## 1. Automatic Stabilizers, Cyclical Policy and Discretionary Policy

## Source details

**Canonical URL:** [1. Automatic Stabilizers, Cyclical Policy and Discretionary Policy](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10111.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10111.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10111.pdf.json)

---

### Introduction and motivation
- Revival of debate on fiscal policy as macroeconomic stabilization tool; micro-founded general equilibrium models with non-Ricardian features corroborate benefits of timely countercyclical fiscal response (Botman and others, 2006; Kumhof and Laxton, 2009).
- Three channels for fiscal stabilization highlighted:
  - Automatic stabilizers.
  - Deliberate discretionary changes.
  - Tax/transfer structure enhancing resilience.
- Scope and sample:
  - 49 industrial and developing countries.
  - Annual data 1970–2006, aggregated into 4 ten-year periods (1970-79, 1980-89, 1990-99, 2000-06).
- Main empirical summary findings:
  - Automatic stabilizers strongly contribute to output stability across advanced and developing economies; effect on private consumption volatility weaker and less robust.
  - Countries with more volatile cyclically-adjusted budget balances exhibit more volatile output and private consumption (reverse causality concerns noted).
  - Access of individual consumers to credit exerts a stabilizing influence on output and private consumption.

### Data and stylized facts: measurement and patterns
- Fiscal dimensions and measures:
  - Automatic stabilizers approximated by the ratio of general government expenditure to GDP (public expenditure-to-GDP ratio).
  - Cyclical fiscal policy measured by coefficient β from regression of cyclically-adjusted balance (CAB) on output gap (equation (3)): CAByCABμγβ α tttt 1 − + + + = .
  - Exogenous discretionary fiscal policy measured by ζ = standard deviation of residuals from IV-estimated CAB regression.
- Key measurement rules and instruments:
  - Under Rη = 1 and Gη = 0, government size g equals the semi-elasticity of the budget balance (in percent of GDP) to the output gap (equation (2)).
  - Output gap computed from HP trend.
  - IV instruments for output gap: lagged output gap, log-differenced terms of trade and oil prices, energy use per capita; for oil exporters: lagged output gap, US output gap, and its lagged value.
  - If β̂ not statistically different from zero at 10 percent, cyclical response is set to zero.
- Stylized empirical observations:
  - OECD countries: larger automatic stabilizers and more countercyclical fiscal responses (positive β) on average.
  - Many non-OECD countries: negative β in many cases.
  - Conditional correlations (Table 1) — selected signs and associations preserved:
    - GDP per capita: positive correlation with automatic stabilizers and countercyclicality.
    - Openness: generally associated with larger automatic stabilizers; also linked to greater discretionary fiscal policy in OECD-20.
    - Debt: generally constraining on fiscal policy (negative signs).
    - Political variables (presidential system, proportional electoral rule, political constraints, government fragmentation/stability) exhibit expected correlations with fiscal policy indicators.

### Cyclical and discretionary fiscal patterns: selected country and period entries
- Cyclical dimension (selected β estimates, Table A1):
  - Argentina: -0.0188
  - Australia: 0.7556*
  - Brazil: -0.7710*
  - Canada: 0.2818*
  - Germany: -0.2414*
  - India: -0.2407*
  - Ireland: -0.5471*
  - Italy: 0.3390*
  - Mexico: -0.5670*
  - United States: 0.5361*
  - Venezuela, Rep. Bol.: -0.5570*
- Discretionary dimension (selected ζ entries, Table A2) — representative country-period values:
  - United States: 1970-79 1.247; 1980-89 0.748; 1990-99 1.178; 2000-06 0.889
  - United Kingdom: 1970-79 0.847; 1980-89 0.935; 1990-99 1.827; 2000-06 1.053
  - Norway: 1970-79 2.884; 1980-89 3.455; 1990-99 3.015; 2000-06 5.072
  - Mexico: 1970-79 1.000; 1980-89 7.080; 1990-99 2.420; 2000-06 1.690
  - Venezuela: 1990-99 5.429; 2000-06 4.604

### Determinants of fiscal dimensions (selected coefficient estimates)
- Determinants of cyclical fiscal policy (Table A3) — selected coefficients:
  - GDP per capita: .104***; .078***; .052**; .061**; .055*; .053*
  - Openness: -.373*; additional openness estimates -.516**, -.611***, -.505**, -.435*, -.457* across specifications
  - Presidential System (dummy): .124**; .104; .110*; .112*
  - Proportional Electoral rule (dummy): -.238***; -.208***; -.184**; -.193**
  - Political Constraints V-Henisz: .334**; .288*; .349*; .338*
  - Observations across columns: 42, 39, 42, 39, 39, 39
- Determinants of automatic stabilizers (Table A4) — selected coefficients:
  - Urbanization Rate: .003*; .005***; .003; .002
  - Openness: .375***; .184; .245*; .204
  - GDP per capita: .128***; .070**; .136***; .130***
  - Proportional Electoral rule (dummy): .203*** (in one specification)
  - Presidential System (dummy): -.222*** (in one specification)
  - Government Fragmentation: .260*** (in one specification)
  - Observations: 133; 152; 127; 151
  - R-squared: .482; .526; .471; .500
- Determinants of discretionary fiscal policy (Table A5) — selected coefficients:
  - Automatic stabilizers: 1.231***; 1.233***; 1.396**; 1.405**; 1.453*; 1.324*
  - Openness: 1.593**; 1.595**; .715; .421; 1.400**; 1.338*
  - Debt: -.781**; -.780**; -2.835**; -2.998***; -.164; -.141
  - Political Constraints V-Henisz: -3.425***; -3.421***; -4.507***; -4.796*** (in specific splits)
  - Observations: 118; 118; 46; 46; 72; 72
  - R-squared: .334; .334; .545; .549; .187; .195

