## 1. Inter-Quartile Range of Tax Revenues Over Time

## Source details

**Canonical URL:** [1. Inter-Quartile Range of Tax Revenues Over Time](https://www.imf.org/-/media/files/publications/wp/wp1704.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/wp1704.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/wp1704.pdf.json)

---

### I. Introduction
- Research questions:
  - How large are tax buoyancies for different country groups (Advanced Economies, Emerging Market Economies, Low Income Countries) and for different revenue types, over the long run and short run?
  - Do tax buoyancies differ across business cycle phases?
  - Which structural characteristics determine tax buoyancy?
- Definitions:
  - Tax buoyancy: percentage change in tax revenue associated with a one percent change in income (includes discretionary policy changes).
  - Elasticity: responsiveness of tax revenue to changes in national income keeping other parameters constant.
  - A buoyancy of one implies an extra one percent of GDP increases tax revenue by one percent; buoyancy > 1 implies revenue increases more than GDP.
- Study contributions:
  - Estimate short- and long-run tax buoyancies using FMOLS and Mean Group estimators for PIT, CIT, TGS, SSC across 107 countries (1980–2014 depending on group).
  - Analyze influence of business cycle turning points with focus on the Global Financial Crisis (GFC).
  - Assess macro and structural determinants of cross-country differences in tax buoyancy.

### II. Estimates of Buoyancy and Elasticity (Methodology and Data)
- Methodology:
  - Main regression: unrestricted error correction ARDL(p,q) linking ln(tax revenue) and ln(GDP) with country fixed effects and distributed lags; long-run coefficients θ = (β)/(1−φ).
  - Estimators: country-by-country FMOLS; panel MG and PMG estimators.
- Data:
  - Unbalanced panel: 31 AEs, 38 EMEs, 38 LICs, years between 1980 and 2014; coverage varies by group and revenue component.
  - Revenue components: Personal Income Tax (PIT), Corporate Income Tax (CIT), Taxes on Goods and Services (TGS), Social Security Contributions (SSC).
  - Tax rates from IMF Tax Policy Division (highest marginal PIT, base CIT, base VAT/TGS). GDP, output gap, inflation from IMF IFS.

### III. Country-by-country FMOLS Results (Long-run and Short-run)
- Long-run FMOLS means (aggregate total revenue):
  - AE 1.06
  - EME 1.15
  - LIC 1.15
- Long-run interpretation:
  - Long-run buoyancy generally not statistically different from one for the majority of countries in each group.
  - Component patterns: CIT on average > 1 with substantial heterogeneity; PIT, TGS, SSC slightly > 1 on average but often not significantly different from one.
- Short-run FMOLS summary:
  - Average short-run buoyancy range across groups: between 0.96 and 1.21.
  - Short-run buoyancy not significantly different from one in majority of countries; tax systems mostly neither strong nor weak output stabilizers on average.
  - CIT short-run buoyancy on average substantially > 1; significant in about half the countries in each group.
  - Speed of adjustment (error correction) generally negative, consistent with convergence to long-run equilibrium.

### IV. Panel Regression Results — Total Tax Revenue (MG and PMG, Table 4 exact values)
- AE:
  - MG long-run buoyancy: 0.997*** (0.024)
  - MG (constraint) long-run: 1.000 (0.000)
  - PMG long-run: 1.000*** (0.008)
  - MG short-run: 1.010*** (0.071)
  - MG (constraint) short-run: 1.015*** (0.054)
  - PMG short-run: 0.975*** (0.053)
  - Speed of adjustment: AE MG -0.355*** (0.034); AE PMG -0.236*** (0.025)
  - Observations: 922; # countries: 31
- EME:
  - MG long-run: 1.060*** (0.039)
  - MG (constraint) long-run: 1.000 (0.000)
  - PMG long-run: 1.070*** (0.061)
  - MG short-run: 1.235*** (0.090)
  - MG (constraint) short-run: 1.221*** (0.090)
  - PMG short-run: 1.214*** (0.091)
  - Speed of adjustment: EME PMG -0.339*** (0.033)
  - Observations: 726; # countries: 38
- LIC:
  - MG long-run: 1.210*** (0.012)
  - MG (constraint) long-run: 1.000 (0.000)
  - PMG long-run: 1.201*** (0.012)
  - MG short-run: 1.200*** (0.172)
  - MG (constraint) short-run: 0.839*** (0.136)
  - PMG short-run: 1.196*** (0.164)
  - Speed of adjustment: LIC PMG -0.325*** (0.037)
  - Observations: 716; # countries: 38

### V. Panel Regression by Tax Component (PMG estimates, Table 5 exact values)
- Advanced Economies (AE):
  - Long-run buoyancy:
    - PIT 0.907*** (0.015)
    - CIT 1.516*** (0.034)
    - TGS 0.951*** (0.008)
    - SSC 0.941*** (0.009)
  - Short-run buoyancy:
    - PIT 0.798*** (0.110)
    - CIT 2.936*** (0.240)
    - TGS 0.873*** (0.051)
    - SSC 0.600*** (0.102)
  - Speed of adjustment: PIT -0.249*** (0.026); CIT -0.344*** (0.034); TGS -0.256*** (0.034); SSC -0.251*** (0.036)
- Emerging Market Economies (EME):
  - Long-run buoyancy:
    - PIT 1.036*** (0.016)
    - CIT 1.014*** (0.011)
    - TGS 1.000*** (0.009)
    - SSC 1.668*** (0.029)
  - Short-run buoyancy:
    - PIT 1.211*** (0.204)
    - CIT 1.590*** (0.272)
    - TGS 1.027*** (0.120)
    - SSC 1.210*** (0.272)
  - Speed of adjustment: PIT -0.407*** (0.062); CIT -0.379*** (0.054); TGS -0.359*** (0.047); SSC -0.024 (0.210)
- Low Income Countries (LIC):
  - Long-run buoyancy:
    - PIT 0.487*** (0.043)
    - CIT 0.887*** (0.048)
    - TGS 1.075*** (0.013)
    - SSC 1.024*** (0.034)
  - Short-run buoyancy:
    - PIT 0.932** (0.470)
    - CIT 1.491*** (0.493)
    - TGS 1.030*** (0.211)
    - SSC 1.041 (0.929)
  - Speed of adjustment: PIT -0.162** (0.063); CIT -0.251*** (0.056); TGS -0.353*** (0.035); SSC -0.416*** (0.087)

### VI. Robustness Checks: Controlling for Tax Rates and Inflation
- Controlling for tax rates (Table 6 selected exact comparisons, AE examples):
  - PIT AE long-run: no control 0.907*** (0.015); control 0.966*** (0.014).
  - PIT AE short-run: no control 0.798*** (0.110); control 0.974*** (0.135).
  - CIT AE long-run: no control 1.516*** (0.034); control 1.585*** (0.047).
  - CIT AE short-run: no control 2.936*** (0.240); control 3.611*** (0.399).
  - TGS AE long-run: no control 0.951*** (0.008); control 0.867*** (0.012).
  - TGS AE short-run: no control 0.873*** (0.051); control 0.890*** (0.104).
  - General: Controlling for tax rates generally increases estimated buoyancies (especially CIT and PIT in some groups).
- Controlling for inflation (Table 7):
  - AE long-run buoyancy: no control 1.000*** (0.008); control 0.996*** (0.009).
  - EME long-run: no control 1.070*** (0.006); control 1.073*** (0.007).
  - LIC long-run: no control 1.201*** (0.012); control 1.188*** (0.011).
  - Long-run price effects significant and positive: AE 0.493*** (0.146); EME 0.821*** (0.144); LIC 0.323* (0.192).
  - Inclusion of inflation lowers buoyancy coefficients: nominal buoyancy > real buoyancy.

