Which Fiscal Rules Are in Place?

3. Fiscal rules can be grouped into three main categories depending on their objectives.³

• Budget balance rules are the most common rules in advanced economies (Charts). They are present in 28 advanced economies, including five federations.⁴ These rules can support debt sustainability, help with economic stabilization, and are relatively easy to monitor and implement depending on their specification (overall balance, primary balance, structural balance, etc.). Structural balance rules account for economic shocks, such as cyclical and/or commodity price fluctuations for commodity exporters, enhancing the economic stabilization role of fiscal policy. However, inherent uncertainties in estimating output and commodity price gaps make these rules difficult to monitor and communicate.

• Debt rules are the second most common rules in advanced economies (22 countries, including four federations). These rules are theoretically the most effective rules in ensuring convergence to a debt target and they are easy to communicate. However, debt rules do not provide short-term guidance for fiscal policy because policy slippages and budgetary measures impact debt ratios with a lag. An additional operational challenge with debt rules is the difficulty of setting the appropriate debt target or objective.

• Expenditure rules, usually defined as caps to nominal spending or real expenditure growth, are the third most common rules in advanced economies. Relatively easy to monitor and communicate, these rules are present in 13 advanced economies, including three federations (Australia, Germany, and the United States). Expenditure rules can also enhance the stabilization role of fiscal policy by constraining spending during booms, when windfall revenues, particularly commodity-related revenues, are temporarily high, but allowing tax revenues to adjust to cyclical or discretionary changes during downturns. While most cyclically-sensitive items are on the revenue side, expenditure rules could potentially constraint automatic stabilizers on the spending side (e.g., unemployment benefits) during downturns. Excluding cyclically-sensitive expenditures from target variables is often discussed as a solution but this may complicate monitoring. By setting spending levels, expenditure rules can also provide operational guidance in choosing fiscal targets. However, expenditure rules alone do not provide a direct anchor for debt sustainability.

View Full Size

Number of Countries with Fiscal Rules, 2013 ^1/

(Count)

Citation: IMF Staff Country Reports 2015, 023; 10.5089/9781498385305.002.A003

^1/ Advanced economies with national and supranational rules.Source: IMF Staff estimates.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Number of Federations with Fiscal Rules, 2013 ^1/

(Count)

Citation: IMF Staff Country Reports 2015, 023; 10.5089/9781498385305.002.A003

^1/ Federal advanced economies with national and supranational rules.Source: IMF Staff estimates.

• Download Figure
• Download figure as PowerPoint slide

4. Because of the trade-offs across different rules with respect to their properties, many countries have combined elements of various fiscal rules. To minimize the gaps associated with each specific rule, the most common combinations among advanced economies associate a balance budget rule with a debt rule or an expenditure rule (Chart). Only Australia combines an expenditure rule with a debt anchor.

View Full Size

Combination of Fiscal Rules Across Advanced Economies, 2013 ^1/

(Count)

Citation: IMF Staff Country Reports 2015, 023; 10.5089/9781498385305.002.A003

ER = Expenditure Rule; DR = Debt Rule; BBR = Balanced Budget Rule.^1/ Advanced economies with national and suprational rules.Sources: FAD Fiscal Rules datasets.

• Download Figure
• Download figure as PowerPoint slide

How Has Compliance Been with Fiscal Rules?

5. Spending-based rules tend to be associated with effective control as countries comply more often with expenditure rules than with other rules. Drawing from Cordes and others (2014), this section assesses countries’ compliance rate with respect to the three main types of fiscal rules. The analysis highlights that, across advanced economies, expenditure rules have a better compliance record than budget balance and debt rules (Chart). The higher compliance rate of expenditure rules is even more striking when focusing on federal nations. Indeed, federal nations have complied with their expenditure rules close to 90 percent of the time. This higher compliance rate of expenditure rules is consistent with the fact that they are easy to monitor and immediately enforceable through the annual budget. In opposite to the budget balance and debt rules that are more exposed to shocks out of the government’s control (interest rate-growth differential, realization of contingent liabilities, etc.), expenditure rules are more directly connected to instruments effectively under the policymakers’ control.

