## wpiea2022011-print-pdf ## Source details **Canonical URL:** [wpiea2022011-print-pdf](https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022011-print-pdf.pdf) ## Other formats - [Markdown version](/-/media/files/publications/wp/2022/english/wpiea2022011-print-pdf.pdf.md) - [Structured JSON version](/-/media/files/publications/wp/2022/english/wpiea2022011-print-pdf.pdf.json) --- ### Definitions and scope: What is considered a fiscal rule and dataset coverage - Fiscal rules defined as long-lasting constraints on fiscal policy through numerical limits on broad budget aggregates. - Dataset covers specifics and institutional features for over 100 countries during the period 1985–2021. - As of end-2021: - about 105 economies have adopted at least one fiscal rule, 11 countries more than the last update in 2015 and 96 countries more than 1985. - 53 countries have supranational rules, many complemented with national rules. - Inclusion principles: - Only rules with targets fixed in legislation and revisable only on a low-frequency basis and binding for at least three years are considered fiscal rules. - Dataset considers rules that set numerical targets on aggregates covering a large share of public finances and at a minimum central government level; subnational rules and fiscal sub-aggregates are excluded. - Focus is on de jure arrangements, not adherence in practice. - Temporal coverage: indicators cover rules that took effect by end-December 2021 or with transition regimes in place at that time; adopted but not yet implemented rules are described but not included in charts and tables. ### Types of fiscal rules and common combinations - Four main rule types in dataset: expenditure rules (ER), revenue rules (RR), budget balance rules (BBR), and debt rules (DR). - Typical configurations and prevalence: - About 70 percent of countries with fiscal rules have a debt rule combined with operational limits on annual budget aggregates. - Out of economies with fiscal rules in 2021: - one third had a debt rule together with a deficit limit and an expenditure ceiling; - another quarter had a debt rule combined with a budget balance rule. - Expenditure rules: - Often set as a ceiling on annual expenditure growth; about three quarters of advanced economies have expenditure rules. - Less than a third of emerging markets and developing economies adopted expenditure rules. - Revenue rules: - Less used than other types; often set as a ceiling on revenue-to-GDP ratio in advanced countries or as a floor in low-income countries. - Debt rules: - Over 80 percent of EMDEs have adopted them. - Majority set as a debt limit or ceiling; about 10 percent use a medium-term anchor concept. - Most expressed in percent of GDP; occasionally set in net present value terms for low-income countries. ### Design features: flexibility, legal basis, enforcement, coverage, exclusions - Flexibility and escape clauses: - Greater flexibility a key trend; escape clauses and formal correction mechanisms are increasingly common. - Before the pandemic, two thirds of countries with fiscal rules had included escape clauses. - Over 30 countries invoked escape clauses during the COVID-19 pandemic. - Legal basis and enforcement: - In 2000, only 30 percent of countries established fiscal rules in legislation. - Currently more than 60 countries have fiscal rules featured at or above statutory levels. - As of 2021, over 40 percent of fiscal rules were supported by fiscal responsibility or budget framework laws (doubled from a decade ago). - Formal enforcement: 72 out of 104 countries with fiscal rules have formal enforcement mechanisms. - Coverage: - Majority of supranational rules cover general government; less than half of national rules do so. - Coverage weights in strength index: general government weight = 1; central government weight = 0.75. - Exclusions and multi-year ceilings: - Common exclusions from expenditure rules: capital spending, interest payments, pensions, nondiscretionary unemployment benefits. - As of 2021, over a third of countries with fiscal rules had established multi-year ceilings. ### Fiscal councils: definition, growth, mandates, and independence - Definition and role: - Independent non-partisan agencies with an official mandate to assess fiscal policy, plans, and rule compliance. - Tasks: assess public finances; review annual or medium-term budgets; evaluate macroeconomic and budgetary forecasts; monitor fiscal rule implementation; cost government measures. - Growth and forms: - There were 51 fiscal councils in 49 countries as of 2021. - About twice the number in 2010 and a third more relative to 2016. - Institutional forms: attached to legislature (parliamentary budget offices), attached to executive, stand-alone entities, or attached to the supreme audit institution. - Mandates and activities (2021): - 80 percent of fiscal councils were tasked to monitor implementation of fiscal rules (a 25-percentage points increase from 2010). - Nearly half involved in costing policy measures. - Many prepare or assess macroeconomic and budget forecasts; share preparing forecasts has gradually decreased. - Independence: - Over 80 percent of fiscal councils in advanced economies had de-jure operational independence in 2021. - Operational independence lower in EMDEs; examples of resource constraints or political interference cited. ### Trends before and during the pandemic (aggregate findings) - Pre-pandemic: - Growing adoption of rules-based fiscal frameworks over past two decades, with advanced economies first but increasing adoption among EMDEs since the late 2000s. - As of end-2021, there are more than twice EMDEs than advanced economies with fiscal rules. - Expansion occurred in waves driven by supranational rules and adoption after large shocks (e.g., early 1990s linked to 1992 Maastricht Treaty). - Pandemic impacts: - COVID-19 led to widespread activation of escape clauses, temporary suspensions, and ad-hoc modifications to permit extraordinary fiscal support. - About 90 percent of countries had deficits larger than the rule limits in 2020. - Public debt exceeded limits or anchor levels in over half of countries in 2020. - Deviations from deficit rule limits are associated with higher financing costs and tend to occur during faltering economic growth. - Evidence indicates countries with good pre-crisis rule track records responded more forcefully to large shocks. - Key policy challenge: how and when to return to fiscal rules after crisis-driven suspensions. ### Deviations from fiscal rule limits: patterns, magnitudes, and projections - Definitions: - Formal deviation: D_i,t^X = X_i,t − X_i,t^limit, where X is deficit or debt as percent of GDP. - BBR deviations use IMF WEO budget balance concepts; DR deviations use gross debt versus rule limit or anchor. - Some rules specify NPV debt (typically about 10 percent smaller than nominal debt); NPV debt not available in WEO so nominal debt used as approximation. - Historical frequency (2004–21 averages): - Countries exceeded deficit limits about 50 percent of the time. - Countries exceeded debt limits about 42 percent of the time. - Pandemic-era magnitudes (2020): - About 90 percent of countries with BBRs saw deficits exceed rule limits in 2020; median positive deviation about 4 percent of GDP. - Over half of countries with DRs had debt exceeding limits or anchors in 2020. - Median deviation of debt: advanced economies = 50 percent of GDP; emerging market and