VIII Summary and Conclusions
- Ahsan Mansur, Richard Haas, and Peter Heller
- Published Date:
- May 1986
This paper has focused on several aspects of the methodology used by the Fund in its calculation of the fiscal impulse and has considered how this methodology could be extended to include estimates of the structural budget balance. This section provides a summary of the principal conclusions of the above discussion and presents a measure of the overall effect of introducing the above-mentioned changes.
Calculation of Potential GDP
The current estimates of the growth of potential GDP are based on estimates provided by Fund economists working on particular countries. One alternative approach would use estimates of the growth rates of potential GDP that are consistent with the notion that the level of potential GDP is “attainable” in a given medium-term target year and start at the present rate of capacity utilization. The medium-term targets would be based on projections made in the context of the annual medium-term scenario analysis contained in the World Economic Outlook paper. Adoption of this alternative does affect the measured fiscal impulse, leading in the present case to a more expansionary assessment of the thrust of fiscal policy for all countries. However, such a method raises several conceptual issues. These include the biases to the potential growth rate that would arise if an exogenous shock lowered the level of potential output during the period under analysis, the issue of whether cyclically neutral expenditure should be related to “potential” or “attainable” GDP, the implied variability in the historical cyclically neutral expenditure series arising from changes in the medium-target year, and the desirability of a smooth linear growth path of potential output from the base year to the target year.
Treatment of Unemployment Compensation
The present methodology assumes that the entire increase in unemployment compensation benefits since the base year is due to cyclical developments in the economy. If there has been an increase in the “normal” unemployment rate that would prevail in the target year relative to the base year, and to the extent that the per capita level of unemployment insurance benefits (UIB) has increased in real terms, the expansionary impact of the fiscal impulse, as currently measured, would be understated. In examining alternatives, a distinction is made between the change in UIB payments which is attributable to cyclical factors, as opposed to what might be termed “structural” factors. A methodology has been examined whereby increases in unemployment compensation payments that reflect either a change in the real base or structural unemployment rate or a change in the real benefit level are treated as noncyclical elements of the budget. While the alternative approach is a conceptual improvement over the existing method, the empirical results barely differ.
Adjustments for Inflation
Currently, the fiscal impulse measure is not adjusted for the effects of inflation on the budget balance. For example, a deterioration in the government’s fiscal position owing to an inflation-induced rise in the nominal interest rate and ensuing interest payments would, under the existing methodology, be assumed to affect economic activity in the same way as any other increase in spending. An alternative school of thought contends that this approach ignores the net reduction in the government’s real liabilities arising from inflation and that this bias impairs the utility of the current indicator. Two alternative inflation adjustments are considered. These include an adjustment for the actual inflation rate, and an adjustment based on an assumed constant historical real rate of interest. Estimates of the deficit and impulse, adjusted for inflation, are useful as supplementary indicators to the existing measures.
National Accounts-Basis Data Versus Cash-Basis Data
In its analysis of the fiscal policies of the central government, the Fund’s World Economic Outlook exercise presents the budget balance on a cash basis. rather than on a national accounts basis. The latter is used by the Fund and other agencies in the appraisal of the fiscal policy stance of general government. Both conceptual and practical factors enter into the decision to use one or the other of the two accounting frameworks. On conceptual grounds, strong arguments can be made for both approaches, with the argument hinging on the economic impact of net lending transactions and on the timing and the relative impact on aggregate demand of specific transactions. Practical considerations relating to the ready availability of current and prospective budgetary data on the central government budget suggest that it would be difficult to shift to a national accounts basis for all of the central governments of the major industrial countries.
Structural Budget Balance
The level of the structural budget balance—that surplus or deficit which would prevail if the economy were at a “normal” level—has become an increasingly important measure. Section VII has suggested that the ratio of the structural balance to normal GDP may be estimated by adding the cumulative fiscal impulse shares since the base year to the share of the structural balance in GDP that prevailed in the base year. When adjustments are made to the initial structural budget balance and to the fiscal impulse to correct for inflation, one observes that, with the exception of the United States, all of the Group of Seven countries have reduced their structural deficits since the base period.
