Fiscal Rules in Israel1
This annex provides an assessment of the fiscal rules currently in place in Israel.2We argue that reasons for the historic weak compliance with the rules include: the absence of a medium-term budget framework; a loose link between the ultimate objective of the rules and the operational targets; insufficient flexibility of the rules in the face of shocks; and weak enforcement. Measures to improve the current arrangements could include: establishing a medium-term budget framework; strengthening the link between the objective and the immediate target; introducing some flexibility to account for shocks in the rule design; and establishing an independent body to monitor compliance.
Current fiscal rules arrangements in Israel
1. Fiscal policy in Israel is regulated by a budget rule and an expenditure rule, with the purported goal being to reduce public debt. The Deficit Reduction Law (DRL) in 1991 established a budget rule under which the Government sets multiannual targets for the deficit of the central government. In 2004, the DRL was amended to include an expenditure rule, which initially limited central government expenditure to one per cent real growth. Causing a reduction in general government expenditure (from 50 to 40.3 percent of GDP between 2002 and 2012) to levels well below the OECD average of 47 percent, the ceiling was revised in 2010 with a new rule setting real expenditure growth as a function of the deviation of public debt from 60 percent of GDP and of the average GDP growth rate over the past ten years. When the two targets conflict, the most restrictive of the two applies.3
2. Deficit targets have been missed and reset on several occasions. Over the last twenty years, targets were revised about fifteen times (Table 1 and Figure 1). In addition to revisions to the level of the targeted deficit, the targeted aggregate was also modified and changed from domestic deficit to overall deficit (domestic and external) in 1996, and to exclude the BOI’s profits in 1999. Targets were also often missed largely owing to growth shocks, which caused revenue shortfalls.
|Israeli Central Government Deficit Targets, by year established|
3. The expenditure ceiling has been adjusted several times. Set initially to 1 percent, the expenditure ceiling was modified to 1.7 percent in 2007 (excluding war-related and disengagement transitory expenditure) to better reflect population growth and then to 3 percent in 2009. The ceiling was also modified by the Knesset every year from 2005 to 2010 to allow for higher spending due to unforeseen events (like disengagement from Gaza in 2005–08; the second Lebanon War in 2006–07; and the global financial crisis 2008–09). An important factor for difficulties in complying with the ceiling is that the budget has been conducted on an annual and, since 2009 biennial basis, while medium-term expenditure strategies of line ministries are conducted on a horizon of 3 to 5 years. As a consequence, in recent years many policy decisions have increased expenditure over the medium-term, creating commitments which are not consistent with the deficit targets and expenditure ceiling. Also, the general practice is that deficit targets are modified soon after the elections; and almost every government sets multi-year deficit targets, with implied sharp declines in the future, while setting higher deficit targets for the current year (BOI 2013).
Actual Deficits vs. Deficit Targets
Sources: Brender (2011), Bank of Israel Reports
4. In spite of the weak compliance with the rules, debt as a share of GDP declined substantially over the last twenty years. Since the early 1990s debt decreased from almost 140 percent of GDP to below 70 percent in 2012. This debt reduction was mostly due to high growth (figure 2) and partly to large privatizations (Fisher and Flug, 2007). Only between 2004 and 2008 deficit reductions played a role for debt reduction, reflecting the consolidation efforts taken between 2003 and 2007 which led to a sharp contraction in debt to GDP. Despite this declining trend, debt ratios remain high in Israel and interest payments are among the highest in the OECD countries (4 percent of GDP). Moreover, with inflation at relatively low levels and economic growth expected to be stable but moderate over the medium-term, further progress on reducing debt will need to come from maintaining low deficits.
Debt dynamics (1992-2012)
Source: IMF Staff’s Calculations on Bank of Israel’s Data
Assessment and diagnostic
5. Generally, the literature identifies several possible reasons for weak compliance with fiscal rules: (i) weak medium-term budgeting and forecasting, such as unrealistic medium-term targets or systematically biased forecasts; (ii) a lack of clarity in the fiscal framework; (iii) an overly rigid design of the fiscal rules; (iv) and, weak enforcement and lack of political will.
Medium-term budgeting and forecasting
Forecasting error, from budget to actual
Source: Bank of Israel’s Data
6. Short-term budgetary forecasts in Israel tend to be conservative and unbiased. Preliminary evidence on the biannual budget forecasts and actual realization for fiscal aggregates and GDP suggests that the authorities adopt a conservative approach. GDP has been better than forecast in 2007 and 2008, but lower than budgeted in the following years; hence no systematic bias is apparent. Revenue forecasts have been mostly pessimistic, while expenditure has been lower than forecast in most years (figure 3).
