## _wp1349 - Section IV outlines some of the basic features and calibration of GIMF, and presents its

## Source details

**Canonical URL:** [_wp1349 - Section IV outlines some of the basic features and calibration of GIMF, and presents its](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp1349.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp1349.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp1349.pdf.json)

---

### Literature review: scope and datasets
- Majority of fiscal multiplier studies focus on advanced economies.
- Baunsgaard and others (2012) review a total of 37 studies including both model based (DSGE) and vector autoregressive (VAR) approaches.
- Baunsgaard and others (2012) report:
  - Government spending multipliers range between 0 and 2.0, with a mean of 0.8 during the first year after fiscal measures are taken.
  - Government revenue multipliers range from about –1.5 to 1.4, with a mean of 0.3.
- Coenen and others (2012) compare 7 models used at policy making institutions, finding that most models (6 are DSGE models, including GIMF) produce similar short-run multipliers for temporary changes:
  - Government spending: roughly 0.6 to 1.5 under the normal conduct of monetary policy.
  - Revenues: roughly 0.1 to 0.5 under the normal conduct of monetary policy.
- Sources used in empirical comparisons include IMF Fiscal Affairs Department Fiscal Rules database and Fiscal Transparency database; Organization for Economic Cooperation and Development (OECD); and IMF staff estimates.
- Note on empirical measures:
  - Openness is measured by import penetration, the 2008–11 average of Imports/(GDP – Exports + Imports)*100.
  - Automatic stabilizers are measured as the semielasticity of the budget balance and are taken from André and Girouard (2005).
  - Negative correlations reported are robust to outliers removed using an automated Stata procedure based on leverage and residual.

### Cross-country patterns and state dependence
- General empirical patterns:
  - Fiscal multipliers tend to be smaller in more open economies and in countries with larger automatic stabilizers.
  - Spilimbergo and others (2009) suggest a rule of thumb for advanced economies:
    - Government consumption multipliers are 0.5 or less in small open economies.
    - Smaller values for revenue and transfers.
    - Slightly larger multipliers for investment.
  - Recent studies indicate multipliers are significantly larger during recessions than during expansions (Auerbach and Gorodnichenko, 2012; Batini and others, 2012; Baum and others, 2012).

### Evidence for emerging economies
- Fewer papers estimate multipliers in emerging economies; prevailing hypotheses and findings include:
  - Fiscal multipliers may be lower because financial markets are less developed, sovereign risk premia are higher, and fiscal stimulus can raise interest rates, partly offsetting the impulse.
  - Panel-data empirical studies tend to validate lower multipliers in emerging economies (IMF, 2008; Mendoza and others, 2011; Ilzetzki, 2011).
  - Some studies find multipliers generally negative, particularly in the longer term (IMF, 2008) and when public debt is high (Ghosh and Rahman, 2008).
  - Revenue-based stimulus measures may be more effective at boosting output in the short term than expenditure-based measures, in contrast to advanced economies.
  - Ilzetzki and others (2011) and Ilzetzki (2011) find:
    - Spending multipliers are very small and not generally significant.
    - Revenue multipliers are positive and significant and lie around 0.3 in the short term.
  - Mendoza and others (2010) suggest:
    - Low overall spending multiplier in emerging economies could reflect a negative government consumption multiplier combined with a positive response of output to government investment with a multiplier of around 0.6 in the short-term.

### Model-based approaches and ranking of fiscal instruments
- DSGE and other model-based approaches facilitate instrument-level analysis without data constraints.
- Established ranking for short-term negative growth impact:
  - Spending instruments (from most to least negative short-term impact):
    - Investment has the highest negative short-term multiplier.
    - Government consumption (wages and government purchases).
    - Transfers to liquidity-constrained households.
    - Transfers to households associated with the lowest output impact among spending instruments.
  - Revenue instruments (ranking reflects distortionary effects):
    - Corporate income taxes and personal income taxes have the most negative effects on GDP.
    - Consumption taxes perform relatively better.
    - Property taxes appear to be the most growth-friendly instrument (OECD 2009, OECD 2010, and EC 2010).

### Calibration context and related GIMF work
- Klyuev and Snudden (2011) calibrate the IMF’s GIMF to assess fiscal consolidation in the Czech Republic.
  - That study finds fiscal multipliers are quite small, ranging from virtually zero to 0.5 in the first year. However, the effect of fiscal consolidation on output depends on the fiscal...

