## 18. Alternative Database: Effect of Tax Reforms on Fiscal Shock Smoothing

## Source details

**Canonical URL:** [18. Alternative Database: Effect of Tax Reforms on Fiscal Shock Smoothing](https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019113.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2019/wpiea2019113.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2019/wpiea2019113.pdf.json)

---

### Introduction and research questions
- Research questions:
  - To what extent and which tax policy reforms affect the response of disposable income (consumption) to GDP shocks (fiscal shock smoothing)?
  - Through which channels do tax reforms affect fiscal shock smoothing (fiscal automatic stabilizers and the impact of taxes on disposable income/consumption)?
- Data and sample:
  - Major tax reform episodes from the IMF Tax Policy Reform Database (TPRD) and Dabla-Norris and Lima (working paper version, September 2018).
  - Sample: panel of 13 OECD countries over the period 1980-2016 (exogenous reforms identified via Dabla-Norris and Lima).
- Identification:
  - Focus on policy changes not implemented for stabilization purposes (exogenous reforms).
  - Baseline: major tax reforms with expected revenue yield (loss) above ½ percent of GDP (robustness: ¼ percent of GDP for CIT and PIT in some checks).
  - Cumulative indicator R increments (+1 or −1) in the year of each major reform to capture past regulatory changes and reversals.
- Episodes:
  - 77 major tax reform episodes in total.
  - About 70 percent of these reforms are related to direct tax policy changes.
  - Within direct tax reforms: 23 base reforms and 26 rate reforms.
  - Within VAT reforms: 8 base reforms and 20 rate reforms.

### Empirical framework and channels
- Core regression (extended Asdrubali et al. (1996) decomposition for net transfers):
  - ∆log GNN − ∆log(GGNN) = αi + τt + β1 ∆log GG + β2 ∆log GG × R + γ R + θ X_{t−1} + ε
  - β1: average shock smoothing through fiscal policy.
  - β2: incremental smoothing following tax reforms.
- Additional specifications:
  - Tax elasticity regressions: nominal tax revenue change (∆Y) regressed on ∆ nominal GDP and interactions with R; controls expanded to include changes in domestic prices.
  - Progressivity and collection-efficiency: dynamic specifications estimated with GMM controlling for lagged dependent variables.
- Channels and indicators:
  - Tax progressivity: 1 - (100 - marginal tax rate)/(100 - average tax rate) (Arnold 2008 approach).
  - VAT collection-efficiency (C-efficiency): actual VAT revenue divided by theoretical revenue (final aggregate consumption × VAT standard rate) under perfect enforcement (Ueda, 2017).
- Controls:
  - Country and time fixed effects; spending shocks/reforms; financial crises; other structural reforms; lagged GDP; public debt; EU adhesion dummy; time trends.

### Main headline findings
- Baseline magnitudes:
  - The magnitude of shock smoothing through fiscal policy channel increases from about 2 percent to 3-3½ percent, on average.
- Reform-type heterogeneity:
  - Reforms that broaden the tax base have economically and statistically significant effects on fiscal shock smoothing.
  - Reforms increasing tax rates have smaller and less precisely estimated impacts; direct tax rate reforms are not statistically significant on average.
  - Effects are generally larger and more precisely estimated for indirect (VAT) base reforms than for direct tax changes.
- Symmetry:
  - Effects on shock smoothing are symmetric: coefficients for reforms that enhance revenue and those that worsen revenue are not statistically different.
  - Example: direct tax base reforms that raise revenue increase fiscal shock smoothing from about 1.9 to 2.9 percent; tax base reforms that reduce revenue decrease fiscal stabilization to about 0.9 percent.
- Key channels:
  - Tax elasticity, tax progressivity, and VAT collection efficiency are the primary channels through which base-broadening reforms enhance fiscal shock smoothing.

