## _wp13136

## Source details

**Canonical URL:** [_wp13136](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13136.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13136.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13136.pdf.json)

---

### Major findings and policy-relevant results
- Fiscal multipliers are generally significantly different from zero (at least in the first three years following fiscal policy shocks).
- Medium-term effects of fiscal consolidation:
  - Positive for most Central American and low-income countries, particularly when consolidation is based on either increases in tax revenues or cuts in government current spending.
  - Cuts in capital spending are, in general, bad for short- and medium-term output.
  - Advanced and emerging market economies post better medium-term output effects when consolidation focuses on current spending cuts, although the medium-term effects of fiscal consolidation in these countries remain negative.
- The analysis highlights “contractionary fiscal expansions” as possible: wider interest-rate spreads triggered by debt accumulation can depress activity despite an initial positive demand impulse.
- The adequacy of fiscal stimulus depends on public debt levels and external financing needs; feedback from debt accumulation to growth is crucial for evaluating stimulus versus consolidation trade-offs.

### Contributions to methodology
- Develops a methodology to calculate fiscal multipliers using annual data while explicitly accounting for feedback effects from debt accumulation to growth.
- Employs a structural vector error-correction model (SVECM) approach that:
  - Nests short-run restrictions and uses cointegrating relationships to exploit long-run properties of the data.
  - Extends Blanchard and Perotti (2002) and Perotti (2005) by incorporating a government budget constraint and debt feedback following Favero and Giavazzi (2007).
- Uses annual data from 1972 to 2010 to estimate multipliers across a wide range of countries and groupings.
- Tailors the SVECM model to each country to account for idiosyncratic factors (monetary, exchange rate, trade, and fiscal policies), vulnerabilities, and structural breaks.

### Country coverage and groupings
- Regional focus: Central America, Panama and the Dominican Republic (CAPDR).
- Comparator groups include: low-income countries (LIC), heavily indebted countries (HIPC), sub-Saharan Africa (SSA), emerging market economies (EME), oil producing countries, and advanced economies (AE).
- Annual data window used for comparability: 1972 to 2010.

### Model specification and identification (key elements preserved)
- Core SVAR/SVECM representation:
  - Equation [1]: 01(  )(  ) tttt A YA L YF  L dB ε− = + + 
  - Government budget constraint (debt evolution) shown in Equation [2]:
    1 (1)(    )(  ) (1)(1)(   ) ttt tt t iExp  gExp t dd pyExp  y − + − = + + Δ   + Δ
  - Observed fiscal policy reactions (expenditure and tax) captured by [3] and [4] as functions of automatic responses, systematic discretionary responses, and random discretionary fiscal shocks.
- Vector of endogenous variables:
  - Y_t = [g_t, t_t, y_t, reer_t, i_t] where:
    - g_t: government spending (sum of government consumption and investment, excluding interest payments)
    - t_t: net revenue excluding interest receipt on government debt
    - y_t: real output
    - reer_t: real effective exchange rate
    - i_t: yield on government securities
- Structural shocks vector:
  - ε'_t = [ε_g,t, ε_t,t, ε_y,t, ε_reer,t, ε_i,t]
  - Innovations vector: e'_t = [e_g,t, e_t,t, e_y,t, e_reer,t, e_i,t]
- Identification strategy:
  - Uses cointegrating relationships and assumed policy decision lags to identify fiscal shocks with annual data.
  - Accounts for contemporaneous parameter restrictions in the 0A matrix; a just-identification of [5] requires 5(5   1) 15 2 restrictions on 0A (as stated in the source).
  - Imposes restriction [6]: ,,,0 g i t i α α = = on grounds that interest payments on government debt are excluded from the definition of expenditures and taxes used in the model and the contemporaneous impact of interest rates on taxes is likely to be small or close to zero in practice.

### Policy implications and interpretation
- When public debt dynamics and external financing constraints are significant, fiscal consolidation that prioritizes tax revenue increases or cuts to current spending is less disruptive to growth than cuts to capital spending.
- Country-specific modeling is important: multipliers and the growth impact of fiscal consolidation vary by country group and idiosyncratic policy frameworks.
- Annual-data SVECM identification provides a tool to estimate multipliers in regions lacking high-frequency fiscal data, allowing informed assessment of growth-friendly consolidation strategies.

### Identification and structural VECM / SVECM restrictions
- Contemporaneous restriction reported in the source:
  - ,0 iy     [7]
- Common high-frequency identification assumption (may not hold for annual data):
  - ,0 gy  and ,0 ty 
- Cointegration strategy:
  - Use cointegration analysis to impose missing identification assumptions when variables are I(1).
  - With at least one cointegrating relationship, estimates of the automatic response of taxes to changes in the economic environment and the exchange rate, or the automatic response of government spending to economic or exchange rate shocks, can be obtained.
- Structural VECM representation (one cointegrating vector) given as:
  - '0111 () ttttt A yayA yF L dB          [8]
  - where 0 aA  and  is the loading parameter.
- With cointegration between government spending, output growth, and exchange rate, remaining coefficients can be obtained by the system in [9]; parameters of [3’] can be proxied by estimates from [9]. This strategy identifies all fiscal shocks and allows full estimation of matrix 0 A.

