## wpiea2020143-print-pdf

## Source details

**Canonical URL:** [wpiea2020143-print-pdf](https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020143-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2020/english/wpiea2020143-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2020/english/wpiea2020143-print-pdf.pdf.json)

---

### Introduction: purpose and context
- Fragile and Conflict-Affected States (FCS) face interconnected economic and structural challenges including poor governance, weak institutional capacity, and political and social instability.
- By 2030, half of the world’s poor are estimated to live in fragile countries (Commission on State Fragility, Growth and Development, 2018).
- The inability of FCS to raise sufficient tax revenue is strongly associated with constraints in building resilience (Deléchat et al., 2018).
- Building tax capacity is critical for state building and development; increasing tax collection is important for economic growth and development (Besley and Persson, 2009; Gaspar et al., 2016a, 2016b).
- IMF (2017a) finding: 90 percent of countries that graduated from fragility had substantially improved their CPIA sub-score for “efficiency of revenue mobilization.”

### Key findings and motivations
- Enhancing tax collection is critical for post-COVID-19 recovery: public health expenditure is the number one priority (IMF Fiscal Monitor, April, 2020); many countries face falling fiscal revenues and rising public debt.
- IMF (2019b) estimate: delivering on the SDG agenda will require additional spending in 2030 of US$0.5 trillion (15 percentage points of GDP) for the low-income country group, which includes most FCS.
- Declining trend of aid flows to FCS increases the importance of domestic revenue mobilization to reduce aid dependency.

### Stylized facts on tax performance in FCS
- Overall tax performance in FCS has been constantly weaker than in other developing countries.
  - For the last decade, tax collection for FCS has on average hovered around 11.5–12.5 percent of GDP.
  - Compared with other developing countries, FCS tax revenues have been consistently lower by 4–6 percentage points of GDP.
- Tax composition differences:
  - FCS collect far less taxes on goods and services compared with non-FCS; this holds for both resource rich and non-resource rich countries.
  - Goods and services taxes are separated into VAT, excise taxes, and general (other) goods and services taxes in the analysis.
- Cross-country distribution by income:
  - About 80 percent of FCS are placed below the trend line of tax-to-GDP ratios for non-FCS peers of the same income level.
  - Some FCS deviate from the trend by more than 10 percentage points of GDP; some FCS are above the trend line, indicating fragility is not the sole determinant of tax performance.
- COVID-19 impact (projection):
  - Tax revenue in FCS is projected to decline from 12 percent of GDP to 11.3 percent of GDP in 2020, before recovering in 2021.
  - COVID-19 affects tax collection through slower economic activity, obstruction of administrative processes (many FCS lack automation), tax waivers and relief measures, and deteriorating taxpayer compliance.

### Recovery dynamics of tax revenues in FCS
- Tax recovery after the weakest revenue performance tends to be sustained up to 2–3 years.
- On average, the tax collection of FCS steadily improves by 3.5 percentage points of GDP from the lowest level, over a period of three years.
- Momentum for tax recovery is short-lived: after three years, FCS recover the level of tax collection, though the momentum of tax increases is not sustained over time.
- Some of the observed revenue increase may be associated with improved macroeconomic environment; causality between reforms and revenue increases cannot reliably be established given lack of additional micro-evidence or natural control groups.
- Confidence intervals around the recovery estimates are wide and results should be interpreted with caution.
- Historical shock evidence:
  - Ebola (2014–15): infected over 28,000 people and caused more than 11,000 deaths in Guinea, Liberia, and Sierra Leone.
  - Following Ebola, tax collections declined by about 1½ percentage points of GDP in these countries.
  - Recovery pace differed: Sierra Leone returned to pre-Ebola tax levels in 2016 (one year after the end of the outbreak), while Liberia and Guinea did not.

### Empirical analysis: scope, approach, and data
- Objectives:
  - Examine whether FCS are recovering tax revenues after troughs and identify determinants of tax revenues.
- Event study approach:
  - Fixed-effects panel used to capture dynamics of tax revenues during an 11-year window around tax revenue troughs (dummy variables for t+j where t is trough year).
  - Specification follows Gourinchas and Obstfeld (2012) methodology; conditional effects βj measure tax performance over the event window relative to tranquil times.
- Data and sample:
  - Focus on developments since the early 2000s; troughs defined as the year of the lowest tax-to-GDP ratio during 1995–2017.
  - Conflict periods (and three years after) removed where conflict incidents reported to impact tax revenue.
  - Annual data from 2000 to 2017.
  - Main data sources: IMF’s World Economic Outlook (WEO), FAD Revenue Analysis Tool (RAT), World Revenue Longitudinal Database (WoRLD).
  - Tax revenue data exclude grants and non-tax revenue.
  - Given small number of FCS (19), GMM was avoided.
  - Use of General government (GG) data where available; otherwise central government (CG) tax revenue used as proxy.

### Determinants of tax revenues — variables and rationale
- Structural variables: real GDP per capita growth, trade openness, net official aid, inflation (lagged), share of agriculture in total output.
- Social/public variable: government expenditure on education (lagged) as proxy for social spending and tax morale.
- Institutional variables: control of corruption, government effectiveness, political stability (Worldwide Governance Indicators).
- Income inequality: GINI index (lagged).
- Rationale:
  - Per capita GDP growth controls for business cycle impacts.
  - Trade openness captures degree of liberalization.
  - Lagged inflation gauges effect of prior-year macroeconomic policy.
  - Large agriculture shares and informality imply different tax regimes and narrower tax bases.
  - Government education spending relates to tax morale via higher education levels.