### Bulgaria — empirical VAR approach (Section III)
- Data and methodology:
  - Structural VAR (Blanchard and Perotti, 2002) with three variables: real GDP, real net revenue, real net expenditure.
  - Revenue defined as total revenues excluding transfers and subsides and interest payments. Expenditure defined as government consumption and investment.
  - Quarterly accrual-based fiscal data: 1999–2011, GDP deflator used to deflate series.
  - Monthly cash-based data: 2003–mid-2012, industrial production used as proxy for GDP.
  - Also estimate model with data prior to 2007 to account for EU accession structural break.
  - Threshold VAR used to assess regime-dependent multipliers; output gap chosen as threshold variable. Tsay (1998) multivariate threshold approach employed.
  - Discretionary fiscal policy shocks identified using exogenous revenue and expenditure elasticities to account for automatic stabilizers:
    - Elasticities used: direct taxes 1.1, indirect taxes 1, social security contributions 0.7, social benefits -0.1.
  - Caveat: structural identification filters cyclical movements in revenues but not changes due to asset price cycles.
- Empirical results — VAR estimates (historical multipliers in Bulgaria, Table 1):
  - Quarterly, 1999–2011:
    - Revenue 0.33
    - Spending 0.04
  - Monthly, 2003–011:
    - Revenue 0.32*
    - Spending 0.31
  - Monthly, 2003–2006:
    - Revenue 0.42*
    - Spending 0.16*
  - Notes: * denotes significance on impact according to a 95 percent confidence interval.
- Key findings from VAR:
  - First year spending multipliers around zero and first year revenue multipliers about 0.3 using quarterly accrual-based data (1999–2011); statistically insignificant in that specification.
  - Using monthly data over whole sample: revenue impact significant with an increase in tax collections decreasing economic activity; expenditure multipliers remain insignificant.
  - Pre-2007 (pre-EU accession) sample: both revenue and expenditure multipliers significant and positive with first-year multipliers of 0.4 (revenue) and 0.2 (spending).
- State-dependent multipliers (threshold at output gap = -1.73 percent; regimes: downturn and expansion):
  - In a downturn:
    - Revenue first-year multiplier 0.5
    - Expenditure first-year multiplier 0.3
  - In an expansion:
    - Revenue first-year multiplier 0.4
    - Expenditure first-year multiplier 0.2
- Empirical caveats:
  - Cash-based data may reflect fiscal actions with a lag.
  - Estimated multipliers are overall multipliers, not instrument-specific (spending multiplier averages productive and unproductive spending effects).
  - Results are averages over time; structural changes in Bulgaria suggest a model-based approach is complementary.

### GIMF model and calibration for Bulgaria (Section IV)
- Model features:
  - GIMF: multi-country DSGE model with optimizing households/firms, full inter-temporal stock-flow accounting, sticky prices/wages, real adjustment costs, liquidity-constrained households, finite planning horizons.
  - Non-Ricardian features:
    i) OLG households value government debt as wealth and consume out of wealth.
    ii) Liquidity constrained households consume only post-tax labor income and transfers.
    iii) Multiple distortionary taxes: labor income, corporate income, consumption (VAT).
- Fiscal instruments modeled: government consumption, government investment, lump-sum transfers (general and targeted), taxes on labor, corporate, consumption, lump-sum taxes. Government investment augments public infrastructure with constant depreciation.
- Calibration details:
  - 3-region model: Bulgaria, euro area, rest of world.
  - Data sources: national account ratios, tax revenue components, lump-sum transfers for 2011 (WEO), trade decomposition (WEO, direction of trade, U.N. COMTRADE).
  - Debt-to-GDP ratio set to 16 percent.
  - Share of liquidity-constrained households set to 50 percent.
  - Bulgaria faces nominal rigidities in line with rest of world; euro area nominal rigidities 50 percent higher.
- Consolidation experiment:
  - Permanent 1 percent of GDP decrease in the budget balance, financed by one of seven fiscal instruments.
  - On impact, the fiscal instrument changes by 1 percent of GDP.
  - Lower government debt reduces interest payments; government uses additional fiscal space to reverse as much as possible the change in the fiscal instrument.
  - Assume consolidation is credible in baseline; non-credible scenarios also examined.

