## 1. Economy’s Structure and Export Diversification

## Source details

**Canonical URL:** [1. Economy’s Structure and Export Diversification](https://www.imf.org/-/media/files/publications/selected-issues-papers/2026/english/sipea2026067.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/selected-issues-papers/2026/english/sipea2026067.pdf.md)
- [Structured JSON version](/-/media/files/publications/selected-issues-papers/2026/english/sipea2026067.pdf.json)

---

### Introduction and key policy needs
- Growth averaged around 5.4 percent since 2017, following the democratic transition in 2017.
- GDP per capita increased from about US$600 in 2015 to approximately US$940 in 2024.
- Key policy needs identified:
  - strengthen governance and institutional capacity;
  - maintain macroeconomic stability through prudent fiscal and monetary policies;
  - raise agricultural productivity;
  - foster private sector–led growth via improvements in the business environment and access to finance;
  - promote economic diversification;
  - invest in human capital;
  - address critical infrastructure gaps.

### Background: structural features and vulnerabilities
- Human development and demographics:
  - 2024 UN HDI rank: 171st out of 191 countries; HDI approximately 0.52 versus SSA average about 0.55.
  - Population growth: 2.5 percent per annum; one third below age 15; 6 percent over age 65.
  - Average family size increased relative to a decade ago.
- Economic structure:
  - Agriculture share in GDP value added: from 35 percent in 2010 to 22 percent in 2023.
  - High volatility: standard deviation of nominal and real GDP growth between 2000-2024 is among the highest in the region.
- Climate and environment:
  - Severe flooding July‑August 2022 affected more than 50,000 people, displaced over 7,000 in Banjul Area.
  - 2024 rainy season floods killed at least 11 people, displaced over 5,000.
  - Climate shocks constrained agriculture and tourism, increased fiscal demands for disaster response and resilience.

### Empirical analysis — Growth accounting (1995–2024)
- Methodology:
  - Cobb-Douglas production: Yt = At Kt^α Lt^(1−α).
  - Capital stock via perpetual inventory: Kt = (1−δ)Kt−1 + It.
  - Initial capital stock set to equal 0.7 times real GDP.
  - Depreciation rate δ set to 5 percent per year.
  - Elasticity of capital to output α assumed 0.3.
  - Data: real GDP, total investment, and labor from the World Bank.
- Main findings:
  - Real GDP growth over 1995-2024 was mostly driven by factor accumulation (labor and physical capital); TFP contributed little on average.
  - Human capital accumulation consistently provided the most important contribution to real GDP growth.
  - Physical capital’s contribution was low in the 1990s, increased in the 2000s with large public investments.
  - TFP contributions have been volatile and often negative, indicating lack of sustained productivity gains.

### Empirical analysis — VAR (2003Q1–2024Q4)
- Model setup:
  - Endogenous variables: bank credit to private sector, total remittances, foreign aid, FDI, real GDP, inflation rates, lending rates.
  - Exogenous variable: real effective exchange rate.
  - Data frequency: quarterly; lag order chosen as two.
  - Variables: FDI, remittances, and credit expressed as ratios to GDP where appropriate; log transformations applied as needed.
- Impulse response results (accumulated IRFs, generalized ordering):
  - Growth response to FDI shocks:
    - Positive but statistically insignificant on impact.
    - Becomes positive after approximately four quarters, peaks between six and eight quarters, then dissipates.
  - Growth response to remittance shocks:
    - Positive and immediate.
    - Strongest within first two quarters; fades after approximately four to six quarters.
  - Growth response to credit shocks:
    - Positive but delayed; noticeable after two to four quarters and modest in magnitude.
  - Transmission channels:
    - Remittance shocks are associated with an increase in private sector credit.
    - FDI shocks have weaker and less immediate effects on credit, consistent with enclave nature of some foreign investment.