### VII. Tax Buoyancy Around Business Cycle Turning Points
- Event-study (recessions t, t±1, t±2):
  - Overall tax revenues decrease immediately from t-1 to t across all country groups: short-run buoyancy appears > 1 during a recession.
  - EMEs and LICs: tax revenues recover by t+1; AEs: tax revenues remain subdued and recover more slowly.
  - Tax-category patterns:
    - In AEs, lagged and slow recovery mainly in PIT and TGS.
    - SSC tends to increase in the year of the recession across all groups.
- Global Financial Crisis (2008–2012) findings (Table A4/A5 highlights):
  - Buoyancy estimates generally higher during 2008–2012 than average.
  - In AEs, PIT and TGS long-run buoyancy coefficients larger during GFC and significantly > 1.
  - Short-run buoyancy of all taxes enhanced during the Great Recession in both AEs and EMEs; in LICs the contribution came from CIT and TGS.
- Asymmetric short-term buoyancy over the business cycle (PMG, Table 8 exact values):
  - Total tax revenue short-run buoyancy:
    - AE Expansion 1.087*** (0.050); Contraction 0.950*** (0.051)
    - EME Expansion 1.213*** (0.075); Contraction 1.183*** (0.080)
    - LIC Expansion 1.485*** (0.249); Contraction 0.652*** (0.182)
  - Component examples:
    - AE CIT: Expansion 1.096*** (0.293); Contraction 2.038*** (0.305)
    - EME CIT: Expansion 0.886*** (0.244); Contraction 2.348*** (0.321)
    - EME TGS: Expansion 0.754*** (0.109); Contraction 1.253*** (0.123)
  - Interpretation:
    - In AEs only CIT buoyancy is larger during contractions than expansions.
    - In EMEs both CIT and TGS buoyancy increase during contractions.

### VIII. Determinants of Tax Buoyancy (Cross-country regressions, Table 9 and Table 10)
- Approach:
  - Regress estimated short-run buoyancies on country-specific average determinants using Weighted Least Squares (weights inverse of standard errors).
  - Determinant groups: Structural, Demographic, Macroeconomic, Institutions.
- Key findings — total taxes (Table 9 selected exact coefficients):
  - Trade openness increases tax buoyancy: specification (4) trade openness 0.349*** (0.087); specification (7) trade openness 0.447*** (0.078).
  - Human capital increases tax buoyancy: specification (6) human capital 0.190*** (0.064).
  - Inflation decreases tax buoyancy: specification (5) inflation -0.102* (0.059).
  - Output volatility decreases tax buoyancy: specification (5) output volatility -0.055*** (0.018); specification (7) -0.064*** (0.018).
  - Population density and share of agriculture coefficients vary by specification; share of agriculture generally negative but not always significant.
- Determinants by tax category (Table 10 selected exact coefficients):
  - PIT buoyancy:
    - Human capital positive: specification (2) human capital 1.044*** (0.339).
    - Output volatility negative: specification (1) -0.045*** (0.017).
    - Civil liberties 0.313** (0.121); political rights 0.243* (0.124) in some specifications.
  - CIT buoyancy:
    - Human capital 1.272*** (0.355).
    - Output volatility -0.192** (0.093).
    - Trade openness 0.818* (0.450) (marginal).
  - TGS buoyancy:
    - Population density 0.033*** (0.010).
    - Human capital 0.162* (0.084).
    - Share of agriculture -0.118* (0.070).
    - Output volatility -0.076*** (0.021).

### IX. Conclusion and Policy Implications
- Aggregate empirical conclusions:
  - Long-run buoyancy of total revenue is not different from one across country groups on average.
  - Short-run buoyancy:
    - Not statistically different from one in Advanced Economies.
    - Statistically larger than one in Emerging Market and Low Income Countries.
  - By tax type:
    - Long-run buoyancy > 1 for CIT in AEs; PIT and SSC in EMEs; TGS in LICs.
    - CIT exhibits strong short-run buoyancy across country groups (stabilizing role).
  - Nominal vs real:
    - Tax buoyancy is not neutral to inflation: nominal buoyancy > real buoyancy.
  - During downturns:
    - In AEs, only CIT buoyancy increases during contractions relative to expansions.
    - In EMEs, both CIT and TGS buoyancy increase during contractions.
  - Determinants:
    - Trade openness and human capital increase tax buoyancy.
    - Inflation and output volatility decrease tax buoyancy.
- Policy implications (derived from empirical results):
  - Because long-run tax-to-GDP shares tend to remain constant, permanent increases in spending-to-GDP that do not improve structural conditions or human capital should be accompanied by revenue mobilization reforms to avoid permanent fiscal deterioration.
  - Reliance on CIT improves the stabilization role of the revenue system, especially during recessions, but declines in CIT collection as a share of GDP can be large and recovery may take longer.
  - Policies that enhance human capital and trade openness can raise buoyancy and thereby strengthen revenue responsiveness to growth.
  - Macroeconomic stabilization (reducing inflation and output volatility) supports higher tax buoyancy and hence more reliable revenue mobilization.

