## wpiea2020021-print-pdf

## Source details

**Canonical URL:** [wpiea2020021-print-pdf](https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020021-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2020/english/wpiea2020021-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2020/english/wpiea2020021-print-pdf.pdf.json)

---

### Fiscal sustainability and vulnerability
- Fiscal sustainability identified as a paramount challenge for small countries with high debt and greater vulnerability to climate change.
- Many Caribbean countries still face significant fiscal vulnerabilities.
- Average annual cost of weather-related damage in the Caribbean is 2.4 percent of GDP, about six times more than that of larger counties.
- The state of public finances in the Caribbean has improved in recent years, with more prudent policies and reforms aimed at enhancing revenue mobilization and rationalizing expenditures.
- The fiscal stance, as measured by the primary budget balance, moved to a surplus of 0.8 percent of GDP, on average, during the period 2015–2018 from a deficit of 0.6 percent between 2009 and 2014.
- The average level of government debt stood at 73.3 percent of GDP in 2018, compared to 60.4 percent a decade earlier before the onset of the global financial crisis.
- Key comparative statistic: an average of 73.3 percent of GDP government debt in 2018 in the Caribbean compared to 50.8 percent among other developing countries.

### Scope and objective
- Scope: 16 Caribbean countries over the period 1980–2018.
- Main objective: Analyze fiscal sustainability and cyclicality using the cyclically adjusted primary balance (CAPB) and the model-based fiscal reaction function approach (Bohn, 1998).

### Empirical approach and data
- Dependent variable: CAPB (cyclically adjusted primary balance) scaled by potential GDP.
- Main regressors: lagged CAPB, lagged debt-to-GDP ratio, output gap, plus controls (real GDP per capita, consumer price inflation, trade openness, financial development).
- Additional robustness controls: real effective exchange rate, terms-of-trade index, binary for natural disasters, presence of national and supranational fiscal rules.
- Sample construction: unbalanced annual panel; countries include Anguilla, Antigua and Barbuda, Aruba, The Bahamas, Barbados, Belize, Dominica, Grenada, Guyana, Haiti, Jamaica, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, Suriname, and Trinidad and Tobago.
- Potential output estimation: Hodrick-Prescott (HP) filter with λ = 6.25; series extended through 2024 using projections to mitigate end-point problem.
- Cyclical adjustment: follows Hagemann (1999) and Fedelino, Ivanova, and Horton (2009); aggregate revenue elasticity with respect to the output gap assumed to be 1 and aggregate expenditure elasticity assumed to be 0; interest payments removed to compute CAPB.
- Estimation methods: dynamic panel (preferred) using one-step system GMM (Arellano-Bover/Blundell-Bond), with minimal instrument set (collapsed instruments) and robust standard errors clustered at the country level.
- Validation: Arellano-Bond AR tests and Hansen J-test.

### Key findings and quantitative results
- Persistence of discretionary fiscal policy:
  - Lagged CAPB coefficient frequently positive and highly significant.
  - Representative estimates: CAPB_t-1 = 0.661*** [0.036]; 0.629*** [0.039] in baseline specifications.
- Sustainability response to debt:
  - Coefficient on government debt is positive and statistically significant.
  - Magnitude: fiscal policy reaction to an increase of one percentage point in lagged debt-to-GDP is limited to about 0.02 percentage point (period 1980–2018).
  - Representative debt coefficients: Debt_t-1 = 0.028** [0.012]; 0.049*** [0.012]; 0.015*** [0.004]; 0.023*** [0.005].
- Cyclicality:
  - Output gap coefficient positive and statistically significant, indicating a countercyclical fiscal stance on average for the sample over 1980–2018.
  - Reported output gap estimates: 0.185* [0.101]; 0.263*** [0.089]; 0.212*** [0.063]; 0.262*** [0.069].
- Composition of adjustment:
  - Fiscal adjustment to debt accumulation is revenue-based on average: cyclically adjusted revenues respond positively to debt increases while expenditures show an insignificant (or negative) response.
  - Representative composition estimates:
    - Revenue lagged dependent variable = 0.739*** [0.0379].
    - Primary spending lagged dependent variable = 0.626*** [0.0410].
    - Debt_t-1 in revenue equation = 0.0174* [0.009].
    - Debt_t-1 in primary spending = -0.001 [0.0101].
    - Output gap in spending equation = -0.453*** [0.147] (indicating procyclical bias in spending).
- Nonlinearities and fiscal fatigue:
  - Quadratic specification finds CAPB response to debt-to-GDP is nonlinear, turning negative at higher levels of government debt—evidence consistent with fiscal fatigue.
  - Representative nonlinear estimates: Debt_t-1 = 0.033*** [0.012]; Debt2_t-1 = 0 -0.000* [8.000][9.000]; output gap = 0.213*** [0.063].
- Control variables:
  - Real GDP per capita: CAPB improves by about 0.5 percentage point with each percentage point increase in income in some specifications (consistent with estimated income elasticity interval 0.3–0.6).
  - Inflation: negative coefficient but not statistically significant in dynamic regressions.
  - Trade openness and financial development: significant negative effects on discretionary fiscal policy.

