## wpiea2023260-print-pdf

## Source details

**Canonical URL:** [wpiea2023260-print-pdf](https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023260-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2023/english/wpiea2023260-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2023/english/wpiea2023260-print-pdf.pdf.json)

---

### Overview of the policy challenge
- Global headline inflation reached 8.7 percent in 2022.
- Core inflation has risen significantly above central bank targets in many countries and remains stubbornly high.
- Pandemic-era monetary and fiscal expansion, supply chain disruptions, and energy shocks from Russia’s invasion of Ukraine contributed to the inflation surge.
- Fiscal support during the pandemic and the recent energy crisis produced record government debt levels in many economies.
- Higher-for-longer policy rates raise debt service costs, lower economic activity and tax revenues, and put further upward pressure on government debt.

### Analytical approach and models used
- Stylized closed-economy New Keynesian model: variant of Woodford (2003)/Erceg and Lindé (2014) to assess whether fiscal policy can achieve the same inflation-reducing effect as monetary policy.
- Workhorse open-economy TANK DSGE model: two-agent open-economy framework for small open advanced and emerging market economies, including financial frictions in capital accumulation, liquidity-constrained households, endogenous UIP risk-premiums, and real rigidities.
- Rationale: closed-economy framework is a starting point for global tandem policy moves; TANK model provides greater empirical realism and cross-country interaction channels.

### Key findings — stylized closed-economy model
- Fiscal tightening can reduce inflation as much as monetary policy for a given decline in the output gap because both raise the wedge between the actual real rate and the potential real interest rate, lowering domestic demand and inflation via a Phillips curve.
- Central bank response matters:
  - If inflation is at target prior to fiscal intervention and the central bank does not “welcome” a tighter stance, fiscal tightening can be offset by a more stimulative monetary policy.
- Fiscal tightening may be less efficacious at reducing inflation for a given decline in the level of output because:
  - Monetary tightening does not affect potential output.
  - Fiscal tightening that triggers the same-sized output gap decline is associated with a larger decline in potential output, implying a smaller inflation reduction for the same level-output drop.
- Inflation-output elasticities in the stylized model:
  - The model delivers the same inflation-output gap elasticity for monetary and fiscal policy.
  - Measured as inflation response per 1 percent reduction in average output over a given horizon, fiscal policy displays notably lower inflation-output elasticity than monetary policy (example result: elasticity twice as high for monetary policy compared with fiscal policy in the canonical NK calibration).

### Key findings — workhorse open-economy TANK model
- Advanced small open economies with flexible exchange rates:
  - Monetary tightening leads to domestic currency appreciation, which lowers inflation in addition to the direct demand effect.
  - Fiscal tightening strengthens the external position and reduces the domestic policy rate differential vis-à-vis the rest of the world, weakening the exchange rate and reducing fiscal tightening’s ability to fight inflation.
- Emerging market economies (EMEs) with shallow FX markets:
  - Fiscal tightening can improve the net foreign asset position and lead to large declines in the UIP risk premium, producing currency appreciation.
  - Fiscal tightening in vulnerable EMEs and LICs can be as effective in battling inflation as monetary policy.
- When many countries tighten simultaneously:
  - Exchange rate channel is muted (exchange rates do not move much), removing monetary policy’s exchange-rate edge even in advanced economies.
  - Both monetary and fiscal tightening are effective because weaker global demand helps lower inflation.
  - A smaller monetary and fiscal tightening is needed to curb inflation relative to a one-country tightening scenario.
- TANK model elasticities and drivers:
  - TANK implies similar inflation-output elasticity for monetary and fiscal policy (contrast to stylized model).
  - Frictions that boost fiscal policy’s relative effect: financial accelerator, hand-to-mouth households, endogenous capital accumulation, habit formation for optimizing households.
  - These frictions amplify fiscal policy’s effect on inflation relative to output more than monetary policy.

### Spillovers and cross-border effects
- Spillovers from disinflations in large economies to EMDEs can be more severe from monetary policy tightening in the large economies compared with a policy package that includes both tighter monetary and fiscal policy.
- Mechanisms for recipient small open economies:
  - Tighter monetary policy in major economies appreciates the major currency, allowing recipient small open economies’ currencies to depreciate and cushion output via larger net exports.
  - Tighter fiscal policy in major economies reduces foreign demand more and thus reduces inflation in recipient economies more, increasing pressure on foreign central banks to hike.
- For EMDEs with shallow currency markets and large exchange rate pass-through:
  - Adverse spillovers on output can be notably severe, especially if recipient economies must defend their currency; financial conditions tighten more under major-economy monetary tightening.