### X. Appendix Highlights (selected country-level and robustness statistics)
- Overall tax buoyancy — Advanced Economies (Table A1a) selected country entries (exact formatting preserved):
  - Australia: 0.953*** 1.004*** -0.153
  - Austria: 1.000*** 0.565** -0.382***
  - Canada: 0.550** 1.179*** -0.095***
  - Cyprus: 1.446*** 2.415*** -0.635***
  - Germany: 1.422*** 1.921*** -0.136**
  - New Zealand: 7.792 2.016*** 0.006
  - Norway: 3.872 1.590*** 0.021
  - United States: 1.4072.652***-0.024
  - Aggregate AE statistics: Mean (Long run) 1.262; Mean (Short run) 1.203; Median (Long run) 1.007; Median (Short run) 1.095; SD (Long run) 1.437; SD (Short run) 0.438
- Overall tax buoyancy — Emerging Market Economies (Table A1b) selected entries:
  - Algeria: 1.211*** 1.849*** -0.494***
  - Brazil: 0.976*** 1.306*** -0.206
  - China: 1.238*** 0.656*** -0.065
  - India: 1.104*** 1.668*** -0.669***
  - Mexico: 1.001*** 0.624* -0.481**
  - Turkey: 1.176***1.554***-0.931***
  - Aggregate EME statistics: Mean (Long run) 1.06; Mean (Short run) 1.31; Median (Long run) 1.07; Median (Short run) 1.23; SD (Long run) 0.25; SD (Short run) 0.65
- Overall tax buoyancy — Low Income Countries (Table A1c) selected entries:
  - Bolivia: 1.281***1.732***-0.337
  - Bangladesh: 1.198*** 0.777 -0.544***
  - Cameroon: 1.001*** 3.814** -0.749***
  - Nepal: 1.409*** 1.515*** -0.742**
  - Rwanda: 1.275*** 2.200*** -0.074
  - Zambia: 1.756 2.565** -0.135
  - Aggregate LIC statistics: Mean (Long run) 1.23; Mean (Short run) 1.27; Median (Long run) 1.16; Median (Short run) 0.91; SD (Long run) 0.42; SD (Short run) 1.19
- Panel and unit-root robustness tests:
  - First-generation IPS and MW tests and second-generation Pesaran CIPS reported for GDP, Tax revenue, PIT, CIT, TGS, SSC across AE, EME, LIC panels (exact statistics and p-values preserved in source tables).
- GFC-specific panel estimates (Table A4 and Table A5 selected exact values):
  - Overall (Table A4): AE PMG long-run 1.102*** (0.002); EME PMG long-run 1.705*** (0.065); LIC PMG long-run 0.596*** (0.003)
  - GFC short-run examples: AE PMG 1.190*** (0.113); EME PMG 1.734*** (0.207); LIC PMG 1.573*** (0.295)
  - Component-level GFC (Table A5) AE long-run: PIT 1.698***; CIT 0.799***; TGS 1.625***; SSC 0.820***. AE short-run CIT 3.541***. LIC GFC component example: CIT long-run -6.163 (7.499); CIT short-run 2.339*** (0.694).

*Source: wp1704 - 1. Inter-Quartile Range of Tax Revenues Over Time (IMF working paper appendix tables).*

### 1. Inter-Quartile Range of Tax Revenues Over Time ................................................................9

### 1. Inter-Quartile Range of Tax Revenues Over Time

### Figures (included in this content unit)
- 1. Inter-Quartile Range of Tax Revenues Over Time ................................................................9
- 2. Kernel Density of FMOLS Estimates of Long-Run Buoyancies.........................................11
- 3. Kernel Density of FMOLS Estimates of Short-Run Buoyancies ........................................12
- 4. Tax Revenue and Categories (Percent of GDP) Around Recession Episodes .....................17

### Tables (included in this content unit)
- 1. Summary Statistics.................................................................................................................9
- 2. Summary Statistics of FMOLS Estimates of Long-Run Buoyancies ..................................11
- 3. Summary Statistics of FMOLS Estimates of Short-Run Buoyancies ..................................12
- 4. Overall Tax Buoyancy by Country Group, Alternative Estimators .....................................13
- 5. Buoyancy of Tax Revenue Components ..............................................................................14
- 6. Buoyancy of PIT, CIT and TGS with and without Controlling for Tax Rates ....................15
- 7. Overall Tax Buoyancy with and without Controlling for Inflation .....................................16
- 8. Asymmetric Short-term Buoyancy over the Business Cycle ...............................................19
- 9. Determinants of Tax Buoyancy: Total Taxes ......................................................................21
- 10. Determinants of Tax Buoyancy: Alternative Tax Categories ............................................22

*Source: wp1704 - 1. Inter-Quartile Range of Tax Revenues Over Time (PDF).*

### Appendix Tables ........................................................................................................

### wp1704 - Appendix Tables

### I. Introduction
- Research questions:
  - How large are tax buoyancies for different country groups (Advanced Economies, Emerging Market Economies, Low Income Countries) and for different revenue types, over the long run and short run?
  - Do tax buoyancies differ across business cycle phases?
  - Which structural characteristics determine tax buoyancy?
- Definitions and policy relevance:
  - Tax buoyancy: percentage change in tax revenue associated with a one percent change in income (includes discretionary policy changes).
  - Elasticity: responsiveness of tax revenue to changes in national income keeping other parameters constant.
  - A buoyancy of one implies an extra one percent of GDP increases tax revenue by one percent; buoyancy > 1 implies revenue increases more than GDP.
- Study contributions:
  - Estimate short- and long-run tax buoyancies (and test equality to one) using FMOLS and Mean Group estimators for PIT, CIT, TGS, SSC across 107 countries (1980–2014 depending on group).
  - Analyze influence of business cycle turning points with focus on the Global Financial Crisis (GFC).
  - Assess macro and structural determinants of cross-country differences in tax buoyancy.