### GIMF results — instrumental multipliers (Table 2)
- Fiscal Multipliers based on 1 Percent of GDP Permanent Change in the Budget Balance (Percent Deviation from Baseline):
  - Government Investment:
    - Year 1 0.61
    - Year 2 0.55
    - Year 3 0.49
    - Year 4 0.54
    - Year 5 0.71
  - Government Consumption:
    - Year 1 0.51
    - Year 2 0.33
    - Year 3 0.13
    - Year 4 0.02
    - Year 5 0.01
  - Targeted Lumpsum Transfers:
    - Year 1 0.45
    - Year 2 0.23
    - Year 3 -0.07
    - Year 4 -0.28
    - Year 5 -0.35
  - General Lumpsum Transfers:
    - Year 1 0.30
    - Year 2 0.13
    - Year 3 -0.12
    - Year 4 -0.28
    - Year 5 -0.31
  - Corporate Income Taxes:
    - Year 1 0.53
    - Year 2 0.56
    - Year 3 0.51
    - Year 4 0.58
    - Year 5 0.80
  - Labor Income Taxes:
    - Year 1 0.40
    - Year 2 0.44
    - Year 3 0.42
    - Year 4 0.41
    - Year 5 0.43
  - Consumption Taxes (VAT):
    - Year 1 0.38
    - Year 2 0.32
    - Year 3 0.17
    - Year 4 0.07
    - Year 5 0.05
- Interpretations:
  - Among spending instruments, government investment has the highest multiplier (largest impact on productivity and GDP), followed by government consumption, then transfers (targeted transfers larger short-run effect than general lump-sum transfers).
  - Among revenue instruments, corporate income taxes have the most negative long-run effect on GDP (largest distortionary effect on investment), followed by labor income taxes, with consumption taxes (VAT) having the lowest impact on GDP.
  - Even with permanent consolidation in debt, change in fiscal instrument is temporary because of evolving government debt dynamics in the baseline.
  - Credibility matters: non-credible consolidations can produce different short-run and cumulative costs.

### Credibility scenarios — government consumption consolidation (Table 3)
- Fiscal Multipliers based on 1 Percent of GDP Permanent Change in the Budget Balance, Considering Issues of Credibility (Percent Deviation from Baseline):
  - Government Consumption — Immediately Credible:
    - Year 1 0.51
    - Year 2 0.33
    - Year 3 0.13
    - Year 4 0.02
    - Year 5 0.01
  - Credible in Year 2:
    - Year 1 0.48
    - Year 2 0.42
    - Year 3 0.25
    - Year 4 0.10
    - Year 5 0.04
  - Credible in Year 3:
    - Year 1 0.48
    - Year 2 0.38
    - Year 3 0.34
    - Year 4 0.20
    - Year 5 0.10
  - Credible in Year 4:
    - Year 1 0.48
    - Year 2 0.38
    - Year 3 0.29
    - Year 4 0.29
    - Year 5 0.21
- Key insight:
  - If consolidation is initially non-credible (perceived as temporary) but later becomes credible, the eventual costs can be higher than for an always-credible consolidation because agents must adjust behavior more abruptly when credibility is established.
  - For Bulgaria, government overperformance on fiscal targets in 2011 and 2012 makes prolonged non-credibility less likely.