developing economies = 26 percent of GDP. - Post-pandemic prospects: - Returning to rule limits likely to take time; WEO projections suggest many countries will continue to have deficits exceeding rule limits in the next 2–3 years. - EU intended to deactivate escape clause by end 2022 but many member states will need transitional periods. - Debt projected to remain significantly higher than limit or anchor levels for most advanced economies; deviations projected to decline gradually over medium term. ### Persistence and dynamics of deviations (econometric results) - Empirical specification: - Panel regression: D_i,t^X = β_0 D_i,t−1^X + β_1 D_i,t−1^X * I(D_i,t−1^X < 0) + β_2 D_i,t−1^(−X) * I(D_i,t−1^(−X) ≥ 0) + β_3 D_i,t−1^(−X) * I(D_i,t−1^(−X) < 0) + β_4 g_i,t + δ_i + u_i,t - Sample: about 90 countries for 1990–2021 (smaller than number with rules in 2021 because not all rules have simple measurable limits). Data from IMF Fiscal Rules and WEO; 2021 data are IMF projections. - Key persistence estimates (Annex II summary): - For countries with deficits exceeding the rule limit, estimated autocorrelation coefficient ≈ 0.55. - Interpretation: a country with the typical deviation (1.9 percent) would on average take 3–4 years to converge close to its means. - For observations below the BBR limit in previous period, autocorrelation coefficient ≈ 0.81. - Interpretation: a typical average gap of 1.9 percent of GDP below the mean would remain so in 10 years absent a large shock. - Debt rule deviations very persistent: autocorrelation coefficient ≈ 0.9, without significant differences between countries above or below debt ceilings/anchors. - Regression dynamics after breaching BBR limits: - BBR deviations typically coincide with a large decline in real GDP growth in the initial period; growth tends to stay lower in subsequent years. - When deficits cross rule limits, countries typically face persistent debt buildup and higher financing costs; CDS typically rise gradually for about 3–4 years to about 1 percentage points higher. - Interaction effects: - Negative and statistically significant coefficient on DR deviations when interacting BBR and DR deviations — higher debt tends to associate with smaller deficit deviation. - Real GDP growth coefficient: slightly negative for deficit deviations and large and statistically significant for DR deviations. ### Regression robustness and selected coefficients (Arellano-Bond and IV results) - Arellano-Bond (lag persistence reported in Table A2.2): - lag BBR_deviation (baseline): 0.807*** (0.055) - lag BBR_deviation (including small countries): 0.765*** (0.054) - lag BBR_deviation (instrumenting GDP growth): 0.736*** (0.099) - lag BBR_deviation (Big Recessions): 0.707*** (0.113) - lag BBR_deviation (with year FE): 0.822*** (0.059) - lag BBR_deviation (exceeding) baseline: -0.254*** (0.074) - lag DR_deviation baseline: 0.910*** (0.029) - lag DR_deviation (including small countries): 0.832*** (0.043) - RGDP growth coefficients reported (BBR baseline): -0.404*** (0.030); (year FE) -0.178*** (0.057) - N reported across specifications: 732 - IV first-stage instruments (Table A2.3): - lag Real GDP growth coefficient for 2009: 0.572*** (0.189) - lag Real GDP growth coefficient for 2020: 0.596* (0.321) - lag Construction Sector share (2009): -71.435* (38.629) - Stringency Covid measures (2020): -0.160*** (0.040) - N: 3065; R squared: 0.243 (2009), 0.264 (2020) - Fiscal impulse regressions (selected values from Table A2.4): - Average Deviation in Past Years: - 2009 baseline: 0.450 (0.389) - 2020 baseline: 0.368** (0.182) - lag Debt/GDP examples: - 2009 baseline: -0.016* (0.009) - 2020 baseline: -0.016** (0.007) - lag Primary Balance examples: - 2009 baseline: 0.046 (0.179) - 2020 baseline: -0.336*** (0.115) - 2009 IV: -0.507*** (0.131) - 2020 IV: -0.376*** (0.123) - Robustness: OLS and IV yield similar main results; alternative measures and instruments produce similar findings though sample sizes vary. ### Selected country and supranational examples of pandemic adjustments - Escape clause activations and suspensions: - Over 30 countries invoked escape clauses (examples include Armenia, Georgia, Grenada, United Kingdom). - Jamaica activated an escape clause in 2020 for one year and postponed achieving the debt target by two years to FY2027/28. - Supranational activations sometimes automatically activated national clauses (examples: Cote d’Ivoire, France, Italy, Portugal, Senegal). - Temporary ad-hoc suspensions and modifications: - Countries suspending rules temporarily: Azerbaijan, Colombia, Iceland, Indonesia, Peru, Russia. - Nearly 20 countries modified national rule limits during the pandemic (examples: Chile, Ecuador, Malaysia, Mexico, Mongolia, Namibia, Panama, Vietnam). - Ecuador (2020): introduced an expenditure growth rule, new annual budget balance targets, and a debt rule capping nonfinancial public sector debt at 40 percent of GDP by 2032 with interim targets. - Panama (late 2020): raised the deficit target (above its 2 percent of GDP limit) and anchored budget deficit at 1.5 percent of GDP from 2025. - Country highlights: - Brazil (2020): declared a “public calamity”; amended constitution to create a “war budget” easing of 8.4 per cent of GDP for COVID-19 measures and exempted prohibition to borrow for current expenditures. - Colombia: Social Investment Law (September 2021) introduced a medium-term debt anchor and revised structural net primary balance ceiling; transition path of deficits during 2022–25; fiscal council to gain greater operational independence. - India: activated escape clause in February 2020 allowing a temporary deviation of deficit not exceeding ½ percentage points of GDP a year; raised FY2019/20 and FY2020/21 deficits to 3.8 and 3.5 per cent of GDP, respectively. - Poland: escape clause allowed higher expenditure limits; scope to raise deficit up to 8 per cent of GDP introduced in 2020. - United Kingdom: activated escape clause March 2020; Oct 2021 reinstated fiscal rules including a balanced current budget by the third year of the rolling forecast period and a ceiling on government net investment of 3 per cent of GDP on average during the rolling period. ### Measuring the strength of fiscal rules (Annex III) - Strength index framework: - Based on five institutional criteria: (i) statutory or legal basis; (ii) room for setting/revising rules; (iii) nature of monitoring entity; (iv) correction mechanism; (v) resilience against shocks. - Strength scored for each rule type using 11 indicators; indicator scores standardized between 0 and 1; escape clauses and cyclical adjustment carry [0,2] scoring in mapping. - Coverage and weighting: - Coverage weights: general government = 1; central government = 0.75. - Multiple-rule weighting: highest rule weight = 1; second = 1/2; third = 1/3; fourth = 1/4. - Composite index standardized by sample mean and standard errors. - Reported ranges: - European Commission style index range: -0.99 to 3.5. - Paper’s compiled index standardized range: -0.7 to 3.6. - Theoretical maximum score per rule before coverage weights: 10. - Validation and robustness: - Correlation with European Commission index for EU countries: 0.72. - Alternative indices (count supranational and national separately; penalize too many/amended rules) have correlation ≥ 0.9 with baseline. - Empirical associations: - Strength index improved over time across income groups (2010–2020), with notable drivers: greater resilience