### II. Estimates of Buoyancy and Elasticity
- Methodology:
  - Main regression framework: unrestricted error correction ARDL(p,q) (equations (1)–(3) in source) linking ln(tax revenue) and ln(GDP) with country fixed effects and distributed lags.
  - Long-run coefficients derived as θ = (β)/(1−φ) (notation from source).
  - Estimators used:
    - Country-by-country: Fully Modified OLS (FMOLS).
    - Panel: Mean Group (MG) and Pooled Mean Group (PMG) estimators.
  - Short-run vs long-run interpretation: long-run buoyancy expected ≈ 1; short-run buoyancies may differ by tax type and reflect automatic stabilizers and discretionary policy.
- Data:
  - Unbalanced panel: 31 AEs, 38 EMEs, 38 LICs, years between 1980 and 2014; coverage varies by group and revenue component.
  - Revenue components: Personal Income Tax (PIT), Corporate Income Tax (CIT), Taxes on Goods and Services (TGS), Social Security Contributions (SSC).
  - Sources: OECD when available; IMF WEO otherwise. Tax rates from IMF Tax Policy Division (highest marginal PIT, base CIT, base VAT/TGS). GDP, output gap, inflation from IMF IFS.
- Country-by-country FMOLS results (long-run):
  - Aggregate total revenue averages: AE 1.06, EME 1.15, LIC 1.15 (long-run buoyancy means).
  - Long-run buoyancy generally not statistically different from one for the majority of countries in each group.
  - For components, CIT is on average larger than one, with substantial heterogeneity; PIT, TGS, SSC slightly > 1 on average but often not significantly different from one.
- Country-by-country FMOLS results (short-run):
  - Average short-run buoyancy range across groups: between 0.96 and 1.21.
  - Short-run buoyancy not significantly different from one in majority of countries; tax systems mostly neither strong nor weak output stabilizers on average.
  - CIT short-run buoyancy is on average substantially greater than one: contributes to output stabilization; significant in about half the countries in each group.
  - Speed of adjustment (error correction) generally negative, consistent with convergence to long-run equilibrium.
- Panel regression (MG and PMG) — total tax revenue (Table 4 exact values):
  - AE MG long-run buoyancy: 0.997*** (std. err. (0.024))
  - AE MG (constraint) long-run: 1.000 (0.000)
  - AE PMG long-run: 1.000*** (0.008)
  - EME MG long-run: 1.060*** (0.039)
  - EME MG (constraint) long-run: 1.000 (0.000)
  - EME PMG long-run: 1.070*** (0.061)
  - LIC MG long-run: 1.210*** (0.012)
  - LIC MG (constraint) long-run: 1.000 (0.000)
  - LIC PMG long-run: 1.201*** (0.012)
  - Short-run buoyancies (selected exact values from Table 4):
    - AE MG short-run: 1.010*** (0.071)
    - AE MG (constraint) short-run: 1.015*** (0.054)
    - AE PMG short-run: 0.975*** (0.053)
    - EME MG short-run: 1.235*** (0.090)
    - EME MG (constraint) short-run: 1.221*** (0.090)
    - EME PMG short-run: 1.214*** (0.091)
    - LIC MG short-run: 1.200*** (0.172)
    - LIC MG (constraint) short-run: 0.839*** (0.136)
    - LIC PMG short-run: 1.196*** (0.164)
  - Speed of adjustment values (exact from Table 4): AE MG -0.355*** (0.034); AE PMG -0.236*** (0.025); EME PMG -0.339*** (0.033); LIC PMG -0.325*** (0.037).
  - Observations and # Countries as in Table 4: AEs obs 922, # countries 31; EMEs obs 726, # countries 38; LICs obs 716, # countries 38.
- Panel regression by tax component (PMG estimates, Table 5 exact values):
  - AE long-run buoyancy:
    - PIT 0.907*** (0.015)
    - CIT 1.516*** (0.034)
    - TGS 0.951*** (0.008)
    - SSC 0.941*** (0.009)
  - AE short-run buoyancy:
    - PIT 0.798*** (0.110)
    - CIT 2.936*** (0.240)
    - TGS 0.873*** (0.051)
    - SSC 0.600*** (0.102)
  - AE speed of adjustment: PIT -0.249*** (0.026); CIT -0.344*** (0.034); TGS -0.256*** (0.034); SSC -0.251*** (0.036)
  - EME long-run buoyancy:
    - PIT 1.036*** (0.016)
    - CIT 1.014*** (0.011)
    - TGS 1.000*** (0.009)
    - SSC 1.668*** (0.029)
  - EME short-run buoyancy:
    - PIT 1.211*** (0.204)
    - CIT 1.590*** (0.272)
    - TGS 1.027*** (0.120)
    - SSC 1.210*** (0.272)
  - EME speed of adjustment: PIT -0.407*** (0.062); CIT -0.379*** (0.054); TGS -0.359*** (0.047); SSC -0.024 (0.210)
  - LIC long-run buoyancy:
    - PIT 0.487*** (0.043)
    - CIT 0.887*** (0.048)
    - TGS 1.075*** (0.013)
    - SSC 1.024*** (0.034)
  - LIC short-run buoyancy:
    - PIT 0.932** (0.470)
    - CIT 1.491*** (0.493)
    - TGS 1.030*** (0.211)
    - SSC 1.041 (0.929)
  - LIC speed of adjustment: PIT -0.162** (0.063); CIT -0.251*** (0.056); TGS -0.353*** (0.035); SSC -0.416*** (0.087)
  - Observations and # Countries by component as in Table 5.
- Robustness checks and elasticity estimates:
  - Controlling for tax rates (Table 6):
    - PIT (AE) long-run no control 0.907*** (0.015); control 0.966*** (0.014). PIT short-run no control 0.798*** (0.110); control 0.974*** (0.135).
    - CIT (AE) long-run no control 1.516*** (0.034); control 1.585*** (0.047). CIT short-run no control 2.936*** (0.240); control 3.611*** (0.399).
    - TGS (AE) long-run no control 0.951*** (0.008); control 0.867*** (0.012). TGS short-run no control 0.873*** (0.051); control 0.890*** (0.104).
    - Controlling for tax rates generally increases estimated buoyancies (especially CIT and PIT in some groups).
  - Controlling for inflation (Table 7):
    - AE long-run buoyancy no control 1.000*** (0.008); control 0.996*** (0.009).
    - EME long-run no control 1.070*** (0.006); control 1.073*** (0.007).
    - LIC long-run no control 1.201*** (0.012); control 1.188*** (0.011).
    - Long-run price effects (inflation) significant and positive: AE 0.493*** (0.146); EME 0.821*** (0.144); LIC 0.323* (0.192).
    - Inclusion of inflation lowers buoyancy coefficients: tax buoyancy in real terms is smaller than in nominal terms.

### III. Tax Buoyancy Around Business Cycle Turning Points
- Event-study findings (recessions t, t±1, t±2; Figure 4 observations):
  - Overall tax revenues decrease immediately from t-1 to t across all country groups: short-run buoyancy appears > 1 during a recession.
  - EMEs and LICs: tax revenues recover by t+1; AEs: tax revenues remain subdued and recover more slowly.
  - Across tax categories:
    - In AEs, lagged and slow recovery mainly in PIT and TGS.
    - SSC tends to increase in the year of the recession across all groups.
- GFC (2008–2012) specific findings:
  - Buoyancy estimates generally higher during 2008–2012 than average (Table A4/A5 referenced in source).
  - In AEs, PIT and TGS long-run buoyancy coefficients larger during GFC and significantly > 1.
  - Short-run buoyancy of all taxes enhanced during the Great Recession in both AEs and EMEs; in LICs the contribution came from CIT and TGS.
- Asymmetric short-term buoyancy over the business cycle (PMG estimates, Table 8 exact values):
  - Total tax revenue short-run buoyancy:
    - AE Expansion 1.087*** (0.050); Contraction 0.950*** (0.051)
    - EME Expansion 1.213*** (0.075); Contraction 1.183*** (0.080)
    - LIC Expansion 1.485*** (0.249); Contraction 0.652*** (0.182)
  - Component-level asymmetric results (selected exact values):
    - AE CIT: Expansion 1.096*** (0.293); Contraction 2.038*** (0.305)
    - EME CIT: Expansion 0.886*** (0.244); Contraction 2.348*** (0.321)
    - EME TGS: Expansion 0.754*** (0.109); Contraction 1.253*** (0.123)
  - Interpretation:
    - In AEs only CIT buoyancy is larger during contractions than expansions: CIT better stabilizes output during bad times.
    - In EMEs both CIT and TGS show larger buoyancy during contractions.