### Expanded model: splitting total expenditure into capital and current spending
- Expanded equation with 6 endogenous variables splits total expenditure (excluding interest payment on debt) into:
  - capital expenditure (“ca”) and current spending (“cu”).
- The expanded SVECM requires 6(6 + 1)/2 = 21 restrictions in which 15 are already obtained.
- Additional contemporaneous restrictions imposed following [6] and [7]:
  - ,, ,0 ca i cu it i     
  - ,0 iy 
- Remaining two coefficients are determined using cointegration properties.

### Empirical setup and estimation details
- Data:
  - Annual data for 1973–2011 from IMF International Financial Statistics and IMF World Economic Outlook.
- Country and group coverage includes:
  - Costa Rica (CRI), Dominican Republic (DOM), El Salvador (ELS), Honduras (HON), Guatemala (GUA), Nicaragua (NIC), Panama (PAN), Low-Income Countries (LICs), Heavily Indebted Poor Countries (HIPC), Oil producing countries (OIL), Emerging Market Economies (EMEs), Advanced Economies (AEs), and Sub-Saharan Africa excluding South Africa (SSA).
- SVAR / SVECM estimation steps:
  - (i) testing for optimal lag length;
  - (ii) testing for model stability;
  - (iii) analyzing partial autocorrelograms for serial correlation and heteroscedasticity;
  - (iv) testing for unit roots;
  - (v) performing cointegration analysis and setting up ECM equations;
  - (vi) transforming the SVAR into SVECM;
  - (vii) producing impulse response functions from which impact and cumulative multipliers are calculated.
- Multiplier definitions (as used):
  - Impact Multiplier = change in output at t=0 / exog. and temp. change in fiscal outcome (expenditure or taxes) at t=0
  - Cumulative multipliers computed from impulse responses; i is the median interest rate in the sample (following Ilzetzki et al. (2010)).
- Time-series diagnostics and choices:
  - Unit-root tests indicate most variables are nonstationary.
  - Optimal lag tests (AIC, SC, Hannan-Quinn) suggest lags vary between 1 and 2; model generalized to allow for 2 (when test indicates 1 lag) to 3 (when test indicates 2 lag) lags because of limited observations.
  - Unrestricted VAR(2) or VAR(3) diagnostics confirm absence of serial autocorrelation and heteroscedasticity.
  - Johansen’s LR trace test generally indicates one cointegrating equation at 5 percent significance.
  - Monte Carlo simulations with bootstrapped standard errors and 500 repetitions used to estimate 90 percent confidence intervals.

### Estimated automatic stabilizers and elasticities
- Overall estimated output elasticities of net taxes:
  - range from 0.17 (HIPC) to 1.85 (AE) and 1.96 (EME).
- Interpretation and caveats:
  - Low tax elasticity for HIPC and LICs may reflect consistent discretionary tax relief, lower tax bases, widespread corruption in tax collection, or poor data quality.

### Main empirical findings (overall and by country/group)
- General result:
  - Fiscal consolidation in less developed countries hurts output only in the short-term.
  - Negative short-term effect is largest for advanced economies (AEs), significant for emerging markets (EMEs), and small for less developed economies.
- Statistical significance:
  - Impact and cumulative multipliers are generally statistically significant from zero at the 90 percent confidence level through each horizon, but often not at the 95 percent confidence level; error bands widen at longer horizons.
- Cumulative and medium-run results:
  - The positive long-run effect of tax increases on output tends to be higher than the positive effect of cutting spending in HIPCs, LICs, and SSA.
  - This suggests that fiscal stabilization focused on spending cuts having better growth outcomes may not apply to poorer economies—possible explanations include inefficient tax administration and that observed tax revenue increases reflect improved efficiency with little distortive impact.
- Central America (CAPDR) specifics:
  - Most Central American countries have larger tax and spending multipliers in the medium term than other countries at similar development stages.

### Spending multipliers (detailed figures)
- Impact multipliers for spending cuts in Central America:
  - range from -0.01 (Nicaragua) to -0.44 (Panama).
- Impact multipliers across country groupings:
  - between -0.01 (HIPCs) and -0.43 (AEs).
  - Interpretation example: a one percentage-point decline in government spending as a ratio to GDP is associated with:
    - a 0.01 percent output contraction in HIPCs and Nicaragua in the same year,
    - a 0.44 percent output contraction in Panama in the same year,
    - a 0.43 percent output contraction in AEs in the same year.
- Two- to four-year (medium-term) effects:
  - Spending cut multipliers in Central America range from -0.94 (Panama) to 0.43 (Nicaragua).
  - Multipliers are positive for some poorer economies (for instance, 0.41 for SSA excluding South Africa) but negative in EMEs (-1.03) and AEs (-0.8).