### Econometric analysis: output and consumption volatility — framework
- Baseline panel regression (equation (4)):
  - titij J j jti XGPY , , 1 , 4 2 ,  + + + + =
  - Yi,t = real GDP volatility (standard deviation of real GDP growth per period); G = log public expenditure percent of GDP.
- Extended specification (equation (5)) adds cyclical and discretionary measures:
  - titij J j jti iti t t ttti XDiscrCycGPY , , 1 , 3 2 , 4 2 ,  + + + + + + =

### Parsimonious and pooled results (selected exact coefficients preserved)
- Parsimonious (OECD-20, Table 2) examples:
  - Automatic stabilizers coefficients: -1.248**, -2.224***, -.244 (period splits in table).
- Parsimonious (All 49 countries, Table 3) — selected entries:
  - Automatic stabilizers -1.614*** (All, std dev of real GDP growth, Table 3, column (1)).
  - Automatic stabilizers -1.383*** (All, std dev of first differenced output gap).
  - Non-OECD adjusted gap: Automatic stabilizers .983* (Table 3, column (4)).
- Adding monetary policy and financial development (Table 4) — selected entries:
  - Financial development (credit to private sector as percent of GDP) stabilizing: -.428* (OECD-20), -1.217*** (All).
  - Automatic stabilizers remain negative and significant: -1.401**, -1.523*** across specifications.
  - CBI impact: ambiguous and often statistically insignificant.
- Including cyclical and discretionary fiscal measures (Table 5) — selected entries:
  - Automatic stabilizers: -1.409***; -2.013***; -2.257*** (assorted specifications).
  - Discretionary fiscal policy (ζ): .911***; .672***; .877*** for All and Non-OECD samples in various columns.
  - Cyclical fiscal policy coefficient: generally statistically insignificant.
  - Interaction term (discretion × CBI): 2.118*** (Table 5, column (6)).
- 2SLS (Table 6) — selected instrumented results:
  - Instruments: electoral rule, political system, number of veto points, ideological distribution, GDP per capita (PPP, log), dependency ratio, urbanization rate, oil-producer dummy.
  - Wu-Hausman tests: exogeneity rejected mainly for discretionary fiscal policy; CBI borderline (p-value 0.11).
  - 2SLS example (Table 6, column (1), instrumenting Automatic stabilizers):
    - Automatic stabilizers -2.271***.
    - Financial Development -.817***.
    - Discretionary Fiscal Policy .671***.
  - Weak-instrument issues reported in some IV estimations (first-stage F-statistics concerns).
- Robustness, omitted variables and fixed-effects (Table 7):
  - Adding controls (country size, mean real GDP growth, GDP per capita, terms-of-trade volatility, oil dummy, government stability) does not materially change core pooled coefficients for automatic stabilizers and discretionary policy.
  - Country fixed-effects: financial development’s stabilizing coefficient loses significance — suggests financial development may proxy for deep country-specific features.
  - Automatic stabilizers and discretionary policy retain respective stabilizing and destabilizing signs even with extensive controls.
- Consumption volatility (Table 8) — selected entries:
  - Financial development strongly stabilizing consumption: -1.429***; -1.196***; -1.394*** in OLS variants.
  - Automatic stabilizers stabilize consumption but effect smaller and less precise: -1.14* (OLS); instrumented result -2.046*** when instrumenting automatic stabilizers.
  - Cyclical fiscal policy: tendency toward negative impact on consumption volatility in some specifications (e.g., -.875*), but estimates noisy.
  - Discretionary fiscal policy generally shows destabilizing signs for consumption (e.g., .525*** in one specification), significance dependent on instrumenting.

### Endogeneity, instruments, and identification issues
- Endogeneity concerns:
  - Discretionary fiscal policy often endogenous (Wu-Hausman exogeneity rejected).
  - CBI borderline endogenous (p-value 0.11 in Wu-Hausman test).
- IV/2SLS instruments used (Table 6): electoral rule, political system, number of veto points, ideological distribution, GDP per capita (PPP, log), dependency ratio, urbanization rate, oil-producer dummy.
- IV results generally confirm automatic stabilizers and financial development are stabilizing, but weak instruments limit inference for some variables (e.g., CBI, cyclical fiscal policy).