### IV. Determinants of Tax Buoyancy
- Approach:
  - Use country-specific average determinants and cross-country regression with estimated short-run buoyancies as dependent variable (Weighted Least Squares; weights inverse of standard errors of buoyancy estimates).
  - Determinant groups: Structural (share of agriculture, share of mining, trade openness), Demographic (population density, human capital), Macroeconomic (inflation, output volatility), Institutions (political rights, civil liberties).
  - Cross-section sample up to 107 observations.
- Main findings (Table 9 — total taxes, key coefficient signs and significance):
  - Trade openness increases tax buoyancy: e.g., specification (4) trade openness 0.349*** (0.087); specification (7) trade openness 0.447*** (0.078).
  - Human capital increases tax buoyancy: specification (6) human capital 0.190*** (0.064).
  - Inflation decreases tax buoyancy: specification (5) inflation -0.102* (0.059).
  - Output volatility decreases tax buoyancy: specification (5) output volatility -0.055*** (0.018); specification (7) output volatility -0.064*** (0.018).
  - Population density and share of agriculture coefficients vary by specification; share of agriculture generally negative but not always significant.
- Determinants by tax category (Table 10):
  - PIT buoyancy:
    - Human capital positive and significant: specification (2) human capital 1.044*** (0.339).
    - Output volatility negative: specification (1) -0.045*** (0.017).
    - Civil liberties and political rights positive in some specifications: civil liberties 0.313** (0.121); political rights 0.243* (0.124).
  - CIT buoyancy:
    - Human capital positive and significant: specification (4) human capital 1.272*** (0.355).
    - Output volatility negative and significant: specification (4) output volatility -0.192** (0.093).
    - Trade openness positive and marginal: specification (4) trade openness 0.818* (0.450).
  - TGS buoyancy:
    - Population density positive and significant: specification (6) population density 0.033*** (0.010).
    - Human capital positive and significant: specification (6) human capital 0.162* (0.084).
    - Share of agriculture negative and significant in specification (6): -0.118* (0.070).
    - Output volatility negative and significant: specification (6) -0.076*** (0.021).

### V. Conclusion and Policy Implications
- Aggregate empirical findings:
  - Long-run buoyancy of total revenue is not different from one across country groups on average.
  - Short-run buoyancy:
    - Not statistically different from one in Advanced Economies.
    - Statistically larger than one in Emerging Markets and Low Income Countries.
  - By tax type:
    - Long-run buoyancy > 1 for CIT in AEs; PIT and SSC in EMEs; TGS in LICs.
    - CIT exhibits strong short-run buoyancy across country groups (stabilizing role).
  - Tax buoyancy is not neutral to inflation: nominal buoyancy > real buoyancy.
  - During downturns:
    - In AEs, only CIT buoyancy increases during contractions relative to expansions.
    - In EMEs, both CIT and TGS buoyancy increase during contractions.
  - Determinants:
    - Trade openness and human capital increase tax buoyancy.
    - Inflation and output volatility decrease tax buoyancy.
- Policy implications (derived from empirical results):
  - Because long-run tax-to-GDP shares tend to remain constant, permanent increases in spending-to-GDP that do not improve structural conditions or human capital should be accompanied by revenue mobilization reforms to avoid permanent fiscal deterioration.
  - Reliance on CIT improves the stabilization role of the revenue system, especially during recessions, but declines in CIT collection as a share of GDP can be large and recovery may take longer.
  - Policies that enhance human capital and trade openness can raise buoyancy and thereby strengthen revenue responsiveness to growth.
  - Macroeconomic stabilization (reducing inflation and output volatility) supports higher tax buoyancy and hence more reliable revenue mobilization.

*Source: wp1704 - Appendix Tables (IMF working paper content provided).*

### Appendix Tables

### Appendix Tables

### Overall tax buoyancy — Advanced Economies (Table A1a)
- Long run buoyancy and short run buoyancy reported for each country with associated speed of adjustment. Examples (preserve exact reported entries):
  - Australia: 0.953*** 1.004*** -0.153
  - Austria: 1.000*** 0.565** -0.382***
  - Belgium: 1.013*** 0.965*** -0.209*
  - Canada: 0.550** 1.179*** -0.095***
  - Cyprus: 1.446*** 2.415*** -0.635***
  - Czech Republic: 1.016*** 1.230*** -0.660**
  - Denmark: 1.017*** 0.898*** -0.219**
  - Estonia: 1.019*** 0.841*** -0.931***
  - Finland: 1.007*** 0.947*** -0.180**
  - France: 1.054*** 1.317*** -0.093**
  - Germany: 1.422*** 1.921*** -0.136**
  - Greece: 1.953*** 1.048*** -0.332**
  - Iceland: 0.929*** 0.828*** -0.125
  - Ireland: 0.911*** 1.045*** -0.365***
  - Israel: 0.696*** 1.471** -0.614**
  - Italy: 0.807*** 0.593** -0.170***
  - Japan: 0.864*** 1.054*** -0.154
  - Korea: 1.098*** 1.347*** -0.792***
  - Luxembourg: 0.459 0.809*** -0.053
  - Netherlands: -0.757 0.957*** -0.021
  - New Zealand: 7.792 2.016*** 0.006
  - Norway: 3.872 1.590*** 0.021
  - Portugal: 1.295*** 1.993*** -0.308***
  - Singapore: 0.840*** 1.164*** -0.209
  - Slovak Republic: 0.798*** 1.095*** -0.547**
  - Slovenia: 1.017*** 1.290*** -0.854***
  - Spain: 0.491 1.396*** -0.056
  - Sweden: 0.947*** 1.177*** -0.181*
  - Switzerland: 1.090*** 0.752** -0.531***
  - United Kingdom: 1.006*** 1.234*** -0.226**
  - United States: 1.4072.652***-0.024
- Aggregate statistics:
  - Mean (Long run buoyancy): 1.262
  - Mean (Short run buoyancy): 1.203
  - Median (Long run buoyancy): 1.007
  - Median (Short run buoyancy): 1.095
  - Standard Deviation (Long run buoyancy): 1.437
  - Standard Deviation (Short run buoyancy): 0.438
- Note: Significance notation preserved from source: Bold italic means statistically greater than one at 5 percent level; bold means statistically not different from one at 5 percent level. *, **, *** denote statistical significance at the 10, 5 and 1 percent levels, respectively.