### Selected empirical results and quantitative evidence
- Episodes and sample:
  - 77 major reform episodes across 13 OECD countries; N reported in regressions: 322 (direct tax columns) and 375 (indirect tax columns).
- Table 2 (Effect of Tax Reforms on Fiscal Shock Smoothing (Percent)) — selected coefficients (notation preserved):
  - Net tax and transfers (β1):
    - Direct tax: 1.560* (t-stat 1.75)
    - Direct tax base: 1.898* (1.78)
    - Direct tax rate: 1.877* (1.76)
    - Indirect tax: 1.634** (2.45)
    - Indirect tax base: 1.328* (1.85)
    - Indirect tax rate: 1.681** (2.11)
  - Net tax and transfers after tax reform (β2):
    - Direct tax: 0.953* (1.84)
    - Direct tax base: 0.994*** (3.14)
    - Direct tax rate: 0.749 (0.74)
    - Indirect tax: 0.632** (2.80)
    - Indirect tax base: 2.115* (2.02)
    - Indirect tax rate: 0.443* (1.85)
  - Tax reform (γ):
    - Direct tax: -0.049* (-1.89)
    - Direct tax base: -0.018 (-0.66)
    - Direct tax rate: -0.040 (-1.52)
    - Indirect tax: -0.041* (-1.92)
    - Indirect tax base: 0.017 (0.48)
    - Indirect tax rate: -0.054** (-2.86)
  - R2 values reported: 0.22, 0.22, 0.22, 0.24, 0.24, 0.23.
- Tax elasticity results:
  - Major direct tax base reforms increase tax elasticity with respect to GDP from 1.1 percent to 1.3 percent.
  - Major indirect (VAT) base reforms increase tax elasticity from 0.5 percent to 1 percent.
  - Major direct and indirect tax rate reforms show no statistically significant impact on tax elasticity on average.
- Tax progressivity:
  - Major PIT base reforms increase tax progressivity by 6 percentage points (60 percent of the standard deviation of the indicator).
  - PIT rate reforms increase progressivity by 2.5 percentage points (less precisely estimated).
- VAT collection-efficiency (C-efficiency):
  - Major changes in both VAT base and rate increase VAT collection efficiency; Figure 12 reports impacts of 1.5 and 1.2 percentage points, respectively (Figure 12).

### Robustness checks and alternative specifications
- Expanded controls:
  - Adding lagged GDP, public debt, financial crises, other structural reforms, EU adhesion dummy, and time trends yields estimates similar and not statistically different from baseline.
- Addressing potential endogeneity and omitted variables:
  - Controlling for contemporaneous and lagged TPRD tax reform indicators: results broadly in line with baseline.
  - IV approach: instrument change in tax-to-GDP ratio using the exogenous tax reform indicator; IV results broadly in line with baseline (Figure 14).
- Alternative estimators:
  - GMM estimator (including lagged GDP and net transfers): results similar and not statistically different from baseline (Figure 15).
  - GLS estimator to address heteroscedasticity: similar to baseline (available upon request).
- Alternative reform identification:
  - Threshold increased from ½ percent to 1 percent of GDP: results in line with baseline; impact tends to be larger for larger revenue yield (loss) reforms (Figure 16).
  - Binary ‘0-1’ codification of reforms (reform indicator = 1 in a year after a tax reform yielding at least 0.5 percent of GDP): broadly in line with baseline but less precisely estimated (Figure 17).
  - Use of Alesina and others (2015) reform identification: similar number of major direct and indirect reforms identified; results broadly in line with baseline.

### Policy implications and recommendations
- Prioritize base-broadening reforms (direct and indirect):
  - Increase fiscal shock smoothing materially (examples: direct base reforms raise net-transfer smoothing from 2 percent to about 3 percent; VAT base reforms raise smoothing by about 2 percentage points to about 3.5 percent).
  - Improve tax elasticity, tax progressivity, and VAT collection efficiency — the three channels identified as key for enhancing shock smoothing.
  - Are revenue-raising while being less detrimental for short-run growth (consistent with Dabla-Norris and Lima 2018).
- Caution on rate changes:
  - Reforms that change tax rates generally have smaller and less robust effects on shock smoothing; direct tax rate reforms are not statistically significant on average.
- Symmetry consideration:
  - Because effects are symmetric, reversals of past reforms can undo stabilization gains; policymakers should assess combined package effects since rate and base changes can offset each other’s impact on stabilization.
- Practical guidance:
  - Design tax reform packages that prioritize broadening bases and improving collection efficiency and progressivity to strengthen fiscal automatic stabilizers and enhance macroeconomic stabilization.