### Revenue-specific institutions and relation to general institutions
- Revenue institution index constructed from ISORA indicators related to risk management, public accountability, third party information, and limits on discretionary power; indicators averaged with equal weights.
- Cross-section analysis confirms a positive relationship between general institutions (government effectiveness, control of corruption) and revenue institutions.
- Two-thirds of FCS in 2015 were among countries with the highest corruption levels, per the Corruption Perceptions Index (Transparency International).
- Corruption reduces revenue collection through mechanisms such as tax loopholes in exchange for bribes and undervaluation or underdeclaration of goods at customs.

### Regression results (fixed-effect estimates — highlights and exact statistics)
- Improving institutional variables is more relevant in FCS than in non-FCS:
  - Control of corruption, government effectiveness, and political stability are important determinants of tax collection in FCS.
  - In non-FCS, these institutional variables are not significant.
- Other significant determinants:
  - Real GDP per capita growth is positive and statistically significant in both groups.
  - Share of agriculture in value added is statistically significant and negative in both groups, with larger negative effects in FCS.
  - Lagged consumer prices (inflation) have a slightly negative and statistically significant effect in FCS; inflation has no impact in non-FCS.
  - Government expenditure on education (lagged) is positive and significant in non-FCS but insignificant in FCS.
- Variables not significant in FCS:
  - Trade openness (positive sign but not statistically significant).
  - GINI index (income inequality) is not significant.
  - Net official aid is not significant.
- Selected exact regression outputs:
  - Observations: 298 (Fragile states), 1,201 (Nonfragile states).
  - Number of ifscode: 19 (Fragile states), 78 (Nonfragile states).
  - R-squared: 0.455 (Fragile states), 0.320 (Nonfragile states).
  - Example coefficient estimates (with significance and z-statistics as reported):
    - Control of corruption 0.183** (2.138) for fragile.
    - Government effectiveness 0.150* (1.691) for fragile.
    - Political stability 0.0713** (2.142) for fragile.
    - Real GDP per capita growth 0.00800*** (3.969) fragile and 0.00306*** (2.642) nonfragile.
    - Agriculture sector -0.0122*** (-3.787) fragile and -0.00480*** (-3.247) nonfragile.
    - Consumer prices (lag) -0.00239*** (-4.641) fragile and -0.000331 (-0.973) nonfragile.
  - Significance legend: *** p<0.01, ** p<0.05, * p<0.1.
- Robustness checks:
  - Hausman specification test rejected random effects in favor of fixed effects.
  - High correlation between institutional variables and GDP per capita addressed by estimating models without GDP per capita and with different control combinations; main results remain unchanged.
  - Interaction terms of institutional variables with FCS dummies confirm results: most interaction terms statistically significant while institutional variables without FCS dummies are not.

### Case evidence: sizable tax increases and lessons (summary)
- Four FCS examples (Liberia, Malawi, Nepal, Solomon Islands) achieved sizable increases in tax revenues:
  - Each achieved tax revenue increases of 7–20 percentage points of GDP over about a decade.
- Common lessons:
  - Target high-potential sectors (e.g., logging sector in Solomon Islands).
  - Start with reform steps that have immediate effects to build momentum.
  - Tax reforms can be pursued with initially weak institutions, but political commitment is crucial to sustain reform efforts (e.g., over ten years) for long-lasting, sizable gains.
- Table and figure references used in the source: Table 1 and Figures 1–10 (figures document tax revenues by type, income level, recovery patterns, and country-specific tax revenues and institutional indicators).

### Country case highlights

- Liberia
  - Post-conflict reforms: total tax revenues more than doubled from 6.7 percent of GDP in 2003 to 14.2 percent in 2012 following decade-long reforms.
  - Real GDP per capita (in 2005 prices) declined from US$1,269 in 1980 to US$163 in 2005.
  - In 2006, unemployment reportedly stood at over 80 percent; a similar percentage subsisted on less than US$1 a day.
  - Key measures: broadened GST base, reduced duty and tax exemptions on rice and petroleum, uniform excise tax of 15 percent on alcoholic beverages, excise on cigarettes increased to 15 percent, higher excises on cell phone usage, air travel, and hotels, reorganization by taxpayer segmentation, TIN re-registration under SIGTAS, Integrated Tax Administration System (ITAS), rollout of ASYCUDA at major ports.
  - Revenue outcome: Liberia’s tax revenues increased by 7.5 percent of GDP over nine years.
  - Drivers: increase in trade taxes by 3.33 percentage points of GDP in nine years, curbing exemptions, inflows from extractive industry.
  - Increase not fully sustained due to changes in income tax reporting base, legal issues in forestry sector (moratorium on timber exports end-2012), and lower imports.
  - Institutional improvements: control of corruption strengthened substantially initially; political stability index improved following election of new president.

- Malawi
  - Tax-to-GDP ratio increased by 11 percentage points during 2002 to 2013.
  - Creation of an LTO focused on around 400 large taxpayers who pay 70 percent of domestic taxes.
  - Modernization transformed MRA into an integrated, function-based, and segment-focused organization.
  - Political commitment and public support under the government of President Mutharika emphasized improving competitiveness and equity of the tax system.

- Nepal
  - Tax-to-GDP ratio increased from 9 percent of GDP in 2006 to nearly 22 percent of GDP in 2019.
  - Remittances from an estimated 2.2 million workers abroad (8 percent of the population) increased from 9 percent of GDP in 2000 to almost 30 percent of GDP in recent years, contributing to higher import-related taxes.
  - VAT outcomes: net VAT collection rose from 3½ percent of GDP in 2005/06 to 5¼ percent of GDP in 2013/14.
  - VAT C-efficiency improved from 0.28 in FY2005/06 to 0.42 in FY2012/13.
  - Key reforms: strengthened LTO, automation and e-filing (95 percent of LTO taxpayers filing electronically in 2010/2011), expanded registration, ETDS compulsory for all large taxpayers effective January 2007, customs automation (ASYCUDA), selectivity modules, Post Clearance Audit (PCA), presumptive turnover tax at 1.5 percent for firms between NPR 2 million and VAT threshold.