View Full Size

Average Compliance Rates with Fiscal Rules, 1985 - 2012

(Percent)

Citation: IMF Staff Country Reports 2015, 023; 10.5089/9781498385305.002.A003

ER = Expenditure Rule ; DR = Debt Rule ; BBR = Balanced Budget Rule. Compliance rates are for national and supranational rules for yearas in which an assessment was possible.Sources: FAD Fiscal Rules datasets and IMF Staff estimates.

• Download Figure
• Download figure as PowerPoint slide

Key Considerations

6. An appropriate fiscal rule framework should address the need for short-term stabilization while ensuring long-term fiscal sustainability. Given the importance of natural resources for the Canadian economy, the design of a fiscal framework needs to internalize the higher likelihood of amplified business cycle that could stem from commodity price fluctuations. Balanced budget rules for headline fiscal balances are relatively simple and easy to communicate but can lead to procyclical fiscal policies. Among existing fiscal rules, the structural balance and expenditure rules better accommodate business cycle fluctuations by enabling automatic stabilizers to function. The technical complexity associated with the structural balance rule has, however, raised serious concerns for its implementation, including in Europe where this rule constitutes a centerpiece of the fiscal surveillance framework (Eyraud and Wu, 2014). Expenditure rules, on the other hand, allow a full play of automatic stabilizers on the revenue side and are relatively simple to implement.

7. A broader issue with expenditure-based rules is that they do not provide an anchor for long-term fiscal policy. While expenditure rules provide a proper operational target for fiscal policy in the short term, they do not offer a longer-term anchor. Expenditure rules are often directly related to the formulation of the annual or multi-year budget. These rules directly target expenditure pressures, often at the origin of excessive deficits, but still have a weaker link to debt sustainability than other rules with a broader coverage of fiscal aggregates (budget balance or debt rules).

8. Combining a short-term operational target with a long-term anchor could be a sensible option for Canada. A hybrid fiscal rule in a federal nation, the Swiss “debt brake,” has attracted a lot of attention thanks to its successful implementation. This arrangement combines an explicit correction mechanism through which the rule’s operational target (expenditure) is adjusted to reflect developments in the anchor (structural balance). Operationally, the Swiss “debt brake” implies that deviations from the structural balance target are recorded in an “amortization account” and need to be eliminated through spending cuts (Kinda and others, 2013). While fulfilling many criteria of an adequate rule, the Swiss type “debt brake” relies on the structural balance, which as explained above is a potential drawback because of the uncertainty in measuring the output gap.

9. A valuable alternative to the Swiss-type rule would be an expenditure rule anchored in a debt objective (Debrun, Epstein, and Symansky, 2008). Such combination would also be consistent with the authorities’ objective to reduce their debt-to-GDP ratio in the medium-term and contain spending, as put forward in the 2013 Speech from the Throne and confirmed in subsequent budgets. The new rule would trigger a correction (via spending) when debt deviates from a defined desirable debt path. By not relying on a measure of the business cycle, the proposed rule would be relatively simple and easy to communicate. It would also foster economic stabilization by preventing pro-cyclical spending during booms and encouraging the use of revenue windfalls for debt reduction. The rule would include three key elements: (i) a specific debt path (debt norm) leading to the medium or long-run debt objective; (ii) medium-term expenditure caps consistent with the deficits derived from the debt norm; and (iii) a debt-feedback or error-correction mechanism (ECM). In implementing the rule, the government will: (i) first define a desirable path for debt and a date to reach a debt target; and (ii) every three years set annual nominal growth ceilings for total government spending that are binding and met every single year to ensure a constant reduction in debt (given reasonable revenue forecasts). The ECM will imply progressive elimination of past deviations when expenditure ceilings are revised at the end of each planning horizon (see Box 1 for a conceptual framework of the expenditure rule with a debt-based ECM).⁵