### Event study: before and after 2017 regime change (pre: 2013–2016; post: 2018–2024; post* 2018–2019)
- Method:
  - Exclude event year 2017.
  - Data from World Bank 2026 Economic Development Indicators.
  - Statistical significance via Welch's two-sample t-test.
- Key comparative results (Difference = Post − Pre; t-stat and p-val reported):
  - Real Sector:
    - Real GDP growth (%) 1.87 → 5.18; Post* = 6.73; Difference +3.31*; t-stat −2.33; p-val 0.060.
    - Real GDP per capita (USD) 621.3 → 700.9; Post* = 670.5; Difference +79.6***; t-stat −5.34; p-val 0.001.
    - Gross investment (% GDP) 19.16 → 27.75; Post* = 22.77; Difference +8.59**; t-stat −3.20; p-val 0.021.
  - External Sector:
    - FDI net inflows (% GDP) 4.19 → 8.65; Post* = 4.41; Difference +4.46**; t-stat −3.14; p-val 0.012.
    - Remittances (% GDP) 10.05 → 20.63; Post* = 13.95; Difference +10.59***; t-stat −5.23; p-val 0.001.
    - Aid received (% GDP) increases noted below in Fiscal Sector.
  - Fiscal Sector:
    - Aid received (% GDP) 7.70 → 13.39; Post* = 12.26; Difference +5.70***; t-stat −5.68; p-val 0.001.
    - Education spending (% GDP) 2.06 → 2.80; Post* = 2.59; Difference +0.74***; t-stat −5.68; p-val 0.001.
    - Health spending (% GDP) 1.12 → 1.40; Post* = 1.17; Difference +0.29*; t-stat −2.03; p-val 0.082.
  - Governance Indicators (World Bank WGI scores; higher = better):
    - Government effectiveness −0.60 → −0.52; Difference +0.08**; t-stat −3.18; p-val 0.019.
    - Corruption control −0.90 → −0.32; Difference +0.58***; t-stat −23.25; p-val 0.000.
    - Rule of law −0.89 → −0.39; Difference +0.51***; t-stat −13.25; p-val 0.000.
    - Regulatory quality −0.64 → −0.69; Difference −0.05*; t-stat +2.09; p-val 0.073.
    - Voice & accountability −1.24 → −0.24; Difference +1.00***; t-stat −29.59; p-val 0.000.
    - Political stability −0.31 → +0.05; Difference +0.36***; t-stat −5.62; p-val 0.002.
  - Social Indicators:
    - School enrollment (%) 81.84 → 96.34; Difference +14.50***; t-stat −7.30; p-val 0.001.
    - Population growth (%) 2.78 → 2.34; Difference −0.44***; t-stat +6.45; p-val 0.005.
- Interpretation:
  - Post-2017 period associated with stronger economic growth, higher investment, increased external inflows (remittances, FDI, aid), greater spending on education and health, and significant governance improvements.
  - Inflation (GDP deflator) remained broadly stable: 7.17 → 7.30; Difference +0.13; t-stat −0.08; p-val 0.940.

### Potential output and growth scenarios
- Definitions and methods:
  - Potential growth estimated using growth accounting and multiple statistical filters (Hodrick-Prescott two-sided and one-sided, Christiano-Fitzgerald, Baxter-King, Hamilton, Butterworth, Bayesian).
  - Cross‑filter average used as summary measure.
- Growth accounting scenarios (labor growth assumed constant at 3 percent per year; public and private investment consistent with macroeconomic forecasts):
  - Productivity growth scenario → Estimated real GDP growth (average 2025-2030):
    - Very pessimistic −2%/year → 3.6
    - Pessimistic −1%/year → 4.1
    - Neutral 0%/year → 4.6
    - Optimistic +1%/year → 5.1
    - Very optimistic +2%/year → 5.6
- Statistical filters:
  - Cross-filter average potential growth estimated at about 4.7 percent on average over the last five years.
  - The projected 6 percent growth outturn in 2025 would imply growth temporarily above estimated potential, consistent with a cyclical upswing or favorable transitory factors.
  - Sustaining growth at or above 6 percent would require structural reforms and productivity gains to raise potential growth above recent estimated range.