### Calibration, empirical alignment, and validation targets
- TANK parameterization aims to match US and euro area empirical evidence of monetary and fiscal impacts on inflation and output.
- Empirical calibration targets cited:
  - IMF (2023): a 1 percent of GDP increase in public spending leads to roughly 0.5 percent higher CPI for a group of OECD countries and for the US.
  - Coenen et al. (2012): a temporary fiscal stimulus of 1 percent of GDP in one country raises inflation by 0.1 to 0.3 percentage point depending on measure and monetary accommodation.

### Policy-relevant quantitative results (stylized model simulations)
- Stylized closed-economy Dynare experiments:
  - Decline in potential output y^{pot}_t is about ¾ of a percent for a cut in government consumption equal to 3 percent of baseline GDP.
  - If the central bank perfectly stabilizes inflation and the output gap (aggressive rate cuts), fiscal consolidation produces nil elasticity (zero) because inflation is perfectly stabilized.
  - In a more realistic central-bank reaction, monetary policy’s inflation-to-actual-output elasticity is about twice that of fiscal policy in the canonical NK setup.

### Policy-relevant quantitative results (TANK baseline and robustness)
- Joint global tightening (full TANK with endogenous capital, financial frictions, distortionary taxes):
  - Average output falls 1 percent the first two years (experiment normalization used throughout).
  - Exchange rate does not move measurably under global simultaneous tightening.
  - Monetary and fiscal policy are equally effective tools; less fiscal tightening is needed than in a single-country tightening case.
  - A central bank can raise its policy rate by ¾ as much when the rest of the world is tightening as well (to achieve the same domestic output effect).
  - A global cut in government consumption about ½ as large as a single-country consolidation suffices to reduce output by 1 percent for two years.
- Spending cuts versus tax hikes:
  - Spending cuts are more effective at reducing inflation than a mix of tax hikes when conditioned on the same-sized output contraction because tax hikes have adverse supply-side effects.
  - Empirical caveat: recent US evidence suggests a personal income tax increase can produce a sizeable decline in inflation; corporate tax increases have small effects when instruments are sized to imply an average output contraction of 1 percent the first two years.
- Distributional effects and mitigation:
  - Fiscal tightening widens consumption inequality more than monetary tightening because FP tightening affects labor income more and hurts hand-to-mouth households.
  - Targeted transfers combined with general spending cuts can offset adverse inequality effects while achieving the same macro outcomes; such a package requires larger spending cuts to offset the demand effect of transfers.

### Key table of inflation-output elasticities (selected exact values)
- Simple Model: FP 0.31; MP 0.66; Relative Effectiveness (FP/MP) 0.47
- TANK Model: FP 0.19; MP 0.23; Relative Effectiveness (FP/MP) 0.83
- TANK Model — No Financial Frictions (FF): FP 0.17; MP 0.25; Relative Effectiveness 0.68
- TANK Model — No FF and Keynesian households (HTM): FP 0.17; MP 0.26; Relative Effectiveness 0.65
- TANK Model — No Capital, HTM and Habit: FP 0.10; MP 0.22; Relative Effectiveness 0.46

### Model structure highlights (TANK Appendix)
- Two-country open-economy TANK:
  - Endogenous investment, HM and FL households, sticky wages and prices, trade adjustment costs, incomplete international financial markets.
  - Financial accelerator implemented following Bernanke, Gertler and Gilchrist (1999) / Christiano, Motto, and Rostagno (2008); special-case abstraction from accelerator described in appendix.
- Key features in firms, households, policy, and resource constraints preserved in formal equations (price setting, export LCP deviations, production CES, household budget and wage Calvo settings, government debt law of motion, fiscal rule, net foreign assets dynamics).
- Monetary and fiscal rules: labor-income-tax adjustment rule targets debt/GDP and deficit stabilization.

### Key calibration parameters and exact numerical values (selected)
- Relative country size: SOE: 0.001, Large Home: 1/3
- Trade share (SOE / Large Home): 0.35 / 0.14; 0.48 / 0.15; 0.82 / 0.22
- Price markup θ_p = 0.1
- Financial intermediation parameter Γ: AE: 0.001, EME: 0.1
- Monitoring cost μ = 0.12
- Default rate for entrepreneurs = 3 percent / year
- Variance of idiosyncratic productivity shock = 0.28
- Relative risk aversion σ = 2
- Habit persistence α = 0.8
- Share of HM agents ϖ = 0.55
- Investment adjustment cost φ_I = 3
- Capital depreciation δ = 0.025
- Elasticity of substitution between intermediate goods ρ = -1
- Quasi-capital share ω_K = 0.3
- Calvo durations: prices and imports ξ_p = ξ_m = 0.67; wage ξ_w = 0.75
- Price indexation ι_p = 1; wage/import indexation AE: 0.5, EME: 1
- Wage markup θ_W = 1/3
- Interest rate smoothing γ = 0.92; employment growth γ_ΔE = 0.25
- Inflation gap coefficient = 1.5
- Change in exchange rate (Managed Floating EME) = 5
- Discount factor β = 0.99875; steady state nominal interest rate i = 2.5
- Steady state gov’t debt/GDP (%) Home AE 129, EME: 65, Foreign 98
- Gov’t consumption/GDP Home: 0.19, Foreign: 0.25
- Steady state transfers/GDP: 6.75 and 8.25
- Steady state labor income tax rate (%) Home: 0.3, Foreign: 0.32
- Steady state consumption tax rate (%) Home: 0.07, Foreign: 0.2
- Steady state capital tax rate (%) Home: 0.21, Foreign: 0.262
- Bond maturing probability ϖ = 0.055
- Tax adjustment rule υ_1 = 0.985, υ_2 = υ_3 = 0.015