### Analytical mapping to fiscal multipliers (Appendix D): calibration and implied ranges
- Model setup: log-linear, backward-looking IS with parameters α, ρ and notation in Appendix.
- Calibration assumptions:
  - Fiscal policy can stabilize about one third of shocks to ϑ.
  - Set d̄(ϑ) = 1.5 × sample measure of output variability.
  - Assume ρ = 0.6.
  - Assume θ1 spans [0.1; 1.5].
  - Assume government size between 0.2 and 0.6.
- Implied ranges and mappings:
  - Implied values for estimated parameter γ̂1 lie between -2.64 and -0.48.
  - Using sample average government size 0.38 and assuming discretionary fiscal policy acyclical (β̃ = 0), the 95 percent confidence interval of γ̂1 ([-2.81;-1.22]) maps into “fiscal multipliers” [(θ1)(θ1+θ̃+α)ρ]1/5 between 0.4 and 1.5.
  - Using standard deviation of the output gap as volatility measure (95 percent CI for γ̂1 [-2.29;-0.92]) yields implied multipliers between 0.4 and 1.0.
- Policy-relevant analytical interpretations:
  - Larger automatic stabilizers reduce output-gap volatility, with diminishing marginal stabilizing effects.
  - Automatic stabilizers positively associated with discretionary fiscal activism (Automatic stabilizers coefficients 1.231*** etc. in Table A5).
  - Political-institution variables significantly influence cyclical and discretionary fiscal behavior.
  - Openness exhibits mixed effects: negative association with cyclical responsiveness (Table A3) but positive association with automatic stabilizer size (Table A4).

### Key quantitative points and representative coefficients (exact values preserved)
- Sample and periods: 49 countries; annual data 1970-2006; periods 1970-79, 1980-89, 1990-99, 2000-06.
- Representative estimated coefficients (selected exact entries from tables):
  - Parsimonious All sample: Automatic stabilizers -1.614*** (Table 3, column (1)).
  - Parsimonious Non-OECD (adjusted gap): Automatic stabilizers .983* (Table 3, column (4)).
  - With monetary and financial controls (Table 4, column (3)): Financial development -1.217***; Automatic stabilizers -1.523***.
  - Full fiscal controls (Table 5, column (3)): Automatic stabilizers -2.013***; Discretionary Fiscal Policy .672***; Financial Development -.788***.
  - Interaction (Table 5, column (6)): discretion × CBI 2.118***.
  - 2SLS (Table 6, column (1) instrumenting Automatic stabilizers): Automatic stabilizers -2.271***; Financial Development -.817***; Discretionary Fiscal Policy .671***.
  - Consumption (Table 8, column (1)): Financial Development -1.429***; Automatic stabilizers -1.14*; Discretionary Fiscal Policy .525***.
- Statistical notation: significance stars retained (* 10 percent, ** 5 percent, *** 1 percent).

### Conclusions and policy implications (preserved from source)
- Main conclusions:
  - Fiscal stabilization works mainly through automatic stabilizers; discretionary cyclical responses show little robust stabilizing impact on output.
  - Greater variability in cyclically-adjusted balances (discretionary shocks) tends to increase output and consumption volatility, particularly outside core OECD-20 economies.
  - Financial development is an important stabilizer in pooled regressions, though country fixed-effects attenuate this result.
  - Central bank independence can be beneficial only conditional on limited fiscal discretion; discretionary shocks interacting with CBI are associated with higher volatility, signaling potential monetary-fiscal coordination failures.
- Policy recommendations emphasized:
  - Favor timely, predictable, and symmetric fiscal arrangements that operate like automatic stabilizers.
  - Subject other fiscal objectives (redistribution, efficiency) to a “stability test” to reduce conflicts among public finance objectives and between monetary and fiscal policies.
  - Consider budget preparation rules with quantitative objectives or binding constraints (e.g., structural balance or expenditure ceilings).
  - Explore mechanisms to enhance stabilizers without increasing government size (e.g., automatic adjustments in selected tax rates or expenditure programs).
- Research agenda highlighted:
  - Better measures of automatic stabilizers accounting for expenditure and revenue composition.
  - Deeper investigation of monetary-fiscal conflicts and alternative measures of monetary policy quality.
  - Improved identification of fiscal discretion and disentangling it from shocks with direct budgetary effects (commodity, exchange rate, inflation, asset price shocks).

*Source: _wp10111 - 1. Automatic Stabilizers, Cyclical Policy and Discretionary Policy.*

### 1. Automatic Stabilizers, Cyclical Policy and Discretionary Policy  ......................................10

### 1. Automatic Stabilizers, Cyclical Policy and Discretionary Policy  ......................................10

### Major sections (contents)
- 1. Automatic Stabilizers, Cyclical Policy and Discretionary Policy  ......................................10
- 2. Government Size and Cyclical Sensitivity of the Budget Balance (OECD) .......................14
- 3. Automatic Stabilizers and Output Volatility (1970-2006) ...................................................15
- 4. Output Volatility over Time.................................................................................................17

### Tables listed in the content unit
- Table 1. Fiscal Policy Indicators: Selected Conditional Correlations ...............................................11
- Table 2. Parsimonious Model and Restricted Sample (OECD-20) ...................................................19
- Table 3. Parsimonious Model—Full Sample and Non-OECD-20 .....................................................20
- Table 4. Introducing Monetary Policy and Access to Credit .............................................................21
- Table 5. Introducing Cyclical and Discretionary Dimensions of Fiscal Policy .................................23
- Table 6. Two-Stage-Least-Squares (2SLS) Estimates .......................................................................26
- Table 7. Adding control variables ......................................................................................................27
- Table 8. Fiscal Policy and Consumption Volatility  ..........................................................................29

*Source: _wp10111 - 1. Automatic Stabilizers, Cyclical Policy and Discretionary Policy  ......................................10*

### References .............................................................................................................