### Growth-friendly rebalancing — scenario analysis (Section V)
- Scenario setup:
  - With adjustment to balanced budget by 2015 largely completed, priority is making budget composition more growth friendly.
  - GIMF scenario: permanent 2 percent of current GDP increase in public investment, with 85 percent financed by EU grants and 15 percent co-payment.
  - Two financing variants for the 15 percent co-payment:
    (i) Co-payment financed in a budget neutral way by broadening the tax base.
    (ii) Co-payment financed by a higher deficit.
- Scenario results and mechanisms:
  - Long-run effects of a permanent 2 percent of nominal GDP increase in government investment:
    - Long run: 30 percentage point increase in stock of public infrastructure.
    - Real GDP increases about 3 percent relative to baseline in the long run.
  - Channels:
    - Higher public infrastructure raises productivity of factors of production.
    - Private investment begins rising immediately.
    - Consumption increases as higher productive capacity lowers prices.
    - Under currency board: nominal interest rates unchanged, real interest rates eventually fall, providing stimulus.
    - Short-run: greater domestic inflationary pressures, real exchange rate appreciation.
    - Long-run: significant real depreciation (~3 percent) as productivity increase makes Bulgarian goods cheaper.
  - Financing comparison:
    - Budget neutral (broadening indirect tax base) vs deficit financing:
      - Short-run: consumption lower under broader tax base due to VAT impact.
      - Medium term: deficit financing leads to higher government debt demand in local markets, crowding out private investment and larger current account deficit.
      - Broadening tax base avoids downward pressure on investment from higher debt, enabling stronger expansion of productive capacity, higher wage income, household wealth, and consumer spending (despite higher VAT drag).
      - Broad tax base also leads to greater real exchange rate depreciation due to less foreign funding demand for government debt.

### Conclusions and policy implications (Section VI)
- This is the first attempt to estimate fiscal multipliers for Bulgaria, using both an econometric VAR approach and calibration of IMF’s GIMF.
- Overall findings:
  - As a small open economy, Bulgaria experienced modest historical fiscal multipliers.
  - Fiscal multipliers are state-dependent: larger in downturns than in expansions (consistent with higher share of credit-constrained agents during negative output gaps).
  - GIMF establishes a clear ranking of instruments by growth impact:
    - Spending side (largest to smallest multiplier): capital spending (government investment), government consumption, transfers.
    - Revenue side (largest to smallest impact on output): corporate income taxes, labor income taxes, consumption taxes.
  - Credibility of consolidation materially affects multiplier magnitude and timing; non-credible consolidations can produce higher eventual costs once credibility is achieved.
- Policy implications and recommendations:
  - Make budget composition more growth friendly by prioritizing higher capital spending and addressing underperformance on capital spending.
  - Optimal adjustment favoring higher public investment financed by EU funds and, for required co-payments, prefer budget-neutral financing via tax base broadening rather than deficit financing.
  - Specific steps to support growth-friendly budget:
    - Address barriers to EU funds absorption to increase capital spending on infrastructure.
    - Tackle the large shadow economy to broaden the tax base and create resources for productive infrastructure spending.
    - Review subsidies and non-targeted government transfers (general lump-sum transfers) to make room for growth-enhancing expenditure; note subsidies of SOEs have been growing and are associated with small multipliers.
    - Further pension reforms to increase length of working lives to raise potential growth; such reforms reduce permanent pension spending and have relatively small short-term multipliers while helping contain projected spending increases.
  - Overall recommendation: avoid raising government consumption; pursue higher capital spending financed by broadening indirect tax revenue collection through base broadening to achieve sizeable medium- and long-term growth gains.

*Source: _wp1349 - Section IV outlines some of the basic features and calibration of GIMF, and presents its*

### Section IV outlines some of the basic features and calibration of GIMF, and presents its

### _wp1349 - Section IV outlines some of the basic features and calibration of GIMF, and presents its

### Literature review: scope and datasets
- Majority of fiscal multiplier studies focus on advanced economies.
- Baunsgaard and others (2012) review a total of 37 studies including both model based (DSGE) and vector autoregressive (VAR) approaches.
- Baunsgaard and others (2012) report:
  - Government spending multipliers range between 0 and 2.0, with a mean of 0.8 during the first year after fiscal measures are taken.
  - Government revenue multipliers range from about –1.5 to 1.4, with a mean of 0.3.
- Coenen and others (2012) compare 7 models used at policy making institutions, finding that most models (6 are DSGE models, including GIMF) produce similar short-run multipliers for temporary changes:
  - Government spending: roughly 0.6 to 1.5 under the normal conduct of monetary policy.
  - Revenues: roughly 0.1 to 0.5 under the normal conduct of monetary policy.
- Sources used in empirical comparisons include IMF Fiscal Affairs Department Fiscal Rules database and Fiscal Transparency database; Organization for Economic Cooperation and Development (OECD); and IMF staff estimates.
- Note on empirical measures:
  - Openness is measured by import penetration, the 2008–11 average of Imports/(GDP – Exports + Imports)*100.
  - Automatic stabilizers are measured as the semielasticity of the budget balance and are taken from André and Girouard (2005).
  - Negative correlations reported are robust to outliers removed using an automated Stata procedure based on leverage and residual.