### Policy implications and considerations
- Monetary policy should remain the primary tool to maintain price stability, but tighter fiscal policy can lower inflation via reduced domestic demand and may stabilize government debt if real rates rise.
- Country-specific and international factors determine fiscal consolidation potency and side effects:
  - Central bank response function, trade openness, exchange-rate regime, FX market frictions, UIP risk-premium behavior, synchronization across countries, financial frictions, capital dynamics, hand-to-mouth households, and habit persistence.
- Coordinated tightening across countries reduces the required magnitude of tightening and mitigates exchange-rate-driven asymmetries between monetary and fiscal effectiveness.
- Distributional trade-offs: fiscal tightening tends to widen inequality relative to monetary tightening; targeted transfers and reallocation of spending can preserve redistribution while achieving macro objectives.
- Areas for further research: stronger sovereign-fiscal-financial linkages, nonlinearities in high-inflation environments, and game-theoretic incentives across countries pursuing differing monetary-fiscal mixes.

*Source: IMF staff — "1. INTRODUCTION", "2. Results with a Stylized Model", "5.1 Baseline Results", and Appendix A (wpiea2023260-print-pdf).*

### 1.    INTRODUCTION .....................................................................................................

### 1. INTRODUCTION

### Overview of the policy challenge
- Global headline inflation reached 8.7 percent in 2022.
- Core inflation has risen significantly above central bank targets in many countries and remains stubbornly high.
- Pandemic-era monetary and fiscal expansion, supply chain disruptions, and energy shocks from Russia’s invasion of Ukraine contributed to the inflation surge.
- Fiscal support during the pandemic and the recent energy crisis produced record government debt levels in many economies.
- Higher-for-longer policy rates raise debt service costs, lower economic activity and tax revenues, and put further upward pressure on government debt.

### Analytical approach and models used
- The paper uses:
  - A variation of Woodford’s (2003) textbook closed-economy New Keynesian model (stylized model) to examine whether fiscal policy can achieve the same inflation-reducing effect as monetary policy.
  - A workhorse open-economy two-agent New Keynesian (TANK) DSGE model to analyze small open advanced and emerging market economies, including channels from financial frictions in capital accumulation, liquidity-constrained households, endogenous movements in UIP risk-premiums, and real rigidities.
- Rationale: closed-economy framework is a useful starting point when fiscal and monetary policy move in tandem globally; TANK model provides a more empirically realistic framework for open economies and cross-country interactions.

### Key findings from the stylized closed-economy model
- Fiscal tightening can reduce inflation as much as monetary policy for a given decline in the output gap because both raise the wedge between the actual real rate and the potential real interest rate, lowering domestic demand and inflation via a Phillips curve.
- The transmission of fiscal policy depends crucially on the central bank policy response:
  - If inflation is at target prior to fiscal intervention and the central bank does not “welcome” a tighter stance, fiscal tightening can be offset by a more stimulative monetary policy.
- Fiscal policy may be less efficacious at reducing inflation for a given decline in the level of output because:
  - Monetary tightening does not affect potential output.
  - Fiscal tightening that triggers the same-sized output gap decline is associated with a larger decline in potential output, implying a smaller inflation reduction for the same level-output drop.
- The stylized model delivers the same inflation-output gap elasticity for monetary and fiscal policy, but implies a notably lower inflation-output elasticity for fiscal policy when measured as inflation response per 1 percent reduction in average output over a given horizon.

