDP. For revenues, the projections incorporate, through the three-year budget horizon, the announced commitment to raise excises on tobacco and road fuels each year in real terms; thereafter, real tax rates are assumed to remain constant. Canada: Federal government outlays for departmental spending and business subsidies conform to the medium-term commitments announced in the March 1996 budget. Other outlays and revenues are assumed to evolve in line with projected macroeconomic developments. The unemployment insurance premium, however, is assumed to fall in 1998/99 to a level that is consistent with an unchanged surplus in the unemployment insurance account. The fiscal situation of the provinces is assumed to be consistent with their stated medium-term deficit targets. Spain: Projections for 1996–97 incorporate the multiyear expenditure control agreements in the areas of public sector wages and employment, health care, and regional and local governments; additional cuts announced in May; and the economic liberalization, tax measures, and wage freeze announced in June and July. For 1997, the income tax schedule and excise taxes are assumed to be adjusted in line with inflation. The projections do not take into account the implications for 1996–97 of the recently discovered large public expenditure overruns in 1995. Netherlands: Projections assume continued adherence to the government's medium-term expenditure path for the central government and social security. While the authorities have not specified a general government deficit target for 1997, their 1997 expenditure ceilings for central government and social security that were decided in the spring of 1996 provide scope for a deficit well under 3 percent of GDP. Belgium: Projections for 1996 are based on the budget and the spring budget control exercise. For 1997, IMF staff estimates reflect the general government deficit ceiling of 3 percent of GDP included in the framework law passed by Parliament in July 1996 that gives the government broad powers to achieve participation in EMU (recent pronouncements by the authorities have signaled a desire for a deficit of 2.8 percent of GDP). Sweden: The medium-term projections are based on the government's multiyear consolidation program approved by Parliament in 1995 and additional measures taken in April 1996. Greece: Projections for 1996 reflect the IMF staff's assessment of the outcome of the official budget. Measures to meet the targets of the convergence plan are still being defined in the 1997 budget under preparation, and the projections at this stage are based on IMF staff estimates of current services. Portugal: Projections for 1996 are based on implementation to date of the official budget. Preparation of the 1997 budget (which will aim at a deficit of not more than 3 percent of GDP) is currently under way, and the projections for 1997 and beyond are at this stage based on an unchanged structural primary balance. Switzerland: Projections for 1996–99 are based on official estimates for current services. Thereafter, the general government structural primary balance is assumed to remain constant. Australia: Projections for the Commonwealth government are based on the 1996/97 budget, adjusted for any differences between the economic projections of the staff and the authorities. Unchanged policies are assumed for the state and local government sector from 1996. * * * Monetary policy assumptions are based on the established framework for monetary policy in each country, which in most cases implies a nonaccommodative stance over the business cycle. Hence, it is generally assumed that official interest rates will firm when economic indicators, including monetary aggregates, suggest that inflation will rise above its acceptable rate or range and ease when the indicators suggest that prospective inflation does not exceed the acceptable rate or range and that prospective output growth is below its potential rate. For the exchange rate mechanism (ERM) countries, which use monetary policy to adhere to exchange rate anchors, official interest rates are assumed to move in line with those in Germany, except that progress on fiscal consolidation may influence interest differentials relative to Germany. On this basis, it is assumed that the London interbank offered rate (LIBOR) on six-month U.S. dollar deposits will average 5.6 percent in 1996 and 6 percent in 1997; that the six-month LIBOR rate in Japan will average 1 percent in 1996 and 2.4 percent in 1997; and that the three-month interbank deposit rate in Germany will average 3.3 percent in 1996 and 3.8 percent in 1997.