### Overall tax buoyancy — Emerging Market Economies (Table A1b)
- Country-level entries (selected examples, exact formatting preserved):
  - Algeria: 1.211*** 1.849*** -0.494***
  - Angola: 1.140*** 1.471*** -0.750**
  - Argentina: 1.144*** 1.407*** -0.546**
  - Azerbaijan: 0.981*** 1.459*** -0354**
  - Belarus: 0.751 1.131** -0.08
  - Brazil: 0.976*** 1.306*** -0.206
  - Chile: 1.157*** 1.939*** -0.579***
  - China: 1.238*** 0.656*** -0.065
  - Colombia: 1.249*** 1.558*** -0.720***
  - Croatia: 1.033*** 1.302*** -0.791**
  - Dominican Republic: 1.068*** 0.887*** -0.473**
  - Ecuador: 1.383*** 0.861*** -0.689***
  - Egypt: 0.974*** 1.628*** -0.650**
  - Hungary: 0.922*** 0.431 -0.780***
  - India: 1.104*** 1.668*** -0.669***
  - Indonesia: 1.109*** 0.998*** -0.491**
  - Iran: 1.028*** 0.33 -0.509***
  - Kazakhstan: 1.111*** 2.023*** -0.663***
  - Kuwait: 0.958*** 1.997*** -0.469**
  - Libya: 0.072 0.933*** -0.214
  - Mexico: 1.001*** 0.624* -0.481**
  - Morocco: 1.197*** 0.604* -0.484**
  - Oman: 1.272*** 0.241 -0.479***
  - Pakistan: 0.981*** 0.909*** -0.815***
  - Peru: 1.142** 2.049*** -0.430**
  - Philippines: 1.064*** 2.267*** -0.224
  - Poland: 1.071*** 2.205*** -0.661
  - Qatar: 1.666*** 0.389 -0.983***
  - Romania: 1.068*** 1.185*** -0.678
  - Russia: 1.109*** 1.326*** -0.394
  - Saudi Arabia: 0.799*** 0.919*** -0.838***
  - South Africa: 1.072*** 1.610*** -0.434*
  - Thailand: 1.208*** 1.780*** -0.371**
  - Turkey: 1.176***1.554***-0.931***
  - Ukraine: 1.120*** 1.114*** 0.536***
  - United Arab Emirates: 1.347*** 2.888*** -0.186
  - Uruguay: 1.007*** 1.231*** -0.329
  - Venezuela: 1.039*** 0.897*** -0.582**
- Aggregate statistics:
  - Mean (Long run buoyancy): 1.06
  - Mean (Short run buoyancy): 1.31
  - Median (Long run buoyancy): 1.07
  - Median (Short run buoyancy): 1.23
  - Standard Deviation (Long run buoyancy): 0.25
  - Standard Deviation (Short run buoyancy): 0.65
- Note: Significance notation preserved as in source.

### Overall tax buoyancy — Low Income Countries (Table A1c)
- Country-level entries (selected examples, exact formatting preserved):
  - Bolivia: 1.281***1.732***-0.337
  - Bangladesh: 1.198*** 0.777 -0.544***
  - Benin: 1.206*** 0.932*** -0.345**
  - Burkina Faso: 1.435*** 0.301 -0.572**
  - Cameroon: 1.001*** 3.814** -0.749***
  - Chad: 1.638*** 0.021 -0.670***
  - Congo, Dem Rep: 1.360*** 0.524*** -0.760***
  - Congo, Rep: 1.089*** 0.372* -0.321*
  - Cote Ivoire: 1.036*** 1.874*** -0.931***
  - Ethiopia: 1.320*** 0.455* -0.284**
  - Ghana: 1.156*** 0.913 -0.521**
  - Guinea: 1.233*** 1.350*** -0.446***
  - Haiti: 1.301*** 0.974*** -0.405**
  - Honduras: 1.031*** 0.977*** -0.815***
  - Kenya: 1.133*** 1.078* -0.361**
  - Kyrgyz: 1.181*** 0.442 -0.162
  - Lao: 1.349*** 0.489 -0.311**
  - Madagascar: 1.039*** 2.252*** -0.681***
  - Mali: 0.925*** 0.325 -0.403**
  - Moldova: 1.022*** 0.779** -0.467**
  - Mozambique: 1.089*** 0.878*** -0.18
  - Myanmar: 1.358*** 0.212 -0.26
  - Nepal: 1.409*** 1.515*** -0.742**
  - Nicaragua: 1.223*** 1.421*** -0.742*
  - Niger: 1.528*** 0.889*** -0.751***
  - Nigeria: 0.864*** 0.228 -0.900***
  - Papua New Guinea: 1.121*** 1.093** -0.520**
  - Rwanda: 1.275*** 2.200*** -0.074
  - Senegal: 1.193*** 1.987*** -0.942**
  - Sudan: 0.598** 0.350** -0.209
  - Tajikistan: 1.134*** 1.279*** -0.931***
  - Tanzania: 1.194*** 2.312*** -0.683***
  - Uganda: 1.184*** 0.258 -0.285
  - Uzbekistan: 0.904*** 0.697*** -0.894***
  - Vietnam: 0.906*** 1.309*** -0.247
  - Yemen: 0.957*** 0.809*** -0.703***
  - Zambia: 1.756 2.565** -0.135
  - Zimbabwe: 3.011* 5.451*** -0.567
- Aggregate statistics:
  - Mean (Long run buoyancy): 1.23
  - Mean (Short run buoyancy): 1.27
  - Median (Long run buoyancy): 1.16
  - Median (Short run buoyancy): 0.91
  - Standard Deviation (Long run buoyancy): 0.42
  - Standard Deviation (Short run buoyancy): 1.19
- Note: Significance notation preserved as in source.

### First-generation panel unit root tests (Table A2)
- Im, Pesaran and Shin (2003) Panel Unit Root Test (IPS):
  - Reported t-bar statistics (average of country-specific “ideal” lag-augmentation via AIC). Selected reported entries (preserve exact formatting):
    - AE: GDP lags 0.90 [t-bar] 2.36; Tax revenue 1.13 1.09; PIT 0.76 -1.33*; CIT 0.90 -0.69; TGS 1.13 0.65; SSC 0.78 -0.95
    - EME: 1.45 -9.52*** 1.34 -2.80*** 1.45 -2.53*** 0.80 -0.76 1.40 (other entries as in source)
    - LIC: 0.82 -9.61*** 0.95 -4.61*** 1.47 -1.31* 0.94 -2.17** 0.91 -1.85** 1.71
  - Note: Critical values (-1.73 for 5%, -1.69 for 10%) referenced from IPS paper; cases where null is rejected indicated with *** in source.
- Maddala and Wu (1999) Panel Unit Root Test (MW):
  - MW statistics and p-values reported for different lag augmentations (0, 1, 2). Selected reported entries:
    - AE (lags 0): GDP 97.06 (0.00); Tax revenue 154.67 (0.00); PIT 113.46 (0.00); CIT 52.14 (0.46); TGS 124.54 (0.00); SSC 132.27 (0.00)
    - AE (lags 1): GDP 42.740 (0.81); Tax revenue 56.95 (0.296); PIT 98.41 (0.00); CIT 68.69 (0.06); TGS 51.71 (0.48); SSC 73.99 (0.02)
    - EME and LIC panels: full MW tables with exact lambda and p-values preserved in source (examples: EME lags 0 GDP 101.74 (0.00); LIC lags 0 GDP 0.30 (0.85), etc.)
  - Notes: All variables are in logarithms. MW statistic constructed via sum of log p-values; distribution )2(2 Nχ under null. P-values reported in parentheses in source.