### Key findings from the workhorse open-economy TANK model
- In advanced small open economies with flexible exchange rates:
  - Monetary tightening leads to domestic currency appreciation, which lowers inflation in addition to the direct demand effect.
  - Fiscal tightening strengthens the external position and reduces the domestic policy rate differential vis-à-vis the rest of the world, weakening the exchange rate and reducing the ability of fiscal tightening to fight inflation.
- In emerging market economies (EMEs) with shallow FX markets:
  - Fiscal tightening can improve the net foreign asset position and lead to large declines in the UIP risk premium, producing currency appreciation.
  - As a result, fiscal tightening in vulnerable EMEs and LICs can be as effective in battling inflation as monetary policy.
- When many countries tighten simultaneously:
  - The exchange rate channel is muted (exchange rates do not move much), removing the monetary policy edge from the exchange rate channel even in advanced economies.
  - Both monetary and fiscal tightening are effective because weaker global demand helps lower inflation.
  - A smaller monetary and fiscal tightening is needed to curb inflation relative to a one-country tightening scenario.
- The TANK model implies a similar inflation-output elasticity for monetary and fiscal policy (contrast to the stylized model). The larger impact of fiscal policy on inflation relative to output in the TANK model arises from:
  - Financial frictions and endogenous capital accumulation.
  - Presence of hand-to-mouth households.
  - Habit formation for optimizing households.
  - These frictions boost the impact of monetary and fiscal policy on output, but amplify fiscal policy’s effect on inflation relatively more than monetary policy.

### Spillovers and cross-border effects
- Spillovers from disinflations in large economies to EMDEs can be more severe from monetary policy tightening in the large economies compared with a policy package that includes both tighter monetary and fiscal policy.
- Mechanism for recipient small open economies:
  - Tighter monetary policy in major economies appreciates the major currency, which allows recipient small open economies’ currencies to depreciate and cushion output via larger net exports.
  - Tighter fiscal policy in major economies reduces foreign demand more and thus reduces inflation in recipient economies more, increasing pressure on foreign central banks to hike.
- For EMDEs with shallow currency markets and large exchange rate pass-through:
  - Adverse spillovers on output can be notably severe, especially if recipient economies must defend their currency; financial conditions tighten more under major-economy monetary tightening.

### Calibration and empirical alignment
- The TANK model is parameterized to match US and euro area empirical evidence of monetary and fiscal policy impacts on inflation and output.
- Empirical literature and calibration targets:
  - IMF (2023) suggests a 1 percent of GDP increase in public spending leads to roughly 0.5 percent higher CPI for a group of OECD countries and for the US.
  - Coenen et al. (2012) find a temporary fiscal stimulus of 1 percent of GDP in one country raises inflation by 0.1 to 0.3 percentage point depending on type of fiscal measures and degree of monetary accommodation.

### Conceptual implications for policy
- The effectiveness of fiscal consolidation in fighting inflation depends on:
  - Central bank response function.
  - Trade openness and exchange rate regimes.
  - FX market frictions and extent of UIP risk premium adjustments.
  - The number of economies engaged in synchronized tightening.
  - Presence of financial frictions, capital dynamics, hand-to-mouth households, and habit persistence.
- Policy consideration:
  - Fiscal consolidation can complement monetary policy in disinflation, but its potency and side effects (e.g., on potential output and debt dynamics) depend on country-specific and international factors.
  - Coordinated tightening across countries reduces the required magnitude of tightening and mitigates exchange-rate-driven asymmetries between monetary and fiscal policy effectiveness.

*Source: IMF staff — "1. INTRODUCTION" (wpiea2023260-print-pdf).*

### 2. Results with a Stylized Model

### 2. Results with a Stylized Model

### Model setup and key equations
- Framework: Erceg and Lindé (2014) variant of Eggertsson and Woodford (2003) with zero steady state government debt and lump-sum taxes; similar in spirit to Clarida, Gali and Gertler (1999).
- Core model features:
  - Standard closed-economy New Keynesian structure with gradual price adjustment and expectations terms.
  - Monetary policy follows a policy rule with interest-rate smoothing and responses to inflation and the output gap (see eq. (3) and eq. (11) for the more elaborate model).
  - Government spending shock follows an AR process (eq. (10)): g_t = ρ_g g_{t-1} + ε_{g,t}.
  - Steady-state financing: steady-state level of g financed by labor income taxes; deviations financed by lump-sum taxes (Ricardian equivalence holds under this assumption).
- Composite parameter definitions (preserved notation):
  - σ̃ = σ(1 − g_y). (eq. 7)
  - κ_price = κ_menu φ_menu, with κ_menu = (1 − ξ_p)(1 − β ξ_p)/ξ_p. (eq. 8)
  - φ_menu = χ/(1 − α) + 1/σ̃ + α/(1 − α). (eq. 9)
- Shocks:
  - ε_{i,t} is an i.i.d. monetary policy shock.
  - g_t is a government spending shock following AR(1) in the stylized model; in the workhorse model government spending growth follows an AR(1) with error-correction leading to an AR(2) representation (eq. (12)).