in advanced economies and stronger legal basis and monitoring in EMDEs. - Higher strength index associated with stronger primary balance (R-squared reported = 0.1953 for fitted relation shown). - Countries with stronger BBRs typically have smaller and less frequent breaches; relationship stronger in EMDEs. No strong correlation between debt deviations and strength index. - Tradeoffs: - Improved flexibility/resilience (escape clauses, correction mechanisms) coincided with greater complexity and multiple rules, reducing simplicity scores. - Sustainability measured by (i) existence of debt rule; (ii) frequency of changes in debt rule; (iii) deviations from debt rule limits. ### Policy implications and recommendations - Returning to fiscal rule limits will likely take time, especially where large debt deviations exist; historical evidence shows debt deviations are more persistent than deficit deviations. - When charting transition paths (especially in supranational contexts), consider long-standing divergence among member states and the feasibility of uniform pace rules (example: EU 1/20 rule; Italy example where debt ≈ 160 percent of GDP in 2020 would imply ~5 percentage points of GDP reduction each year under 1/20). - Recommended focus areas to strengthen fiscal frameworks: - Strong political commitment and a better record of compliance. - Incentives to build buffers during good times. - Designing effective mechanisms to manage large shocks in bad times (clear escape clauses and formal correction mechanisms). - Strengthening fiscal councils’ independence and capacity to improve credibility and accountability. - Empirical finding: maintaining deficits below limits appears to strengthen fiscal credibility and allows larger fiscal responses when shocks hit; a good track record of rule compliance associated with larger fiscal response in 2020 (statistically significant) and similar sign for 2009 (not statistically significant). *Italic: wpiea2022011-print-pdf — IMF Working Paper "Fiscal Rules and Fiscal Councils: Recent Trends and Performance during the COVID-19 Pandemic" (dataset coverage and key findings as presented in the supplied content).* ### Annex I. Definit ions and Types of Fiscal Rules and Councils ........................................................... ### Annex I. Definit ions and Types of Fiscal Rules and Councils ### What is considered a fiscal rule in this paper and included in the dataset? - Fiscal rules are defined as long-lasting constraints on fiscal policy through numerical limits on broad budget aggregates. - The fiscal rule dataset covers the specifics and institutional features for over 100 countries during the period 1985–2021. - As of end-2021, about 105 economies have adopted at least one fiscal rule, 11 countries more than the last update in 2015 and 96 countries more than 1985. - 53 countries have supranational rules, many of which are complemented with national rules. ### Types of fiscal rules - Common rule types include debt ceilings or debt anchors and operational rules on expenditures or the budget balance. - Most countries have multiple fiscal rules covering different budget components to achieve fiscal objectives. - Typical combinations: a debt ceiling or debt anchor supported by other operational rules (e.g., expenditure rules, budget-balance rules). ### Other key characteristics and design features - Greater flexibility in rules has been a key trend, including increased use of escape clauses. - Features of rules that have evolved include flexibility, enforcement, and monitoring. - Escape clauses and formal correction mechanisms are part of rule designs to accommodate shocks; where escape clauses are absent, countries have used ad-hoc suspensions or modifications of rules, or introduced new fiscal rules. ### Fiscal councils: definition, role, and trends - Fiscal councils are independent non-partisan agencies with an official mandate to assess fiscal policy, plans, and rule compliance. - Fiscal councils have often been tasked to provide fiscal oversight, including monitoring fiscal rules and assessing credibility of budgets and quality of public policies. - The number of fiscal councils has doubled over the last decade. - Many fiscal councils were established to monitor new fiscal rules or in response to external pressures after large shocks. - During the COVID-19 pandemic, fiscal councils provided costing of emergency measures and analyses of the crisis impact on public finances, and assessed governments’ responses and use of escape clauses. ### Trends before the pandemic (aggregate findings) - During the past two decades, a growing number of countries adopted rules-based fiscal frameworks. - Advanced economies were frontrunners in adopting fiscal rules, but rules became increasingly common among emerging market and developing economies (EMDEs), especially since the late 2000s. - As of end 2021, there are more than twice EMDEs than advanced economies with fiscal rules. - The expansion of rules occurred in waves, driven by supranational rules and the adoption of rules after large shocks (example: early 1990s increase linked to the 1992 Maastricht Treaty). ### Pandemic impacts, deviations from rules, and implications - The COVID-19 pandemic led to widespread activation of escape clauses to temporarily suspend rule limits, allowing extraordinary fiscal support to households and firms. - About 90 percent of countries had deficits larger than the rule limits in 2020. - Public debt exceeded the limits or anchor levels in over half of countries, adding to already large pre-COVID deviations. - Deviations from deficit rule limits are associated with higher financing costs and tend to occur in periods of faltering economic growth. - Evidence indicates countries with a good track record in abiding by rules before crises can respond to large shocks more forcefully. - A key policy challenge identified is how and when to return to fiscal rules after crisis-driven suspensions. *Source: Annex I and related excerpts from IMF Working Paper Fiscal Rules and Fiscal Councils: Recent Trends and Performance during the COVID-19 Pandemic (dataset coverage and key findings as presented in the supplied content).* ### 2. Fiscal rules by income groups ### 2. Fiscal rules by income groups ### Overview and recent trends - Sources: IMF Fiscal Rules Dataset: 1985–2021; Davoodi, Elger, Fotiou, Garcia-Macia, Lagerborg, Lam, and Pillai (2022a). - The average number of fiscal rules per country has increased steadily: - Countries now have an average of about 3 fiscal rules, up from about 2 in the early 2000s. - The updated fiscal rule dataset contains 106 economies. - The increase in the number of rules is more pronounced in Europe, where many countries have adopted national rules alongside supranational rules. - Multiple rules can be motivated to ensure greater fiscal discipline or achieve multiple fiscal objectives, but they: - Increase complexity of the fiscal framework. - Create potential conflicts between rules and objectives. - Make compliance more difficult to explain and monitor. - The use of multiple rules has sometimes reflected political difficulties in eliminating existing rules when introducing new ones. ### Types of fiscal rules and common combinations - Four main types of fiscal rules in the dataset: expenditure rules (ER), revenue rules (RR), budget balance rules (BBR), and debt rules (DR). - The most common configuration is a debt rule together with operational