### _wp10111 - References

### Introduction
- Context:
  - Revival of debate on fiscal policy as macroeconomic stabilization tool (van der Ploeg, 2005); micro-founded general equilibrium models with non-Ricardian features corroborate benefits of timely countercyclical fiscal response (Botman and others, 2006; Kumhof and Laxton, 2009).
  - Three channels for fiscal stabilization: automatic stabilizers, deliberate discretionary changes, and tax/transfer structure enhancing resilience.
- Motivation and scope:
  - Revisit contribution of fiscal policy to macroeconomic stability over the last 40 years across industrial and developing economies.
  - Sample: 49 industrial and developing countries with fiscal data covering general government.
  - Time span: annual data between 1970 and 2006, aggregated into 4 ten-year periods (1970-79, 1980-89, 1990-99, 2000-06).
- Main summary findings (as reported):
  - Automatic stabilizers strongly contribute to output stability across advanced and developing economies; effect on private consumption volatility weaker and less robust.
  - Countries with more volatile cyclically-adjusted budget balances also exhibit more volatile output and private consumption (reverse causality concerns noted).
  - Access of individual consumers to credit exerts a stabilizing influence on output and private consumption.

### Data and Stylized Facts
- Fiscal policy dimensions constructed and measured:
  - Automatic stabilizers approximated by the ratio of general government expenditure to GDP (public expenditure-to-GDP ratio).
  - Cyclical fiscal policy: systematic response of the cyclically-adjusted balance (CAB) to the output gap; coefficient β from regression (equation (3)).
  - Exogenous discretionary fiscal policy: standard deviation of residuals from IV-estimated regression of CAB on lagged CAB, output gap and AR(1) term; denoted ζ (calculated as variability of residual 1^).
- Key measurement relationships (preserved equations and rules of thumb):
  - Under Rη = 1 and Gη = 0, government size g equals the semi-elasticity of the budget balance (in percent of GDP) to the output gap (equation (2)).
  - CAB regression used: CAByCABμγβ α tttt 1 − + + + = (equation (3)), with output gap computed from HP trend.
- Sample and estimation notes:
  - IV instruments for output gap: lagged output gap, log-differenced terms of trade and oil prices, energy use per capita; for oil exporters: lagged output gap, US output gap, and its lagged value.
  - For countries where β̂ not statistically different from zero at 10 percent, cyclical response set to zero to reduce noise.
- Stylized empirical observations:
  - On average, fiscal policy is more countercyclical in OECD countries than in non-OECD; OECD countries have larger automatic stabilizers and tend to exhibit stabilizing cyclical responses (positive β), while many non-OECD countries show negative β.
  - Conditional correlations (Table 1) highlight:
    - GDP per capita: positive correlation with automatic stabilizers and countercyclicality (statistically significant).
    - Openness: mixed correlations—generally associated with larger automatic stabilizers; also linked to greater discretionary fiscal policy in OECD-20.
    - Debt: generally constraining on fiscal policy (negative signs reported).
    - Political variables (presidential system, proportional electoral rule, political constraints, government fragmentation/stability) exhibit expected correlations with fiscal policy indicators (see Table 1 signs and significance).
  - Caveats on proxying automatic stabilizers:
    - Comparison with OECD semi-elasticity estimates (Girouard and André, 2005) shows government size proxy aligns closely for OECD (regression statistically indistinguishable from 45-degree line).
    - Outside OECD, the proxy may overestimate stabilizers due to lower progressivity of taxes and weaker social safety nets.