- Solomon Islands
  - Tax-to-GDP ratio almost tripled from 10 percent of GDP in 2001 to nearly 30 percent in 2013.
  - Revenue increases driven by logging-sector reforms, strengthening revenue administration, curbing exemptions, combating evasion, and improved trade revenue.
  - Initial one-off measures: tax amnesty and back taxes payment amounting to 1 percent of GDP.
  - Audit-assessed tax amounts increased from $11.2 million in 2006 to $191 million in 2011.
  - Evidence of compliance gaps: at one point only 30–40 percent of log shipments were inspected, leading to foregone tax revenues of nearly 4 percent of GDP in 2007; revenue forgone from import duty exemptions alone was about 4 percent of GDP; revenue forgone from log duty exemptions in 2004 was 1.5 percent of GDP.
  - Increase not sustained due to economic slowdown and adverse terms-of-trade shocks (decline in logging output and prices).

### Policy implications and recommendations
- Priorities for FCS to improve tax performance:
  - Strengthen institutional quality incrementally; even marginal improvements can yield significant revenue gains.
  - Pursue sustainable reform efforts with strong political commitment to extend reforms beyond short 2–3 year windows.
  - Focus on easier tax handles (e.g., taxes on goods and services and excises) that concentrate collection points—customs border posts and large companies—while working toward broader-based taxation.
  - After COVID-19, lift temporary tax measures and strengthen compliance to rebuild tax bases.
  - Prepare revenue administrations to reinvigorate operations and restore trade and tax processes once containment measures are loosened.
- Fiscal sustainability and development:
  - Mobilizing domestic revenue is essential to fund increased spending needs, meet SDGs, and safeguard debt sustainability given projected additional spending needs (US$0.5 trillion, 15 percentage points of GDP for low-income countries by 2030).
- Implementation guidance:
  - Start with measures with immediate effects to build momentum (reforming indirect taxes, curbing exemptions, focusing on large/high-value taxpayers, establishing an LTO, enhancing risk-based audit).
  - As capacity builds, pursue a Medium-Term Revenue Strategy (MTRS) approach when moving from immediate post-conflict/disaster stage to a more stable institutional stage.
  - Technical assistance should include organizational design, operational frameworks, large taxpayer risk strategies, and IT systems to support taxpayer identification and information exchange.
  - International partners’ technical assistance and capacity development are instrumental; political leadership facilitates coordination and implementation.

*Source: wpiea2020143-print-pdf*

### REFERENCES ___________________________________________________________________________________ 38

### wpiea2020143-print-pdf - REFERENCES ___________________________________________________________________________________ 38

### Introduction: purpose and context
- Fragile and Conflict-Affected States (FCS) face interconnected economic and structural challenges including poor governance, weak institutional capacity, and political and social instability.
- By 2030, half of the world’s poor are estimated to live in fragile countries (Commission on State Fragility, Growth and Development, 2018).
- The inability of FCS to raise sufficient tax revenue is strongly associated with constraints in building resilience (Deléchat et al., 2018).
- Building tax capacity is critical for state building and development; increasing tax collection is important for economic growth and development (Besley and Persson, 2009; Gaspar et al., 2016a, 2016b).
- IMF (2017a) finding: 90 percent of countries that graduated from fragility had substantially improved their CPIA sub-score for “efficiency of revenue mobilization.”

### Key findings and motivations
- Enhancing tax collection is critical for post-COVID-19 recovery: public health expenditure is the number one priority (IMF Fiscal Monitor, April, 2020); many countries face falling fiscal revenues and rising public debt.
- IMF (2019b) estimate: delivering on the SDG agenda will require additional spending in 2030 of US$0.5 trillion (15 percentage points of GDP) for the low-income country group, which includes most FCS.
- Declining trend of aid flows to FCS increases the importance of domestic revenue mobilization to reduce aid dependency.

### Stylized facts on tax performance in FCS
- Overall tax performance in FCS has been constantly weaker than in other developing countries.
  - For the last decade, tax collection for FCS has on average hovered around 11.5–12.5 percent of GDP.
  - Compared with other developing countries, FCS tax revenues have been consistently lower by 4–6 percentage points of GDP.
- Tax composition differences:
  - FCS collect far less taxes on goods and services compared with non-FCS; this holds for both resource rich and non-resource rich countries.
  - Goods and services taxes are separated into VAT, excise taxes, and general (other) goods and services taxes in the analysis.
- Cross-country distribution by income:
  - About 80 percent of FCS are placed below the trend line of tax-to-GDP ratios for non-FCS peers of the same income level.
  - Some FCS deviate from the trend by more than 10 percentage points of GDP; some FCS are above the trend line, indicating fragility is not the sole determinant of tax performance.
- COVID-19 impact (projection):
  - Tax revenue in FCS is projected to decline from 12 percent of GDP to 11.3 percent of GDP in 2020, before recovering in 2021.
  - COVID-19 affects tax collection through slower economic activity, obstruction of administrative processes (many FCS lack automation), tax waivers and relief measures, and deteriorating taxpayer compliance.
- Recovery patterns after shocks:
  - Ebola (2014–15) example: infected over 28,000 people and caused more than 11,000 deaths in Guinea, Liberia, and Sierra Leone.
  - Following Ebola, tax collections in these countries declined significantly (by about 1½ percentage points of GDP).
  - Pace of recovery differed: Sierra Leone returned to pre-Ebola tax levels in 2016 (one year after the end of the outbreak), while Liberia and Guinea did not maintain momentum.
  - One-off factors (e.g., arrears payments from mineral companies, capital gains tax from sale of a mineral company) partly explain Sierra Leone’s 2016 increase.