### Second-generation panel unit root tests (Table A3)
- Pesaran (2007) Panel Unit Root Test (CIPS):
  - CIPS statistics and p-values reported for lags 0, 1, 2 for AE, EME, LIC across variables GDP, Tax revenue, PIT, CIT, TGS, SSC. Selected reported entries:
    - AE (lags 0): GDP 3.627 (1.00); Tax revenue 1.58 (0.94); PIT 1.52 (0.93); CIT -1.11 (0.13); TGS 1.16 (0.87); SSC 1.62 (0.94)
    - AE (lags 1): GDP 2.453 (0.99); Tax revenue 0.47 (0.68); PIT -0.45 (0.32); CIT 0.48 (0.68); TGS 0.92 (0.82); SSC 1.63 (0.94)
    - EME (lags 1): GDP -2.82 (0.00); Tax revenue 0.10 (0.54); PIT 1.55 (0.94); CIT -1.89 (0.03); TGS 1.08 (0.86); SSC 0.53 (0.70)
    - LIC entries: lags 0,1,2 reported with associated p-values (examples: LIC lags 0 GDP 3.17 (0.20); Tax revenue 0.32 (0.85))
  - Notes: All variables in logarithms. Null hypothesis is of nonstationarity.

### Overall tax buoyancy during the Global Financial Crisis (Table A4)
- Panel estimations reported for country groups AE, EME, LIC using MG and PMG estimators; estimation of Equation (3) by PMG estimator noted in source.
- Reported coefficients and standard errors (preserve exact formatting):
  - AE MG Long run buoyancy: -0.129 (0.801)
  - AE PMG Long run buoyancy: 1.102*** (0.002)
  - EME MG Long run buoyancy: 1.193*** (0.171)
  - EME PMG Long run buoyancy: 1.705*** (0.065)
  - LIC MG Long run buoyancy: -1.870 (2.855)
  - LIC PMG Long run buoyancy: 0.596*** (0.003)
- Short run buoyancy (examples):
  - AE MG: 1.249*** (0.124)
  - AE PMG: 1.190*** (0.113)
  - EME MG: 2.171*** (0.291)
  - EME PMG: 1.734*** (0.207)
  - LIC MG: 1.709*** (0.612)
  - LIC PMG: 1.573*** (0.295)
- Speed of adjustment (examples):
  - AE MG: -0.887*** (0.179)
  - AE PMG: -0.765*** (0.172)
  - EME MG: -0.266 (0.302)
  - EME PMG: -0.253*** (0.080)
  - LIC MG: -0.369** (0.186)
  - LIC PMG: -0.126 (0.101)
- Sample sizes:
  - Observations: AE 155 (MG and PMG), EME 190 (MG and PMG), LIC 190 (MG and PMG)
  - # Countries: AE 31, EME 38, LIC 38
- Note: Standard errors in parenthesis. Bold italic means statistically greater than one at 5 percent level; bold means statistically not different from one at 5 percent level. *, **, *** denote statistical significance at the 10, 5 and 1 percent levels, respectively.

### Buoyancy of tax revenue components during the Global Financial Crisis (Table A5)
- Advanced Economies (AE), Emerging Market Economies (EME), and Low Income Countries (LIC) reported separately for PIT, CIT, TGS, SSC with Long run buoyancy, Short run buoyancy, Speed of adjustment, Observations, and # Countries. Selected reported values (exact formatting preserved):
- AE:
  - Long run buoyancy: 1.698*** (PIT), 0.799*** (CIT), 1.625*** (TGS), 0.820*** (SSC)
  - Short run buoyancy: 1.033*** (PIT), 3.541*** (CIT), 1.219*** (TGS), 0.626*** (SSC)
  - Speed of adjustment: -0.553*** (PIT), -0.454*** (CIT), -0.423*** (TGS), -0.439*** (SSC)
  - Observations: PIT 145, CIT 145, TGS 150, SSC 140
  - # Countries: PIT 29, CIT 29, TGS 30, SSC 27
- EME:
  - Long run buoyancy: 1.077*** (PIT), 0.889*** (CIT), 1.196*** (TGS), - (SSC not reported)
  - Short run buoyancy: 1.364*** (PIT), 1.541*** (CIT), 1.347*** (TGS), -
  - Speed of adjustment: -0.722*** (PIT), -0.717*** (CIT), -0.779*** (TGS), -
  - Observations: PIT 110, CIT 125, TGS 140, SSC -
  - # Countries: PIT 22, CIT 25, TGS 28, SSC -
- LIC:
  - Long run buoyancy: 1.632 (PIT), -6.163 (CIT), 0.902*** (TGS), 1.060*** (SSC)
    - Standard errors: (1.601) (7.499) (0.141) (0.209)
  - Short run buoyancy: 0.460 (PIT), 2.339*** (CIT), 1.506*** (TGS), 1.369 (SSC)
    - Standard errors: (1.428) (0.694) (0.421) (0.963)
  - Speed of adjustment: -0.869*** (PIT), -1.203*** (CIT), -0.948*** (TGS), -0.840*** (SSC)
    - Standard errors: (0.294) (0.242) (0.226) (0.224)
  - Observations: PIT 88, CIT 87, TGS 160, SSC 40
  - # Countries: PIT 17, CIT 17, TGS 32, SSC 7
- Note: Estimation of Equation (3) by PMG estimator (see main text for details). Bold italic means statistically greater than one at 5 percent level; bold means statistically not different from one at 5 percent level. Standard errors in parenthesis. *, **, *** denote statistical significance at the 10, 5 and 1 percent levels, respectively.

