## wpiea2025045-print-pdf

## Source details

**Canonical URL:** [wpiea2025045-print-pdf](https://www.imf.org/-/media/files/publications/wp/2025/english/wpiea2025045-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2025/english/wpiea2025045-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2025/english/wpiea2025045-print-pdf.pdf.json)

---

### 1. Introduction — Key findings and framing
- International trade has been a vital engine for economic growth and development; since 1990, international trade volume has more than quadrupled.
- For sub-Saharan Africa (SSA) understanding trade dynamics is critical for long-term economic convergence and macroeconomic policy formulation.
- SSA has been grappling with chronic current account deficits, with the regional average of 5-6 percent.
- In 2022, approximately half of the SSA countries faced current account deficits surpassing 5 percent of their GDP, mostly driven by the trade balance on goods and services.
- Tight global financial conditions (major central banks raising interest rates) increase the challenge of financing these deficits.
- Policy objectives highlighted:
  - Moderate import demand and build substantial foreign exchange (FX) reserve buffers.
  - Attain conventional reserve adequacy of 3-4 months of import coverage.
  - Empirically assess exchange rate transmission to trade given SSA’s commodity-based exports and dollar pricing.
- AfCFTA opportunity: intra-regional trade in SSA accounts for less than 20 percent of total trade; AfCFTA could create a market of over 1.3 billion people and a combined GDP exceeding US$3.4 trillion.

### 2. Sub-Saharan Africa’s Trade Landscape
- Trade openness:
  - SSA exhibits lower trade openness (trade-to-GDP ratio) relative to other regions.
  - Oil-exporting countries show the highest median trade-to-GDP ratio among SSA groups.
  - Seychelles and Lesotho are outliers with significantly higher trade-to-GDP ratios.
- Imports (structure and partners):
  - SSA is heavily dependent on imports due to limited industrialization and infrastructure challenges.
  - Top import product shares (SSA): Machinery and electronics: 20 percent; Fuels: 14 percent.
  - Shift in import sources (2000 to 2020):
    - Share from advanced economies: declined from 50 percent in 2000 to 29 percent in 2020.
    - U.S. share of SSA imports: declined from 9 percent in 2000 to 5 percent in 2020.
    - Share from emerging markets and other low-income countries outside SSA: rose from 18 percent to 30 percent.
    - China’s share of SSA imports: increased from 4 percent in 2000 to around 20 percent in 2020.
  - In 2022, Nigeria, South Africa, Ghana, and Kenya together accounted for more than half of all imports of Chinese goods to Africa.
- Exports (composition and partners):
  - SSA exports are predominantly commodity-based; commodities account for more than 50 percent of total exports.
  - Export product shares: Fuels: 22 percent; Stone and glass: 18 percent; Metals: 14 percent; Minerals: 10 percent; Vegetables: 8 percent.
  - Export destinations include China, India, the United States, Switzerland, Germany, and South Africa.
  - Cross-border trade trends: modest growth, stagnant services trade share, limited merchandise trade growth, limited integration into global value chains (GVCs).

### 3. Literature Review — Determinants identified
- Established determinants: GDP (economic size) and its sub-components, exchange rates, and global demand are critical determinants of trade flows.
- Key prior findings:
  - IMF (2016): weak economic activity, particularly investment, accounts for about three-fourths of the dramatic slowdown in the volume of trade since 2012.
  - Exchange rate volatility generally inversely related to trade flows.
  - Kang and Liao (2016): investment in China has much higher import intensity compared to domestic consumption.
  - Bussière et al. (2013): demand composition central to trade dynamics due to high import intensity of certain expenditure categories.
- SSA-specific findings:
  - Asaana and Sakyi (2021): expenditure components, relative import prices, and FX reserves primarily drive SSA import demand.
  - Ngouhouo et al. (2021): domestic institutions significantly influence SSA trade openness.
  - Sekkat and Varoudakis (2000): exchange rate management influences manufactured exports in SSA.
  - Alege and Osabuohien (2015): export and import are inelastic to changes in exchange rates in SSA.
  - Meniago and Eita (2017): positive but very low responsiveness of imports to exchange rate changes.
- Contribution of this study:
  - Comprehensive empirical examination of imports and exports across 44 SSA countries (1990–2022).
  - Disaggregates SSA into oil exporters, other resource-intensive, and non-resource-intensive countries and examines regional blocs.