### Econometric Analysis: Output and Consumption Volatility
- Empirical framework:
  - Baseline panel regression with period fixed effects: titij J j jti t t tti XGPY , , 1 , 4 2 ,  + + + + = (equation (4)), where Yi,t is real GDP volatility (standard deviation of real GDP growth per period) and G is log public expenditure percent of GDP.
  - Extended specification adding cyclical and discretionary fiscal policy: titij J j jti iti t t ttti XDiscrCycGPY , , 1 , 3 2 , 4 2 ,  + + + + + + = (equation (5)).
- Parsimonious results:
  - For OECD-20 (replicating Fatás and Mihov sample), automatic stabilizers negatively and significantly associated with output volatility (Table 2): automatic stabilizers coefficient examples: -1.248**, -2.224***, -.244 (period splits).
  - Full sample (49 countries) parsimonious regressions (Table 3):
    - Automatic stabilizers coefficient: -1.614*** (All, std dev of real GDP growth) and -1.383*** (All, std dev of first differenced output gap).
    - For non-OECD, government size coefficient sometimes positive (destabilizing) and statistically significant in adjusted measures (e.g., 1.038).
- Adding controls for monetary policy and financial development (Table 4):
  - Financial development (credit to private sector as percent of GDP) is a significant stabilizer: examples include coefficients -.428* (OECD-20), -1.217*** (All).
  - Central bank independence (CBI) impact: ambiguous and statistically insignificant in many specifications; interaction effects investigated later.
  - Automatic stabilizers remain negative and significant: examples -1.401**, -1.523*** across specifications.
- Including cyclical and discretionary fiscal measures (Table 5):
  - Controlling for discretionary volatility yields stronger and more precisely estimated negative effects of government size on output volatility; coefficients of automatic stabilizers around -1.409***, -2.013***, -2.257*** in assorted specifications.
  - Discretionary fiscal policy coefficient (ζ) positive and significant for All and Non-OECD samples (e.g., .911***, .672***, .877***), indicating discretionary fiscal volatility tends to increase output volatility outside core OECD-20.
  - Cyclical fiscal policy coefficient generally statistically insignificant (no robust stabilizing effect detected).
  - Interaction: discretion × CBI is positive and significant in specifications where included (e.g., 2.118***), suggesting discretionary fiscal shocks may undermine benefits of CBI via coordination failures.
- Endogeneity and IV analysis (Table 6):
  - Instruments used: electoral rule, political system, number of veto points, ideological distribution, GDP per capita (PPP, log), dependency ratio, urbanization rate, oil-producer dummy.
  - Wu-Hausman tests: exogeneity rejected mainly for discretionary fiscal policy; CBI borderline (p-value 0.11).
  - 2SLS results: automatic stabilizers and financial development remain stabilizing (examples: Automatic stabilizers -2.271***; Financial development -.817***).
  - Weak-instrument issues in some IV estimations (reported first-stage F-statistics), particularly when instrumenting CBI or cyclical fiscal policy.
- Omitted-variable and fixed-effects checks (Table 7):
  - Adding many control variables (country size, mean real GDP growth, GDP per capita, terms-of-trade volatility, oil dummy, government stability) does not materially change core coefficients for automatic stabilizers and discretionary policy in pooled specifications.
  - Country fixed-effects reduce statistical significance of financial development as stabilizer—suggesting financial development may proxy for deep, country-specific features—but automatic stabilizers and discretionary policy remain respectively stabilizing and destabilizing.
- Consumption volatility (Table 8):
  - Private consumption volatility modeled with same fiscal indicators:
    - Financial development strongly stabilizing for consumption (e.g., -1.429***, -1.196***, -1.394*** across OLS variants).
    - Automatic stabilizers stabilize consumption but effect smaller than for output and less precise (e.g., -1.14*; instrumented results show -2.046*** when instrumenting automatic stabilizers).
    - Cyclical fiscal policy tends to have expected negative impact on consumption volatility (e.g., -.875* in one specification), though estimates noisy.
    - Discretionary fiscal policy has generally destabilizing signs for consumption as well (mixed significance depending on instrumenting).

### Key Quantitative Points and Technical Details
- Sample: 49 countries; annual data 1970-2006; periods aggregated into 4 ten-year periods.
- Government size (sample average) used in Appendix: 0.38 (sample average of government size; used to derive plausible φ1 range).
- Plausible values for φ1 (impact of government size on volatility) lie between -0.5 and -2.6 (Appendix example).
- Representative estimated coefficients (selected from reported tables, preserved exactly):
  - Parsimonious All sample: Automatic stabilizers -1.614*** (Table 3, column (1)).
  - Parsimonious Non-OECD (adjusted gap): Automatic stabilizers .983* (Table 3, column (4)).
  - With monetary and financial controls (Table 4, column (3)): Financial development -1.217***; Automatic stabilizers -1.523***.
  - Full fiscal controls (Table 5, column (3)): Automatic stabilizers -2.013***; Discretionary Fiscal Policy .672***; Financial Development -.788***.
  - Interaction term (Table 5, column (6)): Interaction: discretion x CBI 2.118***.
  - 2SLS (Table 6, column (1) instrumenting Automatic stabilizers): Automatic stabilizers -2.271***; Financial Development -.817***; Discretionary Fiscal Policy .671***.
  - Consumption (Table 8, column (1)): Financial Development -1.429***; Automatic stabilizers -1.14*; Discretionary Fiscal Policy .525***.
- Statistical conventions preserved: stars denote significance (* 10 percent, ** 5 percent, *** 1 percent); robust t-statistics reported in parentheses in original tables.