Simulated Properties of Alternative Fiscal Rules

10. This section assesses the performance of five alternative fiscal rules at the federal level for Canada: (i) an annual balanced budget; (ii) a balanced budget over three years; (iii) a balanced budget over seven years; (iv) a debt-based ECM with a constant error-correction parameter; and (v) a debt-based, non-linear ECM with a variable correction parameter that increases with positive deviations from the debt norm and decreases with negative deviations. The first fiscal rule option targets a balanced budget every single year while options (ii) and (iii) allow for positive and negative deviations (up to ½ percent of GDP) that are corrected within three years (option ii) or seven years (option iii). Consistently with the authorities’ objective, the path under our assumed debt norm would reduce the federal government gross debt ratio in the medium term.⁶

11. The performance of the selected fiscal rules is assessed under three scenarios: a baseline and two deterministic shock scenarios. Under the baseline, nominal growth is assumed to converge to its potential of 4½ percent in 2018, closing the output gap. The first shock scenario assumes a positive output shock. Under this scenario, the output gap widens in 2016 and reaches 4¾ percent of potential GDP in 2018, before gradually converging to zero. The second shock scenario is a boom-bust cycle, relatively common with commodity price fluctuations, where the output gap first increases in 2016 and reaches 4¾ percent of potential GDP in 2017, then drops to -6¾ percent in 2021 before closing by 2023. All simulations are performed under a partial equilibrium approach.⁷

12. The assessment of the relative performance of each fiscal rule is based on its properties for debt reduction along the debt norm and economic stabilization (cyclicality). The root mean squared deviation (RMSD) from the debt norm captures the flexibility of each fiscal rule option while the mean deviation from the norm and the deviation from the norm in 2020 capture the precision in hitting the debt objective. The cumulative procyclical impulses over the period 2016–20 capture the procyclicality associated with each fiscal rule option. Cumulative procyclical impulses correspond to improvements in the primary balance during bad times, and deteriorations of the primary balance in good times.⁸ A higher indicator corresponds to a more procyclical rule with negative number indicating that fiscal impulses have been countercyclical during the period.

Box 1.

A Conceptual Framework for an Expenditure Rule with a Debt-Based Error Correction Mechanism

Following Debrun and others (2008), the debt brake approach illustrated below relies on two main components: an anchor and an operational target. The anchor is a specific debt path (debt norm) leading to the government’s long-run debt objective. The operational target (or implementation tool) is a set of expenditure ceilings that are consistent with the deficits underlying the debt norm. Assuming a planning period of s years (we assume three years in the simulations), the government should set (at the end of year x) nominal expenditure ceilings for each year between x+1 and x+s.

• If actual debt is below the defined normative debt path ${\left\{d_{\tau }^{*}\right\}}_{\tau =t_0}^T$, that is if $(d_x\le d_x^{*})$, but above the long-run debt objective of the government (d^o), the government should define expenditure ceilings that preserve the margin with respect to the norm and aim at reducing debt at a similar pace as the norm (see figure).

• If actual debt is above the norm $(d_x>d_x^{*})$, expenditure ceilings should be defined to ensure a faster reduction in debt during s years (along a transitional path $\{d_{\tau }^{**}{\}}_{\tau =x}^{x+s}$ but not an over correction: $\left|{\mathrm{\Delta }}_{x+1,x+s}{d}_t^{*}\right|<\left|{\mathrm{\Delta }}_{x+1,x+s}{d}^{**}\right|\le |d_{x+s}^{*}-d_x|$.