### Tax multipliers (detailed figures)
- Impact multipliers for tax increases in Central America:
  - small and not statistically different from zero.
- Short-run output effect of a tax increase by group:
  - HIPCs: positive impact (specific impact value not provided in the excerpt);
  - AEs: zero;
  - SSA: zero;
  - LICs: zero;
  - OIL: -0.12;
  - EMEs: -0.10.
- Cumulative (medium-run) effects on output of tax-revenue increases:
  - Positive for Central American countries: range from 0.20 (Dominican Republic) to 0.51 (Guatemala).
  - Positive for HIPCs, LICs, and SSA.
  - Negative for AEs, EMEs, and oil producers (consistent with other literature).
  - Medium-run cumulative multiplier ranges reported:
    - Central America: 0.24 (Nicaragua) and 0.57 (Panama).
    - Other country groups: 0.34 (oil exporter countries) and 0.53 (LIC).

### Capital and current spending multipliers
- Splitting public spending into investment (capital) and current expenditure:
  - Public investment increases produce positive impact multipliers on output (generally higher than that of taxes).
  - Increases in current spending lower output in several countries.
- The multipliers for capital expenditures range on impact from 0.21 to 0.70 and on cumulative/medium-run from 0.42 to 0.91 (with some tables reporting up to 1.00).
- Current spending multipliers:
  - Impact range: -0.24 to 0.19.
  - Cumulative impact range: -1.00 to -0.41.
- Point estimates are generally significantly different from zero in the short-run, while medium-run estimates are associated with large uncertainties for both current and capital investments.
- Government current expenditure tends to respond strongly to public investment shocks; this is confirmed when imposing a zero restriction on government current spending to isolate “pure” government investment effects.
- There is no single fiscal multiplier: multipliers are country- or region-specific and model-specific.

### Key comparative results (selected exact estimates)
- This paper — Central America (incl. Dominican Republic and Panama), SVECM, Annual:
  - Total expenditure impact: 0.01 to 0.44.
  - Total expenditure cumulative: 0.45 to 0.94.
  - Taxes impact: 0.00 to 0.02.
  - Taxes cumulative: -1.13 to -0.20.
  - Current expenditure impact: -0.19 to -0.04.
  - Current expenditure cumulative: -0.40 to 0.79.
  - Capital expenditure impact: 0.21 to 0.70.
  - Capital expenditure cumulative: 0.42 to 1.00.
- Other regional/comparator multipliers (exact table excerpts):
  - AE (Advanced Economies) total expenditure: 0.43 (impact) and 0.78 (cumulative).
  - EME total expenditure: 0.34 (impact) and 1.04 (cumulative).
  - Oil producing countries total expenditure: 0.31 to 0.44 (impact reported).
  - HIPC total expenditure: 0.01 (impact) and -0.26 (cumulative).
  - LIC total expenditure: 0.17 (impact) and -0.33 (cumulative).
  - SSA total expenditure: 0.25 to 0.41 (impact reported).
- Selected literature benchmarks (exact figures as reported in table):
  - Ilzetzti, Mendoza, and Vegh (2010), 44 countries (SVAR, quarterly, developing): Total expenditure -0.21 to 0.18; Capital expenditure 0.57 to 0.75.
  - Blanchard and Perotti (2002), United States (VAR, quarterly): Spending 0.8 to 0.9 (impact) and 0.7 to 1.0 (cumulative); Tax 0.7 to 1.3.
  - Mountford and Uhlig (2002), United States (VAR, quarterly): Spending 0.2 to 0.5; Tax 0.2 to 0.4.
  - Perotti (2002), United States (VAR, quarterly): Spending 0.1 to 0.4 (impact) and -1.3 to 1.0 (cumulative); Tax 0.2 to 0.4.

### Country- and region-specific policy-relevant findings
- CAPDR (Central America including Dominican Republic and Panama):
  - Results point to the importance of moving the composition of public spending toward investment given capital scarcity in the region.
  - Fiscal consolidation paths should preserve (or even increase) public investment in CAPDR.
- Nicaragua:
  - Presents large negative multipliers associated with increases in current spending both in the short- and in the long-run; controlling current spending is particularly relevant.
  - Nicaragua and Honduras show greater benefits from cutting current spending than from raising tax revenues (as reported).
- Panama and Costa Rica:
  - Panama shows a relatively large capital-expenditure multiplier (Panama: capital expenditure cumulative reported as up to 1.1921 in one band; impact and cumulative bands reported in table).
  - Costa Rica and some mid-income countries may experience positive medium-term effects from fiscal consolidation if based on improved tax system efficiency.
- HIPCs and LICs:
  - Fiscal consolidation tends to have a strong positive effect in low income countries (especially HIPCs).
  - Tax revenue increases appear to yield greater medium-term output growth than current spending control in several Central American countries and LICs, including HIPCs, but public investment multipliers are larger than tax multipliers.