### 4. Data and Methodology
- Data:
  - Coverage period: 1990 to 2022.
  - Sample: 44 countries in sub-Saharan Africa.
  - Data sources: IMF World Economic Outlook (WEO) and UNCTAD database.
  - Dependent variables: logarithmic changes in the total volume of exports and imports of goods and services (real terms).
  - Import drivers: consumption, investment, exports, and real effective exchange rate (REER).
  - Export drivers: REER, world real GDP, and foreign direct investment (FDI).
- Econometric approach:
  - Estimators: Fixed effects (FE), Random effects (RE), and Arellano-Bover / Blundell-Bond system GMM (two-step) with Windmeijer (2005) robust standard errors.
  - GMM specifics: lagged differences as instruments for level equations and lagged levels as instruments for difference equations.
  - Diagnostics: Hansen test for over-identifying restrictions and Arellano-Bond test for second-order autocorrelation.
  - Additional estimators: pooled-OLS (Annex II) and Mean-Group (MG) estimator (robustness checks).

### 5. Empirical Results — Imports
- Aggregate import demand elasticities (Table 1, SSA aggregate):
  - Consumption: 0.9
  - Investment: 0.3
  - Exports: 0.5
  - REER: 0.1
  - Imports (Lag): -0.04**
  - Constant: -0.01***
  - Observations: 1,498
  - R-squared: 0.49
  - Number of countrycode: 42
- Group- and bloc-specific findings:
  - Consumption coefficients by commodity groups:
    - Oil exporters: 0.8
    - Other resource-intensive and non-resource-intensive: range 0.87 to 0.99
  - Regional consumption coefficients (EAC, CEMAC, WAEMU): range 0.84 to 1.27
  - Investment impact on imports:
    - Range across country groups: 0.2 to 0.4
    - Oil exporters: 0.4
    - Other resource-intensive: 0.3
    - Regional blocs: range 0.2 to 0.4 (highest in CEMAC)
  - Exports’ correlation with imports:
    - Range across groups: 0.2 to 0.8
    - Oil exporters: 0.8
    - Other resource-intensive: 0.4
    - Non-resource-intensive: 0.3
    - Blocs: between 0.2 and 0.5 (larger in EAC)
  - REER pass-through to imports:
    - Non-resource-intensive (FE model): 0.15 (one percent appreciation → 0.15 percent increase in imports)
    - EAC: about 0.3 (one percent appreciation → ~0.3 percent increase in imports)
    - CEMAC and WAEMU: quantitatively smaller and statistically insignificant
- Interpretation:
  - Consumption is the dominant driver of imports across groups and blocs.
  - Investment and exports are positive and significant determinants; magnitudes vary by resource intensity and region.

### 5. Empirical Results — Exports
- Aggregate export regression (Table 4, SSA aggregate):
  - Real World GDP: 2.67*** (one percent increase in global growth → 2.67 percent increase in exports)
  - REER: -0.28*** (one percent REER depreciation → increase in exports)
  - FDI: 0.00
  - Exports (Lag): -0.11***
  - Constant: -0.03* or -0.02 depending on specification
  - Observations: 1,182 (FE/RE), 1,087 (Arellano Bond)
  - R-squared: 0.06
  - Number of countrycode: 42
- Group- and bloc-specific export sensitivities:
  - World GDP elasticity:
    - SSA average: 2.6 (Table 4 reports 2.67; MG estimate 2.476)
    - Non-resource-intensive countries: 3.7
    - Other country groupings: around 1.6
    - EAC: 2.3
    - CEMAC and WAEMU: much smaller and statistically insignificant
  - REER pass-through to exports:
    - Regionwide: one percent depreciation could increase exports by 0.2–0.5 percent
    - Non-resource countries: 0.2
    - Oil exporters: 0.3
    - Non-oil commodity exporters: 0.5
    - EAC: 0.5
    - WAEMU: 0.2–0.3
    - CEMAC: not significant
  - FDI: aggregate coefficient 0.00; group results small, mixed, largely insignificant
- Model fit and interpretation:
  - Export regressions have much lower overall R2 (0.06) than import regressions (0.49), indicating omitted factors (trade policy, structural change).