limits on expenditures and/or budget balance: - About 70 percent of countries with fiscal rules have a debt rule combined with operational limits on annual budget aggregates. - Out of the economies with fiscal rules in 2021: - one third had a debt rule together with a deficit limit and an expenditure ceiling; - another quarter of economies had a debt rule combined with a budget balance rule. - Expenditure rules: - Increasingly common, often set as a ceiling on annual expenditure growth. - Revenue rules: - Less used than other rule types. - Often set as a ceiling on revenue-to-GDP ratio in advanced countries (example: Belgium). - Often set as a floor in low-income countries (example: WAEMU) to encourage revenue mobilization. - Debt rules: - Particularly common in developing economies: over 80 percent of EMDEs have adopted them. - Majority set as a debt limit or ceiling; about 10 percent use a (medium-term) anchor concept (examples: Finland, United Kingdom). - Most debt rules expressed in percent of GDP; occasionally set in net present value terms for low-income countries. ### Differences across income groups - Expenditure rules: - About three quarters of advanced economies have expenditure rules. - Less than a third of emerging markets and developing economies adopted expenditure rules (examples: Brazil, Mongolia, Paraguay). - Debt rules: - Over 80 percent of EMDEs have adopted debt rules. - Budget balance rules accounting for business cycles: - More predominant in advanced economies (examples: Czech Republic, Estonia) than in emerging markets (examples: Chile, Colombia). - For EMDEs with cycle-adjusted deficit limits, reliance is often on thresholds on actual activity rather than on output gap measures. ### Flexibility, legal basis, and enforcement features - Flexibility enhancements and escape clauses: - Many countries introduced escape clauses over the past two decades; escape clauses became a cornerstone of ‘second generation’ fiscal rules. - Before the pandemic, two thirds of countries with fiscal rules had included escape clauses. - Over 30 countries invoked escape clauses during the COVID-19 pandemic. - Legal basis: - In 2000, only 30 percent of countries established fiscal rules in legislation. - Currently more than 60 countries have fiscal rules featured at or above statutory levels (e.g., fiscal responsibility or budget framework laws, constitutions). - As of 2021, over 40 percent of fiscal rules were supported by fiscal responsibility or budget framework laws (doubled from a decade ago). - Formal enforcement: - A large share of countries with fiscal rules have formal enforcement mechanisms: 72 out of 104. - The share with formal enforcement has largely stayed flat over the past decade. - Examples of enforcement/correction mechanisms: 2012 Fiscal Compact correction mechanisms in Europe; Austria strengthened Court of Auditors and sanctions; Poland established preemptive triggers as debt approaches limits; Peru and Panama have correction mechanisms guiding returns after deviations. - Stabilization (cyclical adjustment): - In 2021, about 40 countries had rules that accounted for fluctuations over business cycles. - Over the last decade, the share of countries with rules adjusted for cyclical conditions has gradually declined. ### Strength of fiscal rules and tradeoffs - A standardized strength index on fiscal rules is constructed based on four institutional criteria: - (i) legal basis; - (ii) presence of a monitoring mechanism; - (iii) enforcement and correction mechanism in place; - (iv) flexibility and resilience against shocks. - Scoring and weighting: - A numerical score is assigned for each type of fiscal rule based on indicators in the four criteria. - Rules are weighted reflecting government coverage, with higher weight for general government than central government. - Scores are summed into a single index; higher score indicates greater strength. - Trends in the index: - The standardized scoring index shows continued improvement on average over time across income groups (2010–2020). - Notable drivers: - Greater resilience against shocks in advanced economies. - Adoption of stronger legal basis and monitoring in EMDEs. - Variation across countries: several countries have seen declines owing to frequent amendments and growing complexity. - Tradeoffs: - Flexibility and resilience have improved, but rules are becoming more complicated. - Multiple rules reduce the scoring of simplicity principles. - Existence of debt rules that are not amended frequently helps contribute to the sustainability principle. ### Fiscal councils (independent fiscal institutions) - Definition and role: - Fiscal councils are nonpartisan public entities with statutory or executive mandates to promote sustainable public finances through assessing fiscal plans and performance, evaluating macroeconomic and budgetary forecasts, monitoring implementation of fiscal rules, and costing government measures. - They raise reputational and electoral costs of undesirable policies and broken commitments. - Growth in number: - There were 51 fiscal councils in 49 countries as of 2021. - This is about twice the number in 2010 and a third more relative to the last update in 2016. - The European Fiscal Board established in 2015 monitors supranational rule implementation. - The rise of fiscal councils extended beyond Europe into emerging markets (examples: Brazil, Costa Rica, Chile, Panama). - Institutional forms: - Most are attached to legislature (parliamentary budget offices), the executive, or are stand-alone entities. - Examples: parliamentary budget offices in Korea and the United States; executive-attached councils in Belgium, Croatia, the Netherlands, United Kingdom; stand-alone entities in Czech Republic, Slovak Republic, Sweden. - Two countries (France and Finland) attach fiscal councils to the supreme audit institution. - Mandates and activities: - Almost all assess public finances and review annual or medium-term budget proposals and/or long-term sustainability. - Half of fiscal councils globally provide normative analysis with recommendations on fiscal policies. - Many are responsible for preparing or assessing macroeconomic and budget forecasts, although the share preparing forecasts has gradually decreased over time. - Examples of forecast roles: - Netherlands: government must use council’s forecasts in the budget. - United Kingdom: use of OBR forecasts required, with ‘comply or explain’ if not used. - Canada and United States: council forecasts serve as comparators to official projections. - Forecast horizons vary: short- to medium-term; some councils use a 10-year horizon (U.S. Congressional Budget Office) or 40–75 years for fiscal sustainability (Canada, Korea, Netherlands, United Kingdom). - In 2021, 80 percent of fiscal councils were tasked to monitor the implementation of fiscal rules (a 25-percentage points increase from 2010). - Nearly half of fiscal councils are involved in costing policy measures. - Independence and access to information: - Improvements over two decades, but challenges remain in ensuring operational independence and access to information. - Over 80 percent of fiscal councils in advanced economies had de-jure operational independence in 2021 (examples of safeguards: appointing own staff, own press channel, long-term appointments). - Operational independence is lower in EMDEs, particularly where budget safeguards and resources are subject to political