### Cross-country patterns and state dependence
- General empirical patterns:
  - Fiscal multipliers tend to be smaller in more open economies and in countries with larger automatic stabilizers.
  - Spilimbergo and others (2009) suggest a rule of thumb for advanced economies:
    - Government consumption multipliers are 0.5 or less in small open economies.
    - Smaller values for revenue and transfers.
    - Slightly larger multipliers for investment.
  - Recent studies indicate multipliers are significantly larger during recessions than during expansions (Auerbach and Gorodnichenko, 2012; Batini and others, 2012; Baum and others, 2012).

### Evidence for emerging economies
- Fewer papers estimate multipliers in emerging economies; prevailing hypotheses and findings include:
  - Fiscal multipliers may be lower because financial markets are less developed, sovereign risk premia are higher, and fiscal stimulus can raise interest rates, partly offsetting the impulse.
  - Panel-data empirical studies tend to validate lower multipliers in emerging economies (IMF, 2008; Mendoza and others, 2011; Ilzetzki, 2011).
  - Some studies find multipliers generally negative, particularly in the longer term (IMF, 2008) and when public debt is high (Ghosh and Rahman, 2008).
  - Revenue-based stimulus measures may be more effective at boosting output in the short term than expenditure-based measures, in contrast to advanced economies.
  - Ilzetzki and others (2011) and Ilzetzki (2011) find:
    - Spending multipliers are very small and not generally significant.
    - Revenue multipliers are positive and significant and lie around 0.3 in the short term.
  - Mendoza and others (2010) suggest:
    - Low overall spending multiplier in emerging economies could reflect a negative government consumption multiplier combined with a positive response of output to government investment with a multiplier of around 0.6 in the short-term.

### Model-based approaches and ranking of fiscal instruments
- DSGE and other model-based approaches facilitate instrument-level analysis without data constraints.
- Established ranking for short-term negative growth impact:
  - Spending instruments (from most to least negative short-term impact):
    - Investment has the highest negative short-term multiplier.
    - Government consumption (wages and government purchases).
    - Transfers to liquidity-constrained households.
    - Transfers to households associated with the lowest output impact among spending instruments.
  - Revenue instruments (ranking reflects distortionary effects):
    - Corporate income taxes and personal income taxes have the most negative effects on GDP.
    - Consumption taxes perform relatively better.
    - Property taxes appear to be the most growth-friendly instrument (OECD 2009, OECD 2010, and EC 2010).

### Calibration context and related GIMF work
- Klyuev and Snudden (2011) calibrate the IMF’s GIMF to assess fiscal consolidation in the Czech Republic.
  - That study finds fiscal multipliers are quite small, ranging from virtually zero to [text truncated in source].

*Source: _wp1349 - Section IV outlines some of the basic features and calibration of GIMF, and presents its*

### 0.5 in the first year. However, the effect of fiscal consolidation on output depends on the fiscal

### III. FISCAL MULTIPLIERS IN BULGARIA—AN EMPIRICAL APPROACH

### Data and methodology
- Structural VAR (Blanchard and Perotti, 2002) with three variables: real GDP, real net revenue, real net expenditure.
- Revenue defined as total revenues excluding transfers and subsides and interest payments. Expenditure defined as government consumption and investment.
- Quarterly accrual-based fiscal data: 1999–2011, GDP deflator used to deflate series.
- Monthly cash-based data: 2003–mid-2012, industrial production used as proxy for GDP.
- Also estimate model with data prior to 2007 to account for EU accession structural break.
- Threshold VAR used to assess regime-dependent multipliers; output gap chosen as threshold variable. Tsay (1998) multivariate threshold approach employed.
- Discretionary fiscal policy shocks identified using exogenous revenue and expenditure elasticities to account for automatic stabilizers:
  - Elasticities used: direct taxes 1.1, indirect taxes 1, social security contributions 0.7, social benefits -0.1.
- Caveat: structural identification filters cyclical movements in revenues but not changes due to asset price cycles.