### Conclusions and Policy Implications
- Main conclusions:
  - Fiscal stabilization operates mainly through automatic stabilizers; discretionary cyclical responses show little robust stabilizing impact on output.
  - Fiscal variability not systematically related to the business cycle (discretionary shocks) tends to increase output and consumption volatility, especially outside core OECD economies.
  - Financial development is an important stabilizing mechanism for consumption and output in pooled regressions, but country fixed-effects mitigate this result, suggesting deep country-specific determinants.
  - Central bank independence can be beneficial conditional on limited fiscal discretion; discretionary shocks interacting with CBI are associated with higher volatility, signaling potential monetary-fiscal coordination failures.
- Policy recommendations emphasized in source:
  - Favor timely, predictable, and symmetric fiscal arrangements that operate like automatic stabilizers.
  - Subject other fiscal objectives (redistribution, efficiency) to a “stability test” to reduce conflicts among public finance objectives and between monetary and fiscal policies.
  - Consider budget preparation rules with quantitative objectives or binding constraints (e.g., structural balance or expenditure ceilings).
  - Explore mechanisms to enhance stabilizers without increasing government size (e.g., automatic adjustments in selected tax rates or expenditure programs).
- Research agenda highlighted:
  - Better measures of automatic stabilizers that account for expenditure and revenue composition.
  - Deeper investigation of monetary-fiscal conflicts and alternative measures of monetary policy quality.
  - Improved identification of fiscal discretion and disentangling it from shocks with direct budgetary effects (commodity, exchange rate, inflation, asset price shocks).

*Italic: Summary based only on content provided in source _wp10111 - References.*

### REFERENCES

### _wp10111 - REFERENCES

### Listed references
- Acemoglu, Daron, Simon Johnson, James Robinson, and Yunyong Thaicharoen (2002), Institutional Causes, Macroeconomic Symptoms: Volatility, Crisis and Growth,” MIT (Cambridge, MA: Massachusetts Institute of Technology).  
- Afonso, Antonio, Luca Agnello, and Davide Furceri (2009), “Fiscal Policy Responsiveness, Persistence, and Discretion,” Public Choice, forthcoming.  
- Alesina, Alberto (2000), “The Political Economy of the Budget Surplus in the US,” Journal of Economic Perspectives, 14: 3-19.  
- Alesina, Alberto and Guido Tabellini (2005), “Why Is Fiscal Policy Often Procyclical?” NBER Working Papers No 11600 (Cambride, MA: National Bureau of Economic Research).  
- Andrés, Javier, Rafael Doménech, and Antonio Fatás (2008), “The Stabilizing Role of Government Size”, Journal of Economic Dynamics and Control 32: 571-93.  
- Baunsgaard, Thomas and Steven Symansky (2009), “Automatic Fiscal Stabilizers,” IMF Staff Position Note No 09/23, (Washington, D.C: International Monetary Fund).  
- Beck, Thorsten, George Clarke, Alberto Groff, Philip Keefer, and Patrick Walsh, 2001. "New Tools in Comparative Political Economy: The Database of Political Institutions." World Bank Economic Review 15: 165-76.  
- Beetsma, Roel, Xavier Debrun, and Frank Klaassen (2001), “Is Fiscal policy Coordination in EMU Desirable?” Swedish Economic Policy Review 8: 57-98.  
- Berger, Helge, and Ulrich Woitek (2005), “Does Conservatism Matter? A Time-Series Approach to Central Bank Behaviour,” The Economic Journal, 115: 745-66.  
- Blanchard, Olivier (2000), “Commentary,” Economic Policy Review, April, (New York: Federal Reserve Bank of New York).  
- Blanchard, Olivier and Roberto Perotti (2002), “An Empirical Characterization of the Dynamic Effects of Changes in Government Spending and Taxes on Output,” Quarterly Journal of Economics 117: 609-57.  
- Blinder, Alan, and Robert Solow (1974), “Analytical Foundations of Fiscal Policy,” in The Economics of Public Finance, (Washington, D.C: The Brookings Institution).  
- Botman, Dennis, Douglas Laxton, Dirk Muir, and Andrei Romanov (2006), “A New-Open-Economy Macro Model for Fiscal Policy Evaluation,” IMF Working Paper No 06/45, (Washington, D.C: International Monetary Fund).  
- Bouthevillain, Carine, Philippine Cour-Thimann, Pablo Hernandez Cos, Matthias Mohr, Mika Tujula, Geert Langenus, Sandro Momigliano and Gerrit Van Den Dool (2001), “Cyclically-adjusted Budget Balances: An Alternative Approach,” ECB Working Papers No 77 (Frankfurt: European Central Bank).  
- Catão Luis and Bennett Sutton (2002), “Sovereign Defaults: The Role of Volatility,” IMF Working Paper No 02/149, (Washington, D.C: International Monetary Fund).  
- Crowe, Christopher and Ellen Meade (2008), “Central Bank Independence and Transparency: Evolution and Effectiveness,” IMF Working Paper No 08/119, (Washington, D.C: International Monetary Fund).  
- Darby, Julia and Jacques Mélitz (2008), “Social Spending and Automatic Stabilizers in the OECD,” Economic Policy 56: 715-56.  
- Debrun, Xavier, Jean Pisani-Ferry, and André Sapir (2008), “Should We Forsake Automatic Stabilization?” European Economy—Economic papers No 316 (Brussels: European Commission); forthcoming in Buti, Marco, Servaas Deroose and Vitor Gaspar (eds.) The Euro—the first decade, Cambridge University Press  
- Debrun, Xavier, David Hauner and Manmohan Kumar (2009), “Independent Fiscal Agencies,” Journal of Economic Surveys 23: 44-81.  
- Dolls, Mathias, Clemens Fuest, and Andreas Peichl (2009), “Automatic Stabilizers and Economic Crisis,” CESifo Working Paper No 2878, (Munich: CESifo).  
- Eichengreen, Barry, Ricardo Hausmann, and Juergen von Hagen (1999) “Reforming Budgetary Institutions in Latin America: the Case for a National Fiscal Council,” Open Economies Review 10: 415–442.  
- Fatás, Antonio, and Mihov, Ilian (2001), “Government Size and Automatic Stabilizers: International and Intranational Evidence,” Journal of International Economics 55: 3–28.  
- Fatás, Antonio, and Mihov, Ilian (2003), “The Case for Restricting Fiscal Policy Discretion,” Quarterly Journal of Economics 118: 1419-47.  
- Fatás, Antonio, and Mihov, Ilian (2009), “The Euro and Fiscal Policy,” NBER Working Paper No 14722, (Cambridge, MA: National Bureau of Economic Research).  
- Fedelino, Annalisa, Anna Ivanova and Mark Horton (2009), “Computing Cyclically-Adjusted Balances and Automatic Stabilizers,” IMF Technical Notes and Manuals No 09/05, (Washington, D.C: International Monetary Fund).  
- Furceri, Davide (2009), “Stabilization Effects of Social Spending: Empirical Evidence from a Panel of OECD Countries,” North American Journal and Economics and Finance 21: 34-48.  
- Gavin Michael and Roberto Perotti (1997), “Fiscal Policy in Latin America,” NBER Macroeconomics Annual 12: 11-72  
- Galì, Jordi (1994), “Government Size and Macroeconomic Stability’, European Economic Review 38: 117–132.  
- Galì, Jordi, and Roberto Perotti (2003), “Fiscal Policy and Monetary Integration in Europe,” Economic Policy 18: 533-72.  
- Girouard, Nathalie and Christophe André, “Measuring Cyclically-Adjusted Budget Balances for OECD Countries,” OECD Economics Department Working Paper No 434 (Paris: Organization for Economic Cooperation and Development)  
- Henisz, Witold (2006), The Political Constraint Index Dataset, Wharton School of the University of Pennsylvania.  
- Horton, Mark, Manmohan Kumar, and Paolo Mauro (2009), “The State of Public Finances: a Cross-Country Fiscal Monitor,” IMF Staff Position Note No 09/21, (Washington, D.C: International Monetary Fund).  
- International Monetary Fund (2007), “The Changing Dynamics of the Global Business Cycle,”