discretion. - Examples of challenges: Canada’s council had limited resources; Hungary’s council faced restructuring and resource cuts in 2021. ### III. Rules-based fiscal frameworks tested by the COVID-19 pandemic - The pandemic tested resilience of rules-based frameworks; many governments deviated from rule limits to provide fiscal support. - Fiscal councils were active in assessing crisis impact, costing emergency measures, and monitoring suspension of rules—supporting transparency and accountability. Adjustments used by countries during the pandemic - Different adaptation methods were used, including activation of escape clauses, temporary suspension, and modification of rule limits. - Activation of escape clauses: - Over 30 countries invoked escape clauses in national and supranational rules (examples: Armenia, Georgia, Grenada, United Kingdom). - Jamaica activated its escape clause in 2020 for one year and postponed achieving the debt target by two years to FY2027/28. - Supranational escape clause activations sometimes automatically activated national clauses (examples: Cote d’Ivoire, France, Italy, Portugal, Senegal). - Temporary ad-hoc suspension: - Examples of countries suspending rules temporarily: Azerbaijan, Colombia, Iceland, Indonesia, Peru, Russia. - Colombia suspended rules for two years to make room for higher deficits. - Indonesia suspended fiscal rules via constitutional provisions allowing a regulation in Lieu of Law under emergency. - Russia requested temporary suspension of the oil price-based rule for 2020–21. - Modified fiscal rule limits: - Nearly 20 countries modified national rule limits during the pandemic (examples: Chile, Ecuador, Malaysia, Mexico, Mongolia, Namibia, Panama, Vietnam). - Ecuador (2020): introduced an expenditure growth rule, new annual budget balance targets, and a debt rule capping nonfinancial public sector debt at 40 percent of GDP by 2032 with interim targets. - Panama (late 2020): raised the deficit target (above its 2 percent of GDP limit) and anchored budget deficit at 1.5 percent of GDP from 2025. Selected country examples (highlights from Table 1) - Brazil: - In 2020 declared a “public calamity” allowing not to meet primary fiscal deficit and other requirements of fiscal responsibility legislation. - Amended constitution to create a “war budget” (an additional easing of 8.4 per cent of GDP for COVID-19 related measures) and exempted government from prohibition to borrow and finance current expenditures. - Colombia: - Social Investment Law sanctioned September 2021 introduced a medium-term debt anchor and revised structural net primary balance ceiling varying with debt level. - Transition path of deficits during 2022–25 established; fiscal council (Autonomous Fiscal Rule Committee) to be tasked with greater operational independence. - India: - Activated escape clause in February 2020 before the pandemic allowing a temporary deviation of deficit not exceeding ½ percentage points of GDP a year. - Raised FY2019/20 and FY2020/21 deficits to 3.8 and 3.5 per cent of GDP, respectively, and suspended the fiscal rule through FY2020/21. - Poland: - Escape clause allowed higher expenditure limits in case of war, national emergency, natural disaster; epidemic added in 2020. - Change provided scope to raise deficit up to 8 per cent of GDP and introduced new national escape clauses in 2020. - United Kingdom: - Government activated escape clause in March 2020. - Office of Budget Responsibility (OBR) tasked to assess fiscal performance against legislated targets. - In October 2021, government reinstated fiscal rules as part of post-pandemic framework requiring: - a balanced current budget by the third year of the rolling forecast period; - a ceiling on government net investment of 3 per cent of GDP on average during the rolling period; - a declining public sector net debt (excluding the Bank of England) by the third year of the rolling five-year forecast period; - a ceiling on welfare spending; - an innovation with a focus on the public sector balance sheet. - Other adjustments and transitions: - Some countries introduced new fiscal rules during the pandemic (examples: Uruguay, Costa Rica). - Several countries announced plans or frameworks to reinstate fiscal rules or transition back (examples: Colombia Social Investment Law; Uganda charter of fiscal responsibility FY2021–26; United Kingdom medium-term fiscal strategy). - For escape clause activations, many were set to expire by end of 2021 or 2022 with specified timelines for return to rules in some countries (examples: Georgia, Panama, Switzerland, Germany). - Supranational rules: - European Union activated the escape clause until end 2022. - WAEMU suspended its convergence pact in April 2020 and committed (June 2021) to converging toward the fiscal deficit anchor over 2024–26. - ECCU extended the timeline for achieving the debt target at or below 60 percent of GDP by five years. *Italic: Source — wpiea2022011-print-pdf, IMF Fiscal Rules Dataset: 1985–2021; Davoodi, Elger, Fotiou, Garcia-Macia, Lagerborg, Lam, and Pillai (2022a, 2022b).* ### 2035. When charting the transition path to reinstate the fiscal rules, the long-standing divergence among ### wpiea2022011-print-pdf - 2035. When charting the transition path to reinstate the fiscal rules, the long-standing divergence among ### Transition path to reinstate fiscal rules - The current framework in the European Union requires an adjustment by one twentieth (1/20) of the difference between the debt level and the debt anchor every year. - Such pace could prove difficult for high debt countries. - Example: in 2020 Italy’s debt stands at about 160 percent of GDP. Under the current 1/20 rule, Italy would need to reduce debt by about 5 percentage points of GDP each year. - Divergence among member states (which became more prominent during the pandemic) must be considered when charting transition paths. - Regional experiences: - ECCU: St. Kitts and Nevis has stayed within the debt target while debt in Dominica exceeded 100 percent of GDP in 2020, making it difficult to achieve the supranational debt target over the medium term. - CEMAC: regional framework requires implementing credible three-year national plans to meet convergence criteria (a deficit limit of 1.5 percent of GDP and public debt limit at 70 percent of GDP), although persistent deviations existed in most member states before the pandemic. ### Experience across supranational rules during COVID-19 - European Union: - Activated the general escape clause for the first time in March 2020 as the European Commission concluded conditions were met in the Stability and Growth Pact (SGP). - The escape clause allowed temporary departure from normal budgetary requirements provided medium-term fiscal sustainability was not endangered. - The escape clause has been extended to end 2022. - ECCU: - No union-wide escape clause; extended the convergence to the 60 percent of GDP debt target date by 5 years to 2035 in February 2021. - Encouraged member countries to enact fiscal frameworks to support post-pandemic efforts. - Member countries, highly reliant on tourism, suffered a large contraction (about 16 per cent of GDP in 2020). - Public debt stood at 84 per cent of GDP in 2020 and would reach about 90 per cent of GDP in 2021. - CEMAC: - Suspended fiscal rules given lack of an escape clause. - In 2020, five countries missed the deficit ceiling (1.5 percent of GDP) and two missed the debt target (70 