### Panel data analysis: institutions and growth (69 LICs, 2001–2024)
- Data and methods:
  - Unbalanced panel for 69 LICs over 2001-2024.
  - Governance measures: government effectiveness, control of corruption, political stability, bureaucratic quality, rule of law, voice and accountability.
  - Estimators: fixed effects and system GMM to address unobserved heterogeneity and endogeneity.
- Results:
  - Positive and statistically significant relationship between government effectiveness and economic growth.
  - Control of corruption (β = 0.0418, p < 0.01) and bureaucratic quality (β = 0.0407, p < 0.01) have the strongest estimated effects.
  - Lagged dependent variable highly significant, indicating persistence.
  - Aid inflows and trade openness consistently improve real GDP per capita.
  - Political stability, investment, FDI, remittances, and inflation do not show statistically significant effects in these specifications.
- Application to The Gambia:
  - Government effectiveness improved from −0.923 in 2015 to −0.536 in 2024: a 0.387-point increase (approximately 42 percent relative to its 2015 level).
  - Based on System GMM estimates, this improvement is associated with an increase in real GDP per capita growth of approximately 0.015 percentage points in the short run.
  - Accounting for growth persistence, the implied long-run effect is approximately 0.68 percentage points (long-run multiplier caveated given high persistence).

### Annex I — Short- and Long-run Impact of Governance Indicators (method and point estimates)
- Estimator and specification details:
  - Blundell-Bond system GMM (xtdpdsys), one-step with robust VCE.
  - All regressors one-period lagged; AR and Sargan/Hansen tests suppressed under vce(robust).
  - Notes on significance: *** p<0.01, ** p<0.05, * p<0.10.
- Short-run illustrative calculation (Model (1)):
  - Government effectiveness coefficient = 0.0384.
  - Example: If government effectiveness rises from −0.5 to 0.5 (Δ = 1) then Δy_SS = 0.0384 × 1 = 0.0384.
- Long-run multiplier formula and example:
  - Lagged dependent variable 훼 = 0.9782.
  - Long-run multiplier = β/(1 − α).
  - For Government Effectiveness: β = 0.0384; 1 − α = 0.0218; β/(1 − α) = 0.0384 / 0.0218 = 1.761.
  - Interpretation: A permanent one-unit improvement in government effectiveness predicts a long-run increase of approximately 1.76 units in the dependent variable.
- Short- and long-run summary (authors' estimates):
  - Government effectiveness: β = 0.038, α = 0.978, Long-run effect = 1.761
  - Corruption: β = 0.042, α = 0.981, Long-run effect = 2.155
  - Rule of law: β = 0.034, α = 0.982, Long-run effect = 1.878
  - Bureaucratic quality: β = 0.041, α = 0.979, Long-run effect = 1.947
  - Voice & accountability: β = 0.020, α = 0.985, Long-run effect = 1.336
- System GMM point estimates (Sub-Saharan Africa sample, selected coefficients):
  - Government effectiveness: 0.0384*** (0.0126)
  - Corruption: 0.0418*** (0.0103)
  - Rule of law: 0.0340** (0.0141)
  - Bureaucratic quality: 0.0407*** (0.0125)
  - Voice & accountability: 0.0195** (0.0088)
  - Lagged dependent variable examples: 0.9782***, 0.9806***, 0.9819***, 0.9791***, 0.9854***.
  - Control example: Aid (% GDP) = 0.0024*** (0.0009) in system GMM (SSA).
  - Observations: 673; Countries: 37; Instruments: 261.
  - Wald chi² examples: 11939.30; 19802.30; 12315.54; 11705.05; 9891.58.