*Source: _wp10111 - REFERENCES*

### Chapter 2 of the October World Economic Outlook, (Washington, D.C: International

### _wp10111 - Chapter 2 of the October World Economic Outlook, (Washington, D.C: International Monetary Fund)

### Key data sources
- General government expenditure as a percentage of GDP, GDP per capita, openness to trade, public debt (percentage of GDP), private consumption, dependency ratio and urbanization rates: IMF World Economic Outlook Database.
- Financial development (total stock of credit by deposit money banks to private sector as percentage of GDP) and indices of oil prices: IMF International Financial Statistics.
- Political and electoral systems: Database of Political Institutions (Beck and others, 2001).
- Political constraint index: POLCON database (Henisz, 2006).
- Index of government stability: International Country Risk Guide database.
- Index of Central Bank Independence: Crowe and Meade (2008).

### Cyclical dimension of fiscal policy (Table A1) — selected country estimates
- Argentina: -0.0188
- Australia: 0.7556*
- Austria: 0.1364
- Brazil: -0.7710*
- Canada: 0.2818*
- Chile: -0.4145
- China: -0.0444
- France: -0.4423
- Germany: -0.2414*
- India: -0.2407*
- Ireland: -0.5471*
- Italy: 0.3390*
- Japan: 0.1712
- Mexico: -0.5670*
- Netherlands: -0.6352
- Norway: -0.9851
- Russia: -0.5091
- Spain: -0.4718*
- Switzerland: 0.5308
- United Kingdom: 0.1062
- United States: 0.5361*
- Venezuela, Rep. Bol.: -0.5570*

(Source: Authors’ calculations.)

### Discretionary dimension of fiscal policy (Table A2) — representative entries
- United States: 1970-79 1.247; 1980-89 0.748; 1990-99 1.178; 2000-06 0.889
- United Kingdom: 1970-79 0.847; 1980-89 0.935; 1990-99 1.827; 2000-06 1.053
- Italy: 1970-79 1.448; 1980-89 1.198; 1990-99 1.899; 2000-06 1.055
- Netherlands: 1970-79 1.714; 1980-89 1.129; 1990-99 1.884; 2000-06 1.918
- Norway: 1970-79 2.884; 1980-89 3.455; 1990-99 3.015; 2000-06 5.072
- Sweden: 1970-79 2.110; 1980-89 2.208; 1990-99 3.522; 2000-06 2.013
- Ireland: 1970-79 4.318; 1980-89 2.110; 1990-99 1.401; 2000-06 2.080
- Mexico: 1970-79 1.000; 1980-89 7.080; 1990-99 2.420; 2000-06 1.690
- United Arab Emirates: 1970-79 4.673; 1980-89 7.839; 1990-99 5.843; 2000-06 8.970
- Venezuela: 1990-99 5.429; 2000-06 4.604

(Entries “...” indicate missing or not reported in table cells. Source: Authors’ calculations.)