### Calibrated parameters in the stylized model
- Production function parameter: 0.3
- Government spending share in steady state: 0.2
- 1/χ Frisch Elasticity: 2.5
- Intertemporal Elasticity of Substitution: 1
- Sticky price Calvo Probability: 0.9
- Discount factor: 0.995
- Response coefficient to π in policy rule: 1.5
- Response coefficient to x in policy rule: 0.125
- Smoothing coefficient in policy rule: 0.9
- AR(1) coefficient for government spending: 0.9

### Closed-economy simulation results (Dynare)
- Experiment design:
  - Two simulations using Dynare: fiscal consolidation (cut in government consumption g_t) and discretionary monetary tightening (ε_{i,t}) sized to match inflation and output-gap paths.
- Key dynamics when government consumption is cut:
  - A cut in government spending (negative ε_{g,t}) lowers the equilibrium real rate r^{pot}_t.
  - If the central bank announces it will cut the policy rate i_t in lockstep with the decline in r^{pot}_t, the central bank can keep the output gap unchanged and inflation at target.
  - To achieve stabilization in this stylized setting, either the response coefficient on inflation (γ_π) or on the output gap (γ_x) must be set arbitrarily large while smoothing coefficient γ_i is small.
- Quantitative magnitudes:
  - The decline in potential output y^{pot}_t is relatively modest: about ¾ of a percent for a cut in government consumption equal to 3 percent of baseline GDP (lower-right panel comment).
  - Modest decline in actual output y_t stems from crowding in of private consumption due to lower real interest rates.
- Inflation-output elasticity:
  - Defined as: 3-year average of the inflation (YoY) response divided by the 2-year average decline in output.
  - In the scenario where the central bank perfectly stabilizes inflation and the output gap (aggressive rate cuts), the elasticity is nil (zero) because inflation is perfectly stabilized — fiscal consolidation has zero traction on inflation in that case.

### Comparison: discretionary monetary vs fiscal tightening in the stylized model
- Experiment design in Figure 3:
  - More realistic central bank reaction function with gradual responses to inflation and output.
  - A discretionary monetary tightening ε_{i,t} is sized so that i_t produces identical inflation and output-gap paths as a 3 percent cut in g_t.
- Isomorphism and potency:
  - Paths of inflation and the output gap can be made identical by construction, creating an isomorphism between monetary and fiscal policy in terms of the inflation/output-gap elasticity (bottom-left panel).
  - Monetary policy remains notably more potent with respect to inflation to actual output elasticity:
    - The lower-right panel in Figure 3 shows the elasticity is twice as high for monetary policy compared with fiscal policy in the basic canonical NK model. The reduction in potential output from fiscal consolidation lowers the elasticity for fiscal policy.
- Implication:
  - When inflation is projected to remain well above the central bank’s target absent further monetary tightening, the central bank welcomes fiscal consolidation (transmission of fiscal policy to inflation is positive).
  - Still, in this stylized closed-economy NK model, fiscal policy’s inflation-output elasticity is notably lower than monetary policy’s.

### Preview of extensions and role of richer models
- The authors note that in Section 5 and later sections they will show that fiscal policy can be almost as effective as monetary policy in battling inflation in a more elaborate model with more realistic monetary and fiscal transmission mechanisms (workhorse open economy TANK model with capital and two countries).

*Source: IMF staff calculations, "2. Results with a Stylized Model", wpiea2023260-print-pdf*

### 5.1 Baseline Results

### 5.1 Baseline Results

### Joint global tightening: output, inflation, and exchange rates
- When all countries tighten simultaneously in the full TANK model with endogenous capital accumulation, financial frictions and distortionary taxes:
  - Average output falls 1 percent the first two years (experiment normalization used throughout).
  - The exchange rate does not move measurably, reducing the ability of monetary policy to curb inflation via exchange-rate appreciation.
  - Monetary and fiscal policy are equally effective tools because weaker global demand also helps to lower inflation.
  - Less fiscal tightening is needed compared to the case where only one country tightens.

- Fiscal versus monetary interactions:
  - A tighter fiscal stance takes pressure off the central bank: interest rates fall (or rise by less) and the government debt outlook improves.

### Spending cuts versus tax hikes
- The TANK model implies:
  - Spending cuts are more effective to reduce inflation than a mix of tax hikes, because:
    - Tax hikes have adverse supply-side effects which tend to keep inflation higher.
    - Spending cuts mainly work through demand and are hence more effective in reducing inflation when conditioning on the same-sized output contraction.
- Empirical caveat:
  - Recent empirical work on US data (Cloyne et al. (2023a, 2023b) and Dabla-Norris and others (forthcoming)) suggests:
    - A sizeable decline in inflation for an increase in the personal income tax.
    - A corporate income tax increase has small (if any) effects on inflation when both instruments are sized to imply an average output contraction of 1 percent the first two years.
  - Implication: a consolidation based on a mix of spending cuts and tax hikes can also be effective.