### Policy implications and recommendations
- Fiscal consolidation should pay attention to composition:
  - Preserve or increase public investment where capital scarcity warrants (e.g., CAPDR).
  - Control current spending where multipliers are negative or contractionary (notably Nicaragua).
- The persistence and composition of fiscal measures matter:
  - The size and sign of multipliers depend on whether consolidation targets current expenditure, capital expenditure, or taxes, on short- versus long-run horizons, and on monetary policy accommodation.
- Debt dynamics and financial constraints:
  - Including debt-to-GDP feedback matters because declining public debt from consolidation may reduce interest rates and potentially offset part of the initial negative impulse of consolidation.

*Source: _wp13136 - References (IMF PDF content).*

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

### _wp13136 - References

### Major findings and policy-relevant results
- Fiscal multipliers are generally significantly different from zero (at least in the first three years following fiscal policy shocks).
- Medium-term effects of fiscal consolidation:
  - Positive for most Central American and low-income countries, particularly when consolidation is based on either increases in tax revenues or cuts in government current spending.
  - Cuts in capital spending are, in general, bad for short- and medium-term output.
  - Advanced and emerging market economies post better medium-term output effects when consolidation focuses on current spending cuts, although the medium-term effects of fiscal consolidation in these countries remain negative.
- The analysis highlights “contractionary fiscal expansions” as possible: wider interest-rate spreads triggered by debt accumulation can depress activity despite an initial positive demand impulse.
- The adequacy of fiscal stimulus depends on public debt levels and external financing needs; feedback from debt accumulation to growth is crucial for evaluating stimulus versus consolidation trade-offs.

### Contributions to methodology
- Develops a methodology to calculate fiscal multipliers using annual data while explicitly accounting for feedback effects from debt accumulation to growth.
- Employs a structural vector error-correction model (SVECM) approach that:
  - Nests short-run restrictions and uses cointegrating relationships to exploit long-run properties of the data.
  - Extends Blanchard and Perotti (2002) and Perotti (2005) by incorporating a government budget constraint and debt feedback following Favero and Giavazzi (2007).
- Uses annual data from 1972 to 2010 to estimate multipliers across a wide range of countries and groupings.
- Tailors the SVECM model to each country to account for idiosyncratic factors (monetary, exchange rate, trade, and fiscal policies), vulnerabilities, and structural breaks.

### Country coverage and groupings
- Regional focus: Central America, Panama and the Dominican Republic (CAPDR).
- Comparator groups include: low-income countries (LIC), heavily indebted countries (HIPC), sub-Saharan Africa (SSA), emerging market economies (EME), oil producing countries, and advanced economies (AE).
- Annual data window used for comparability: 1972 to 2010.

### Model specification and identification (key elements preserved)
- Core SVAR/SVECM representation:
  - Equation [1]: 01(  )(  ) tttt A YA L YF  L dB ε− = + + 
  - Government budget constraint (debt evolution) shown in Equation [2]:
    1 (1)(    )(  ) (1)(1)(   ) ttt tt t iExp  gExp t dd pyExp  y − + − = + + Δ   + Δ
  - Observed fiscal policy reactions (expenditure and tax) captured by [3] and [4] as functions of automatic responses, systematic discretionary responses, and random discretionary fiscal shocks.
- Vector of endogenous variables:
  - Y_t = [g_t, t_t, y_t, reer_t, i_t] where:
    - g_t: government spending (sum of government consumption and investment, excluding interest payments)
    - t_t: net revenue excluding interest receipt on government debt
    - y_t: real output
    - reer_t: real effective exchange rate
    - i_t: yield on government securities
- Structural shocks vector:
  - ε'_t = [ε_g,t, ε_t,t, ε_y,t, ε_reer,t, ε_i,t]
  - Innovations vector: e'_t = [e_g,t, e_t,t, e_y,t, e_reer,t, e_i,t]
- Identification strategy:
  - Uses cointegrating relationships and assumed policy decision lags to identify fiscal shocks with annual data.
  - Accounts for contemporaneous parameter restrictions in the 0A matrix; a just-identification of [5] requires 5(5   1) 15 2 restrictions on 0A (as stated in the source).
  - Imposes restriction [6]: ,,,0 g i t i α α = = on grounds that interest payments on government debt are excluded from the definition of expenditures and taxes used in the model and the contemporaneous impact of interest rates on taxes is likely to be small or close to zero in practice.

### Policy implications and interpretation
- When public debt dynamics and external financing constraints are significant, fiscal consolidation that prioritizes tax revenue increases or cuts to current spending is less disruptive to growth than cuts to capital spending.
- Country-specific modeling is important: multipliers and the growth impact of fiscal consolidation vary by country group and idiosyncratic policy frameworks.
- Annual-data SVECM identification provides a tool to estimate multipliers in regions lacking high-frequency fiscal data, allowing informed assessment of growth-friendly consolidation strategies.