### Empirical analysis: scope and approach
- Objectives:
  - Examine whether FCS are recovering tax revenues after troughs and identify determinants of tax revenues.
- Data and sample considerations:
  - Focus on developments since the early 2000s due to data limitations in earlier years.
  - Troughs defined as the year of the lowest tax-to-GDP ratio during 1995–2017.
  - Conflict periods (and three years after) removed where conflict incidents reported to impact tax revenue.
- Event study approach:
  - Fixed-effects panel used to capture dynamics of tax revenues during an 11-year window around tax revenue troughs (dummy variables for t+j where t is trough year).
  - Specification follows Gourinchas and Obstfeld (2012) methodology; conditional effects βj measure tax performance over the event window relative to tranquil times.
- Key empirical insights (summary of results reported):
  - Many FCS demonstrate some recovery of tax revenues after the peak of fragility, but reform efforts often last only 2–3 years and fail to catch up with non-fragile countries.
  - Changes in the quality of institutions are a key contributory factor to tax performance in FCS—more so than in non-FCS—implying that even marginal institutional improvements could bring large revenue increases.
  - Data challenges: securing credible data for countries under fragile environments is technically challenging; WEO database crosschecked with IMF staff reports as needed.

### Case evidence: sizable tax increases and lessons
- Four FCS examples (Liberia, Malawi, Nepal, Solomon Islands) achieved sizable increases in tax revenues:
  - Each achieved tax revenue increases of 7–20 percentage points of GDP over about a decade.
  - Specific lessons from these cases:
    - Target high-potential sectors (e.g., logging sector in Solomon Islands).
    - Start with reform steps that have immediate effects to build momentum.
    - Tax reforms can be pursued with initially weak institutions, but political commitment is crucial to sustain reform efforts (e.g., over ten years) for long-lasting, sizable gains.
- Table and figure references (analysis draws from):
  - Table 1: Fragile and Conflict-Affected States Achieving Large Tax Increases.
  - Figures documenting tax revenues by type, income level, recovery patterns, and country-specific tax revenues and institutional indicators (e.g., Figures 1–10 and subpanels).

### Policy implications and recommendations (drawn from analysis)
- Priorities for FCS to improve tax performance:
  - Strengthen institutional quality incrementally; even marginal improvements can yield significant revenue gains.
  - Pursue sustainable reform efforts with strong political commitment to extend reforms beyond short 2–3 year windows.
  - Focus on easier tax handles (e.g., taxes on goods and services and excises) that concentrate collection points—customs border posts and large companies—while working toward broader-based taxation.
  - After COVID-19, lift temporary tax measures and strengthen compliance to rebuild tax bases.
  - Prepare revenue administrations to reinvigorate operations and restore trade and tax processes once containment measures are loosened.
- Fiscal sustainability and development:
  - Mobilizing domestic revenue is essential to fund increased spending needs, meet SDGs, and safeguard debt sustainability given projected additional spending needs (US$0.5 trillion, 15 percentage points of GDP for low-income countries by 2030).

*Source: wpiea2020143-print-pdf - REFERENCES ___________________________________________________________________________________ 38*

### 11.      The results suggest that tax

### 11.      The results suggest that tax

### Recovery dynamics of tax revenues in FCS
- Tax recovery after the weakest revenue performance tends to be sustained up to 2–3 years.
- On average, the tax collection of FCS steadily improves by 3.5 percentage points of GDP from the lowest level, over a period of three years.
- Momentum for tax recovery is short-lived: after three years, FCS recover the level of tax collection, though the momentum of tax increases is not sustained over time.
- Some of the observed revenue increase may be associated with improved macroeconomic environment; causality between reforms and revenue increases cannot reliably be established given lack of additional micro-evidence or natural control groups.
- Confidence intervals around the recovery estimates are wide and results should be interpreted with caution.

### Determinants of tax revenues — literature and selection of variables
- Prior studies identified structural, economic, and institutional determinants:
  - Structural factors: per capita GDP, agriculture share in GDP, ratio of imports to GDP, foreign aid (Gupta, 2007; Ghura, 1998).
  - Institutional factors: corruption, government effectiveness, regulatory quality, rule of law, voice and accountability (Imam and Jacobs, 2007; Epaphra and Massawe, 2017; Cukierman et al., 1992).
- This study selects independent variables based on the literature:
  - Structural variables: real GDP per capita growth, trade openness, net official aid, inflation (lagged), share of agriculture in total output.
  - Social/public variables: government expenditure on education (lagged) as proxy for social spending and tax morale.
  - Institutional variables: control of corruption, government effectiveness, political stability (Worldwide Governance Indicators).
  - Income inequality measured by the GINI index (lagged).
- Rationale highlights:
  - Per capita GDP growth controls for business cycle impacts.
  - Trade openness captures degree of liberalization.
  - Lagged inflation gauges effect of prior-year macroeconomic policy.
  - Large agriculture shares and informality in FCS imply different tax regimes and narrower tax bases.
  - Government education spending relates to tax morale via higher education levels.

### Empirical approach and data
- Econometric specification: fixed-effects panel
  - y_{i,t} = α_i + λ_t + β X_{i,t} + ε_{i,t}
  - y_{i,t} is the log of the tax-to-GDP ratio for country i at year t.
  - X_{i,t} is a vector of independent variables; α_i and λ_t are country and time fixed effects.
- Sample: annual data from 2000 to 2017 (data constrained due to limited availability for FCS prior to 2000).
- Main data sources: IMF’s World Economic Outlook (WEO), FAD Revenue Analysis Tool (RAT), World Revenue Longitudinal Database (WoRLD).
- Tax revenue data exclude grants and non-tax revenue.
- Given the small number of FCS (19), GMM was avoided because it can produce biased estimates in this context.
- General government (GG) data used where available; otherwise central government (CG) tax revenue used as proxy.