### Policy size and coordination effects
- Policy actions needed to achieve a 1 percent decline in domestic output:
  - When only the advanced small open economy tightens interest rate policy versus when all economies tighten:
    - The central bank in the small open economy only needs to raise its policy rate by ¾ as much when the rest of the world is tightening as well.
  - When everyone is cutting government consumption:
    - A much smaller cut—about ½ as large as when only a single economy consolidates—is needed to reduce output by 1 percent for two years.
    - This smaller global cut more effectively reduces inflation because the exchange rate does not depreciate.
  - Central banks can lower their policy rate path more in response to a global fiscal consolidation effort (CBs follow Taylor-type rule in eq. 11); fiscal consolidation would be even more effective if central banks welcomed the tightening and kept the policy rate path unchanged.

### Distributional effects and targeted transfers
- Inequality effects in the TANK model:
  - Fiscal policy (FP) tightening has a larger effect on consumption inequality than monetary policy (MP) tightening.
    - Reason: FP tightening affects labor income relatively more, hurting hand-to-mouth households, while financially unconstrained households benefit from lower real interest rates when the central bank eases.
  - The percent decrease in consumption by hand-to-mouth households is lower for a cut in government consumption than for a tightening of the policy rate.
  - The adverse effect on consumption inequality under fiscal consolidation is driven by crowding in of private consumption of financially unconstrained households rather than crowding out.
- Mitigation:
  - An appropriate combination of general spending cuts coupled with targeted transfers to vulnerable households can offset adverse inequality effects (relative to tighter MP) while achieving the same effects on inflation, output, and government debt as spending-based fiscal consolidations.
  - Note: the FP consolidation package with both spending cuts and positive targeted transfers features larger spending cuts to offset aggregate demand effects by the targeted transfers.

### Why fiscal policy is relatively more potent in the workhorse TANK model
- Key summary from Table 2 (inflation-output elasticities; elasticity = average three-year inflation impact divided by the 2-year average decline in output following a discretionary policy intervention):
  - Simple Model: Fiscal Policy (FP) 0.31; Monetary Policy (MP) 0.66; Relative Effectiveness (FP/MP) 0.47
  - TANK Model: FP 0.19; MP 0.23; Relative Effectiveness (FP/MP) 0.83
  - TANK Model — No Financial Frictions (FF): FP 0.17; MP 0.25; Relative Effectiveness 0.68
  - TANK Model — No FF and Keynesian households (HTM): FP 0.17; MP 0.26; Relative Effectiveness 0.65
  - TANK Model — No Capital, HTM and Habit: FP 0.10; MP 0.22; Relative Effectiveness 0.46

- Interpretation and drivers:
  - Inflation-output elasticities are lower for both policies in the TANK model relative to the simple model because of lower transmission to inflation due to slow nominal wage adjustment (the simple model only features slow price adjustment).
  - Fiscal policy is relatively more effective in the TANK model than in the simple model.
  - Removing the financial accelerator (Bernanke, Gertler and Gilchrist (1999) mechanism formulated as in Christiano, Motto, and Rostagno (2008, 2014)) lowers the relative effectiveness of fiscal policy.
    - Without the financial accelerator, the larger persistent fall in output after fiscal tightening moderates inflation less relative to the monetary case.
  - Removing Keynesian hand-to-mouth households produces only slight declines in relative effectiveness.
  - Only when endogenous capital accumulation and habit formation are also removed does the relative effectiveness of fiscal and monetary policy align with results in the stylized simple model (i.e., excluding capital, hand-to-mouth households and habit formation suffices to reconcile differences).
  - Even in the TANK model, effects on inflation relative to output remain smaller due to nominal wage stickiness.

### International spillovers: advanced economies → EMEs and low-income countries
- Modeling setup:
  - EMEs and low-income countries assumed to have shallow FX markets, UIP risk-premiums sensitive to net lending position, managed floating exchange regimes, and use policy rates to lean against exchange movements.
  - The source economy undertaking the policy action is calibrated to mimic the share of advanced economies in the world economy (75 percent) and their trade with the rest of the world.
  - Policy tools in source economies are resized to have the same one percent reduction in annualized core CPI inflation over three years in the advanced economies undertaking the policy action.

- Spillover patterns:
  - EMEs with shallow currency markets and the need to defend their currency can experience more negative output spillovers from monetary policy tightening in advanced economies because policy rates and term premiums rise.
  - Effects on inflation in EMEs are broadly similar under a tighter fiscal stance in advanced economies because fiscal consolidation leads to appreciation of EME currencies.
  - For foreign advanced economies receiving spillovers from the US and other advanced economies:
    - Tighter US fiscal policy can produce larger spillovers to foreign advanced economies than tighter US monetary policy.
    - Tighter US monetary policy cushions spillovers to output in other advanced economies by depreciating their currencies (i.e., the dollar appreciates), which results in larger net exports for those economies.
    - Tighter US fiscal policy reduces inflation more in advanced foreign economies, easing pressure to hike policy rates by central banks there.