*Source: “18. Alternative Database: Effect of Tax Reforms on Fiscal Shock Smoothing,” IMF working paper chapter (wpiea2019113).*

### REFERENCES ________________________________________________________________________________________________ 14

### wpiea2019113 - REFERENCES ________________________________________________________________________________________________ 14

### Tables
- Table 1. Example: Identification of Reform Episodes ________________________________________________________ 16
- Table 2: Effect of Tax Reforms on Fiscal Shock Smoothing  _________________________________________________ 17

### Figures
- 1. Direct Tax Reform Indicators Over Time __________________________________________________________________ 18
- 2. Indirect Tax Reform Indicators Over Time ________________________________________________________________ 19
- 3. Effect of Direct Tax Reforms on Fiscal Shock Smoothing _________________________________________________ 20
- 4. Effect of PIT Reforms on Fiscal Shock Smoothing ________________________________________________________ 20
- 5. Effect of CIT Reforms on Fiscal Shock Smoothing ________________________________________________________ 21
- 6. Effect of Indirect Tax Reforms on Fiscal Shock Smoothing _______________________________________________ 21
- 7. Effect of Direct Tax Reforms on Fiscal Shock Smoothing Depending on the Direction of the Reform  ___ 22
- 8. Effect of Indirect Tax Rate Reforms on Fiscal Shock Smoothing Depending on the Direction 
      of the Reform_____________________________________________________________________________________________ 22
- 9. Effect of Direct Tax Reforms on Tax Elasticity with Respect to Output  ___________________________________ 23
- 10. Effect of Indirect Tax Reforms on Tax Elasticity with Respect to Output  ________________________________ 23
- 11. Effect of PIT Reforms on Tax Progressivity ______________________________________________________________ 24
- 12. Effect of VAT Reforms on C-Efficiency ___________________________________________________________________ 24
- 13. Effect of Direct Tax Reforms on Fiscal Shock Smoothing Controlling for the Initial Level of Public 
      Debt _____________________________________________________________________________________________________ 25
- 14. IV Estimator: Effect of PIT Reforms on Tax Progressivity ________________________________________________ 25
- 15. GMM Estimator: Effect of Direct Tax Reforms on Fiscal Shock Smoothing ______________________________ 
26
- 16. Alternative Reform Threshold: Effect of Tax Reforms on Fiscal Shock Smoothing  ______________________ 26
- 17. Alternative Reform Codification: Effect of Tax Reforms on Fiscal Shock Smoothing ____________________ 27

*Source: wpiea2019113 - REFERENCES (https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019113.pdf)*

### 18. Alternative Database: Effect of Tax Reforms on Fiscal Shock Smoothing  _______________________________ 27

### 18. Alternative Database: Effect of Tax Reforms on Fiscal Shock Smoothing

### Introduction
- Research questions:
  - To what extent and which tax policy reforms affect the response of disposable income (consumption) to GDP shocks (fiscal shock smoothing)?
  - Through which channels do tax reforms affect fiscal shock smoothing (fiscal automatic stabilizers and the impact of taxes on disposable income/consumption)?
- Data sources:
  - Major tax reform episodes from the IMF Tax Policy Reform Database (TPRD) and Dabla-Norris and Lima (working paper version, September 2018).
  - Sample: panel of 13 OECD countries over the period 1980-2016 (exogenous reforms identified via Dabla-Norris and Lima).
- Identification: focus on policy changes not implemented for stabilization purposes (exogenous reforms).
- Main headline findings:
  - Reforms that broaden the tax base have economically and statistically significant effects on fiscal shock smoothing.
  - Reforms increasing tax rates have smaller and less precisely estimated impacts; direct tax rate reforms are not statistically significant on average.
  - Effects of rate and base reforms are symmetric (increases and decreases yield similar-magnitude changes in shock smoothing).
  - Key channels: tax elasticity, tax progressivity, and tax collection efficiency.