7. However, a key weakness remains the lack of a medium-term budget framework. The budget is conducted on a biennial basis and the Ministry of Finance produces and publishes forecasts for the macro economy and for the public finances for only three years ahead, but the spending envelope is binding only for the first two years.4 This is a relatively shorter time horizon than in most OECD economies, where budgets and forecasts typically have a time horizon of about 3 to 5 years. On the other hand, Line ministries in Israel tend to produce medium term strategy documents, approved by the Government and the Knesset, with expenditure commitments which are not consistent with the expenditure ceiling and deficit targets, creating mismatches and causing the breaching of the fiscal rules in subsequent years.
8. Accuracy and a lack of bias in forecasting become more important in a medium-term framework. To improve forecasting, some countries have recently established independent bodies, known as fiscal councils, to assess the government’s forecasts or in some cases produce their own forecasts. This is discussed further in the following chapter.
Does the arrangement set a proper and clear link with the main objective of fiscal policy?
9. A fiscal rule is a permanent constraint on fiscal policy, resilient to economic and political pressures. An arrangement for the rule usually delineates a numerical target over a long lasting time period and specifies operational fiscal indicators that can be used to attain the target (IMF 2009). As such a rule should provide operational guidance to policy makers by establishing a link between targets that are under the control of the government and the ultimate objective of fiscal policy. To this end, the ultimate objective should be clearly specified by the law together with a time frame to attain the target. In this context, several countries have adopted fiscal responsibility laws with explicit numerical targets and associated operational rules (for instance in the late 1990s New Zealand, the UK, Australia).
10. A proper link with the fiscal policy objective is missing. The deficit target in Israel does not meet this definition of a fiscal rule, as the current practice of setting deficit targets does not entail a permanent constraint on fiscal policy. Although the objective of debt-reduction is an aim of Israel’s fiscal policy, the laws setting the targets do not explicitly indicate that these targets are anchored to a specific debt level; while an indication of this type would provide a firmer signaling to the market. Conversely, the expenditure ceiling takes into consideration a level of debt to GDP of 60 percent but does not explicitly indicate that this is a target or stipulate when this debt level should be attained. With the formula incorporating the average of last ten years of GDP growth, expenditure growth could still be relatively rapid even at high levels of debt. Moreover, the path of debt depends on the growth rate of revenue as well as expenditure.
Are the rules designed in a way that allows flexibility in case of shocks or over the cycle?
11. Fiscal policy in Israel is procyclical. The design of fiscal rules should allow flexibility in the case of shocks and over the cycle. By narrowing the scope for discretionary action, fiscal rules may otherwise tend to limit the ability of fiscal authorities to react to business cycle fluctuations, thus potentially exacerbating volatility.5 Looking at simple correlation coefficients between the cyclical components of real spending and real GDP for advanced economies for the period 1995–2012, it emerges that in Israel fiscal policy is pro-cyclical (figure 4).6 A reason for this is that the deficit target rule amplifies pro-cyclical responses to shocks, since in periods of recession, when tax revenues decline and the deficit increases, sticking to the deficit target rule requires raising taxes or cutting expenditures, which would further moderate economic activity, while during booms, the rule enables a reduction in the tax rates.
Correlation Coefficients 1995–2012
Source: IMF Staff’s Calculation from WEO Data
12. To avoid pro-cyclicality and allow immediate responses to shocks, second generation fiscal rules have included more flexibility in the rule design. As expressed in Schaechter et al. (2012), the financial crisis of 2008–09 triggered a series of adjustment, suspensions and relaxations of fiscal rules, which complicated the design but warranted quick adjustment in case of shocks. Broadly speaking, rules can allow for some flexibility by:
Flexibility in the targets. To allow for flexibility over the cycle, budget rules can be specified as structural or cyclically-adjusted balances, which identify the balance that would prevail in economic conditions that reflect the fundamentals, i.e. netting out one-off events (such as bubbles, commodity shocks) for the structural balance, and cyclical fluctuations in the case of the cyclically-adjusted balance. In both cases, the estimation of the true position of the economy, in particular distinguishing between cyclical and underlying elements, might be difficult. In addition, targets can be specified on a medium-term basis allowing for flexibility in the face of near-term shocks as long as a medium-term path consistent with achieving the rules is maintained.