percent of GDP). - Member countries expected to return to the rule by 2023. - WAEMU: - In April 2020, temporarily relaxed fiscal rules (including a deficit ceiling of 3 per cent of GDP and a debt target of 70 per cent of GDP). - No time horizon initially specified to reinstate rules; in June 2021 heads of state committed to converging toward the fiscal deficit anchor over the years 2024–26. - Regional fiscal deficit close to 6 per cent of GDP in 2020. ### Role and actions of fiscal councils during the pandemic - Over three-quarters of fiscal councils provided swift analysis of economic or budgetary impact of COVID-19, especially among advanced economies. - Fiscal councils’ contributions: - Assessed government measures as appropriate despite raising debt levels (example: UK’s Office of Budget Responsibility). - Assessed financing options and implications for long-term fiscal sustainability (examples: Canada, Czech Republic). - Published reports on crisis impact, medium-term macro-fiscal projections, and cost estimates of policy announcements (example: Brazil). - Provided frequent forecast updates when uncertainty was high (examples: Austrian Fiscal Advisory Council, Canada’s parliamentary budget office). - Provided scenario analysis for multiple pandemic outcomes (example: Netherlands Bureau for Economic Policy Analysis). - On activation of escape clauses or suspensions: - Over half of fiscal councils supported appropriate activation of escape clause or temporary suspension (examples: Brazil, Peru, Spain, United Kingdom, European Fiscal Board). - Concerns raised in some cases about size of fiscal response and nonstandard budgetary procedures (e.g., reliance on emergency proceedings rather than parliamentary approval). - On costing and transparency: - Over a third of fiscal councils estimated costing of emergency measures and released findings. - Many fiscal councils stressed need for greater transparency on design, fiscal impact, take-up rates, viability requirements, overlaps, implementation delays, and bureaucracy of support measures. - Some councils called for assessments of quality of fiscal responses and plans to return to budget balance in the post-COVID environment (examples: Peru, Estonia). - Fiscal councils generally rose to mandates during the pandemic but challenges remain to ensure adequate fiscal oversight as countries exit the crisis. ### Deviations from fiscal rule limits: patterns, pandemic impact, and projections - Definitions and measurement: - Deviations from fiscal rule limits are calculated by comparing fiscal aggregate outturns to limits in fiscal rules. - Focus on budget balance rules (BBRs) and debt rules (DRs). - BBR deviations measured as difference between most relevant budget balance concept in IMF WEO database and the limit set in the rules. - DR deviations defined as gross debt minus the rule limit or anchor levels. - Formal deviation: D_i,t^X = X_i,t − X_i,t^limit, where X is deficit or debt as percent of GDP. - Historical frequency: - On average, countries exceeded deficit limits about 50 percent of the time and debt limits about 42 percent of the time during 2004–21. - Just before the pandemic, debt already exceeded the debt limit or anchor in more than half of countries (44 out of 81). - Deviations from BBRs frequent in emerging markets and developing economies and resource-rich countries. - Large deviations from debt limits more prevalent among advanced economies; distribution skewed to the right for advanced economies. - Pandemic-era deviations: - About 90 percent of countries with BBRs saw deficits exceed rule limits in 2020, with the median positive deviation at about 4 percent of the GDP. - Over half of countries with DRs had debt exceeding the limit or anchor levels in 2020. - Median deviation of debt for advanced economies: 50 percent of GDP. - Median deviation of debt for emerging market and developing economies: 26 percent of GDP. - Deviations from debt rules were at historical records in 2020–21 in terms of levels and share of countries. - Several countries remained below debt limits due to fiscal buffers (examples: Denmark and Estonia). - Post-pandemic prospects: - Returning to fiscal rule limits will likely be a challenge; after the global financial crisis countries returned only gradually to BBR limits. - WEO projections suggest many countries will continue to have deficits exceeding rule limits in the next 2–3 years. - European Union intends to deactivate the escape clause by end 2022, but many member states will need transitional periods given significant deficits and debt exceedances. - Debt projected to remain significantly higher than limit or anchor levels for most advanced economies, with deviations projected to decline gradually over the medium term. - Large deviations imply rapid convergence to rule limits may be infeasible or inappropriate given lingering pandemic and uneven recovery. ### Dynamics of deviations and econometric approach - Empirical strategy: - Panel regression approach analyzes persistence of deviations from BBRs and DRs. - Specification allows heterogeneous persistence coefficients conditional on whether a country was exceeding the deficit or debt rule in the previous period. - Cross-interaction terms between deficit and debt deviations included along with other controls. - Regression model (as specified): - D_i,t^X = β_0 D_i,t−1^X + β_1 D_i,t−1^X * I(D_i,t−1^X < 0) + β_2 D_i,t−1^(−X) * I(D_i,t−1^(−X) ≥ 0) + β_3 D_i,t−1^(−X) * I(D_i,t−1^(−X) < 0) + β_4 g_i,t + δ_i + u_i,t - Where: - D indicates distance from rule limits (BBR or DR), - −X is the opposite rule (DR for BBR regression and vice versa), - I(·) is the indicator function, - g is real GDP growth, - δ_i are country fixed effects, - u_i,t is a potentially autocorrelated error term. - Sample and data notes: - Sample of about 90 countries for 1990–2021, smaller than number of countries with rules in 2021 because not all have simple measurable limits. - Data from IMF Fiscal Rules and World Economic Outlook datasets. - Years older than 2004 are dropped in some analyses because number of countries with fiscal rules was substantially smaller earlier. - Data for 2021 are IMF projections. - Notes on debt measurement: some countries specify limits in net present value (NPV) terms (typically about 10 percent smaller than nominal debt); NPV debt not available in WEO, so deviation approximated using nominal debt. *Italic: IMF Working Papers — Fiscal Rules and Fiscal Councils: Recent Trends and Performance during the COVID-19 Pandemic* ### Annex II contains the regression results, as well as robustness tests and alternative econometric specifications. ### wpiea2022011-print-pdf - Annex II contains the regression results, as well as robustness tests and alternative econometric specifications. ### Persistence of Deviations from Fiscal Rule Limits - For countries with deficits exceeding the rule limit, the estimated autocorrelation coefficient is about 0.55. - Interpretation: A country with the typical deviation (1.9 percent) would on average take 3–4 years to converge close to its means. - For observations below the BBR limit in the previous period, the autocorrelation coefficient of the gap is about 0.81. - Interpretation: For a typical average gap of 1.9 percent of GDP below the mean level will remain so in 10 years in the absence of large shock. - Deviations from the debt rule (DR) are very persistent; the autocorrelation coefficient for DR deviations is