### Conclusions and policy implications
- Main findings:
  - While monetary policy should remain the primary tool to maintain price stability, tighter fiscal policy can lower inflation via reduced domestic demand—similar to tighter monetary policy.
  - Twin benefit of tighter fiscal stance: stabilizes and eventually lowers government debt to sustainable levels if real rates rise; tighter monetary policy tends to put further upward pressure on government debt.
- Cross-country heterogeneity:
  - In small open advanced economies, monetary policy is more effective than fiscal policy in reducing inflation because monetary tightening appreciates the domestic currency while fiscal tightening weakens it.
  - In EMDEs with shallow FX markets, fiscal tightening can be as effective as monetary tightening because fiscal consolidation can lead to appreciation of the domestic currency and improve debt sustainability.
- Coordination and timing:
  - Fiscal consolidation is more effective in reducing inflation when many countries tighten simultaneously because exchange rates do not move much and weaker global demand helps lower inflation.
- Distributional considerations:
  - Fiscal tightening tends to widen income inequality relative to monetary tightening; targeted transfers and reallocation of spending can prevent increases in consumption inequality while achieving macro objectives.
- Spillovers and externalities:
  - Monetary and fiscal tightening in advanced economies have significant spillovers for EMDEs:
    - Advanced-economy fiscal contraction reduces output and inflation in EMDEs via trade by causing EMDE currencies to appreciate.
    - Advanced-economy monetary tightening tightens global financial conditions and may cause sizable EMDE exchange rate depreciation, intensifying downside risks.
- Areas for future research:
  - Incorporating stronger sovereign-fiscal-financial linkages and nonlinearities in high-inflation environments.
  - Considering game-theoretic incentives across countries pursuing alternative monetary and fiscal mixes.

*Source: IMF staff calculations.*

### Appendix A.  Details on the TANK Open Economy

### 0.001 in the case for small open economies, or 휁휁=1/3; approximately matching the US’s share in world output

### wpiea2023260-print-pdf - 0.001 in the case for small open economies, or 휁휁=1/3; approximately matching the US’s share in world output

### Model calibration and setup
- Relative country size:
  - SOE: 0.001, Large Home: 1/3
- Trade share calibration:
  - Trade share 휔휔퐶퐶 and 휔휔퐼퐼 set to match openness targets; SOE trade shares: 0.35, 0.48, 0.82; Large Home: 0.14, 0.15, 0.22 (as in Table A.1)
- Price and demand elasticities:
  - Price markup 휃휃푝푝 = 0.1
  - Long-run price elasticity of demand for imported consumption and investment goods [1.5] 휑휑푀푀퐶퐶 = 휑휑푀푀퐼퐼 = 1 휙휙푏푏
  - Price indexation 휄휄푝푝 = 1; wage and import price indexation AE: 0.5, EME: 1
- Financial frictions and accelerator:
  - Financial intermediation parameter Γ: AE: 0.001, EME: 0.1
  - Monitoring cost 휇휇 = 0.12 (proportion of entrepreneurs’ total gross revenue)
  - Default rate for entrepreneurs 퐹(휔휔) = 3 percent / year
  - Variance of idiosyncratic productivity shock = 0.28
- Household and preferences:
  - Relative risk aversion 휎휎 = 2
  - Habit persistence א = 0.8
  - Share of HM agents 휍휍 = 0.55
  - Leisure in utility 휒휒0 = 1; share of work to time endowment = 0.5
  - Labor utility curvature parameter 휒휒 = 4
  - Real money balance utility 휇휇0 = 0.001
- Investment, production, and capital:
  - Investment adjustment cost 휙휙퐼퐼 = 3
  - Capital depreciation 훿훿 = 0.025 (consistent with an annual depreciation rate of 10 percent)
  - Elasticity of substitution in intermediate goods 휌휌 = -1 (Leontief technology)
  - Quasi-capital share 휔휔퐾퐾 = 0.3
- Price/wage stickiness:
  - Calvo domestic price contract duration 휉휉푝푝 = 0.67
  - Calvo import/export contract duration 휉휉푚푚 = 0.67
  - Calvo wage contract duration 휉휉푤푤 = 0.75
  - Wage markup 휃휃푊푊 = 1/3
- Monetary policy rule parameters:
  - Interest rate smoothing 훾훾 = 0.92
  - Employment growth 훾훾∆퐸퐸 = 0.25
  - Employment gap 훾훾푛푛 = 0
  - Inflation gap coefficient = 1.5
  - Change in exchange rate (Managed Floating) for EME: 5
  - Discount factor 훽훽 = 0.99875
  - With inflation target 2 percent, steady state nominal interest rate i = 2.5
- Fiscal parameters and steady state fiscal ratios:
  - Gov’t consumption/GDP (Home): 0.19; (Foreign): 0.25
  - Steady state gov’t debt/GDP (%) Home AE 129, EME: 65, Foreign 98
  - Steady state transfers/GDP (%) 푡푟푡푂 = 6.75, 푡푟푡퐻퐻푀푀 = 8.25
  - Steady state labor income tax rate (%) 휏휏푁푁 Home: 0.3, Foreign: 0.32
  - Steady state consumption tax rate (%) 휏휏퐶퐶 Home: 0.07, Foreign: 0.2
  - Steady state capital tax rate (%) 휏휏퐾퐾 Home: 0.21, Foreign: 0.262
  - Bond maturing probability 휗휗 = 0.055 (consistent with a 4.5 year steady state maturity)
  - Tax adjustment rule parameters 휐휐1 = 0.985, 휐휐2 = 휐휐3 = 0.015