### Summary statistics (selected)
- Full sample (69 countries, 2003–2024), Panel A examples:
  - ln (GDP per capita): Obs = 1,754; Mean = 7.496; Std. Dev. = 0.953; Min = 5.493; Max = 9.748
  - Government effectiveness: Obs = 1,771; Mean = −0.611; Std. Dev. = 0.619; Min = −2.206; Max = 1.101
  - Corruption control: Obs = 1,771; Mean = −0.673; Std. Dev. = 0.519; Min = −1.806; Max = 1.240
  - Voice & accountability: Obs = 1,771; Mean = −0.569; Std. Dev. = 0.642; Min = −2.231; Max = 0.954
  - ln (Investment): Obs = 1,556; Mean = 22.738; Std. Dev. = 2.012; Min = 16.164; Max = 29.681
  - FDI (% GDP): Obs = 1,761; Mean = 3.573; Std. Dev. = 6.112; Min = −37.171; Max = 103.3
  - Trade openness (% GDP): Obs = 1,649; Mean = 68.76; Std. Dev. = 37.71; Min = 2.00; Max = 347.7
- Notes: All governance indicators are World Bank Worldwide Governance Indicators (WGI), ranging from −2.5 to +2.5. ln(Investment) and ln (CPI) are natural logarithms. Remittances (% GDP) computed as 100 × remittances received / nominal GDP.

### Interpretation and policy-relevant conclusions
- Core interpretation:
  - Governance indicators show statistically significant positive short-run coefficients and sizeable long-run multipliers once dynamic persistence in GDP per capita is accounted for.
  - Medium- to long-term growth prospects depend critically on structural transformation; productivity growth must become the main engine as contributions from capital and labor are expected to diminish.
- Specific policy priorities:
  - Maintaining macroeconomic stability:
    - fiscal policy anchored on gradual consolidation;
    - strengthen domestic revenue mobilization;
    - improved expenditure prioritization;
    - prudent public debt management;
    - monetary policy focused on price stability while supporting adequate credit provision to the private sector.
  - Raising agricultural productivity:
    - prioritize irrigation development;
    - improved access to inputs;
    - strengthened agricultural extension services;
    - adoption of climate-resilient practices.
  - Accelerating structural reforms for private sector–led growth:
    - regulatory simplification;
    - strengthened contract enforcement;
    - enhanced access to finance for SMEs.
  - Promoting economic diversification:
    - support value addition in agriculture, fisheries, and light manufacturing;
    - improve trade facilitation and compliance with regional and international standards.
  - Investments in human capital:
    - enhance efficiency and targeting of public spending on education, health, and skills development;
    - emphasize technical and vocational education aligned with labor market needs.
  - Addressing infrastructure gaps:
    - prioritize investments in energy, transport, and digital connectivity;
    - guide investments by rigorous project appraisal and align with medium-term fiscal and debt sustainability objectives.

*Source: sipea2026067 — 1. Economy’s Structure and Export Diversification (PDF).*

### 1. Economy’s Structure and Export Diversification ________________________________________ 4

### 1. Economy’s Structure and Export Diversification

### Chapter headings and figures
- 1. Economy’s Structure and Export Diversification ________________________________________ 4
- 2a. Impulse Response: Accumulated Impulse Responses of Real GDP to Generalized One S.D. Innovations __________________________________________________________________________ 8
- 2b. Impulse Response: Accumulated Impulse Responses of Credit to Private Sector to Generalized One S.D. Innovations  ________________________________________________________ 9

### Tables listed
- TABLES
  - 1. Event Study: Before and After the 2017 Regime Change _____________________________ 10
  - 2. Summary of GMM Estimates of Governance on GDP Per Capita (SSA) _______________ 15

*Source: sipea2026067 - 1. Economy’s Structure and Export Diversification (PDF).*

### References ______________________________________________________________________________ 17

### sipea2026067 - References ______________________________________________________________________________ 17

### Introduction
- The Gambia’s growth averaged around 5.4 percent since 2017, following the democratic transition in 2017.
- GDP per capita increased from about US$600 in 2015 to approximately US$940 in 2024.
- Key policy needs identified: strengthen governance and institutional capacity; maintain macroeconomic stability through prudent fiscal and monetary policies; raise agricultural productivity; foster private sector–led growth via improvements in the business environment and access to finance; promote economic diversification; invest in human capital; address critical infrastructure gaps.