about 0.9, without significant differences between countries above or below debt ceilings/anchors. - Debt deviations are slightly more persistent in countries with debt anchors, but the difference is not statistically significant. - Interaction effects: - Regression results show a negative statistically significant coefficient on DR deviations when interacting BBR and DR deviations—countries with a higher level of debt tend to have a smaller deficit deviation. - The coefficient on real GDP growth is slightly negative for deficit deviations and large and statistically significant for DR deviations (higher real GDP growth tends to correct for DR deviations). - Note: The persistence measures are obtained from autoregression coefficients (훽0 and 훽1) with initial levels set as average deviations from respective fiscal rule limits. ### Dynamics after Breaching Budget Balance Rule (BBR) Limits - Method: Panel regression with country fixed effects using dummies equal to one for periods two years before to four years after the initial BBR breach (year 0 is the year of initial breach); charts show differences relative to year t-1 for some variables. - Findings: - BBR deviations typically coincide with a large decline in real GDP growth rate at the initial period, suggesting adverse growth shocks may contribute to pushing deficits above the rule limit. - Economic growth tends to stay lower in the aftermath of a BBR deviation in subsequent years. - When the deficit crosses the fiscal rule limits, countries typically face persistent debt buildup and higher financing costs; debt continues to drift higher even after the initial deviation. - Credit default swaps typically rise gradually after a deviation for about 3–4 years to about 1 percentage points higher. - Sample and estimation notes: - The bottom quartile of countries by 2019 GDP in USD is excluded from the sample for noisy data considerations. - Standard errors are clustered at the country level. - A repeated breach within 5 years of an initial breach is not treated as a separate episode. ### Effect of Past Rule Compliance on Response to Large Shocks - Hypothesis: A good track record of rule compliance improves credibility and allows governments to respond more in the face of a large shock. - Method: Regress fiscal response (change in the primary balance) in 2009 and 2020 against average prior deviations from fiscal rule limits (observations below the deficit limit censored at zero) and controls (lag primary balance, GDP growth, lagged GDP per capita PPP, lagged debt-to-GDP). GDP growth is instrumented: lagged construction sector size for 2009; Oxford stringency index for 2020. - Results: - A better record of observing fiscal rules has been associated with a larger fiscal response to large shocks (larger declines in the primary balance). - The difference is statistically significant during the pandemic (2020); it is not statistically significant during the Global Financial Crisis (2009) due to smaller sample size widening the confidence interval. - OLS and IV regressions yield similar results (full regression results in Annex II). ### Key Quantitative Summaries and Historical Patterns - About 105 countries had fiscal rules at the end of 2021, up from less than 10 in the early 1990s. - About 50 countries had fiscal councils as of end-2021. - Pre-pandemic, two thirds of countries with fiscal rules had escape clauses. - During 2004–21, on average countries exceeded the deficit limits about 50 percent of the time and exceeded debt limits about 42 percent of the time. - In 2020: - Almost all countries with deficit rules exceeded the limits—by an average of 4 percent of GDP. - Debt deviations reached unprecedented levels: median deviation reached 50 percent of GDP in advanced economies and 26 percent of GDP in emerging market and developing economies. ### Policy Implications and Recommendations - Returning to fiscal rule limits is likely to take time, especially where there have been large debt deviations; historical evidence shows debt is more persistent than deficits. - When transitioning to reinstate fiscal rules, long-standing divergence among member states (in supranational contexts) needs to be considered. - Strengthening fiscal frameworks should focus on: - Strong political commitment and better record of compliance. - Incentives to build buffers during good times. - Designing effective mechanisms to manage large shocks in bad times (e.g., clear escape clauses). - Strengthening fiscal councils’ independence and capacity to improve credibility and accountability. - Maintaining deficits below limits appears to strengthen fiscal credibility and allows larger fiscal responses when shocks hit. ### Annex I — Definitions and Types of Fiscal Rules and Councils (dataset inclusion criteria) - Fiscal rule definition used: imposes a long-lasting constraint on fiscal policy through numerical limits on budgetary aggregates. - Inclusion principles: - Only rules with targets fixed in legislation and revisable only on a low-frequency basis and binding for at least three years are considered fiscal rules. - Dataset considers rules that set numerical targets on aggregates covering a large share of public finances and at a minimum central government level; subnational rules and fiscal sub-aggregates are excluded. - Focus is on de jure arrangements, not adherence in practice. - Types of fiscal rules distinguished: 1. Debt rules: explicit anchor or ceiling for public debt, often in percent of GDP. 2. Budget balance rules: limits on overall balance, primary balance, or structural/cyclically adjusted balance. 3. Expenditure rules: limits on total, primary, or current government expenditures (absolute terms, growth rates, or percent of GDP, often with 3–5 year horizon). - Note: Pay-as-you-go rules are treated as procedural and not counted as numerical fiscal rules in the database. - Temporal coverage: Indicators cover rules that took effect by end-December 2021 or with transition regimes in place at that time; adopted but not yet implemented rules are described but not included in charts and tables. *Source: wpiea2022011-print-pdf - Annex II contains the regression results, as well as robustness tests and alternative econometric specifications.* ### 4. Revenue rules set ceilings or floors on revenues and are aimed at boosting revenue collection and/or ### 4. Revenue rules set ceilings or floors on revenues and are aimed at boosting revenue collection and/or ### Revenue rules: purpose and main challenge - Revenue rules set ceilings or floors on revenues and are aimed at boosting revenue collection and/or preventing an excessive tax burden. - Most of these rules are not directly linked to the public debt or spending. - Revenues are highly cyclical, which makes setting ceilings or floors on revenues challenging. - Revenue rules that restrict certain use of “windfall” revenue are an exception. - Revenue rules alone could result in procyclical fiscal policy, as floors (ceilings) do not generally account for the operation of automatic stabilizers in a downturn (upturn). - Revenue rules, similar to expenditure rules, can be used to target the size of the government. ### Coverage of fiscal rules - In principle, fiscal rules should cover most fiscal aggregates. - Majority of supranational rules have covered fiscal aggregate at the general government levels; it is the case for less than half of the national rules (Figure A1.1). - The higher status of legislation