*Italicized source: _wp13136 - References (IMF PDF content).*

### 2. The interest rate on government debt depends on the fiscal stance and the exchange

### 2. The interest rate on government debt depends on the fiscal stance and the exchange rate, but (contemporaneously) less on output

### Identification and structural VECM / SVECM restrictions
- Contemporaneous restriction reported in the source:
  - ,0 iy     [7]
- Common high-frequency identification assumption (may not hold for annual data):
  - ,0 gy  and ,0 ty 
- Cointegration strategy:
  - Use cointegration analysis to impose missing identification assumptions when variables are I(1).
  - With at least one cointegrating relationship, estimates of the automatic response of taxes to changes in the economic environment and the exchange rate, or the automatic response of government spending to economic or exchange rate shocks, can be obtained.
- Structural VECM representation (one cointegrating vector) given as:
  - '0111 () ttttt A yayA yF L dB          [8]
  - where 0 aA  and  is the loading parameter.
- With cointegration between government spending, output growth, and exchange rate, remaining coefficients can be obtained by the system in [9]; parameters of [3’] can be proxied by estimates from [9]. This strategy identifies all fiscal shocks and allows full estimation of matrix 0 A.

### Expanded model: splitting total expenditure into capital and current spending
- Expanded equation with 6 endogenous variables splits total expenditure (excluding interest payment on debt) into:
  - capital expenditure (“ca”) and current spending (“cu”).
- The expanded SVECM requires 6(6 + 1)/2 = 21 restrictions in which 15 are already obtained.
- Additional contemporaneous restrictions imposed following [6] and [7]:
  - ,, ,0 ca i cu it i     
  - ,0 iy 
- Remaining two coefficients are determined using cointegration properties.

### Empirical setup and estimation details
- Data:
  - Annual data for 1973–2011 from IMF International Financial Statistics and IMF World Economic Outlook.
- Country and group coverage includes:
  - Costa Rica (CRI), Dominican Republic (DOM), El Salvador (ELS), Honduras (HON), Guatemala (GUA), Nicaragua (NIC), Panama (PAN), Low-Income Countries (LICs), Heavily Indebted Poor Countries (HIPC), Oil producing countries (OIL), Emerging Market Economies (EMEs), Advanced Economies (AEs), and Sub-Saharan Africa excluding South Africa (SSA).
- SVAR / SVECM estimation steps:
  - (i) testing for optimal lag length;
  - (ii) testing for model stability;
  - (iii) analyzing partial autocorrelograms for serial correlation and heteroscedasticity;
  - (iv) testing for unit roots;
  - (v) performing cointegration analysis and setting up ECM equations;
  - (vi) transforming the SVAR into SVECM;
  - (vii) producing impulse response functions from which impact and cumulative multipliers are calculated.
- Multiplier definitions (as used):
  - Impact Multiplier = change in output at t=0 / exog. and temp. change in fiscal outcome (expenditure or taxes) at t=0
  - Cumulative multipliers computed from impulse responses; i is the median interest rate in the sample (following Ilzetzki et al. (2010)).
- Time-series diagnostics and choices:
  - Unit-root tests indicate most variables are nonstationary.
  - Optimal lag tests (AIC, SC, Hannan-Quinn) suggest lags vary between 1 and 2; model generalized to allow for 2 (when test indicates 1 lag) to 3 (when test indicates 2 lag) lags because of limited observations.
  - Unrestricted VAR(2) or VAR(3) diagnostics confirm absence of serial autocorrelation and heteroscedasticity.
  - Johansen’s LR trace test generally indicates one cointegrating equation at 5 percent significance.
  - Monte Carlo simulations with bootstrapped standard errors and 500 repetitions used to estimate 90 percent confidence intervals.

### Estimated automatic stabilizers and elasticities
- Overall estimated output elasticities of net taxes:
  - range from 0.17 (HIPC) to 1.85 (AE) and 1.96 (EME).
- Interpretation and caveats:
  - Low tax elasticity for HIPC and LICs may reflect consistent discretionary tax relief, lower tax bases, widespread corruption in tax collection, or poor data quality.

### Main empirical findings (overall and by country/group)
- General result:
  - Fiscal consolidation in less developed countries hurts output only in the short-term.
  - Negative short-term effect is largest for advanced economies (AEs), significant for emerging markets (EMEs), and small for less developed economies.
- Statistical significance:
  - Impact and cumulative multipliers are generally statistically significant from zero at the 90 percent confidence level through each horizon, but often not at the 95 percent confidence level; error bands widen at longer horizons.
- Cumulative and medium-run results:
  - The positive long-run effect of tax increases on output tends to be higher than the positive effect of cutting spending in HIPCs, LICs, and SSA.
  - This suggests that fiscal stabilization focused on spending cuts having better growth outcomes may not apply to poorer economies—possible explanations include inefficient tax administration and that observed tax revenue increases reflect improved efficiency with little distortive impact.
- Central America (CAPDR) specifics:
  - Most Central American countries have larger tax and spending multipliers in the medium term than other countries at similar development stages.