### Background: structural features and vulnerabilities
- Human development and demographics:
  - 2024 UN HDI rank: 171st out of 191 countries; HDI approximately 0.52 versus SSA average about 0.55.
  - Population growth: 2.5 percent per annum; one third below age 15; 6 percent over age 65.
  - Average family size increased relative to a decade ago.
- Economic structure:
  - Agriculture share in GDP value added: from 35 percent in 2010 to 22 percent in 2023.
  - High volatility: standard deviation of nominal and real GDP growth between 2000-2024 is among the highest in the region.
- Climate and environment:
  - Severe flooding July‑August 2022 affected more than 50,000 people, displaced over 7,000 in Banjul Area.
  - 2024 rainy season floods killed at least 11 people, displaced over 5,000.
  - Climate shocks constrained agriculture and tourism, increased fiscal demands for disaster response and resilience.

### Empirical analysis — Growth accounting (1995–2024)
- Methodology:
  - Cobb-Douglas production: Yt = At Kt^α Lt^(1−α).
  - Capital stock via perpetual inventory: Kt = (1−δ)Kt−1 + It.
  - Initial capital stock set to equal 0.7 times real GDP.
  - Depreciation rate δ set to 5 percent per year.
  - Elasticity of capital to output α assumed 0.3.
  - Data: real GDP, total investment, and labor from the World Bank.
- Main findings:
  - Real GDP growth over 1995-2024 was mostly driven by factor accumulation (labor and physical capital); TFP contributed little on average.
  - Human capital accumulation consistently provided the most important contribution to real GDP growth.
  - Physical capital’s contribution was low in the 1990s, increased in the 2000s with large public investments.
  - TFP contributions have been volatile and often negative, indicating lack of sustained productivity gains.

### Empirical analysis — VAR (2003Q1–2024Q4)
- Model setup:
  - Endogenous variables: bank credit to private sector, total remittances, foreign aid, FDI, real GDP, inflation rates, lending rates.
  - Exogenous variable: real effective exchange rate.
  - Data frequency: quarterly; lag order chosen as two.
  - Variables: FDI, remittances, and credit expressed as ratios to GDP where appropriate; log transformations applied as needed.
- Impulse response results (accumulated IRFs, generalized ordering):
  - Growth response to FDI shocks:
    - Positive but statistically insignificant on impact.
    - Becomes positive after approximately four quarters, peaks between six and eight quarters, then dissipates.
  - Growth response to remittance shocks:
    - Positive and immediate.
    - Strongest within first two quarters; fades after approximately four to six quarters.
  - Growth response to credit shocks:
    - Positive but delayed; noticeable after two to four quarters and modest in magnitude.
  - Transmission channels:
    - Remittance shocks are associated with an increase in private sector credit.
    - FDI shocks have weaker and less immediate effects on credit, consistent with enclave nature of some foreign investment.