### Empirical results — VAR estimates
- Historical multipliers in Bulgaria (Table 1):
  - Frequency: Quarterly, Time period: 1999–2011
    - Revenue 0.33
    - Spending 0.04
  - Frequency: Monthly, Time period: 2003–011
    - Revenue 0.32*
    - Spending 0.31
  - Frequency: Monthly, Time period: 2003–2006
    - Revenue 0.42*
    - Spending 0.16*
  - Notes: * denotes significance on impact according to a 95 percent confidence interval.
- Key findings:
  - First year spending multipliers around zero and first year revenue multipliers about 0.3 using quarterly accrual-based data (1999–2011); statistically insignificant in that specification.
  - Using monthly data over whole sample: revenue impact significant with an increase in tax collections decreasing economic activity; expenditure multipliers remain insignificant.
  - Pre-2007 (pre-EU accession) sample: both revenue and expenditure multipliers significant and positive with first-year multipliers of 0.4 (revenue) and 0.2 (spending).
- State-dependent multipliers (threshold at output gap = -1.73 percent; regimes: downturn and expansion):
  - In a downturn:
    - Revenue first-year multiplier 0.5
    - Expenditure first-year multiplier 0.3
  - In an expansion:
    - Revenue first-year multiplier 0.4
    - Expenditure first-year multiplier 0.2
- Empirical caveats:
  - Cash-based data may reflect fiscal actions with a lag.
  - Estimated multipliers are overall multipliers, not instrument-specific (spending multiplier averages productive and unproductive spending effects).
  - Results are averages over time; structural changes in Bulgaria suggest a model-based approach is complementary.

---

### IV. FISCAL MULTIPLIERS IN BULGARIA—USING THE IMF’S GIMF

### The GIMF model and calibration
- GIMF: multi-country DSGE model with optimizing households/firms, full inter-temporal stock-flow accounting, sticky prices/wages, real adjustment costs, liquidity-constrained households, finite planning horizons.
- Non-Ricardian features emphasized:
  i) OLG households value government debt as wealth and consume out of wealth.
  ii) Liquidity constrained households consume only post-tax labor income and transfers.
  iii) Multiple distortionary taxes: labor income, corporate income, consumption (VAT).
- Fiscal instruments modeled: government consumption, government investment, lump-sum transfers (general and targeted), taxes on labor, corporate, consumption, lump-sum taxes. Government investment augments public infrastructure with constant depreciation.
- Calibration:
  - 3-region model: Bulgaria, euro area, rest of world.
  - Data sources: national account ratios, tax revenue components, lump-sum transfers for 2011 (WEO), trade decomposition (WEO, direction of trade, U.N. COMTRADE).
  - Debt-to-GDP ratio set to 16 percent.
  - Share of liquidity-constrained households set to 50 percent.
  - Bulgaria faces nominal rigidities in line with rest of world; euro area nominal rigidities 50 percent higher.
- Consolidation experiment:
  - Permanent 1 percent of GDP decrease in the budget balance, financed by one of seven fiscal instruments.
  - On impact, the fiscal instrument changes by 1 percent of GDP.
  - Lower government debt reduces interest payments; government uses additional fiscal space to reverse as much as possible the change in the fiscal instrument.
  - Assume consolidation is credible in baseline; non-credible scenarios also examined.

### GIMF results — instrumental multipliers (Table 2)
- Fiscal Multipliers based on 1 Percent of GDP Permanent Change in the Budget Balance (Percent Deviation from Baseline):
  - Government Investment:
    - Year 1 0.61
    - Year 2 0.55
    - Year 3 0.49
    - Year 4 0.54
    - Year 5 0.71
  - Government Consumption:
    - Year 1 0.51
    - Year 2 0.33
    - Year 3 0.13
    - Year 4 0.02
    - Year 5 0.01
  - Targeted Lumpsum Transfers:
    - Year 1 0.45
    - Year 2 0.23
    - Year 3 -0.07
    - Year 4 -0.28
    - Year 5 -0.35
  - General Lumpsum Transfers:
    - Year 1 0.30
    - Year 2 0.13
    - Year 3 -0.12
    - Year 4 -0.28
    - Year 5 -0.31
  - Corporate Income Taxes:
    - Year 1 0.53
    - Year 2 0.56
    - Year 3 0.51
    - Year 4 0.58
    - Year 5 0.80
  - Labor Income Taxes:
    - Year 1 0.40
    - Year 2 0.44
    - Year 3 0.42
    - Year 4 0.41
    - Year 5 0.43
  - Consumption Taxes (VAT):
    - Year 1 0.38
    - Year 2 0.32
    - Year 3 0.17
    - Year 4 0.07
    - Year 5 0.05
- Interpretations:
  - Among spending instruments, government investment has the highest multiplier (largest impact on productivity and GDP), followed by government consumption, then transfers (targeted transfers larger short-run effect than general lump-sum transfers).
  - Among revenue instruments, corporate income taxes have the most negative long-run effect on GDP (largest distortionary effect on investment), followed by labor income taxes, with consumption taxes (VAT) having the lowest impact on GDP.
  - Even with permanent consolidation in debt, change in fiscal instrument is temporary because of evolving government debt dynamics in the baseline.
  - Credibility matters: non-credible consolidations can produce different short-run and cumulative costs.