### Revenue-specific institutions and their relation to general institutions
- Revenue institution index constructed from ISORA indicators related to risk management, public accountability, third party information, and limits on discretionary power; indicators averaged with equal weights.
- Cross-section analysis confirms a positive relationship between general institutions (government effectiveness, control of corruption) and revenue institutions.
- In FCS, poor governance of revenue agencies has coincided with weak administrative practices and impeded tax collection.
- Two-thirds of FCS in 2015 were among countries with the highest corruption levels, per the Corruption Perceptions Index (Transparency International).
- Corruption reduces revenue collection through mechanisms such as tax loopholes in exchange for bribes and undervaluation or underdeclaration of goods at customs.

### Regression results (fixed-effect estimates — highlights)
- Improving institutional variables is more relevant in FCS than in non-FCS:
  - Control of corruption, government effectiveness, and political stability are important determinants of tax collection in FCS.
  - In non-FCS, these institutional variables are not significant, suggesting a nonlinear relation and different marginal impacts depending on initial institutional capacity.
- Other significant determinants:
  - Real GDP per capita growth is important for both FCS and non-FCS (positive and statistically significant).
  - Share of agriculture in value added is statistically significant and negative in both groups, with larger negative effects in FCS.
  - Lagged consumer prices (inflation) have a slightly negative and statistically significant effect in FCS; inflation has no impact in non-FCS.
  - Government expenditure on education (lagged) is positive and significant in non-FCS but insignificant in FCS.
- Variables not significant in FCS:
  - Trade openness (positive sign but not statistically significant in FCS).
  - GINI index (income inequality) is not a significant determinant in FCS.
  - Net official aid is not significant in FCS; suggests large aid does not necessarily imply aid dependency that demotivates tax reform.
- Selected exact regression outputs (columns summarized for fragile and nonfragile states):
  - Observations: 298 (Fragile states), 1,201 (Nonfragile states).
  - Number of ifscode: 19 (Fragile states), 78 (Nonfragile states).
  - R-squared: 0.455 (Fragile states), 0.320 (Nonfragile states).
  - Coefficient magnitudes and statistical significance (z-statistics in parentheses) shown in the source table (see variables such as: Control of corruption 0.183** (2.138) for fragile; Government effectiveness 0.150* (1.691) for fragile; Political stability 0.0713** (2.142) for fragile; Real GDP per capita growth 0.00800*** (3.969) fragile and 0.00306*** (2.642) nonfragile; Agriculture sector -0.0122*** (-3.787) fragile and -0.00480*** (-3.247) nonfragile; Consumer prices (lag) -0.00239*** (-4.641) fragile and -0.000331 (-0.973) nonfragile).
  - Significance legend: *** p<0.01, ** p<0.05, * p<0.1.
- Robustness checks:
  - Hausman specification test rejected random effects in favor of fixed effects.
  - High correlation between institutional variables and GDP per capita addressed by estimating models without GDP per capita and with different control combinations; main results remain unchanged.
  - Interaction terms of institutional variables with FCS dummies confirm results: most interaction terms statistically significant while institutional variables without FCS dummies are not.

### Policy implications and next steps
- Improving governance and political stability is crucial to effective tax reforms in FCS; institutional strengthening has larger marginal benefits where initial conditions are very weak.
- Given the strong association between general institutions and revenue-specific institutions, improving general governance can support revenue administration improvements.
- Future investigation should focus on how FCS can strengthen institutions (reflected in governance indicators) to increase tax collection and draw lessons from successful FCS tax reform episodes.

### Case study selection and high-level lessons
- Four FCS episodes with sizable, sustained increases in tax revenues examined: Liberia, Malawi, Nepal, Solomon Islands.
- Selection criteria:
  - Episodes identified in Akitoby et al. (2018) as substantial tax revenue mobilization.
  - Countries achieved increases in tax-to-GDP ratios beyond 7 percentage points and sustained most of it over a long period.
- Two high-level lessons from the case studies:
  - Tax reforms can be pursued even with initially weak institutions.
  - Strong political commitment is important to sustain reform efforts (e.g., over ten years) and realize long-lasting, sizable gains.

### Liberia — background (case study start)
- Post-conflict reform context:
  - Peace agreement signed in 2003 after more than a decade of internal conflicts.
  - Total tax revenues more than doubled from 6.7 percent of GDP in 2003 to 14.2 percent in 2012 following decade-long reforms in tax and customs administration and tax policy, backed by political commitment and technical support.
- Economic and social conditions after conflict:
  - Real GDP per capita (in 2005 prices) declined from US$1,269 in 1980 to US$163 in 2005 (almost 90 percent decline).
  - In 2006, unemployment reportedly stood at over 80 percent.
  - A similar percentage of the population subsisted on less than US$1 a day.
  - Liberia’s human capital had eroded and public institutions were largely dysfunctional.
  - The tax base was severely eroded by tax concessions, poor administration, low tax morale, and governance problems.
  - Tax structure heavily dependent on trade and income taxes; domestic sales and excise taxes contributed very little.