### Event study: before and after 2017 regime change (pre: 2013–2016; post: 2018–2024; post* 2018–2019)
- Method:
  - Exclude event year 2017.
  - Data from World Bank 2026 Economic Development Indicators.
  - Statistical significance via Welch's two-sample t-test.
- Key comparative results (Difference = Post − Pre; t-stat and p-val reported):
  - Real Sector:
    - Real GDP growth (%) 1.87 → 5.18; Post* = 6.73; Difference +3.31*; t-stat −2.33; p-val 0.060.
    - Real GDP per capita (USD) 621.3 → 700.9; Post* = 670.5; Difference +79.6***; t-stat −5.34; p-val 0.001.
    - Gross investment (% GDP) 19.16 → 27.75; Post* = 22.77; Difference +8.59**; t-stat −3.20; p-val 0.021.
  - External Sector:
    - FDI net inflows (% GDP) 4.19 → 8.65; Post* = 4.41; Difference +4.46**; t-stat −3.14; p-val 0.012.
    - Remittances (% GDP) 10.05 → 20.63; Post* = 13.95; Difference +10.59***; t-stat −5.23; p-val 0.001.
    - Aid received (% GDP) increases noted below in Fiscal Sector.
  - Fiscal Sector:
    - Aid received (% GDP) 7.70 → 13.39; Post* = 12.26; Difference +5.70***; t-stat −5.68; p-val 0.001.
    - Education spending (% GDP) 2.06 → 2.80; Post* = 2.59; Difference +0.74***; t-stat −5.68; p-val 0.001.
    - Health spending (% GDP) 1.12 → 1.40; Post* = 1.17; Difference +0.29*; t-stat −2.03; p-val 0.082.
  - Governance Indicators (World Bank WGI scores; higher = better):
    - Government effectiveness −0.60 → −0.52; Difference +0.08**; t-stat −3.18; p-val 0.019.
    - Corruption control −0.90 → −0.32; Difference +0.58***; t-stat −23.25; p-val 0.000.
    - Rule of law −0.89 → −0.39; Difference +0.51***; t-stat −13.25; p-val 0.000.
    - Regulatory quality −0.64 → −0.69; Difference −0.05*; t-stat +2.09; p-val 0.073 (only governance indicator that declined marginally).
    - Voice & accountability −1.24 → −0.24; Difference +1.00***; t-stat −29.59; p-val 0.000.
    - Political stability −0.31 → +0.05; Difference +0.36***; t-stat −5.62; p-val 0.002.
  - Social Indicators:
    - School enrollment (%) 81.84 → 96.34; Difference +14.50***; t-stat −7.30; p-val 0.001.
    - Population growth (%) 2.78 → 2.34; Difference −0.44***; t-stat +6.45; p-val 0.005.
- Interpretation:
  - Post-2017 period associated with stronger economic growth, higher investment, increased external inflows (remittances, FDI, aid), greater spending on education and health, and significant governance improvements.
  - Inflation (GDP deflator) remained broadly stable: 7.17 → 7.30; Difference +0.13; t-stat −0.08; p-val 0.940.

### Potential output and growth scenarios
- Definitions and methods:
  - Potential growth estimated using growth accounting and multiple statistical filters (Hodrick-Prescott two-sided and one-sided, Christiano-Fitzgerald, Baxter-King, Hamilton, Butterworth, Bayesian).
  - Cross‑filter average used as summary measure.
- Growth accounting scenarios (labor growth assumed constant at 3 percent per year; public and private investment consistent with macroeconomic forecasts):
  - Productivity growth scenario → Estimated real GDP growth (average 2025-2030):
    - Very pessimistic −2%/year → 3.6
    - Pessimistic −1%/year → 4.1
    - Neutral 0%/year → 4.6
    - Optimistic +1%/year → 5.1
    - Very optimistic +2%/year → 5.6
- Statistical filters:
  - Cross-filter average potential growth estimated at about 4.7 percent on average over the last five years.
  - The projected 6 percent growth outturn in 2025 would imply growth temporarily above estimated potential, consistent with a cyclical upswing or favorable transitory factors.
  - Sustaining growth at or above 6 percent would require structural reforms and productivity gains to raise potential growth above recent estimated range.