### Robustness checks and MG results
- GMM diagnostics:
  - AR(1) test: presence of first-order autocorrelation (expected).
  - AR(2) test: fails to reject null in all cases (no second-order serial correlation).
  - Sargan and Hansen tests: p-values largely above 0.10, suggesting valid instruments; some samples show potential weak instrument issues.
- Sample-size limitation: N may not be sufficiently larger than T for some sub-samples, risking biased and inefficient GMM AB estimates.
- Pooled Mean Group (MG) selected results:
  - MG imports (SSA):
    - Consumption: 1.103***
    - Investment: 0.337***
    - Exports: 0.410***
    - REER: 0.088
    - Observations: 1,498
    - R2: 0.680
  - MG imports (Oil Exporter): Consumption 0.923***; Investment 0.483***; Exports 0.602***; REER 0.252
  - MG imports (EAC): Consumption 1.511***; Investment 0.403***; Exports 0.424***; REER 0.203**
  - MG exports (SSA): WorldGDP 2.476***; REER -0.201*; FDI -0.001; Observations 1,182; R2 0.261
  - MG exports (Non Resource): WorldGDP 3.488***; REER -0.133
  - MG exports (EAC): WorldGDP 2.713***; REER -0.263**
- Overall robustness conclusion: MG results broadly consistent with FE/RE/GMM AB findings, reinforcing main correlations while acknowledging identification limitations.

### Key policy-relevant findings (quantitative highlights)
- Imports:
  - A one percent increase in consumption correlates with a 0.9 percent increase in imports.
  - A one percent increase in investment correlates with a 0.3 percent increase in imports.
  - A one percent increase in exports correlates with a 0.5 percent increase in imports.
  - Consumption goods constitute about 40 percent of the import basket in SSA, versus 30 percent in other emerging economies.
- Exports:
  - A one percent surge in global growth correlates with a 2.6 percent growth in the region’s exports.
  - A one percent depreciation of the REER is associated with boosting exports in the region by 0.2 percent.
  - Exchange rate pass-through to imports and exports is heterogeneous: stronger in non-resource-intensive countries and the EAC; limited or insignificant in CEMAC and other fixed-rate regimes.
- Model fit disparity:
  - Import regressions R-squared: 0.49.
  - Export regressions R-squared: 0.06.

### 7. Conclusions and Policy Implications — Recommendations
- General strategy:
  - Account for country-specific economic structure and heterogeneity in trade dynamics when designing policy.
- For non-resource-intensive countries (notably EAC):
  - Policies to promote exchange rate flexibility could help restore trade competitiveness.
  - Exchange rate depreciation can be effective to restore trade balance.
- For commodity/commodity-exporting countries (oil exporters):
  - Currency depreciation may be less effective for restoring trade balance; pursue alternative adjustment mechanisms.
- For countries with exports vulnerable to global fluctuations:
  - Implement robust countercyclical macroeconomic policies to mitigate global shocks.
- Given strong correlation between imports and domestic demand:
  - Fiscal consolidation can be used to restore external balance by moderating import demand.
  - Maintain adequate foreign exchange reserve levels (targeting 3-4 months of imports) to guard against import shocks.
- Complementary measures:
  - Structural reforms, trade policy improvements, and trade facilitation to address export determinants not captured by macro variables.
- Research recommendations:
  - Future work should distinguish final versus intermediate components of demand to better capture input-intensity and GVC linkages.

*IMF Working Paper — 1. Introduction; 5. Empirical Results; 7. Conclusion and Policy Implications, "Understanding Trade Dynamics in Sub-Saharan Africa."*

### 1. Introduction

### 1. Introduction

### Key findings and framing
- International trade has been a vital engine for economic growth and development, influencing capital accumulation, employment, and technological progress; since 1990, international trade volume has more than quadrupled.
- For sub-Saharan Africa (SSA) understanding trade dynamics is critical for long-term economic convergence and macroeconomic policy formulation.
- SSA has been grappling with chronic current account deficits, with the regional average of 5-6 percent.
- In 2022, approximately half of the SSA countries faced current account deficits surpassing 5 percent of their GDP, mostly driven by the trade balance on goods and services.
- Tight global financial conditions (major central banks raising interest rates) increase the challenge of financing these deficits.
- To ensure external sustainability, SSA countries need to moderate import demand and build substantial foreign exchange (FX) reserve buffers.
- A conventional reserve adequacy metric cited is to attain 3-4 months of import coverage; this is a common policy objective in IMF lending programs.
- Exchange rate flexibility is a critical mechanism for external adjustment, but its effectiveness depends on factors such as invoicing currency and FX market characteristics; with dominant currency pricing, depreciation may cut imports but not significantly raise exports in the short term.
- The African Continental Free Trade Area (AfCFTA) presents an opportunity: intra-regional trade in SSA accounts for less than 20 percent of total trade; AfCFTA could create a market of over 1.3 billion people and a combined GDP exceeding US$3.4 trillion.