### Spending multipliers (detailed figures)
- Impact multipliers for spending cuts in Central America:
  - range from -0.01 (Nicaragua) to -0.44 (Panama).
- Impact multipliers across country groupings:
  - between -0.01 (HIPCs) and -0.43 (AEs).
  - Interpretation example: a one percentage-point decline in government spending as a ratio to GDP is associated with:
    - a 0.01 percent output contraction in HIPCs and Nicaragua in the same year,
    - a 0.44 percent output contraction in Panama in the same year,
    - a 0.43 percent output contraction in AEs in the same year.
- Two- to four-year (medium-term) effects:
  - Spending cut multipliers in Central America range from -0.94 (Panama) to 0.43 (Nicaragua).
  - Multipliers are positive for some poorer economies (for instance, 0.41 for SSA excluding South Africa) but negative in EMEs (-1.03) and AEs (-0.8).

### Tax multipliers (detailed figures)
- Impact multipliers for tax increases in Central America:
  - small and not statistically different from zero.
  - Short-run output effect of a tax increase by group:
    - HIPCs: positive impact (specific impact value not provided in the excerpt);
    - AEs: zero;
    - SSA: zero;
    - LICs: zero;
    - OIL: -0.12;
    - EMEs: -0.10.
- Cumulative (medium-run) effects on output of tax-revenue increases:
  - Positive for Central American countries: range from 0.20 (Dominican Republic) to 0.51 (Guatemala).
  - Positive for HIPCs, LICs, and SSA.
  - Negative for AEs, EMEs, and oil producers (consistent with other literature).
  - Medium-run cumulative multiplier ranges reported:
    - Central America: 0.24 (Nicaragua) and 0.57 (Panama).
    - Other country groups: 0.34 (oil exporter countries) and 0.53 (LIC).

### Capital and current spending multipliers
- Splitting public spending into investment (capital) and current expenditure:
  - Public investment increases produce positive impact multipliers on output (generally higher than that of taxes).
  - Increases in current spending lower output in several countries.
- The multipliers for capital expenditures range from (text cut off in the provided excerpt; specific numeric range not supplied in the excerpt).

*Source: _wp13136 - 2. The interest rate on government debt depends on the fiscal stance and the exchange rate, but (contemporaneously) less on output*

### 0.21 to 0.70 in the short-run and 0.42 to 0.91 in the medium-run whereas the multipliers for

### _wp13136 - 0.21 to 0.70 in the short-run and 0.42 to 0.91 in the medium-run whereas the multipliers for

### Major empirical findings
- Capital expenditure multipliers:
  - Range on impact: 0.21 to 0.70.
  - Range on cumulative/medium-run: 0.42 to 0.91 (also reported as up to 1.00 in some tables).
- Current spending multipliers:
  - Impact range: -0.24 to 0.19.
  - Cumulative impact range: -1.00 to -0.41.
- Point estimates are generally significantly different from zero in the short-run, while medium-run estimates are associated with large uncertainties for both current and capital investments.
- Government current expenditure tends to respond strongly to public investment shocks; this is confirmed when imposing a zero restriction on government current spending to isolate “pure” government investment effects.
- There is no single fiscal multiplier: multipliers are country- or region-specific and model-specific.

### Methodology and model features
- Estimation approach:
  - Modified structural VAR modeling to account for debt feedback effects and to enable the use of annual data.
  - Use of cointegration techniques and error-correction modeling to identify exogenous fiscal shocks and estimate fiscal multipliers when annual frequency precludes quarterly timing assumptions.
  - Inclusion of debt as a ratio to GDP in the estimation, as suggested by Favero and Giavazzi (2007), to capture attenuating effects from changes in the fiscal deficit.
- Identification:
  - Institutional assumption based on Blanchard and Perotti (2002) used for identification in some specifications.
  - Cumulative multipliers obtained as ratios of net present values as defined in Section IV.B.

### Key comparative results (selected exact estimates from tables)
- This paper — Central America (incl. Dominican Republic and Panama), SVECM, Annual:
  - Total expenditure impact: 0.01 to 0.44.
  - Total expenditure cumulative: 0.45 to 0.94.
  - Taxes impact: 0.00 to 0.02.
  - Taxes cumulative: -1.13 to -0.20.
  - Current expenditure impact: -0.19 to -0.04.
  - Current expenditure cumulative: -0.40 to 0.79.
  - Capital expenditure impact: 0.21 to 0.70.
  - Capital expenditure cumulative: 0.42 to 1.00.
- Other regional/comparator multipliers (exact table excerpts):
  - AE (Advanced Economies) total expenditure: 0.43 (impact) and 0.78 (cumulative).
  - EME total expenditure: 0.34 (impact) and 1.04 (cumulative).
  - Oil producing countries total expenditure: 0.31 to 0.44 (impact reported).
  - HIPC total expenditure: 0.01 (impact) and -0.26 (cumulative).
  - LIC total expenditure: 0.17 (impact) and -0.33 (cumulative).
  - SSA total expenditure: 0.25 to 0.41 (impact reported).
- Selected literature benchmarks (exact figures as reported in table):
  - Ilzetzti, Mendoza, and Vegh (2010), 44 countries (SVAR, quarterly, developing): Total expenditure -0.21 to 0.18; Capital expenditure 0.57 to 0.75.
  - Blanchard and Perotti (2002), United States (VAR, quarterly): Spending 0.8 to 0.9 (impact) and 0.7 to 1.0 (cumulative); Tax 0.7 to 1.3.
  - Mountford and Uhlig (2002), United States (VAR, quarterly): Spending 0.2 to 0.5; Tax 0.2 to 0.4.
  - Perotti (2002), United States (VAR, quarterly): Spending 0.1 to 0.4 (impact) and -1.3 to 1.0 (cumulative); Tax 0.2 to 0.4.