*Source: wpiea2020143-print-pdf.*

### 23.      With the newly elected President Ellen Johnson-Sirleaf, economic reforms gained

### 23.      With the newly elected President Ellen Johnson-Sirleaf, economic reforms gained

### Background and international support
- Until 2005, pace of reform implementation under the former government was slow and governance problems re-emerged.
- In early 2006, the newly elected President Ellen Johnson-Sirleaf expressed commitment for economic reforms and called for IMF support to develop a program to strengthen economic performance.
- A staff-monitored program (SMP) in 2006 paved the way toward a Fund-supported financial program and the eventual HIPC debt relief in 2008.
- Liberia’s key international partners (US, EU, UN, and ECOWAS) jointly formulated the Governance and Economic Management Assistance Program (GEMAP) in 2005 to strengthen economic governance and financial management, and rebuild Liberia’s key economic institutions.
- IMF technical support began after the Accra Peace Agreement in 2003 with a basic short-term and medium-term actions list to increase government revenue.
- In 2011, the IMF launched a multi-year technical support program financed by the Tax Policy and Administration Topical Trust Fund (TPA-TTF).

### Reform sequencing and capacity building
- Initial focus: short-term measures to broaden the tax base, increase indirect taxes, and strengthen revenue administration.
- Technical support was coordinated with other donors and delivered in phases across tax policy and administration topics to create a tax system based on sound economic principles, broad tax bases, and limited discretionary tax policy.
- Capacity-building constraints in early years included lack of facilities, equipment, and logistics (computers, vehicles, office facilities, stationery), especially in large and medium taxpayer divisions.
- Reforms later moved to more fundamental measures to increase tax collection from domestic-oriented activities and further strengthen tax and customs administration.

### Key tax policy measures implemented
- Expanding the taxable base for the goods and services tax (GST) by:
  - applying it on all goods and services;
  - expanding definition of import duty paid value to include c.i.f. value plus import duty, fees, and excises paid;
  - expanding the definition of domestic supplies on the sales price to include excise taxes.
  - While broadening the tax base, annual turnover for entities subject to the GST was set at a high level to ease compliance costs and administrative burden.
- Reducing duty and tax exemptions on rice and petroleum; the government rationalized duty exemptions, drafted legislation for tax exemptions policy in 2007, and passed the revenue authority law.
- Utilizing excise taxes:
  - a uniform excise tax of 15 percent was applied for imported and domestic alcoholic beverages;
  - the excise tax on cigarettes was increased to 15 percent;
  - higher excises were levied on cell phone usage, air travel, and hotels.
- Strengthening capacity to classify and value goods for customs purposes.
- Reorganizing domestic tax revenue administration on the basis of taxpayer segmentation and strengthening the large taxpayers unit, including audit capacity.
- Enhancing transparency of resource flows and the extractive industry (National Port Authority, Roberts International Airport, Liberia Petroleum Refining Corporation, Forestry Development Agency, Bureau of Maritime Affairs).

### Tax administration and IT reforms
- Reorganization by taxpayer segmentation (large, medium, small) with separate operational units for taxpayer services, returns and payments, collection enforcement, and audit.
- Re-registration and issuance of new Tax Identification Numbers (TINs) under the Standard Integrated Government Tax Administration System (SIGTAS) project to improve information sharing and coordination between tax and customs.
- Computerization of Bureau of Internal Revenue (BIR) operations via the Integrated Tax Administration System (ITAS).
- Customs reforms: rollout of ASYCUDA at major entry ports, development of a strategic plan for the Bureau of Customs and Excise (BCE), reorganization of BCE structure, establishment of a one-stop customer care center.

### Revenue outcomes and fiscal impact
- Liberia’s tax revenues increased by 7.5 percent of GDP over nine years.
- Revenue increases were driven by:
  - an increase in trade taxes by 3.33 percentage points of GDP in nine years, attributed to customs reforms;
  - curbing of tax exemptions;
  - inflows from the extractive industry.
- The TIN re-registration exercise improved the capacity to identify taxpayers and process information for risk mitigation strategies.
- The increase in tax revenue was not sustained, partly reflecting:
  - changes in the income tax reporting base from presumptive to actual;
  - legal issues in the forestry sector (the government imposed a moratorium on timber exports at end-2012 following abuses in the allocation of permits in the forestry sector);
  - lower imports.

### Institutional and governance indicators
- During the reform period, control of corruption was strengthened substantially at an initial period, with gradual improvement in government effectiveness.
- There was substantial improvement in the political stability index following the election of the new president.
- Liberia’s experience highlights importance of strong leadership, properly-sequenced reform steps, and capacity building by international partners; IMF (2017b) notes that in FCS, reform can be sustained through properly sequenced and innovative capacity building approaches championed by strong government leadership.

### Lessons and implications
- Strong political drive for tax reforms—supported by technical support from international partners—was a major factor in sustaining reforms.
- Proper sequencing: begin with measures that broaden the tax base and strengthen administration, then progress to systemic policy and IT-driven reforms.
- Technical assistance should include organizational design, operational frameworks, large taxpayer risk strategies, and IT systems to support taxpayer identification and information exchange.
- Addressing governance issues (e.g., reducing exemptions, increasing transparency in revenue-generating agencies) is critical for improving tax performance in fragile and conflict-affected settings.

*Source: wpiea2020143-print-pdf - 23.      With the newly elected President Ellen Johnson-Sirleaf, economic reforms gained*

### 41.      MRA utilized extensive technical support from the IMF. Technical support was

### MRA utilized extensive technical support from the IMF

### Technical support provided
- Technical support was provided in the areas of tax policy and administration.
- Short-term expert assignments assisted the MRA to:
  - build the DTD institutions, including the headquarters and the launch of the LTO;
  - complete an initial business process review (BPR) exercise;
  - institute project management arrangements;
  - develop an IT strategy;
  - plan for implementation of taxpayer self-assessment; and
  - improve capacity for customs administration.