### Determinants of cyclical fiscal policy (Table A3) — selected coefficient estimates (Variance-Weighted Least Squares)
- GDP per capita: .104*** (model 1); .078*** (2); .052** (3); .061** (4); .055* (5); .053* (6)
- Openness: -.373* (in earlier specification); additional openness estimates reported as -.516**, -.611***, -.505**, -.435*, -.457* across specifications
- Presidential System (dummy): .124** in one specification; .104; .110*; .112* in other specifications
- Proportional Electoral rule (dummy): -.238***; -.208***; -.184**; -.193**
- Political Constraints V-Henisz: .334**; .288*; .349*; .338* (in specific specifications)
- Constant terms reported, e.g., -.831***, -.227, -.434*, -.334, -.316, -.449 across specifications
- Observations: 42, 39, 42, 39, 39, 39 (across columns)

(Source: Authors’ calculations.)

### Determinants of automatic stabilizers (Table A4) — selected coefficient estimates
- Dependency ratio: -.263; -.117; -.311; -.197 (not statistically significant)
- Urbanization Rate: .003*; .005***; .003; .002
- Openness: .375***; .184; .245*; .204
- GDP per capita: .128***; .070**; .136***; .130***
- Proportional Electoral rule (dummy): .203*** (in one specification)
- Presidential System (dummy): -.222*** (in one specification)
- Government Fragmentation: .260*** (in one specification)
- Constant: -2.504***; -1.935***; -2.216***; -2.312***
- Observations: 133; 152; 127; 151
- R-squared: .482; .526; .471; .500

(Source: Authors’ calculations.)

### Determinants of discretionary fiscal policy (Table A5) — selected coefficient estimates
- Automatic stabilizers: 1.231*** (1); 1.233*** (2); 1.396** (3); 1.405** (4); 1.453* (5); 1.324* (6)
- Openness: 1.593** (in some specifications); 1.595**; .715; .421; 1.400**; 1.338*
- Volatility in terms of trade: -.009; -.009; -.041; -.039*; .105; .111
- Debt: -.781**; -.780**; -2.835**; -2.998***; -.164; -.141 (varies by sample)
- Presidential System (dummy): .005; -.293; -.283 (in specific specs)
- Political Constraints V-Henisz: -3.425***; -3.421***; -4.507***; -4.796***; 1.573; 1.541 (OECD/non-OECD splits)
- Constant: 5.406***; 5.402***; 7.339***; 7.867***; .925; .853
- Observations: 118; 118; 46; 46; 72; 72
- R-squared: .334; .334; .545; .549; .187; .195

(Source: Authors’ calculations.)

### Analytical link: automatic stabilizers, fiscal multipliers and volatility (Appendix D)
- Model basis: log-linear, backward-looking IS equation with output gap y and government budget deficit d. Cyclical sensitivity parameter denoted α (alpha), persistence parameter denoted ρ (rho) and other standard notation as in the Appendix.
- Calibration assumptions used to map estimated impact of automatic stabilizers into conventional fiscal multipliers:
  - Fiscal policy can stabilize about one third of shocks to ϑ (assumption guided by recent empirical estimates).
  - Set d̄(ϑ) equal to 1.5 times the sample’s measure of output variability.
  - Assume ρ = 0.6.
  - Assume θ1 (denoted ϑ1 in text) spans over [0.1; 1.5].
  - Assume government size can be anywhere between 0.2 and 0.6.
- Implied ranges:
  - Implied values for the estimated parameter γ̂1 (߶ ෠ 1 in text) lie between -2.64 and -0.48.
  - Using sample average government size of 0.38 and assuming discretionary fiscal policy is acyclical (β̃ = 0), the 95 percent confidence interval of γ̂1 (i.e. [-2.81;-1.22]) maps into “fiscal multipliers” [(θ1)(θ1+θ̃+α)ρ]1/5 between 0.4 and 1.5.
  - Replicating the exercise using the standard deviation of the output gap as the measure of volatility (95 percent CI for γ̂1 i.e. [-2.29;-0.92]) yields implied multipliers between 0.4 and 1.0.

Policy-relevant interpretations (from analytical results and empirical estimates)
- Larger automatic stabilizers (greater government size and related fiscal instruments) are associated with lower output-gap volatility, but the marginal stabilizing effect decreases as stabilizers strengthen.
- Automatic stabilizers are positively associated with discretionary fiscal activism (Table A5: Automatic stabilizers coefficients 1.231*** etc.), suggesting countries with larger stabilizers also tend to have larger discretionary responses in the sample.
- Political-institution variables matter: Proportional electoral rules and political constraints are significant determinants of cyclical and discretionary fiscal behavior (e.g., Proportional Electoral rule dummy -.238*** for cyclical policy; Political Constraints V-Henisz -3.425*** for discretionary policy).
- Openness has mixed effects: negative association with cyclical responsiveness in some specifications (openness coefficients reported as negative in Table A3) but positive association with the size of automatic stabilizers (Table A4 openness .375***).

*Chapter 2 of the October World Economic Outlook, (Washington, D.C: International Monetary Fund).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10111.pdf_