### Country- and region-specific policy-relevant findings
- CAPDR (Central America including Dominican Republic and Panama):
  - Results point to the importance of moving the composition of public spending toward investment given capital scarcity in the region.
  - Fiscal consolidation paths should preserve (or even increase) public investment in CAPDR.
- Nicaragua:
  - Presents large negative multipliers associated with increases in current spending both in the short- and in the long-run; controlling current spending is particularly relevant.
  - Nicaragua and Honduras show greater benefits from cutting current spending than from raising tax revenues (as reported).
- Panama and Costa Rica:
  - Panama shows a relatively large capital-expenditure multiplier (Panama: capital expenditure cumulative reported as up to 1.1921 in one band; impact and cumulative bands reported in table).
  - Costa Rica and some mid-income countries may experience positive medium-term effects from fiscal consolidation if based on improved tax system efficiency.
- HIPCs and LICs:
  - Fiscal consolidation tends to have a strong positive effect in low income countries (especially HIPCs).
  - Tax revenue increases appear to yield greater medium-term output growth than current spending control in several Central American countries and LICs, including HIPCs, but public investment multipliers are larger than tax multipliers.

### Policy implications and recommendations (as stated)
- Fiscal consolidation should pay attention to composition:
  - Preserve or increase public investment where capital scarcity warrants (e.g., CAPDR).
  - Control current spending where multipliers are negative or contractionary (notably Nicaragua).
- The persistence and composition of fiscal measures matter:
  - The size and sign of multipliers depend on whether consolidation targets current expenditure, capital expenditure, or taxes, on short- versus long-run horizons, and on monetary policy accommodation.
- Debt dynamics and financial constraints:
  - Including debt-to-GDP feedback matters because declining public debt from consolidation may reduce interest rates and potentially offset part of the initial negative impulse of consolidation.

*Source: IMF working paper content (figures, tables, and text excerpts provided in the supplied PDF content).*