### Robustness checks and sensitivity
- Alternative specifications tested: lagged output gap included (not statistically significant but points toward countercyclical behavior).
- Sample trimming: truncating at 5th and 95th percentiles — similar results.
- Excluding commodity exporters (Guyana, Suriname, Trinidad and Tobago) — marginally larger coefficients on debt and output gap.
- Excluding post-global financial crisis period — positive debt response remains unchanged but becomes stronger in the post-crisis period.
- Adding REER and TOT as controls — no significantly different effect.
- Including binary for natural disasters — not statistically significant, but occurrence appears to lessen fiscal response to debt and extent of countercyclicality.
- Including fiscal rules (national and supranational) — fiscal rules not statistically significant overall; national fiscal rules coefficient positive, supranational fiscal rules coefficient negative; only Grenada (since 2015) and Jamaica (since 2010) have national rules; six ECCU members have followed supranational targets since 1998 without an effective national constraint.
- System GMM diagnostics: AR1 p-values 0.000; AR2 p-values typically non-significant (examples: 0.14, 0.179, 0.135, 0.161); Hansen J-test p-values examples: 0.05, 0.088, 0.037, 0.062.

### Policy implications and recommendations
- Maintain a countercyclical fiscal stance while accounting for long-run solvency concerns.
- Build fiscal buffers and ensure debt sustainability with high probability even when negative shocks occur repeatedly over the long term.
- Recognize nonlinearities: corrective fiscal action becomes more difficult at high levels of indebtedness (fiscal fatigue), so earlier and sustained adjustments are preferable to late, larger corrections that may be untenable.
- Focus on revenue-based adjustments where feasible, while addressing the procyclical bias in government spending through institutional and policy measures.
- Implement growth-enhancing structural reforms to strengthen fiscal positions and create space for buffers against repeated negative shocks and climate-related risks.

*Source: wpiea2020021-print-pdf — References; I. INTRODUCTION*

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

### References — I. INTRODUCTION

### Fiscal sustainability and vulnerability
- Fiscal sustainability is identified as a paramount challenge for small countries with high debt and greater vulnerability to climate change.
- Many Caribbean countries still face significant fiscal vulnerabilities.

### Recent fiscal stance and reforms
- The state of public finances in the Caribbean has improved in recent years, with more prudent policies and reforms aimed at enhancing revenue mobilization and rationalizing expenditures.
- The fiscal stance, as measured by the primary budget balance, moved to a surplus of 0.8 percent of GDP, on average, during the period 2015–2018 from a deficit of 0.6 percent between 2009 and 2014.

### Government debt levels
- The average level of government debt stood at 73.3 percent of GDP in 2018.
- This is substantially higher than 60.4 percent a decade earlier before the onset of the global financial crisis.

*Source: wpiea2020021-print-pdf — References; I. INTRODUCTION*

### 50.8 percent among developing countries. These small open economies are also significantly

### Debt and Fiscal Stance in the Caribbean

### Summary
- Scope: 16 Caribbean countries over the period 1980–2018.
- Main objective: Analyze fiscal sustainability and cyclicality using the cyclically adjusted primary balance (CAPB) and the model-based fiscal reaction function approach (Bohn, 1998).
- Motivation: Small open Caribbean economies are significantly more vulnerable to natural disasters and climate change; average annual cost of weather-related damage in the Caribbean is 2.4 percent of GDP, about six times more than that of larger counties.
- Key comparative statistic: an average of 73.3 percent of GDP government debt in 2018 in the Caribbean compared to 50.8 percent among other developing countries.

### Empirical approach and data
- Dependent variable: CAPB (cyclically adjusted primary balance) scaled by potential GDP.
- Main regressors: lagged CAPB, lagged debt-to-GDP ratio, output gap, plus controls (real GDP per capita, consumer price inflation, trade openness, financial development). Additional robustness controls include real effective exchange rate, terms-of-trade index, binary for natural disasters, presence of national and supranational fiscal rules.
- Sample construction: unbalanced annual panel; countries include Anguilla, Antigua and Barbuda, Aruba, The Bahamas, Barbados, Belize, Dominica, Grenada, Guyana, Haiti, Jamaica, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, Suriname, and Trinidad and Tobago.
- Potential output estimation: Hodrick-Prescott (HP) filter with λ = 6.25 after experimenting with smoothing parameters; series extended through 2024 using projections to mitigate end-point problem.
- Cyclical adjustment: follows Hagemann (1999) and Fedelino, Ivanova, and Horton (2009); aggregate revenue elasticity with respect to the output gap assumed to be 1 and aggregate expenditure elasticity assumed to be 0; interest payments removed to compute CAPB.
- Estimation methods: dynamic panel (preferred) using one-step system GMM (Arellano-Bover/Blundell-Bond), with minimal instrument set (collapsed instruments) and robust standard errors clustered at the country level. Validation via Arellano-Bond AR tests and Hansen J-test.