View Full Size

Illustrative Paths: Norm and Actual Debt

Citation: IMF Staff Country Reports 2015, 023; 10.5089/9781498385305.002.A003

• Download Figure
• Download figure as PowerPoint slide

The required debt reduction could take the following functional form:

${\mathrm{\Delta }}_{x+1,x+s}{d}^{**}=\{\begin{array}{c}{\mathrm{\Delta }}_{x+1,x+s}{d}^{*}-{\lambda }_x(d_x-d_x^{*}),if{d}_x>d_x^{*}and{d}_{x+1}^{*}<d_x^{*},orif{d}_{x+1}^{*}=d_x^{*}=d^{\circ }\\ {\mathrm{\Delta }}_{x+1,x+s}{d}^{*},if{d}_x\le d_x^{*}and{d}_{x+1}<d_x^{*}\hfill \end{array}$

with ${\lambda }_x={\lambda }_0+\lambda (d_x-d_x^{*})$ and 0 < λ_x < 1. The adjustment speed λ_x is set to increase with the actual deviation from the norm but remains below 1, implying that deviations are not fully corrected within a year and the required fiscal efforts do not lead to excessive adjustment.

This mechanism yields to a downward trend in the debt ratio as long as debt is above the government’s long-run debt objective (d^o). When debt becomes lower than the government’s long-run debt objective, the symmetry in the error correction mechanism can be restored, allowing for more expenditure growth under the fiscal rule framework.

The expenditure ceilings can be defined in nominal terms as:

$\{\begin{array}{c}{\overline{G}}_{x+1}=[{\tau }_{x+1}-b_{x+1,x+s}^{*}]E_x{Y}_{x+1},\\ ⋮\hfill \\ {\overline{G}}_{x+s}=[{\tau }_{x+s}-b_{x+1,x+s}^{*}]E_x{Y}_{x+s}\end{array}$

with E_x representing the expectation operator conditional on information available at the end of year x-1 and $b_{x+1,x+s}^{*}$ the required annual budget-balance over the planning horizon [x+1; x+s] to achieve the targeted debt reduction Δ_x+1,x+sd^**. Y represents the nominal GDP and τ the revenue to GDP ratio. Alternative formulations of the expenditure rule include a cap on nominal expenditure growth consistent with ${\overline{G}}_{x+1},\mathrm{\dots },{\overline{G}}_{x+s}$.

The budget-balance can be expressed as:

$b_{x+1,x+s}^{*}=\left[\frac{(1+g{)}^{-(s+1)}-1}{\underset{j=0}{\overset{s}{\mathrm{\Sigma }}}(1+g{)}^{-(s-j)}}\right]d_x-{\mathrm{\Delta }}_{x+1,x+s}{d}^{**}$

with g representing a constant predicted nominal GDP growth rate.

While expenditure ceilings are normally set for s years to allow for automatic stabilizers on the revenue side, they require certain flexibility. Setting a maximum deviation from the debt norm beyond which tighter expenditure ceilings are defined is critical to prevent excessive deviation from the debt norm.¹ If m<0 represents the maximum deviation under the debt norm, a new set of expenditure ceilings covering the years k+1 to k+3 (k ∈ [x+1; x+3]) should be defined if $d_k>d_k^{*}+m$. Exceptional circumstances such as a well-defined large economic contraction or a natural disaster should be accommodated through revisions of expenditure ceilings and if need be the debt norm.

¹ Setting a maximum deviation of about 5 to 10 percent of GDP to avoid significant contractions in bad times would be a sensible choice but this remains a matter for discussion. The framework does not account for the potential impact of fiscal multipliers.

13. Results from the simulations highlight that while balanced budget rules can be flexible, expenditure rules with a debt anchor are consistently associated with countercyclical fiscal policies (Table 1 and Figure 1):

Table 1.

Comparative Performance of Alternative Fiscal Rules for Canada, 2016-2020

Sources: IMF staff estimates and calculations.

^1/Percent of GDP (2016-2020).

Table 1.