### Policy implications highlighted
- Adopt prudent policy adjustments to moderate import demand given financing constraints.
- Build FX reserve buffers consistent with reserve adequacy targets (3-4 months of imports) to bolster resilience.
- Empirically assess exchange rate transmission to trade given SSA’s commodity-based exports and dollar pricing.

### Organization of the paper (as presented)
- Section 2: summary of the trade landscape in SSA.
- Section 3: literature review.
- Section 4: empirical methodology and data.
- Section 5: research findings.
- Section 6: robustness checks.
- Section 7: conclusions and policy implications.

### Source attribution
*IMF Working Paper — 1. Introduction, "Understanding Trade Dynamics in Sub-Saharan Africa."*

---

### 2. Sub-Saharan Africa’s Trade Landscape

### Trade openness
- SSA exhibits lower trade openness compared to other regions (trade-to-GDP ratio).
- Oil-exporting countries show the highest median trade-to-GDP ratio among SSA country groups.
- Trade openness variability:
  - Lower among other resource-intensive countries.
  - Higher among non-resource-intensive countries.
- Seychelles and Lesotho are outliers with significantly higher trade-to-GDP ratios.

### Imports (structure and partners)
- SSA is heavily dependent on imports due to limited industrialization and infrastructure challenges.
- Top import product shares (SSA):
  - Machinery and electronics: 20 percent of all imports.
  - Fuels: 14 percent of imports.
  - (Also noted: chemicals, metals, vegetables as significant import categories.)
- Shift in import sources (2000 to 2020):
  - Share of imports from advanced economies declined from 50 percent in 2000 to 29 percent in 2020.
  - U.S. share of SSA imports declined from 9 percent in 2000 to 5 percent in 2020.
  - Share of imports from emerging markets and other low-income countries outside SSA rose from 18 percent to 30 percent.
  - China’s share of SSA imports increased from 4 percent in 2000 to around 20 percent in 2020.
- In 2022, Nigeria, South Africa, Ghana, and Kenya together accounted for more than half of all imports of Chinese goods to Africa.

### Exports (composition and partners)
- SSA exports are predominantly commodity-based; commodities account for more than 50 percent of total exports.
- Export product shares (2020 / 2021 context as presented):
  - Fuels: 22 percent of total exports.
  - Stone and glass: 18 percent.
  - Metals: 14 percent.
  - Minerals: 10 percent.
  - Vegetables: 8 percent.
- SSA export destinations include China, India, the United States, Switzerland, Germany, and South Africa (illustrating both global and intra-African linkages).
- Cross-border trade trends: modest growth, stagnant services trade share, limited merchandise trade growth, and limited integration into global value chains (GVCs).

---

### 3. Literature Review

### Established determinants and findings from prior studies
- GDP (economic size) and its sub-components, exchange rates, and global demand are critical determinants of trade flows.
- IMF (2016): weak economic activity, particularly investment, accounts for about three-fourths of the dramatic slowdown in the volume of trade since 2012.
- Exchange rate volatility is generally found to have an inverse relationship with trade flows.
- Kang and Liao (2016): investment in China has much higher import intensity compared to domestic consumption; China’s rebalancing lowers overall import intensity.
- Bussière et al. (2013): demand composition is central to trade dynamics due to high import intensity of certain expenditure categories.

### SSA-specific findings from the literature
- Asaana and Sakyi (2021): expenditure components, relative import prices, and FX reserves primarily drive SSA import demand.
- Ngouhouo et al. (2021): domestic institutions significantly influence SSA trade openness.
- Sekkat and Varoudakis (2000): exchange rate management influences the performance of manufactured exports in SSA.
- Alege and Osabuohien (2015): export and import are inelastic to changes in exchange rates in SSA.
- Meniago and Eita (2017): positive relationship between exchange rate changes and imports in SSA but with very low responsiveness.

### Contribution of this study
- Provides a comprehensive empirical examination of trade dynamics (imports and exports) across SSA.
- Disaggregates SSA into sub-groups: oil exporters, other resource-intensive countries (mainly mineral, ore, and metal), and non-resource-intensive countries.
- Investigates trade dynamics across several regional economic blocs for greater granularity.