### Credibility scenarios — government consumption consolidation (Table 3)
- Fiscal Multipliers based on 1 Percent of GDP Permanent Change in the Budget Balance, Considering Issues of Credibility (Percent Deviation from Baseline):
  - Government Consumption — Immediately Credible:
    - Year 1 0.51
    - Year 2 0.33
    - Year 3 0.13
    - Year 4 0.02
    - Year 5 0.01
  - Credible in Year 2:
    - Year 1 0.48
    - Year 2 0.42
    - Year 3 0.25
    - Year 4 0.10
    - Year 5 0.04
  - Credible in Year 3:
    - Year 1 0.48
    - Year 2 0.38
    - Year 3 0.34
    - Year 4 0.20
    - Year 5 0.10
  - Credible in Year 4:
    - Year 1 0.48
    - Year 2 0.38
    - Year 3 0.29
    - Year 4 0.29
    - Year 5 0.21
- Key insight:
  - If consolidation is initially non-credible (perceived as temporary) but later becomes credible, the eventual costs can be higher than for an always-credible consolidation because agents must adjust behavior more abruptly when credibility is established.
  - For Bulgaria, government overperformance on fiscal targets in 2011 and 2012 makes prolonged non-credibility less likely.

---

### V. BUDGET STRUCTURE TO SUPPORT GROWTH IN BULGARIA — SCENARIO ANALYSIS

### Growth-friendly rebalancing and scenario setup
- With adjustment to balanced budget by 2015 largely completed, priority is making budget composition more growth friendly.
- GIMF scenario: permanent 2 percent of current GDP increase in public investment, with 85 percent financed by EU grants and 15 percent co-payment.
- Two financing variants for the 15 percent co-payment:
  (i) Co-payment financed in a budget neutral way by broadening the tax base.
  (ii) Co-payment financed by a higher deficit.

### Scenario results and mechanisms
- Long-run effects of a permanent 2 percent of nominal GDP increase in government investment:
  - Long run: 30 percentage point increase in stock of public infrastructure.
  - Real GDP increases about 3 percent relative to baseline in the long run.
- Channels:
  - Higher public infrastructure raises productivity of factors of production.
  - Private investment begins rising immediately.
  - Consumption increases as higher productive capacity lowers prices.
  - Under currency board: nominal interest rates unchanged, real interest rates eventually fall, providing stimulus.
  - Short-run: greater domestic inflationary pressures, real exchange rate appreciation.
  - Long-run: significant real depreciation (~3 percent) as productivity increase makes Bulgarian goods cheaper.
- Financing comparison:
  - Budget neutral (broadening indirect tax base) vs deficit financing:
    - Short-run: consumption lower under broader tax base due to VAT impact.
    - Medium term: deficit financing leads to higher government debt demand in local markets, crowding out private investment and larger current account deficit.
    - Broadening tax base avoids downward pressure on investment from higher debt, enabling stronger expansion of productive capacity, higher wage income, household wealth, and consumer spending (despite higher VAT drag).
    - Broad tax base also leads to greater real exchange rate depreciation due to less foreign funding demand for government debt.