### Data and Empirical Methodology
- Tax reform episodes:
  - TPRD documents announcement and implementation dates for changes in rates and bases for 23 OECD countries, 1970-2016.
  - Matched with Dabla-Norris and Lima to identify exogenous reforms with expected revenue yields and motivations.
  - Baseline focus: major tax reforms with expected revenue yield (loss) above ½ percent of GDP (for robustness, alternative thresholds applied; for CIT and PIT a threshold of ¼ percent of GDP is applied).
  - Construct cumulative indicator R that increments (+1 or −1) in the year of each major reform; cumulative approach captures past regulatory changes and reversals.
- Episodes identified:
  - 77 major tax reform episodes in total (about six major tax reforms by country over the sample period).
  - About 70 percent of these reforms are related to direct tax policy changes.
  - Within direct tax reforms: 23 base reforms and 26 rate reforms.
  - Within VAT reforms: 8 base reforms and 20 rate reforms.
- Channels and indicators:
  - Tax progressivity indicator: 1 - (100 - marginal tax rate)/(100 - average tax rate) (Arnold 2008 approach).
  - VAT collection-efficiency (C-efficiency): actual VAT revenue divided by theoretical revenue (final aggregate consumption × VAT standard rate) under perfect enforcement (Ueda, 2017).
- Macroeconomic and control data:
  - OECD national accounts (real and local currency), Duval and others (2018) for structural policy changes, IMF WEO, Fiscal Monitor, IFS, Laeven and Valencia (2018) for financial crisis episodes, Furceri and Zdzienicka (2018) for spending shocks.
- Empirical framework:
  - Extension of Asdrubali et al. (1996) decomposition of cross-sectional variance in GDP into international factor flows, capital depreciation, net transfers, private and public saving.
  - Core regression (extended equation (6)) for net transfers (GNN - GGNN):
    - ∆log GNN − ∆log(GGNN) = αi + τt + β1 ∆log GG + β2 ∆log GG × R + γ R + θ X_{t−1} + ε
    - β1 captures average shock smoothing through fiscal policy; β2 is incremental smoothing following tax reforms.
  - Controls include country and time fixed effects, spending shocks/reforms, financial crises, and other tax/structural reforms.
  - For tax elasticity regressions, nominal tax revenue change (∆Y) regressed on ∆ nominal GDP and interactions with R (equation (10)); controls expanded to include changes in domestic prices.
  - For progressivity and collection-efficiency, dynamic specifications estimated with GMM controlling for lagged dependent variables (equation (11)).

### Empirical Results — Baseline
- Impact of major direct tax base reforms:
  - Fiscal shock smoothing through net transfers increases from 2 percent to about 3 percent after major direct tax base reforms (Figure 3 referenced).
- Direct tax rate reforms:
  - Effects are not statistically significant on average.
- PIT and CIT:
  - Separate analyses for PIT and CIT produce similar qualitative results to overall direct taxes (Figures 4 and 5 referenced).
- Indirect (VAT) reforms:
  - VAT base reforms increase fiscal shock smoothing by about 2 percentage point to about 3.5 percent (from roughly 1.5–1.7 percent to about 3–3.5 percent as described).
  - VAT rate reforms increase shock smoothing by 0.4 percentage point (from 1.7 to 2.1 percent).
- Interpretation:
  - Reforms that broaden tax bases (direct or indirect) not only raise revenue and are less detrimental for short-run growth, but are also more effective in enhancing macroeconomic stabilization.

### Symmetric Impact of Tax Reform
- Symmetry result:
  - Effects on shock smoothing are symmetric: coefficients for reforms that enhance revenue and those that worsen revenue are not statistically different.
  - Example: Direct tax base reforms that raise revenue increase fiscal shock smoothing from about 1.9 to 2.9 percent; tax base reforms that reduce revenue decrease fiscal stabilization to about 0.9 percent (Figure 7 referenced).
  - VAT rate reforms show similar symmetry (Figure 8 referenced); sample lacks exogenous VAT base reforms that reduced revenue, so symmetry for VAT base cannot be tested.
- Implication:
  - Symmetric effects across heterogeneous policy changes help explain why fiscal policy stabilization has changed little over time.

### Channels — Quantitative Evidence
- Tax elasticity (responsiveness of tax revenue to GDP):
  - Major reforms that broaden the direct tax base increase tax elasticity with respect to GDP from 1.1 percent to 1.3 percent (Figure 9 referenced).
  - Major reforms that broaden the indirect tax (VAT) base increase tax elasticity from 0.5 percent to 1 percent (Figure 10 referenced).
  - Major direct and indirect tax rate reforms show no statistically significant impact on tax elasticity on average.
- Tax progressivity:
  - Major PIT base reforms increase tax progressivity by 6 percentage points (60 percent of the standard deviation of the indicator) (Figure 11 referenced).
  - PIT rate reforms increase progressivity by 2.5 percentage points and are less precisely estimated.
- VAT collection efficiency:
  - Major changes in both VAT base and rate increase VAT collection efficiency (C-efficiency); magnitudes are described as significant (specific point estimates appear in figures and tables referenced).