Escape clauses are specific dispositions included in the law establishing the rule that indicate when the rule can be relaxed. Best practice suggests that the criteria for triggering the escape clause should be restricted to extraordinary factors, which normally range from natural disasters, economic recessions to banking bailouts (Table 2). Then, the conditions for triggering the clause should be linked to measurable criteria, like declines in GDP or revenues. The activation of the clause requires broad political and public support, like a large parliamentary majority or independent validation. Also, the law or provision on the escape clause should include the maximum number of years during which the rule is suspended, and should specify the path back to the rules (Schaechter et al. 2012).
|Country and Date||Natural|
|Brazil (since 2000)||X||X||-||-||-||-||X||-|
|Colombia (since 2011)||-||X||-||-||-||X||-||-|
|Germany (since 2010)||X||X||-||-||-||X||X||X|
|Jamaica (since 2010)||X||X||-||-||-||X||-||-|
|Mauritius (since 2008)||X||X||-||-||-||X||-||-|
|Mexico (since 2006)||-||X||-||-||-||-||-||-|
|Panama (since 2008)||X||X||-||-||-||X||-||X|
|Peru (since 2000)||X||X||-||-||-||X||-||X|
|Romania (since 2010)||-||X||-||X||X||X||-||X|
|Slovakia (since 2012)||X||X||X||-||-||X||-||-|
|Spain (since 2002)||X||X||-||-||-||X||X||X|
|Switzerland (since 2003)||X||X||-||-||-||X||X||X|
|EU member states/ euro|
|area (since 2005)||-||X||-||-||-||-||-||X|
|WAEMU (since 2000)|
Does the institutional setting sustain the enforcement of the rule?
13. To enhance enforcement, the institutional setting is key. The mere introduction of fiscal rules does not guarantee success (IMF 2009). Evidence suggests that rules that do not have effective enforcement mechanisms tend to fare worse than rules that do and are more likely to be abandoned or reversed (Debrun and others, 2008b). Yet, in practice an effective enforcement based on sanctions is unrealistic and non feasible for national fiscal rules, since it requires an effective third-party enforcer. To enhance enforcement, however, rules can be supported by legal and institutional frameworks.
Legal Enforcement: National fiscal rules
Source: National authorities; and IMF staff assessment from Schaechter et al. 2012.
On the legal front, while most countries adopt rules through statutory laws (figure 5), France, Germany, Poland and Switzerland7 have raised the fiscal rule law to the constitutional level, making it extremely binding, helping prevent the submission or adoption of a budget at odds with the prescription of the rule (Debrun and others, 2008b).
On the institutional front, many countries have established independent monitoring bodies such as the fiscal councils discussed in Chapter III; and/or have adopted formal enforcement procedures (figure 6). Such procedures can range from obligations to publicly explain the deviation, or the obligation to take corrective action, such as the adjustment of certain expenditure or tax parameters. To this end, some countries have incorporated an autocorrection mechanism in the fiscal rule arrangement which roll-over any deviations from the achievement of fiscal targets to subsequent years.8
Source: National authorities; and IMF staff assessment from Schaechter et al. 2012.
14. Israel’s fiscal rules are set by statutory laws, with no formal enforcement procedure nor external or internal monitoring body, although a sort of auto corrective mechanism is embodied in the expenditure rule.9 At present, the government decides on medium-term programs for years beyond the current annual or biennial budget with no formal (or practical) obligation to examine them against the expenditure ceiling or the medium-term deficit targets. In addition, the legislative body can approve new expenditure programs without due consideration to the overall budget framework. Moreover there is no formal external monitoring of progress against the fiscal rules. While the Bank of Israel produces high-quality independent analysis of the fiscal position it does not have a clear and explicit mandate in this area and an expanded remit would risk creating a conflict of interest with its core role on monetary policy. This is discussed further in the following chapter.
Recommendations: a road map
15. To strengthen fiscal institutions in place, priority should be given to setting a medium term budget framework. Subsequently, the current design of the rules should be strengthened, and finally, enforcement should be boosted.
