APPENDIX—Algebra for the Debt-Feedback Expenditure Rule
Beetsma, R., and X. Debrun, 2007, “The New Stability and Growth Pact: A First Assessment,” European Economic Review, 51, pp. 453–77.
Bodmer, F., 2006, “The Swiss Debt Brake: How It Works and What Can Go Wrong,” Schweizerische Zeitschrift für Volkwirtschaft und Statistik 3/142 pp. 307–330.
Brender, A., 2008, “If You Want to Cut, Cut, Don’t Talk: The Role of Formal Targets in Israel’s Fiscal Consolidation Efforts 1985–2007,” Bank of Israel Lectures, Articles and Position Papers.
Catalan, M., 2007, “Fiscal Institutions and the Political Economy in Israel,” Selected Issues, IMF Country Report No. 07/25 (Washington: International Monetary Fund).
Celasun, O., X. Debrun, and J. D. Ostry, 2006, “Primary Surplus Behavior and Risks to Fiscal Sustainability in Emerging Market Countries: A “Fan-Chart” Approach,” IMF Staff Papers, 53 (3), pp. 401–25.
Debrun, X., and N. Epstein, 2008, “Toward a New Fiscal Rule: What Might Work Well for Israel,” Selected Issues, IMF Country Report No. 08/63, Chapter IV, pp. 43–69. (Washington: International Monetary Fund).
Debrun, X., L. Moulin, A. Turrini, J. Ayuso-i-Casals, and M. S. Kumar, 2008, “Tied to the Mast? The Role of National Fiscal Rules in the European Union,” Economic Policy, forthcoming.
Elekdag, S., N. Epstein, and M. Moreno-Badia, 2006. “Fiscal Consolidation in Israel: A Global Fiscal Model Perspective,” IMF Working Paper No. 06/253 (Washington: International Monetary Fund).
Flug, K., 2006, “The Bumpy Road Towards Fiscal Consolidation: What is the Role of Fiscal Rules, Fiscal Policy in Israel since the 1990s,” presented at the Euro-Mediterranean Conference, ECB, and Falk Institute Annual Conference.
Hallerberg, M., and G, Wolff, 2006, “Fiscal Institutions, Fiscal Policy, and Sovereign Risk Premia,” Deutsche Bundesbank Discussion Paper No 35/2006.
Kopits G., and S. Symansky, 1998, “Fiscal Policy Rules,” IMF Occasional Paper No. 162 (Washington: International Monetary Fund).
Lundback, E. J., 2008, “Medium-Term Budgetary Frameworks—What Can Austria Learn from Other Countries?” IMF Working Paper, forthcoming.
This paper builds on an earlier paper, Toward a New Fiscal Rule: What Might Work Well for Israel (Debrun and Epstein, 2008). The authors are grateful to Jörg Decressin for providing comments as well as Jonathan Manning and Arlene Tayas for providing technical assistance. We would also like to thank, without implicating, participants in seminars at the Bank of Israel and the International Monetary Fund for their comments.
The recognition that large increases in public spending—associated with the substantial wave of immigration from the former Soviet Union during the early 1990s—would result in a deterioration of the fiscal position, prompted the adoption of the Deficit Reduction Law in 1991. The law was intended to send a signal to the markets that the rise in the deficit was transitory (Flug, 2006). See also Brender (2008) for review of Israel’s experience with fiscal rules over the past two decades.
They further define a “good” rule as being simple, transparent, coherent with the final goal, but mindful of other goals of public policies.
For instance, under the assumption of 5 percent nominal growth, a 3 percent deficit ceiling ensures that public debt asymptotically converges to a number below 63 percent of GDP. If permanent GDP growth is actually 4 percent per annum, the same deficit could only guarantee that the public debt will stabilize below 78 percent of GDP.
The sub-national governments (Cantons) are not integrated, nor coordinated with the federal government, but generally have balanced budget rules in place already (Lundback, 2008).
The mechanism was supposed to govern central government budgets starting from 2003. In practice, however—due to a large structural deficit during an unexpected recession that year—it began to take effect with the 2004 budget (Bodmer, 2006).
In fact, the fictional account generated a total credit of around 1.3 percent of GDP between 2004 and 2006, which was largely due to under-execution of the expenditure ceilings.
The difference between the central government balance (cash basis) and the general government balance (accrual basis) comprises the deficits of the local governments (averaging about zero percent of GDP over the past five years) and interest accrued on inflation-index-linked government debt. The latter can vary widely as a function of inflation but has averaged around 1 percent of GDP over the past five years.
In Israel’s case, bilateral foreign transfers are predominantly aimed for imports of military items. Fluctuations in this spending category are large and tend to rise at times of military conflicts—hence, their exclusion would help mitigate the need to use related escape clauses. Moreover, as these transfers are fully directed to imports, they do not affect domestic demand.
While, for operational reasons, the central government balance constitutes the appropriate variable to target, all further scenario analysis and simulations are based on the corresponding general government balance, since it is more closely associated with changes in the overall public debt.
Sensitivity analysis suggests the results are robust to (small) changes in the parameters values.
The potential growth series is updated using a moving average of real GDP growth on a rolling eight-year basis. An alternative measure of potential growth using an HP filter shows very similar results.
Another conceptual issue is that the debt brake leads to a debt-to-GDP ratio that is asymptotically zero in the Swiss case of balanced budget, and less than 20 percent of GDP under the assumption of a 1 percent deficit and 6 percent nominal growth. This is the reason why, in the version proposed in the previous section, the rule would have to be revisited once public debt has been reduced to 60 percent of GDP.
In both cases, and to speed up debt reduction, negative deviations from the norm would only trigger an upward correction of the expenditure ceiling when debt is at or below 60 percent of GDP.
Of course, another option is to use an estimate of trend (or cyclically adjusted) revenue levels. That would relax (tighten) expenditure growth ceilings when the output gap is negative (positive). However, this approach amounts to reintroduce the output gap at the center of the framework, which has some drawbacks, as already discussed.
The rule would have to specify the precise events that could trigger an increase in expenditure above the ceiling and the process that would be followed to enact these changes.
For reasons related to simplicity, transparency, and consistency with the ultimate objective, total expenditure should be subject to the ceiling even though some expenditure items are not under the control of the government, including interest payments or investment projects financed by specific external grants.
A detailed discussion of operational aspects of expenditure rules (real vs. nominal ceilings, comprehensiveness of coverage, contingency margins under the ceiling) is beyond the scope of this paper. See Ljungman (2008) for a recent survey.
The results presented here are only suggestive as they depend on specific assumptions about deterministic business cycles. A more robust analysis should be based on stochastic simulations reflecting the joint conditional probability distribution of shocks affecting the determinants of debt dynamics (budget balance, GDP growth, interest rates and exchange rates, as in Celasun, Debrun, and Ostry, 2006). Such analysis is beyond the scope of this paper.
Unlike the Swiss debt-brake rule, where the ECM is triggered based on deviations from a pre-specified level of deficits accumulation, the alternative ECM rule is anchored more explicitly on a targeted debt path, i.e., the correction mechanism is based directly on deviations from a “norm” debt path objective.
See IMF Country Report (2007).
Parameter s (the length of the planning horizon) reflects the authorities’ choice on the trade-off between flexibility in the response to unforeseen events and the timely realization of the debt objective. A reasonable benchmark could be the length of a normal legislature. It could also be shorter given the potentially large forecast errors at longer horizons.
Again, setting a threshold for acceptable deviations is a matter for discussion, but 5 to 10 percent of GDP seems an appropriate range if one wants to avoid significant contractions in bad times.