### Malawi — Revenue outcomes and modernization impact
- Tax-to-GDP ratio increased by 11 percentage points during the period of 2002 to 2013.
- Improvements in tax revenue were partly owing to robust economic growth and progress in delivering the modernization agenda, as well as policy changes.
- Notable improvements were in income tax and indirect tax (VAT and excise) collections, with improved VAT efforts helping to broaden the tax base.
- Tax administration modernization transformed the MRA into an integrated (direct and indirect taxes), function-based, and segment-focused organization.
- Creation of an LTO provided focus on around 400 large taxpayers who pay 70 percent of domestic taxes and their particular needs and risks, including those from international trade and investment and the growing importance of mining.
- Political commitment and public support under the government of President Mutharika emphasized improving the competitiveness and equity of the tax system and strengthening domestic revenue mobilization.
- The MRA had a strong and well-skilled internal affairs directorate that supported revenue mobilization efforts.
- Managing the political environment, ensuring stakeholder commitment, and addressing governance issues helped Malawi pursue revenue reforms that yielded gains in tax collection.

### Nepal — Background
- Since a major political transition in 2006, Nepal has been undergoing a gradual transition to a federal democratic state following the end of the decade-long civil conflict in 2006.
- Nepal’s history of civil conflict, political instability, poor governance, and weak structural reform reduced economic growth; per capita GDP has barely increased and remains the lowest in the region.
- Remittances from an estimated 2.2 million workers abroad (8 percent of the population) increased from 9 percent of GDP in 2000 to almost 30 percent of GDP in recent years.
- The dramatic increase in remittances contributed to higher tax collection on imports and domestic taxes.
- Poverty declined from over 50 percent of the population in 2003/04 to 24 percent in 2010/11.
- Enhanced revenue collection was deemed critical to ensure budget sustainability and helped keep public debt to GDP on a declining path.
- Nepal received extensive capacity development support for tax policy and administration from the IMF and other donors for the IRD and DOC.
- Nepal experienced marked improvement in its political stability index during the reform period, while indices for control of corruption and government effectiveness remained somewhat stable (but declined compared to the early 2000s).

### Nepal — Reform strategy
- Donors (including DANIDA, GIZ, USAID, ADB, and World Bank) and the IMF provided capacity development to address weaknesses in revenue administration.
- The IRD implemented the Strategic Plan (2012/13 to 2016/17) and the Reform Plan (2012/13 to 2014/15), followed by the Second Strategic Plan for 2018/19 to 2022/23.
- The DOC implemented the Customs Reform and Modernization Strategies and Action Plan (CRMSAP) for 2013 to 2017; the current (fifth) CRMSAP covers 2017 to 2021.
- Key reform actions included:
  - Strengthening the Large Taxpayer Office (LTO) through automation, computerization for data sharing, and improved audit approaches for large taxpayers.
  - Strengthening taxpayer records and information management systems: ETDS made compulsory for all large taxpayers effective January 2007; computerization led to 95 percent of LTO taxpayers filing electronically for the year 2010/2011; expanded registration base during 2010–2011; expanded e-services for registration and filing; internal monitoring system with annual targets.
  - Modernizing tax administration and collecting tax arrears: introduction of self-assessment; new Taxpayer Service Offices (implemented 2010–2012); automation including e-filing and more payment options; performance measurement and monitoring systems; staff incentive scheme; targeted arrears collection increased collections although overall debt outstanding rose.
  - Enhancing computerization and strengthening customs operations: WAN between Kathmandu headquarters and five largest customs offices by mid-March 2007; automation within ASYCUDA; streamlined customs clearance at Birgunj; selectivity module (risk-based clearance in six offices); amendments to the Customs Act; strengthened customs valuation processes and Post Clearance Audit (PCA).
  - Addressing excise leakages by fully operationalizing a new monitoring regime.
- On tax policy, contributing measures included higher customs duty rates on selected imports, higher excise taxes (alcohol, beer, cigarettes), an increase in the VAT threshold from NPR 2 million to NPR 5 million, and a presumptive turnover tax at 1.5 percent for companies with taxable turnover above NPR 2 million but below the VAT threshold.

### Nepal — Revenue outcome and VAT performance
- Tax-to-GDP ratio increased from 9 percent of GDP in 2006 to nearly 22 percent of GDP in 2019.
- Revenue gains were driven by higher customs duty rates on selected imports, a broadening of the excise tax base, reforms in customs administration that increased import-related taxes, and a significant rise in imports fueled by remittances.
- The LTO’s full operationalization and increased audits strengthened enforcement for large taxpayers; the IRD adopted a risk-based audit approach.
- VAT collection increased: net VAT collection rose from 3½ percent of GDP in 2005/06 to 5¼ percent of GDP in 2013/14.
- VAT C-efficiency improved from 0.28 in FY2005/06 to 0.42 in FY2012/13, likely reflecting enhanced tax administration capacity.

### Solomon Islands — Background and revenue increase
- Tax collection increased from 10 percent of GDP in 2001 to nearly 30 percent in 2013.
- Economic and social conditions deteriorated after ethnic violence in 1999; real GDP shrank for four consecutive years through 2002, domestic revenues declined by 10 percent of GDP between 1999 and 2002, and the overall budget deficit increased from 4 percent of GDP in 1999 to 11 percent of GDP in 2002.
- From 2003 economic activity recovered driven by the logging industry: real GDP growth grew by 5 percent in 2003; by 2007 the logging industry accounted for 65 percent of exports and almost 20 percent of tax revenue.
- The Regional Assistance Mission to the Solomon Islands (RAMSI) helped maintain peace and order after 2003 and supported institutional capacity building, including assistance to the Solomon Islands Inland Revenue Division (IRD).
- Indices for control of corruption and political stability improved during the reform period; governance effectiveness improved after 2001 but later fluctuated.