### Panel data analysis: institutions and growth (69 LICs, 2001–2024)
- Data and methods:
  - Unbalanced panel for 69 LICs over 2001-2024.
  - Governance measures: government effectiveness, control of corruption, political stability, bureaucratic quality, rule of law, voice and accountability.
  - Estimators: fixed effects and system GMM to address unobserved heterogeneity and endogeneity.
- Results:
  - Positive and statistically significant relationship between government effectiveness and economic growth.
  - Control of corruption (β = 0.0418, p < 0.01) and bureaucratic quality (β = 0.0407, p < 0.01) have the strongest estimated effects.
  - Lagged dependent variable highly significant, indicating persistence.
  - Aid inflows and trade openness consistently improve real GDP per capita.
  - Political stability, investment, FDI, remittances, and inflation do not show statistically significant effects in these specifications.
- Application to The Gambia:
  - Government effectiveness improved from −0.923 in 2015 to −0.536 in 2024: a 0.387-point increase (approximately 42 percent relative to its 2015 level).
  - Based on System GMM estimates, this improvement is associated with an increase in real GDP per capita growth of approximately 0.015 percentage points in the short run.
  - Accounting for growth persistence, the implied long-run effect is approximately 0.68 percentage points (long-run multiplier caveated given high persistence).

*Source: sipea2026067 - References (June 22, 2026).*

### Annex I).

### Annex I. Short- and Long-run Impact of Governance Indicators

### Methodology and Estimation Details
- Estimator: Blundell-Bond system GMM (xtdpdsys), one-step with robust VCE.
- All regressors are one-period lagged.
- Robust standard errors reported in parentheses in tables.
- AR and Sargan/Hansen tests suppressed under vce(robust).
- Notes on significance: *** p<0.01, ** p<0.05, * p<0.10.

### Short-run Impact (illustrative calculation using Model (1))
- Using Model (1): Government effectiveness coefficient = 0.0384.
- Interpretation: A one-unit increase in Government Effectiveness is associated with an immediate increase of 0.0384 units in the dependent variable.
- Example: If a country's government effectiveness score rises from −0.5 to 0.5: Δ퐺퐺 = 1 then Δ푦푦_푆푆푆푆 = 0.0384 × 1 = 0.0384.

### Long-run Impact (formula and example)
- Lagged dependent variable coefficient reported as 훼 = 0.9782 (from Model (1)).
- Long-run multiplier formula: β/(1 − α).
- For Government Effectiveness:
  - β = 0.0384
  - 1 − α = 1 − 0.9782 = 0.0218
  - β/(1 − α) = 0.0384 / 0.0218 = 1.761
- Interpretation: A permanent one-unit improvement in government effectiveness predicts a long-run increase of approximately 1.76 units in the dependent variable.

### Short- and Long-run Effects: Summary Table (authors' estimates)
- Government effectiveness: β = 0.038, α = 0.978, Long-run effect = 1.761
- Corruption: β = 0.042, α = 0.981, Long-run effect = 2.155
- Rule of law: β = 0.034, α = 0.982, Long-run effect = 1.878
- Bureaucratic quality: β = 0.041, α = 0.979, Long-run effect = 1.947
- Voice & accountability: β = 0.020, α = 0.985, Long-run effect = 1.336

### System GMM Point Estimates (Sub-Saharan Africa sample, selected coefficients)
- Government effectiveness: 0.0384*** (0.0126)
- Corruption: 0.0418*** (0.0103)
- Rule of law: 0.0340** (0.0141)
- Bureaucratic quality: 0.0407*** (0.0125)
- Voice & accountability: 0.0195** (0.0088)
- Lagged dependent variable (examples across specifications): 0.9782***, 0.9806***, 0.9819***, 0.9791***, 0.9854*** (standard errors reported in tables).
- Control variable example: Aid (% GDP) = 0.0024*** (0.0009) in system GMM (SSA).
- Observations: 673; Countries: 37; Instruments: 261 (in system GMM SSA specifications).
- Wald chi² examples: 11939.30; 19802.30; 12315.54; 11705.05; 9891.58 (across governance indicator specifications).