### REFERENCES

### _wp13136 - REFERENCES

### References

- Auerbach, Alan, and Yuriy Gorodnichenko, 2012a, “Fiscal Multipliers in Recession and Expansion,” in Fiscal Policy after the Financial Crisis, edited by Alberto Alesina and Francesco Giavazzi, Chicago: University of Chicago Press.
- ———, 2012b, “Measuring the Output Responses to Fiscal Policy,” American Economic Journal – Economic Policy, Vol. 4, pp. 1–27.
- ———, 2012c, “Output Spillovers from Fiscal Policy,” NBER Working Paper, No. 18578, Cambridge, Massachusetts: National Bureau of Economic Research.
- Batini, Nicoletta, Giovanni Callegari, and Giovanni Melina, 2012, “Successful Austerity in the United States, Europe and Japan,” IMF Working Paper, No. 12/190, Washington: International Monetary Fund.
- Blanchard, Olivier and Roberto Perotti, 2002, “An Empirical Characterization of the Dynamic Effects of Changes in Government Spending and Taxes on Output,” The Quarterly Journal of Economics, 117(4), pages 1329-1368.
- Blanchard, Olivier ; Leigh, Daniel, 2013, “Growth Forecast Errors and Fiscal Multipliers”, IMF Working Paper, 13/1, International Monetary Fund, Washington, DC.
- Baum, Anja, Marcos Poplawski-Ribeiro, and Anke Weber, 2012, “Fiscal Multipliers and the State of the Economy,” IMF Working Paper, No. 12/286, Washington: International Monetary Fund.
- Baunsgaard, T., A. Mineshima, M. Poplawski-Ribeiro, and A. Weber, 2013, “Fiscal Multipliers”, in Post-crisis Fiscal Policy, ed. by C. Cottarelli, P. Gerson, and A. Senhadji, forthcoming, Washington: International Monetary Fund.
- Caldara, Dario; and Kamps, Christophe, 2008, “What Are The Effects of Fiscal Policy Shocks? A VAR-Based Comparative Analysis” Working Paper Series 877, European Central Bank.
- Dungrey, Mardi and Fry, Renée, 2009, “Identification of Fiscal and Monetary Policy in a Structural VAR,” Economic Modeling, 26 (2009) 1147–1160.
- Favero, Carlo and Francesco Giavazzi, 2007, “Debt and the Effects of Fiscal Policy”, NBER Working Paper, 12822.
- –––––, 2010, “Reconciling VAR-based and Narrative Measures of the Tax-Multiplier,” CEPR Discussion Papers, 7769.
- –––––, 2012, “Measuring Tax Multipliers: The Narrative Method in Fiscal VARs” American Economic Journal: Economic Policy, 2012, vol. 4, issue 2, pages 69–94.
- Fatás, Antonio, and I. Mihov, 2001, “The Effects of Fiscal Policy on Consumption and Employment: Theory and Evidence,” CEPR Discussion Paper, 2760, Center for Economic Policy Research, London.
- Feve, Patrick; Matheron, Julien; and Sahuc, Jean-Guillaume, 2012, “A Pitfall with DSGE-Based, Estimated, Government Spending Multipliers,” Institut d’Economie Industrielle (IDEI), Working Paper Series No. 708, April.
- Hemming, Richard, M. Kell, and Selma Mahfouz, 2002, “The Effectiveness of Fiscal Policy in Stimulating Economic Activity: A Review of the Literature” IMF Working Paper, 02/208, International Monetary Fund, Washington, DC.
- Ilzetzki, Ethan, Enrique G. Mendoza and Carlos A. Végh, 2010, “How Big (Small?) Are Fiscal Multipliers?” CEP Discussion Papers, DP1016, Centre for Economic Performance, LSE.
- Ilzetzki, Ethan, 2011, “Fiscal Policy and Debt Dynamics in Developing Countries”, London School of Economics, May.
- International Monetary Fund, 2010, World Economic Outlook, October.
- Kaminsky, Graciela, Carmen Reinhart and Carlos A. Végh, 2004, “When It Rains It Pours: Procyclical Capital Flows and Macroeconomic Policies” in NBER Macroeconomics Annual, edited by Mark Gertler and Kenneth Rogoff, Cambridge, MA: MIT Press.
- Kirchner, Markus, 2011, “Expectations-Based Identification of Government Spending Shocks under Policy Foresight” in Ph.D dissertation: Fiscal policy and the business cycle: the impact of government expenditures, public debt, and sovereign risk on macroeconomic fluctuations, Faculty of Economics and Business, Amsterdam.
- Kraay, Aart, 2010, “How Large is the Government Spending Multiplier? Evidence from World Bank Lending”,. World Bank Policy Research Working Paper No. 5500.
- Mountford, A., and Harald Uhlig, 2005, “What Are the Effects of Fiscal Policy Shocks?” SFB 649 Discussion Paper 2005-039, Humboldt University, Berlin.
- Perotti, Roberto, 2001, “What Do We Know About the Effects of Fiscal Policy?” in M. Bordignon and D. Da Empoli (eds.), Politica .scale, .essibilità dei mercati e crescita. Milano: Angeli.
- –––––, 2004, “Estimating the effects of fiscal policy in OECD countries,” Working Papers 276, IGIER Bocconi University.
- –––––, 2006, “Comparing Alternative Methodologies to Estimate the Effects of Fiscal Policy”, IGIER, Università Bocconi, Milan (mimeo).
- –––––, 2011, “The Effects of Tax Shocks on Output: Not So Large, But Not Small Either”, NBER Working Paper, No. 16786.
- Phillips, Peter.C.B.,1998, “Impulse Response and Forecast Error Variance Asymptotics in Nonstationary VARs,” Journal of Econometrics, 83 (1-2): 21-56.
- Piñón, Marco, Alejandro López-Mejía, Mario Garza, and Fernando L. Delgado (eds.), 2012, Central America, Panama, and the Dominican Republic: Challenges Following the 2008-09 Global Crisis, IMF, Washington, DC.
- Ramey, Valery A., and Mathew D. Shapiro, 1998, “Costly Capital Reallocation and the Effects of Government Spending,” Carnegie-Rochester Conference Series on Public Policy, 48 (June): 145.194.
- Romer, Christina and David H. Romer, 2010, “The Macroeconomic Effects of Tax Changes: Estimates Based on a New Measure of Fiscal Shocks", American Economic Review, 100, 763-801
- –––––, 2007, “The Macroeconomic Effects of Tax Changes: Estimated Based on a New Measure of Fiscal Shocks,” NBER Working Paper, No.132641, Cambridge, MA.
- Reinhart, Carmen M., and Kenneth S. Rogoff, 2010, “Growth in a Time of Debt,” American Economic Review, 100(2).
- Senhadji, Abdelhak S ; Espinoza, Raphael A., 2011, “How Strong are Fiscal Multipliers in the GCC? An Empirical Investigation” Working Paper No. 11/61
- Sims, Christopher A., 1980, “Macroeconomics and Reality,” Econometrica, 48 (1): 1.48.
- –––––, James H. Stock, and Mark W. Watson, 1990, “Inference in Linear Time Series Models with Some Unit Roots,” Econometrica, 58 (1): 113.144.
- –––––, and T. Zha, 1999, “Error Bands for Impulse Responses,” Econometrica, 67 (5): 1113.1155.
- Uhlig, Harald, 2005, “What Are the Effects of Monetary Policy on Output? Results from an Agnostic Identification Procedure,” Journal of Monetary Economics, 52 (2): 381.419.

*Source: _wp13136 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13136.pdf_