### Robustness and Additional Notes
- Reform threshold and indicator construction:
  - Baseline uses major reforms with expected revenue yield (loss) above ½ percent of GDP (¼ percent for CIT and PIT in some checks); robustness checks relax this threshold and compare cumulative indicator to a 0-1 approach.
- Controls and identification:
  - Country fixed effects control for initial differences in taxation systems (tax-to-GDP ratio, measures of progressivity and collection efficiency).
  - Time fixed effects capture global shocks.
  - Spending-side measures and spending shocks are systematically controlled for.
  - GMM used for dynamic specifications to minimize endogeneity bias when investigating progressivity and collection-efficiency outcomes.
- Aggregate counts and composition of episodes:
  - 77 major reform episodes identified across 13 OECD countries, ~70 percent direct tax changes.
  - Within direct taxes: 23 base reforms, 26 rate reforms.
  - Within VAT: 8 base reforms, 20 rate reforms.

### Policy Implications and Conclusions
- Reforms broadening tax bases (direct or indirect):
  - Increase fiscal shock smoothing materially (examples: direct base reforms raise net-transfer smoothing from 2 percent to about 3 percent; VAT base reforms raise smoothing by about 2 percentage points to about 3.5 percent).
  - Improve tax elasticity, tax progressivity, and VAT collection efficiency — the three channels identified as key for enhancing shock smoothing.
- Reforms that change tax rates:
  - Generally have smaller and less robust effects on shock smoothing; direct tax rate reforms are not statistically significant on average.
- Symmetry of impacts:
  - Policymakers should note that expansions and contractions of bases/rates appear to have symmetric effects on shock smoothing, implying that reversal of past reforms may undo stabilization gains.
- Overall recommendation:
  - Policy designs that prioritize broadening tax bases and improving collection efficiency and progressivity are more effective at strengthening fiscal automatic stabilizers and enhancing macroeconomic stabilization.

*Source: “18. Alternative Database: Effect of Tax Reforms on Fiscal Shock Smoothing,” IMF working paper chapter (content as provided).*

### 1.5 and 1.2 percentage points, respectively (Figure 12).

### wpiea2019113 - 1.5 and 1.2 percentage points, respectively (Figure 12).

### Robustness checks: additional controls and specification tests
- Concern: tax reform episodes, while exogenous to macroeconomic fluctuations, might be endogenous to other developments or reforms that influence fiscal shock smoothing.
- Controls added: lagged GDP; public debt level; financial crises; other structural reforms (e.g., labor and product market); EU adhesion dummy; time trends.  
  - Result: estimates with this extensive set of controls are similar and not statically different from the baseline.
- Concern: omission of spending policy reforms and minor tax reforms (attenuation bias).
  - Approach 1: control for contemporaneous and lagged tax reform indicators from the TPRD database.  
    - Result: broadly in line with baseline (results available upon request).
  - Approach 2: instrument change in tax-to-GDP ratio using the exogenous tax reform indicator (IV estimator).  
    - Rationale: instrumented change captures narrative reform information and observed tax changes; IV accounts for measurement error in announced reform sizes.  
    - Result (Figure 14): results broadly in line with baseline—potential biases not a concern.

### Robustness checks: alternative estimators and identification
- Alternative estimators:
  - GMM estimator: include lagged GDP and net transfers as regressors and estimate equation (9) with GMM.  
    - Result (Figure 15): similar and not statically different from baseline.
  - GLS estimator (Asdrubali and others 1996 approach) to address heteroscedasticity.  
    - Result: similar to baseline (available upon request).
- Alternative identification of reform episodes:
  - Threshold change: increase threshold from ½ percent of GDP to 1 percent of GDP.  
    - Result (Figure 16): results in line with baseline; impact tends to be larger for larger revenue yield (loss) reforms.
  - Binary ‘0-1’ codification: reform indicator = 1 in a year after a tax reform yielding at least 0.5 percent of GDP is announced, zero otherwise.  
    - Result (Figure 17): broadly in line with baseline but less precisely estimated (cumulative baseline indicator produces more precise estimates by accounting for number, level impact, and reversals).
  - Use Alesina and others (2015) reform identification.  
    - Result (Figure 18): similar number of major direct and indirect reforms identified; results broadly in line with baseline.