Comparative Performance of Alternative Fiscal Rules for Canada, 2016-2020

	Balanced buget			Expenditure rule with debt anchor
	Annual	Over three years	Over seven years	Constant error correction	Variable error correction
Baseline
Root mean square deviation with respect to the norm path 1/	0.32	1.70	1.59	0.73	0.71
Mean deviation with respect to the norm path 1/	0.17	1.54	1.44	0.73	0.71
Difference from 20 percent debt in 2020	0.55	2.61	2.63	0.71	0.68
Procyclicality (+ is procyclical; -is countercyclical)	0.07	-0.61	-1.13	-0.29	-1.34
Positive output shock
Root mean square deviation with respect to the norm path 1/	0.76	0.54	0.35	0.73	0.74
Mean deviation with respect to the norm path 1/	-0.74	0.21	0.11	-0.63	-0.64
Difference from 20 percent debt in 2020	-0.65	1.11	0.61	-0.64	-0.65
Procyclicality (+ is procyclical; -is countercyclical)	0.36	1.71	0.18	-0.89	-0.89
Boom-bust cycle
Root mean square deviation with respect to the norm path 1/	1.03	2.07	1.96	1.43	1.43
Mean deviation with respect to the norm path 1/	0.02	1.41	1.31	0.63	0.62
Difference from 20 percent debt in 2020	1.67	3.84	3.84	2.74	2.74
Procyclicality (+ is procyclical; -is countercyclical)	0.16	-0.51	-1.04	-0.80	-0.80

Sources: IMF staff estimates and calculations.

^1/Percent of GDP (2016-2020).

View Table

Table 1.

Comparative Performance of Alternative Fiscal Rules for Canada, 2016-2020

	Balanced buget			Expenditure rule with debt anchor
	Annual	Over three years	Over seven years	Constant error correction	Variable error correction
Baseline
Root mean square deviation with respect to the norm path 1/	0.32	1.70	1.59	0.73	0.71
Mean deviation with respect to the norm path 1/	0.17	1.54	1.44	0.73	0.71
Difference from 20 percent debt in 2020	0.55	2.61	2.63	0.71	0.68
Procyclicality (+ is procyclical; -is countercyclical)	0.07	-0.61	-1.13	-0.29	-1.34
Positive output shock
Root mean square deviation with respect to the norm path 1/	0.76	0.54	0.35	0.73	0.74
Mean deviation with respect to the norm path 1/	-0.74	0.21	0.11	-0.63	-0.64
Difference from 20 percent debt in 2020	-0.65	1.11	0.61	-0.64	-0.65
Procyclicality (+ is procyclical; -is countercyclical)	0.36	1.71	0.18	-0.89	-0.89
Boom-bust cycle
Root mean square deviation with respect to the norm path 1/	1.03	2.07	1.96	1.43	1.43
Mean deviation with respect to the norm path 1/	0.02	1.41	1.31	0.63	0.62
Difference from 20 percent debt in 2020	1.67	3.84	3.84	2.74	2.74
Procyclicality (+ is procyclical; -is countercyclical)	0.16	-0.51	-1.04	-0.80	-0.80

Sources: IMF staff estimates and calculations.

^1/Percent of GDP (2016-2020).

View Full Size

Simulated Debt Path Under Various Fiscal Rules, 2013 - 2023 (percent of GDP)

Citation: IMF Staff Country Reports 2015, 023; 10.5089/9781498385305.002.A003

Sources: IMF staff estimates and calculations.

• Download Figure
• Download figure as PowerPoint slide

• All rules, as specified, lead to a convergence towards the medium-term debt objective. The rule aiming for an annual balanced budget allows for a slightly better targeting of the 2020 debt objective under the baseline and the boom-bust scenarios. However, this rule is the least flexible option, as indicated by its lower root mean square deviation. The balanced budget rules defined over three or seven years stand out as the most flexible rules, in particular when economy faces large fluctuations as in the boom-bust scenario. Expenditure rules with a debt anchor rank in the middle range in terms of flexibility in response to shocks.

• The two variants of the expenditure rules with a debt anchor tend to consistently foster more countercyclical fiscal policies. The expenditure rule with a debt anchor and a variable error correction, which allows for faster correction of larger deviations from the debt norm, is particularly countercyclical under baseline. The balanced budget rules defined over a multi-year horizon, particularly seven years, can also support countercyclical fiscal policies (baseline and boom-bust scenarios). However, setting the balanced budget rule in annual term is consistently associated with a procyclical stance of fiscal policy (under the baseline and the shock scenarios).