basis in supranational rules makes it more likely to extend to general government. - For national expenditure and deficit rules, the coverage often focuses on central government, reflecting: - less availability of timely data, and - the fact that local governments are often subject to subnational rules or restrictions that they cannot borrow or maintain a deficit (Eyraud and others 2020). ### Exclusion of budget components from the rules - Countries with structural budget balance rules tend to exclude the cyclical components in the revenues and expenditures to assess whether the fiscal rules are in compliance (Chile, Colombia, European Union). - Countries with expenditure rules often exclude certain budget components. Common exclusions include: - capital spending from the expenditure rule or set a floor of capital spending (Costa Rica, Kosovo 2006–08, Peru 2012, Thailand), - interest payments, - pension (France), - nondiscretionary unemployment benefits (Spain). - Mexico expenditure rule is defined as current primary expenditure but excludes the outlays governed by automatic rules (pensions and subsidies for electricity). - The coverage of expenditure rule in the Netherlands was revised during the global financial crisis, excluding interest payments 2007–10 and unemployment and social assistance benefits (2009–10). - Some countries linked the expenditure rules with the debt level (Armenia, Israel) or the level of financial assets (oil funds in Russia). - In many cases, multi-year expenditure ceilings were established as part of the medium-term fiscal framework. This strengthens the predictability of fiscal policies and supports the implementation of fiscal rules. - As of 2021, over a third of countries with fiscal rules had established multi-year ceilings. ### Correction mechanisms - Correction mechanisms stipulate what governments would need to do in case of breaching the fiscal rules (or at risks of being breached). - Most European countries have introduced correction mechanisms to specify actions and path back towards the budget balance rule following a deviation as part of the 2012 Fiscal Compact. The design varies across countries although it follows principles of the European Commission. - The mechanism is triggered ex-post after the fiscal rules is deemed to be breached. - Poland has established several triggers preemptively as debt approaches its fiscal rule limits with increasing degree of fiscal adjustments. - Outside Europe, correction mechanisms are less common but included in some fiscal responsibility laws such as in Jamaica, Georgia, and Grenada. - Some correction mechanisms specify the precise path of adjustments after the noncompliance of the fiscal rules (Slovak Republic) while others require the government to submit corrective plans for fiscal council assessment and/or parliamentary approval (Germany, Ireland). ### Dataset definitions and coverage notes - To be included in the dataset, a fiscal council (independent fiscal institution) must fulfill the following conditions: - (i) align with the IMF definition of fiscal council specified above, - (ii) be consistent with the main OECD Principles for Independent Fiscal Institutions (von Trapp, Lienert, and Wehner, 2016), - (iii) be functional and visible institutions, such as maintaining a regularly updated website or other forms of public communication. - While the dataset is cross-sectional by nature, the dates of establishment and of major reforms are included. - Sources: IMF Fiscal Rules Dataset: 1985–2021 and IMF Fiscal Council Databases: 2021. - Data as of end-December 2021. - The income group classification is based on IMF World Economic Outlook database. - Notes in table: - 1/ Countries also have supranational rules. United Kingdom was a member state in European Union until January 2020. - 2/ Countries with fiscal councils as of end-2021. - 3/ Korea and South Africa have fiscal councils but do not adopt formally numerical fiscal rules. They have indicative targets in their medium-term budget frameworks. ### Measuring deviations from rules (methodology) - For most countries, the budget balance target has a direct counterpart variable in the WEO database that can be used to calculate deviations from the rule limit. - The most common targets are the overall balance and the structural primary balance. - For countries with more specific targets, deviations need to be approximated with the closest combination of WEO variables. - Table A.2.1 mapping of targets to WEO variable codes (asterisks indicate an imperfect match): - Overall Balance: GGXOFB - Overall Balance excluding Foreign Financing and Grants: GGXOFB-GGRG * - Overall Balance excluding Oil Revenues: GGXOFB-TXGO * - Overall Balance excluding Investment: GGXOFB-NFIG* - Primary Balance: GGXONLB - Primary Balance excluding Oil Revenues: GGXONLB-TXGO * - Primary Balance excluding Investment: GGXOFB-NFIG * - Cyclically-Adjusted Primary Balance: GGCBP - Structural Primary Balance: GGSBP - Structural Overall Balance: GGSBP-GGEI * - Figure A.1.1 shows aggregate positive deviations from budget balance rules or debt rules across country groups and over time. - Note: BBR deviations are not calculated for Argentina and Panama (the latter during 2015–2019) as no close approximation is available. ### Dynamics of deviations—regression and robustness results - Regression framework: Arellano-Bond estimator (Generalized Methods of Moments) including lagged dependent variables. - Robustness tests include: - adding year fixed effects, - including small countries (bottom quartile in GDP levels in USD in 2019), - instrumenting real GDP growth with its own lag, - adding an interaction term between fiscal rule deviations and recessions (output gap < -2 percent). - BBR regressions: - Results similar across robustness tests. - Including small countries or instrumenting GDP growth makes the difference in persistence between countries exceeding the BBR limit and those that do not insignificant. - BBR deviations in countries exceeding the limit are significantly more persistent amid a big recession. - DR regressions: - Results mostly unchanged under robustness checks. - Persistence of deviations is not significantly higher in large recessions. - Countries that set debt limits as anchors rather than ceilings feature slightly more persistence (not shown), but the difference is not statistically significant. - Selected reported coefficients and statistics (exact reported values from Table A2.2): - lag BBR_deviation (baseline): 0.807*** (0.055) - lag BBR_deviation (including small countries): 0.765*** (0.054) - lag BBR_deviation (instrumenting GDP growth): 0.736*** (0.099) - lag BBR_deviation (Big Recessions): 0.707*** (0.113) - lag BBR_deviation (with year FE): 0.822*** (0.059) - lag BBR_deviation (exceeding) baseline: -0.254*** (0.074) - lag DR_deviation baseline: 0.910*** (0.029) - lag DR_deviation (including small countries): 0.832*** (0.043) - RGDP growth coefficients reported (BBR baseline): -0.404*** (0.030); (year FE) -0.178*** (0.057) - N reported across specifications: 732 - Estimator: Arellano-Bond ### Implications of deviations in major crises—regression results - Regression assesses fiscal response to crises using cross-section of countries in two episodes: the global financial crisis (2009) and the pandemic (2020). - Regression equation variables and definitions: - ∆PB: change in primary balance in percent of GDP, - PB_{i,t-1}: lag primary balance, - D̄_{BBR,s