---

### 4. Data and Methodology

### Data
- Coverage period: 1990 to 2022.
- Sample: 44 countries in sub-Saharan Africa.
- Data sources: IMF World Economic Outlook (WEO) and United Nations Conference on Trade and Development (UNCTAD) database.
- Dependent variables: logarithmic changes in the total volume of exports and imports of goods and services (all variables expressed in real terms and logarithmic changes).
- Import drivers included: aggregate demand components (consumption and investment), exports, and the real effective exchange rate (REER).
- Export drivers included: REER, world real GDP (proxy for global demand), and foreign direct investment (FDI).

### Econometric approach
- Three estimation methods employed:
  - Fixed effects model (within-group estimator): y_it = α + β X_it + μ_i + u_it.
  - Random effects model: y_it = α + β X_it + ω_it, ω_it = E_i + V_it.
  - Dynamic panel data model estimated with Arellano-Bover / Blundell-Bond system GMM (two-step) to address endogeneity concerns.
- System GMM specifics:
  - Uses lagged differences as instruments for level equations and lagged levels as instruments for difference equations.
  - Suitable for datasets with relatively short time dimension and large cross-sectional dimension.
  - Two-step system GMM estimator used with Windmeijer (2005) robust standard errors.
  - Diagnostic tests implemented: Hansen test for over-identifying restrictions and Arellano-Bond test for second-order autocorrelation.
- Additional estimators referenced: pooled-OLS reported in Annex II; Mean-Group (MG) estimator used in Section 6 robustness checks.

*IMF Working Paper — 1. Introduction, "Understanding Trade Dynamics in Sub-Saharan Africa."*

### 5. Empirical Results

### 5. Empirical Results

### Determinants of SSA Imports
- Aggregate import demand elasticities (Table 1, SSA aggregate):
  - Consumption: 0.9
  - Investment: 0.3
  - Exports: 0.5
  - REER: 0.1
  - Imports (Lag): -0.04**
  - Constant: -0.01***
  - Observations: 1,498
  - R-squared: 0.49
  - Number of countrycode: 42
- Group- and bloc-specific findings:
  - Consumption coefficients by commodity groups:
    - Oil exporters: 0.8
    - Other resource-intensive and non-resource-intensive: range 0.87 to 0.99
  - Regional consumption coefficients (EAC, CEMAC, WAEMU): range 0.84 to 1.27
  - Investment impact on imports:
    - Range across country groups: 0.2 to 0.4
    - Oil exporters: 0.4
    - Other resource-intensive: 0.3
    - Regional blocs: range 0.2 to 0.4 (highest in CEMAC)
  - Exports’ correlation with imports:
    - Range across groups: 0.2 to 0.8
    - Oil exporters: 0.8
    - Other resource-intensive: 0.4
    - Non-resource-intensive: 0.3
    - Blocs: between 0.2 and 0.5 (larger in EAC)
  - REER pass-through to imports:
    - Non-resource-intensive countries (FE model): 0.15 (one percent appreciation → 0.15 percent increase in imports)
    - EAC: about 0.3 (one percent appreciation → ~0.3 percent increase in imports)
    - CEMAC and WAEMU: quantitatively smaller and statistically insignificant
- Heterogeneity and policy implication:
  - Consumption is a dominant driver of imports across all country groups and regional blocs.
  - Investment and exports are positive and significant determinants of imports, with magnitudes varying by resource intensity and region.
  - Policymakers should consider domestic consumption patterns, investment composition, and export structure when addressing import dynamics and trade balance sustainability.

### Determinants of SSA Exports
- Aggregate export regression (Table 4, SSA aggregate):
  - Real World GDP: 2.67*** (one percent increase in global growth → 2.67 percent increase in exports)
  - REER: -0.28*** (one percent REER depreciation → increase in exports; negative coefficient reported)
  - FDI: 0.00
  - Exports (Lag): -0.11***
  - Constant: -0.03* or -0.02 depending on specification
  - Observations: 1,182 (FE/RE), 1,087 (Arellano Bond)
  - R-squared: 0.06
  - Number of countrycode: 42
- Group- and bloc-specific export sensitivities:
  - World GDP elasticity:
    - SSA average: 2.6 (reported as 2.67 in Table 4; MG estimate 2.476)
    - Non-resource-intensive countries: 3.7
    - Other country groupings: around 1.6
    - EAC (regional bloc): coefficient reported 2.3 (strongest sensitivity among blocs)
    - CEMAC and WAEMU: much smaller and statistically insignificant
  - REER pass-through to exports:
    - Regionwide: one percent depreciation could increase exports by 0.2–0.5 percent
    - Non-resource countries: 0.2
    - Oil exporters: 0.3
    - Non-oil commodity exporters: 0.5
    - EAC: 0.5
    - WAEMU: 0.2–0.3
    - CEMAC: not significant (reflecting fixed exchange rate regime)
  - FDI:
    - Aggregate coefficient reported as 0.00 (Table 4); group results show small, mixed, and largely insignificant effects (see Tables 5–6)
- Model fit and interpretation:
  - Export regressions have much lower overall R2 (0.06) than import regressions (0.49), suggesting additional factors (trade policy, structural change) may significantly affect exports beyond the macro variables included.