*Source: wp1704 - Appendix Tables.*

### References

### References

### Works cited

- Agbeyegbe, T. D., J. G. Stotsky, and A. Wolde-Mariam (2004). Trade liberalization, Exchange Rate Changes, and Tax Revenue in Sub-Saharan Africa. IMF Working Paper 04/178, International Monetary Fund.  
- Ahmad, E and N. Stern (1991), The Theory and Practice of Tax Reforms in Developing Countries, Cambridge University Press. 
- Ahsan, S.M. and Wu, S. (2005). Tax Structure and Reform in China, 1979-2002. Mimeo Department of Economics, Concordia University, Canada. 
- Ariyo, A. (1997), “Productivity of the Nigerian Tax System: 1970 – 1990”, Research paper No. 67. Nairobi: African Economic Research Consortium. 
- Bahl, R.W. (2003), “Reaching the Hardest to Tax: Consequences and Possibilities”, paper presented at the “Hard to Tax: An International Perspective” conference, Andrew Young School of Policy Studies, Georgia State University, May 15-16  
- Belinga V., D. Benedek, R. A. de Mooij and J. Norregaard (2014), “Tax Buoyancy in OECD Countries”, IMF Working Paper 110  
- Bilquees, F. (2004), “Elasticity and Buoyancy of the Tax System in Pakistan”, Pakistan Development Review, 43(1), 73-93  
- Blanchard, O., Dell'Aricia, G., and Mauro P. (2010), “Rethinking Macroeconomic Policy”. IMF Staff Position Note, SPN/10/03. 
- Bouthevillan, C., P. Cour-Thimann, G. Van den Dool, P. Hernandez de Cos, G. Langenus, M. Mohr, S. Momigliano and M. Tujula (2001) “Cyclically Adjusted Budget Balances: An Alternative Approach”, ECB Working Paper No. 77. 
- Bruce, Donald, William F. Fox, and M.H. Tuttle (2006). “Tax Base Elasticities: A Multi-State Analysis of Long-Run and Short-Run Dynamics,” Southern Economic Journal, 73(2), 315-341. 
- Chelliah, R.J., Baas, H.J., and Kelly, M. R. (1975). Tax Ratios and Tax Effort in Developing Countries, 1969-71. IMF Staff Papers, 22(1), 187-205  
- Chipeta, C. (1998). “Tax Reform and Tax yield in Malawi”, AERC Research Paper No. 81. Nairobi: AERC 
- Choudhry, N.N. (1979), “Measuring the Elasticity of Tax Revenue: A Divisia Index Approach”. IMF Staff Paper, 26. 
- Cotton, J.J. (2012), “The Buoyancy and Elasticity of Non-Oil Tax Revenues in Trinidad and Tobago (1990-2009)”, Central Bank of Trinidad and Tobago Working Paper. 
- Ehrhart, H. (2009). Assessing the relationship between Democracy and Domestic taxes in developing countries. CERDI, Etudes et Documents, E.30  
- Evans, P. (1997), “How fast do economies converge”, Review of Economics and Statistics, 46, 1251-1271. 
- Furceri, D., Jalles, J. T. (2016), “Fiscal Stabilization: Determinants and Effects”, IMF Working Paper (forthcoming) 
- Ghura, D. (1998), ‘Tax revenue in Sub-Saharan Africa: Effects of Economic Policies and Corruption’, IMF Working Paper, WP/98/135. September. Internet version 
- Giorno, C., Richardson, P. and P. Van den Noord (1995), “Estimating Potential Output, Output Gaps and Structural Budget Balances”, OECD Economics Department Working Papers 24 
- Girouard, N. and C. Andre (2005), Measuring Cyclically-Adjusted Budget Balances for the OECD Countries, OECD Economics Department Working Papers 434. 
- Granger, C.W.J. and T. Terasvirta (1993). “Modelling Nonlinear Economic Relationships”. Oxford University Press. 
- Hindrichs, H.H. (1966) “A General Theory of Tax Structure Change During Economic Development”, Harvard Law School International Tax Programme.  
- Im, K. S., Pesaran, M. H. and Shin, Y. (2003), “Testing for Unit Roots in Heterogeneous Panels,” Journal of Econometrics, 115, 53– 74. 
- International Monetary Fund (2015), “Current Challenges in Revenue Mobilization: Improving Tax Compliance”, IMF Board Paper. 
- Khattry, Barsha and J. Mohan Rao (2002), “Fiscal Faux Pas?: An Analysis of the Revenue Implications of Trade Liberalization,” World Development, Vol. 30, No. 8, pp. 1431–1444.  
- Kusi, N. K. (1998), “Tax Reform and Revenue Productivity in Ghana”, AERC Research Paper Number Seventy-four. Nairobi: AERC 
- Laeven, L. and Valencia, F. (2008), “Systemic Banking Crises: A New Database,” IMF Working Paper No. 08/224. 
- Lee, K., Pesaran, M. H. and Smith, R. P. (1997), “Growth and Convergence in a Multi-Country Empirical Stochastic Solow model”, Journal of Applied Econometrics, 12, 357-392. 
- Leuthold, J.H. (1991). Tax Shares in Developing Economies: A Panel Study, Journal of Development Economics, 35,173-185 
- Lotz, J.R. and Morss, E.R. (1967). Measuring ‘Tax Effort’ in Developing Countries. IMF Staff Papers, 14, 479-497  
- Maddala, G. S. and Wu, S. (1999), “A Comparative Study of Unit Root Tests with Panel Data and a New Simple Test”, Oxford Bulletin of Economics and Statistics, 61, 631-652. 
- Mahdavi, S. (2008). The Level and Composition of Tax Revenue in Developing Countries: Evidence from Unbalanced Panel Data. International Review of Economics and Finance, 17,607-617 
- Osoro, N.E. (1993), “Revenue Productivity Implications of Tax Reform in Tanzania”. Research Paper No. 20, Nairobi; African Economic Research Consortium.  
- Osoro, N.E. (1995), “Tax Reforms in Tanzania: Motivations, Directions and Implications”. Research paper No. 38. Nairobi: African Economic Research Consortium. 
- Pesaran, M. H. (2007), “A simple Panel Unit Root Test in the Presence of Cross Section Dependence”, Journal of Applied Econometrics, 22(2), 265-312. 
- Pesaran, M. H. and Smith, R. P. (1995), “Estimating Long-Run Relationship from Dynamic Heterogeneous Panels”, Journal of Econometrics, 68, 79-113. 
- Pesaran, M. H., Shin, Y. and Smith, R. P. (1999), “Pooled Mean Group Estimation of Dynamic Heterogeneous Panels”, Journal of American Statistical Association, 94, 31-54. 
- Reinhart, C and K. Rogoff (2011), “From Financial Crash to Debt Crisis,” American Economic Review, 101, 1676–1706. 
- Sancak, Cemile, Ricardo Velloso, and Jing Xing (2010), “Tax Revenue Response to the Business Cycle,” IMF Working Paper 10/71. 
- Skeete, R., Coppin, K. and Boamah, D. (2003), “Elasticities and Buoyancies of the Barbados Tax System, 1977 – 1999”. Central Bank of Barbados. 
- Stotsky, J.G. and A. Wolde-Mariam (1997), ‘Tax effort in Sub-Saharan Africa’, IMF Working Paper, WP/97/107, September. 
- Tanzi, V. (1987). Quantitative Characteristics of the Tax System of Developing Countries. The Theory of Taxation for Developing Countries, New York: Oxford University Press  
- Tanzi, V. and Zee, H., 2000, “Tax Policy for Emerging Markets: Developing Countries,” Working Paper WP/00/35, Washington, D.C.: International Monetary Fund 
- Teera, J. (2002): “Tax Performance: A Comparative Study”, Department of Economics, University of Bath, Working Paper No. 01/2002. 
- Upender, M. (2008), “Degree of Tax Buoyancy in India: An Empirical Study”, International Journal of Applied Econometrics and Quantitative Studies, 5(2)

*Source: wp1704 - References (IMF Working Paper references list).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/wp1704.pdf_