Tables 14 and 15 examine the effects on the fiscal impulse of introducing various combinations of the methodological changes discussed above for both central and general governments. These include (i) the combined effect of the alternative approaches to potential output (alternative I) and unemployment insurance, and (ii) the combined effects of (i) and the two alternative inflation-adjustment procedures. The results are contrasted with those obtained under the Fund’s existing World Economic Outlook methodology. Comparison of fiscal impulses estimated under various approaches indicates that, for any given year, the qualitative shifts in the thrust of the government’s fiscal policy tend to be broadly similar for all countries, regardless of the approach chosen. Based on the available observations for the seven countries, reversal in the measured direction of policy occurs about 7 percent of the time for the central government and about 4.5 percent of the time for general government, although the periods in which the signs differ are not the same in the two cases.
|Conventional WEO1 method||−0.04||1.21||0.98||−0.65||−0.37||−1.09||1.36||0.71||−0.08||1.19|
|All proposed changes and no inflation adjustment||0.18||1.39||1.21||−0.73||−0.41||−1.13||1.37||0.74||0.08||−1.18|
|All proposed changes and full inflation adjustment||…||1.63||1.14||−1.69||−0.72||−1.08||1.25||1.55||0.30||−1.27|
|All proposed changes and real interest adjustment||…||1.38||0.94||−0.99||−0.62||−1.85||1.00||1.04||0.20||−1.00|
|Conventional WEO1 method||−0.87||0.16||—||−0.79||0.35||0.16||0.52||1.63||0.25||0.19|
|All proposed changes and no inflation adjustment||−0.89||0.08||−0.01||−0.77||0.49||0.10||0.68||1.96||−0.02||−0.10|
|All proposed changes and full inflation adjustment||…||−0.21||−0.43||−1.05||0.36||0.12||1.43||2.36||−0.09||−0.29|
|All proposed changes and real interest adjustment||…||0.08||−0.23||−1.02||0.19||−0.45||0.42||1.90||−0.23||−0.28|
|Conventional WEO1 method||0.63||0.19||0.21||1.04||−0.01||−0.31||−0.64||−0.45||−0.23||−0.68|
|All proposed changes and no inflation adjustment||1.06||0.50||0.51||1.29||0.18||−0.17||−0.57||−0.39||−0.15||−0.61|
|All proposed changes and full inflation adjustment||…||0.40||0.41||1.60||−0.03||−0.18||−0.45||0.11||−0.74||−0.66|
|All proposed changes and real interest adjustment||…||0.34||0.42||1.20||−0.07||−0.39||−0.78||−0.64||−0.25||−0.65|
|Conventional WEO1 method||−1.07||−0.35||1.87||−0.85||−0.68||0.96||0.02||−0.32||−0.34||−0.23|
|All proposed changes and no inflation adjustment||…||…||…||−0.68||−0.52||1.20||0.24||−0.20||−0.29||−0.06|
|All proposed changes and full inflation adjustment||…||…||…||−0.84||−0.77||1.11||0.21||0.17||0.02||−0.11|
|All proposed changes and real interest adjustment||…||…||—||−0.84||−0.51||0.71||0.23||−0.28||−0.32||−0.01|
|Germany, Federal Republic of|
|Conventional WEO1 method||0.29||−0.37||0.12||—||−0.38||−0.69||−1.75||−0.11||−0.21||−0.07|
|All proposed changes and no inflation adjustment||0.40||−0.20||0.55||0.68||0.03||−0.38||−1.65||−0.26||0.04||0.04|
|All proposed changes and full inflation adjustment||…||−0.29||0.45||0.65||−0.09||−0.38||−1.83||−0.01||0.05||−0.03|
|All proposed changes and real interest adjustment||…||−0.26||0.59||0.63||−0.08||−0.55||−1.80||−0.45||0.05||0.04|
|Conventional WEO1 method||−0.65||−0.77||5.30||−2.98||−0.07||0.59||0.65||−0.07||−0.09||…|
|All proposed changes and no inflation adjustment||0.13||−0.06||6.02||−2.28||0.62||1.34||1.44||0.58||0.48||…|
|All proposed changes and full inflation adjustment||…||−0.88||7.70||−3.98||−1.95||2.25||1.28||1.17||0.56||…|
|All proposed changes and real interest adjustment||…||−1.75||5.09||−2.37||0.15||0.34||0.27||0.63||−0.06||…|
|Conventional WEO1 method||−2.49||−2.39||2.74||0.53||−2.36||−2.46||−1.25||2.43||−1.22||−0.54|
|All proposed changes and no inflation adjustment||−1.87||−2.10||3.06||0.90||−1.90||−2.14||−0.99||2.74||−0.87||−0.17|
|All proposed changes and full inflation adjustment||…||−2.25||3.93||−0.31||−3.27||0.34||0.37||3.41||−0.94||−0.17|
|All proposed changes and real interest adjustment||…||−2.01||3.04||0.66||−2.12||−2.41||−0.98||3.07||−0.82||−0.15|
World Economic Outlook.