---

### VI. CONCLUSIONS AND POLICY IMPLICATIONS

- This is the first attempt to estimate fiscal multipliers for Bulgaria, using both an econometric VAR approach and calibration of IMF’s GIMF.
- Overall findings:
  - As a small open economy, Bulgaria experienced modest historical fiscal multipliers.
  - Fiscal multipliers are state-dependent: larger in downturns than in expansions (consistent with higher share of credit-constrained agents during negative output gaps).
  - GIMF establishes a clear ranking of instruments by growth impact:
    - Spending side (largest to smallest multiplier): capital spending (government investment), government consumption, transfers.
    - Revenue side (largest to smallest impact on output): corporate income taxes, labor income taxes, consumption taxes.
  - Credibility of consolidation materially affects multiplier magnitude and timing; non-credible consolidations can produce higher eventual costs once credibility is achieved.
- Policy implications and recommendations:
  - Make budget composition more growth friendly by prioritizing higher capital spending and addressing underperformance on capital spending.
  - Optimal adjustment favoring higher public investment financed by EU funds and, for required co-payments, prefer budget-neutral financing via tax base broadening rather than deficit financing.
  - Specific steps to support growth-friendly budget:
    - Address barriers to EU funds absorption to increase capital spending on infrastructure.
    - Tackle the large shadow economy to broaden the tax base and create resources for productive infrastructure spending.
    - Review subsidies and non-targeted government transfers (general lump-sum transfers) to make room for growth-enhancing expenditure; note subsidies of SOEs have been growing and are associated with small multipliers.
    - Further pension reforms to increase length of working lives to raise potential growth; such reforms reduce permanent pension spending and have relatively small short-term multipliers while helping contain projected spending increases.
- Overall recommendation: avoid raising government consumption; pursue higher capital spending financed by broadening indirect tax revenue collection through base broadening to achieve sizeable medium- and long-term growth gains.

*Source: IMF staff estimates and analysis as presented in the provided content.*

### chapter III of Public Finances in EMU,

### chapter III of Public Finances in EMU

### References on Fiscal Multipliers and Fiscal Policy
- Ilzetzki, E., 2011, “Fiscal Policy and Debt Dynamics in Developing Countries,” Policy Research Working Paper Series 5666 (Washington: The World Bank).
- Ilzetzki E., E. G. Mendoza and C. A. Vegh, 2011. "How Big (Small?) Are Fiscal Multipliers?" IMF Working Papers 11/52(Washington: International Monetary Fund).
- Spilimbergo, A., S. Symansky, and M. Schindler, 2009, “Fiscal Multipliers,” IMF Staff Position Note, SPN/09/11, May (Washington: International Monetary Fund).
- International Monetary Fund, 2010, World Economic Outlook, October (Washington: International Monetary Fund).

### References on Pension Reform, Fiscal Consolidation, and Country Case Studies
- Karam, P.D., D. Muir, D., J. Pereira, and A. Tuladhar, 2010, “Macroeconomic Effects of Public Pension Reforms,” IMF Working Paper 10/297 (Washington: International Monetary Fund).
- Klyuev, V. and S. Snudden, 2011, “Effects of Fiscal Consolidation in the Czech Republic,” IMF Working Paper 11/65 (Washington: International Monetary Fund).
- ———, 2012. Bulgaria: 2012 Article IV Consulation – Country Report No. 12/328 (Washington: International Monetary Fund).

### References on Models, Indicators, and OECD Analyses
- Kumhof, M., D. Laxton, D. Muir and S. Mursula, 2010, “The Global Integrated Monetary and Fiscal Model: Theoretical Structure”, IMF Working Paper No. 10/34 (Washington: International Monetary Fund).
- OECD, 2002, “An Updated of the OECD Composite Leading Indicators”, http://www.oecd.org/dataoecd/6/2/2410332.pdf.
- OECD, 2006, “Composite Leading Indicators for Major OECD Non-Member Countries: Brazil, China, India, Indonesia, Russian Federation, South Africa”, Statistics Working Paper N. 16 - 2006/1.
- OECD, 2009, “The Effectiveness and Scope of Fiscal Stimulus,” OECD Economic Outlook Interim Report (Paris: Organization for Economic Cooperation and Development).
- OECD, 2010, “Fiscal Consolidation: Requirements, Timing, Instruments and Institutional Arrangements” OECD Economic Outlook 2010/2 (Paris: Organization for Economic Cooperation and Development).

*Source: chapter III of Public Finances in EMU (excerpt of references).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp1349.pdf_