### Key empirical findings and magnitudes
- Overall effect on fiscal stabilization:
  - "The magnitude of shock smoothing through fiscal policy channel increases from about 2 percent to 3-3½ percent, on average."
- Heterogeneity by reform type:
  - Effect larger and more precisely estimated for reforms that broaden the tax base than for those that increase the tax rate.
  - Effects considerably higher for indirect tax changes than for direct tax changes.
- Channels associated with increased fiscal stabilization:
  - Reforms that increase fiscal shock smoothing are those that improve:
    - tax elasticity,
    - tax progressivity,
    - tax collection efficiency.
  - Mechanism: tax reforms strengthen fiscal automatic stabilizers and amplify how outcomes (disposable income, consumption) react to tax policy changes.
- Symmetry of effects:
  - Effects are symmetric: tax reforms increasing and decreasing tax rates or bases are not statistically different from each other in terms of magnitude.
  - Implication: average shock smoothing through fiscal policy has changed little over time because tax rate and base reforms have often had offsetting effects.

### Empirical evidence highlights (selected numerical results from tables/figures)
- Table 2 (Effect of Tax Reforms on Fiscal Shock Smoothing (Percent)): selected coefficients and statistics (notation preserved):
  - Net tax and transfers (β1): 1.560* (t-stat 1.75) for Direct tax; 1.898* (1.78) for Direct tax base; 1.877* (1.76) for Direct tax rate; 1.634** (2.45) for Indirect tax; 1.328* (1.85) for Indirect tax base; 1.681** (2.11) for Indirect tax rate.
  - Net tax and transfers after tax reform (β2): 0.953* (1.84) for Direct tax; 0.994*** (3.14) for Direct tax base; 0.749 (0.74) for Direct tax rate; 0.632** (2.80) for Indirect tax; 2.115* (2.02) for Indirect tax base; 0.443* (1.85) for Indirect tax rate.
  - Tax reform (γ): -0.049* (-1.89) for Direct tax; -0.018 (-0.66) for Direct tax base; -0.040 (-1.52) for Direct tax rate; -0.041* (-1.92) for Indirect tax; 0.017 (0.48) for Indirect tax base; -0.054** (-2.86) for Indirect tax rate.
  - Sample sizes and fit reported: N 322 (direct tax columns) and 375 (indirect tax columns); R2 values: 0.22, 0.22, 0.22, 0.24, 0.24, 0.23.
- Figures summarized (directional results preserved as reported):
  - Figure 3–6: Blue bars (β1 + β2) show fiscal shock smoothing in years of reform; grey bars show average fiscal shock smoothing (β1); significance indicated with ***/**/* at 1, 5, 10 percent.
  - Figure 9–10: Tax elasticity with respect to GDP increases after reforms; blue bars (β1 + β2) and grey bars (β1) show significant improvements for base and rate reforms (direct and indirect).
  - Figure 11: PIT reforms increase tax progressivity; effects denoted in percentage points with significance levels.
  - Figure 12: VAT reforms affect C-Efficiency by 1.5 and 1.2 percentage points, respectively (these numbers correspond to VAT tax base and VAT tax rate impacts shown in Figure 12).

### Policy implications and recommendations
- Design implication:
  - Reforms that broaden the tax base (direct or indirect) not only:
    - generate substantial tax revenue,
    - are less detrimental for growth in the short-run (Dabla-Norris and Lima 2018),
    - but also foster macroeconomic stabilization and higher growth over the medium term.
- Practical guidance:
  - Prioritize tax base-broadening measures (both direct and indirect) to simultaneously improve revenue mobilization and fiscal automatic stabilizers.
  - Consider the distributional and administrative effects since improvements in tax progressivity and collection efficiency are key channels for enhanced stabilization.
- Interpretation for policy makers:
  - Given the symmetry of effects from increases and decreases in rates or bases, assess combined package effects—rate and base changes can offset each other’s impact on stabilization.

*Source: IMF working paper content (wpiea2019113).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019113.pdf_