### Solomon Islands — Reform strategy and measures
- Initial revenue increases were driven by temporary measures, including a tax amnesty and payment of back taxes amounting to 1 percent of GDP.
- Tax reforms focused on better targeting logs, curbing tax exemptions, and combating tax evasion.
- Measures included:
  - Enhancing tax collection from logging through increasing reference prices and improving tax compliance; an automatic adjustment mechanism for log prices was adopted at end-September 2010.
  - Improving the tax exemptions process: requiring public disclosure of exemptions and revenue loss; moving exemption grants to a formalized public service committee in 2007–2008; from 2013 the public service committee reviewed exemption requests and recommended to the Minister of Finance who cannot overrule recommendations—this reduced the number and value of exemptions.
  - Selected increases in import duties and excises; revenue from new gold production also contributed.
  - Laying foundations for a functional revenue administration by increasing staff in IRD and customs, implementing a targeted amnesty, focusing on arrears collection, strengthening taxpayer registration, and enforcement actions against large taxpayers.
  - Enhancing auditing and enforcement: audit-assessed tax amounts increased from $11.2 million in 2006 to $191 million in 2011; the Customs Valuation Act (approved in 2010) strengthened inspection procedures, implemented WTO valuations, introduced post-clearance audit, and increased penalties for duty evasion.
- Evidence indicated that only 30–40 percent of log shipments were inspected at one point, resulting in foregone tax revenues of nearly 4 percent of GDP in 2007; authorities made efforts to enhance monitoring and curb revenue losses.
- Revenue forgone from import duty exemptions alone was about 4 percent of GDP; revenue forgone from log duty exemptions alone in 2004 was 1.5 percent of GDP.

*Source: wpiea2020143-print-pdf - 41.      MRA utilized extensive technical support from the IMF. Technical support was*

### 62.      Solomon Islands tax-to-GDP ratio almost tripled from 10 percent of GDP in 2001 to

### Solomon Islands: Tax Revenue Mobilization and Reforms (excerpt)

### Overview and outcomes
- Solomon Islands tax-to-GDP ratio almost tripled from 10 percent of GDP in 2001 to nearly 30 percent in 2013.
- The substantial increase in tax revenue was mainly driven by:
  - Tax reforms in the logging sector.
  - Strengthening revenue administration and enforcement actions, which led to higher trade revenue.
  - Curbing tax exemptions, combatting tax evasion, and improving transparency.
- The increase in tax revenue was not sustained as a result of the economic slowdown and adverse terms-of-trade shocks, in particular a decline in logging output and prices (logging tax revenue fell).

### Tax administration actions and compliance gains
- Tax administration efforts made gains through:
  - Bringing businesses back into the tax net.
  - Expansion in taxpayer registration.
  - Collection of tax arrears.
  - Aggressive enforcement, audit, and collection efforts in customs administration.
- These measures supported improved tax compliance and higher collection during the reform period.

### Institutional and political factors
- Tax reform efforts were sustained with strong political commitment and support under RAMSI.
- Government commitment focused on improvements in revenue administration and continual measures to improve the tax system and revenue collection.
- Reforms in the logging industry and improved monitoring increased tax compliance in that sector.
- An improvement in tax exemptions supported revenue mobilization efforts.
- Political commitment, together with these measures, helped secure gains in tax revenue.

### Policy lessons and recommendations
- Sustained tax reform efforts are critical, especially for Fragile and Conflict-affected States (FCS):
  - Most FCS fail to sustain tax increases beyond three years; exiting a fragility trap requires sustained efforts over a longer period.
  - Case studies show sizable gains over a decade, achievable even with changes in administration (example: Malawi).
- Institutional improvements have particularly large impacts on tax performance in FCS:
  - Even marginal improvements in quality of institutions (as reflected in institutional indices) could lead to large revenue increases.
  - Tax reforms can be pursued with initially weak institutions but require political commitment to sustain efforts (e.g., over ten years).
- Key high-level lessons for tax revenue reforms:
  - Reforms in both revenue administration and tax policy are important. Initial focus often on building administrative capacity; a well-designed tax system plus effective administration ensures stable yields.
  - Collecting taxes from a country’s high potential sector leads to efficient tax collection (example: consumption and remittances in Nepal; logging sector in Solomon Islands).
  - Start with reform steps with immediate effects to build momentum (examples: reforming indirect taxes such as excises; curbing exemptions; focusing on large/high-value taxpayers; establishing an LTO; enhancing risk-based audit).
  - As capacity is built, a more structured Medium-Term Revenue Strategy (MTRS) approach can be applicable—especially when a state has moved from immediate post-conflict/disaster stage to a more stable (but still vulnerable) institutional stage.
- Sustaining reforms requires strong political commitment supported by international partners:
  - Political leadership facilitates coordination and implementation (examples: Liberia and Nepal).
  - Technical assistance (TA) and capacity development from IMF and development partners were instrumental in building tax capacity.
- Post-COVID-19 emphasis:
  - Following the COVID-19 outbreak, many FCS experienced a drop in domestic revenue through multiple channels.
  - Authorities should use this period to enhance and sustain reform efforts to more than offset the decline in domestic revenue and achieve steady gains over a sustained period.

### Notable quantitative references from the chapter
- Tax-to-GDP increase example: from 10 percent of GDP in 2001 to nearly 30 percent in 2013.
- Many FCS that sustained reforms achieved tax collection increases as large as 7.5–20 percent of GDP over 9–13 years.
- Most FCS fail to sustain tax increases beyond three years.
- Medium-Term Revenue Strategy (MTRS) concept referenced in footnote tied to a July 2016 G20 Finance Ministers meeting report.

*Source: IMF Working Paper (chapter excerpt).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020143-print-pdf.pdf_