### Fixed-effects Estimations: Key Results (Sub-Saharan Africa and Full sample)
- FE (SSA): Government effectiveness = 0.02** (0.0093); Lagged dependent variable ≈ 0.91***; Observations = 673; Countries = 37; Within R² ≈ 0.936–0.937.
- FE (Full sample): Government effectiveness = 0.0185*** (0.0067); Lagged dependent variable ≈ 0.9284***; Observations = 1,312; Countries = 69; Within R² ≈ 0.961.

### Summary Statistics (selected)
- Panel A: Full sample (69 countries, 2003–2024)
  - ln (GDP per capita): Obs = 1,754; Mean = 7.496; Std. Dev. = 0.953; Min = 5.493; Max = 9.748
  - Government effectiveness: Obs = 1,771; Mean = −0.611; Std. Dev. = 0.619; Min = −2.206; Max = 1.101
  - Corruption control: Obs = 1,771; Mean = −0.673; Std. Dev. = 0.519; Min = −1.806; Max = 1.240
  - Voice & accountability: Obs = 1,771; Mean = −0.569; Std. Dev. = 0.642; Min = −2.231; Max = 0.954
  - ln (Investment): Obs = 1,556; Mean = 22.738; Std. Dev. = 2.012; Min = 16.164; Max = 29.681
  - FDI (% GDP): Obs = 1,761; Mean = 3.573; Std. Dev. = 6.112; Min = −37.171; Max = 103.3
  - Trade openness (% GDP): Obs = 1,649; Mean = 68.76; Std. Dev. = 37.71; Min = 2.00; Max = 347.7
- Notes: All governance indicators are World Bank Worldwide Governance Indicators (WGI), ranging from −2.5 to +2.5. ln(Investment) and ln (CPI) are natural logarithms. Remittances (% GDP) computed as 100 × remittances received / nominal GDP.

### Interpretation and Implications from the Analysis
- Governance indicators (Government effectiveness, Corruption control, Rule of law, Bureaucratic quality, Voice & accountability) show statistically significant positive short-run coefficients in several specifications, and sizeable long-run multipliers when accounting for high persistence in GDP per capita (high lagged dependent variable coefficients).
- Long-run multipliers imply that permanent improvements in governance have amplified effects on the dependent variable (log of real GDP per capita) once dynamic persistence is accounted for.

### Policy-relevant Conclusions (from the paper's Conclusion)
- Medium- to long-term growth prospects depend critically on structural transformation; productivity growth must become the main engine as contributions from capital and labor are expected to diminish.
- Strengthening governance and institutions is required alongside complementary reforms in human capital, infrastructure, financial sector development, and economic diversification.
- Specific policy priorities highlighted:
  - Maintaining macroeconomic stability: fiscal policy anchored on gradual consolidation; strengthen domestic revenue mobilization; improved expenditure prioritization; prudent public debt management; monetary policy focused on price stability while supporting adequate credit provision to the private sector.
  - Raising agricultural productivity: prioritize irrigation development; improved access to inputs; strengthened agricultural extension services; adoption of climate-resilient practices.
  - Accelerating structural reforms for private sector–led growth: regulatory simplification; strengthened contract enforcement; enhanced access to finance for SMEs.
  - Promoting economic diversification: support value addition in agriculture, fisheries, and light manufacturing; improve trade facilitation and compliance with regional and international standards.
  - Investments in human capital: enhance efficiency and targeting of public spending on education, health, and skills development; emphasize technical and vocational education aligned with labor market needs.
  - Addressing infrastructure gaps: prioritize investments in energy, transport, and digital connectivity; guide investments by rigorous project appraisal and align with medium-term fiscal and debt sustainability objectives.

*Source: Annex I. Short- and Long-run Impact of Governance Indicators, SIPEA2026067 (authors' estimates and tables).*

---


_Source: https://www.imf.org/-/media/files/publications/selected-issues-papers/2026/english/sipea2026067.pdf_