### Solution method and linearization
- The model is log-linearized around the non-stochastic steady state; nominal variables rendered stationary by suitable transformations.
- The unconstrained model solution uses Anderson and Moore (1985) algorithm, implementing Blanchard and Kahn (1980) solution method.

### Robustness and policy scenario findings
- Monetary-fiscal mix when central bank (CB) "welcomes" consolidation:
  - Scenario: CB leaves policy rate path unchanged while fiscal consolidation occurs (CB "welcomes" consolidation).
  - Because the model is linear, the combined monetary-fiscal response (green bar) is a weighted average of pure monetary tightening (blue) and pure fiscal tightening (red).
  - The weight on fiscal policy to compute the combined response is slightly above 0.5 because the decline in the policy rate under fiscal tightening is slightly lower than the increase under monetary tightening.
  - Key implication: If monetary policy does not cut the policy rate when fiscal policy consolidates, fiscal tightening has an even larger effect on core inflation (i.e., less fiscal tightening is required to affect core inflation when policy rate is unchanged).
- Tax- versus spending-based fiscal tightening:
  - Comparative scenario: spending-based fiscal tightening (lower government consumption) versus combined tax hikes (labor income and VAT) that generate the same-sized decline in output in the first two years.
  - Main result: Spending-based fiscal tightening is notably more effective at curbing inflation than tax hikes in the TANK model.
  - Rationale: Tax hikes produce direct supply-side effects; spending cuts operate mainly through demand reduction.
  - Caveat: Credible revenue-based consolidations can lower debt and, if tax collection is effective, may have smaller output costs than modeled because they can restore fiscal credibility, reduce credit spreads, and protect growth-enhancing spending that may crowd in private investment.
  - Note: Sizeable cuts in government investment may transmit differently than cuts in government consumption.

### Key quantitative parameters (selected exact values, as reported)
- Relative country size: SOE: 0.001, Large Home: 1/3
- Trade share (SOE / Large Home): 0.35 / 0.14; 0.48 / 0.15; 0.82 / 0.22
- Price markup 휃휃푝푝 = 0.1
- Financial intermediation cost abroad Γ AE: 0.001, EME: 0.1
- Relative risk aversion 휎휎 = 2
- Habit persistence א = 0.8
- Share of hand-to-mouth agents 휍휍 = 0.55
- Investment adjustment cost 휙휙퐼퐼 = 3
- Capital depreciation 훿훿 = 0.025
- Elasticity of substitution between intermediate goods 휌휌 = -1
- Quasi-capital share 휔휔퐾퐾 = 0.3
- Monitoring cost 휇휇 = 0.12
- Default rate for entrepreneurs = 3 percent / year
- Variance of idiosyncratic productivity shock = 0.28
- Calvo durations: prices and imports 휉휉푝푝 = 휉휉푚푚 = 0.67; wage 휉휉푤푤 = 0.75
- Price indexation 휄휄푝푝 = 1; wage/import indexation AE: 0.5, EME: 1
- Wage markup 휃휃푊푊 = 1/3
- Interest rate smoothing 훾훾 = 0.92; employment growth 훾훾∆퐸퐸 = 0.25
- Inflation gap coefficient = 1.5
- Change in exchange rate (Managed Floating EME) = 5
- Discount factor 훽훽 = 0.99875; steady state nominal interest rate i = 2.5
- Steady state gov’t debt/GDP (%) Home AE 129, EME: 65, Foreign 98
- Gov’t consumption/GDP Home: 0.19, Foreign: 0.25
- Steady state transfers/GDP: 6.75 and 8.25
- Steady state labor income tax rate (%) Home: 0.3, Foreign: 0.32
- Steady state consumption tax rate (%) Home: 0.07, Foreign: 0.2
- Steady state capital tax rate (%) Home: 0.21, Foreign: 0.262
- Bond maturing probability 휗휗 = 0.055
- Tax adjustment rule 휐휐1 = 0.985, 휐휐2 = 휐휐3 = 0.015

*Source: wpiea2023260-print-pdf (IMF staff calculations and model appendix).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023260-print-pdf.pdf_