Developing a medium-term budget framework. The authorities should develop a comprehensive medium-term fiscal and expenditure framework by:
(i) developing and publishing in the annual budget medium-term macroeconomic and fiscal forecasts for the budget years plus two or three additional years, and providing an analysis of the risks around these forecasts. This will enhance the credibility of the budget by showing how the medium-term aggregate expenditure ceiling and revenue projections (including announced policy changes) are consistent with meeting the government’s medium-term debt and deficit targets;
(ii) initially agreeing on broad spending priorities to guide the allocation of spending within the aggregate ceiling, and over time establishing a process for setting medium-term ceilings for individual line ministries that are consistent with the aggregate ceiling. This will increase fiscal discipline by ensuring that ministry spending commitments are consistent with the overall ceiling over the medium-term, and should improve efficiency by allowing ministries to more effectively plan expenditure;
(iv) setting up a mechanism and capacity for monitoring medium term spending plans and execution to ensure that the expenditure ceilings are not breached. This will improve fiscal discipline and enhance the credibility of the budget plans;
Modifying the design of the rules by:
(i) clearly specifying their objective. Having a clearly-defined ceiling on public debt would better help reduce debt-to-GDP ratio; the choice of the ceiling should be consistent with fiscal sustainability and the need for fiscal space. Staff is of the view that a ceiling of 60 percent in 2020 could be appropriate for Israel and given the current macroeconomic outlook it could be feasible.
(ii) strengthening the link between the expenditure and deficit targets and the debt objective. The Government’s current deficit path could indeed put debt on the downward path to achieved a level of 60 percent of GDP by 2020. Yet, the current expenditure growth ceiling is too elevated and should be tightened to avoid the adjustment to fall entirely on revenue.
(iii) enhancing flexibility by targeting a cyclically-adjusted balance and including an escape clause. A cyclically-adjusted balance would help meeting the debt objective allowing fiscal stabilizers to work during economic fluctuations around the business cycle. However, assessing the permanent (or non-cyclical) position of the economy is challenging. As in many cases deficit targets were not met for sharp drop in revenue, a shock that seriously reduces revenue could be considered as a possible trigger for an escape clause. The suspension of the rule could be authorized by a large majority in Parliament, and it could be set to last for a year following the recovery from the shock, i.e. when revenue is back to pre-shock level.
Enhancing enforcement. Establishing an independent body, such as a fiscal council, could further sustain compliance with the rules. A council could help ensure robust medium-term forecasting, by assessing the government’s forecasts and plans, and enhance enforcement by independently monitoring the government’s progress in meeting the rules. This is discussed further in the following chapter.
Brender, A. (2012), “The Story of Israel’s New Fiscal Rule: Theoretical Design Meets Politics”, in Rules and Institutions for Sound Fiscal Policy after the Crisis. Banca d’Italia, (2012), pp. 611–629.
Debrun, X., N.Epstein, and S.Symansky (2008a), “A New Fiscal rule: Should Israel ‘Go Swiss?’”, IMF Working Paper n.87.
Debrun, X., L.Moulin, A.Turrini, J.Ayuso-i-Casals and M. S.Kumar (2008b), “Tied to the Mast? National Fiscal Rules in the European Union,” Economic Policy, pp. 299–362.
Fatàs, A. and I.Mihov (2004), “The Macroeconomic Effects of Fiscal Rules in the US States,” Journal of Public Economics 90, 101–117.
Fisher, S. and K.Flug (2007), “The Role of Rules in Fiscal Consolidation: Fiscal Rules in Israel since the 1990s”, mimeo, Bank of Israel.
Frankel, J. A., C.A.Végh and G. Vuletin, 2011, “On Graduation from Fiscal Procyclicality,” National Bureau of Economic Research Working Papers 17169, National Bureau of Economic Research, Inc.
International Monetary Fund, 2009, “Fiscal Rules—Anchoring Expectations for Sustainable Public Finances,” Prepared by the Fiscal Affairs Department (Washington: International Monetary Fund).
Levinson, A., 1998, “Balanced Budgets and Business Cycles: Evidence from the States,”
University of Wisconsin—Madison.
Prepared by Elva Bova.
For a detailed discussion on the expenditure rule see Brender (2012).
For a detailed review of medium term forecast in Israel see IMF (2011).
A positive correlation coefficient indicates that government spending is procyclical, while a negative correlation coefficient indicates that it is countercyclical. In Israel the coefficient is procyclical and amounts to 0.28, while in most G8 countries it is negative, hence countercyclical. For the estimation of correlation coefficients we followed the technique by Frankel and others (2011).
Constitutional laws were set for the revenue rule in France, the budget balance rule in Germany and Switzerland and the debt rule in Poland.
For more information on automatic correction mechanisms see Schaechter and others (2012).
The expenditure ceiling formula incorporates a sort of automatic correction mechanism through two adjustments. First, any deviation from the expenditure ceiling at time t-1 will be corrected at time t, since real expenditure growth is calculated with respect to the previous budgeted level and not realization of expenditure. Second, higher expenditure than planned would entail higher debt at time t and this would lower the coefficient for expenditure growth at time t.