World Economic Outlook.
|Conventional WEO1 method||0.17||−0.16||0.62||−0.87||0.09||−0.96||0.38||0.42||0.13||−1.06|
|All proposed changes and no inflation adjustment||0.65||0.31||1.12||−0.83||0.14||−0.98||0.40||0.54||0.51||−0.97|
|Conventional WEO1 method||−1.01||−0.18||−0.13||−0.48||0.56||−0.17||1.06||0.38||0.75||0.39|
|All proposed changes and no inflation adjustment||−0.97||−0.21||−0.10||−0.44||0.73||−0.20||1.25||0.68||0.72||0.09|
|All proposed changes and full inflation adjustment||…||−0.54||−0.61||−0.79||0.60||−0.12||2.28||1.24||0.70||−0.04|
|Conventional WEO1 method||0.92||0.23||1.76||−0.54||−0.54||−0.33||−0.64||−0.42||−0.19||−0.65|
|All proposed changes and no inflation adjustment||1.43||0.61||2.13||−0.25||−0.31||−0.16||−0.55||−0.35||−0.10||−0.56|
|All proposed changes and full inflation adjustment||…||0.50||2.05||0.18||−0.56||−0.16||−0.39||0.25||−0.79||−0.62|
|All proposed changes and real interest adjustment||…||0.32||1.98||−0.52||−0.72||−0.46||−0.77||−0.55||−0.19||−0.58|
|Conventional WEO1 method||−1.14||0.02||1.42||−1.03||−1.68||0.56||0.13||−0.50||−1.29||−0.34|
|All proposed changes and no inflation adjustment||…||…||…||−0.58||−1.12||1.35||0.80||−0.10||−1.01||−0.02|
|Germany, Federal Republic of|
|Conventional WEO1 method||−0.64||−0.79||0.44||0.76||—||−0.91||−2.42||−0.91||−0.56|
|All proposed changes and no inflation adjustment||−0.34||−0.46||1.02||1.59||0.55||−0.49||−2.25||−0.96||−0.22||0.12|
|All proposed changes and full inflation adjustment||…||−0.60||0.85||1.57||0.33||−0.50||−2.61||−0.44||−0.18||0.03|
|All proposed changes and real interest adjustment||…||−0.58||1.07||1.54||0.39||−0.73||−2.55||−1.24||−0.32||0.07|
|Conventional WEO1 method||−1.61||2.22||1.35||0.42||−1.29||1.89||−1.54||−2.11||0.57||…|
|All proposed changes and no inflation adjustment||−0.76||−1.00||2.24||1.27||−0.51||2.86||−0.66||−1.38||0.57||…|
|Conventional WEO1 method||0.27||−1.65||1.81||−0.77||−1.73||−2.68||−0.70||1.95||−0.50||−0.50|
|All proposed changes and no inflation adjustment||0.96||−1.37||2.17||−0.37||−1.23||−2.32||−0.41||2.30||0.11||−0.09|
|All proposed changes and full inflation adjustment||…||−1.33||3.46||−1.78||−2.87||0.83||1.30||3.14||0.13||−0.04|
|All proposed changes and real interest adjustment||…||…||…||−0.72||−1.57||−2.47||−0.32||2.79||0.05||—|
World Economic Outlook.
World Economic Outlook.
The fiscal impulse is primarily sensitive to changes in the average potential growth rate and in the inflation adjustment. The higher the potential output growth rate, the lower the fiscal impulse, expressed as a percentage of GDP. The proposed treatment of unemployment insurance leads to only marginal quantitative and qualitative changes compared with the World Economic Outlook estimates. Adjustments for inflation significantly influence the magnitude of the fiscal impulse measure, though not the qualitative assessment of fiscal policy. Under the full-inflation-adjustment approach, the impulse measure tends to vary more widely over time, compared with the real-interest-adjustment approach, for those economies with higher, more volatile rates of inflation (e.g., those of the United Kingdom, Italy, and France). The reverse relationship between the two approaches holds for economies with low and stable rates of inflation (e.g., those of Japan and the Federal Republic of Germany).
In conclusion, the fiscal impulse measure is useful, but is subject to limitations. Ideally, the assessment of the stance of fiscal policy should be viewed in the context of the degree of monetary policy accommodation, inflationary consequences, and the possibility of financial crowding out in conditions of monetary restraint. There is a need to determine the extent of monetary accommodation under alternative forms of financing the budgetary gap and the consequent real-government-expenditure multipliers, taking into account the induced changes in imports and private savings, as well as the possibilities of crowding out private sector expenditure. These potentially important issues would be at the heart of any medium-term research effort to provide a revised approach to evaluating the stance of fiscal policy.