### Key findings and quantitative results
- Persistence of discretionary fiscal policy:
  - Lagged CAPB coefficient frequently positive and highly significant. Example reported estimates: CAPB_t-1 = 0.661*** [0.036], 0.629*** [0.039] in baseline specifications.
- Sustainability response to debt:
  - Coefficient on government debt is positive and statistically significant.
  - Magnitude: fiscal policy reaction to an increase of one percentage point in lagged debt-to-GDP is limited to about 0.02 percentage point (period 1980–2018).
  - Representative reported debt coefficients: Debt_t-1 = 0.028** [0.012]; 0.049*** [0.012]; 0.015*** [0.004]; 0.023*** [0.005] across various specifications.
- Cyclicality:
  - Output gap coefficient positive and statistically significant, indicating a countercyclical fiscal stance on average for the sample over 1980–2018.
  - Reported output gap estimates: 0.185* [0.101]; 0.263*** [0.089]; 0.212*** [0.063]; 0.262*** [0.069].
- Composition of adjustment:
  - Fiscal adjustment to debt accumulation is revenue-based on average: cyclically adjusted revenues respond positively to debt increases while expenditures show an insignificant (or negative) response.
  - Representative estimates for composition: Revenue lagged dependent variable = 0.739*** [0.0379]; Primary spending lagged dependent variable = 0.626*** [0.0410]; Debt_t-1 in revenue equation = 0.0174* [0.009]; Debt_t-1 in primary spending = -0.001 [0.0101]; Output gap in spending equation = -0.453*** [0.147] (indicating procyclical bias in spending).
- Nonlinearities and fiscal fatigue:
  - Quadratic specification finds CAPB response to debt-to-GDP is nonlinear, turning negative at higher levels of government debt—evidence consistent with fiscal fatigue.
  - Representative nonlinear estimates: Debt_t-1 = 0.033*** [0.012]; Debt2_t-1 = 0 -0.000* [8.000][9.000]; output gap = 0.213*** [0.063]; with controls included in some specifications.
- Control variables:
  - Real GDP per capita: CAPB improves by about 0.5 percentage point with each percentage point increase in income in some specifications (consistent with estimated income elasticity interval 0.3–0.6).
  - Inflation: negative coefficient but not statistically significant in dynamic regressions.
  - Trade openness and financial development: significant negative effects on discretionary fiscal policy (may reflect greater financial capacity for deficit financing).

### Robustness checks and sensitivity
- Alternative specifications tested:
  - Lagged output gap included (not statistically significant but points toward countercyclical behavior).
  - Sample truncated at 5th and 95th percentiles — similar results.
  - Excluding commodity exporters (Guyana, Suriname, Trinidad and Tobago) — marginally larger coefficients on debt and output gap.
  - Excluding post-global financial crisis period — positive debt response remains unchanged but becomes stronger in the post-crisis period.
  - Adding REER and TOT as controls — no significantly different effect.
  - Including binary for natural disasters — not statistically significant, but occurrence appears to lessen fiscal response to debt and extent of countercyclicality.
  - Including fiscal rules (national and supranational) — fiscal rules not statistically significant overall; national fiscal rules coefficient positive, supranational fiscal rules coefficient negative; only Grenada (since 2015) and Jamaica (since 2010) have national rules; six ECCU members have followed supranational targets since 1998 without an effective national constraint.
- System GMM diagnostics: AR1 p-values 0.000; AR2 p-values typically non-significant (e.g., 0.14, 0.179, 0.135, 0.161 in reported tables) and Hansen J-test p-values generally acceptable (examples: 0.05, 0.088, 0.037, 0.062).

### Policy implications and recommendations
- Maintain a countercyclical fiscal stance while accounting for long-run solvency concerns.
- Build fiscal buffers and ensure debt sustainability with high probability even when negative shocks occur repeatedly over the long term.
- Recognize nonlinearities: corrective fiscal action becomes more difficult at high levels of indebtedness (fiscal fatigue), so earlier and sustained adjustments are preferable to late, larger corrections that may be untenable.
- Focus on revenue-based adjustments where feasible, while addressing the procyclical bias in government spending through institutional and policy measures.
- Implement growth-enhancing structural reforms to strengthen fiscal positions and create space for buffers against repeated negative shocks and climate-related risks.

*Source: IMF working paper analysis of fiscal sustainability and cyclicality for 16 Caribbean countries (1980–2018).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020021-print-pdf.pdf_