### Robustness Checks and Pooled Mean Group (MG) Results
- GMM Arellano-Bond diagnostics (Appendix Table B.13; summarized in text):
  - AR(1) test: presence of first-order autocorrelation (expected in differenced GMM)
  - AR(2) test: fails to reject null in all cases (no second-order serial correlation)
  - Sargan and Hansen tests: p-values largely above 0.10, suggesting valid instrument sets and limited overidentification concerns
  - Caveats: some samples (non-resource-intensive; EAC) show potential weak instrument issues or marginal AR(1) insignificance
- Sample-size limitation:
  - N may not be sufficiently larger than T for some sub-samples, risking biased and inefficient GMM AB estimates
- Pooled Mean Group (MG) estimation results (Tables 7–8) — long-run coefficients broadly consistent with FE/RE/GMM AB:
  - MG imports (selected SSA and subgroup coefficients, Table 7):
    - Consumption (SSA): 1.103***
    - Investment (SSA): 0.337***
    - Exports (SSA): 0.410***
    - REER (SSA): 0.088
    - Observations (SSA imports MG): 1,498
    - R2 (SSA imports MG): 0.680
  - MG imports (selected subgroup examples):
    - Oil Exporter: Consumption 0.923***; Investment 0.483***; Exports 0.602***; REER 0.252
    - EAC: Consumption 1.511***; Investment 0.403***; Exports 0.424***; REER 0.203**
  - MG exports (selected SSA and subgroup coefficients, Table 8):
    - WorldGDP (SSA): 2.476***
    - REER (SSA): -0.201*
    - FDI (SSA): -0.001
    - Observations (SSA exports MG): 1,182
    - R2 (SSA exports MG): 0.261
  - MG exports (selected subgroup examples):
    - Non Resource: WorldGDP 3.488***; REER -0.133
    - EAC: WorldGDP 2.713***; REER -0.263**
- Overall robustness conclusion:
  - MG results are broadly consistent with FE, RE, and GMM AB findings, reinforcing the main correlations reported for imports and exports while acknowledging sample-size and identification limitations.

### Key Policy-Relevant Findings and Recommendations
- Imports:
  - Domestic consumption is the primary driver of import demand across SSA and across commodity and regional groupings (consumption elasticities often near or above 0.8).
  - Investment and export growth also raise import demand (investment elasticities typically 0.2–0.4; export elasticities vary by group).
  - Exchange-rate pass-through to imports is limited overall, but more pronounced in non-resource-intensive countries and in the EAC.
- Exports:
  - Global demand (Real World GDP) is a major determinant of export growth (SSA point estimates around 2.6; non-resource-intensive up to 3.7).
  - REER depreciation generally supports export growth, with heterogeneity by country group and exchange-rate regime (EAC shows relatively strong pass-through; CEMAC shows limited impact).
  - FDI has limited or mixed direct correlation with export growth in these specifications.
- Policy implications:
  - Trade and macroeconomic policies should account for heterogeneity across resource intensity and regional blocs.
  - Policies that influence domestic consumption and investment will have significant implications for import dynamics and balance-of-payments management.
  - Exchange-rate policy effects vary by country characteristics and exchange-rate regime; policymakers in non-resource-intensive and EAC countries should be particularly mindful of REER movements.
  - Strengthening trade policy, structural reforms, and trade facilitation could address export determinants not captured by macro variables (given low R2 for export regressions).

*Source: https://www.imf.org/-/media/files/publications/wp/2025/english/wpiea2025045-print-pdf.pdf*

### 7. Conclusion and Policy Implications

### 7. Conclusion and Policy Implications

### Key empirical findings on import dynamics
- Domestic demand and exports are highly correlated with SSA imports.
- A one percent increase in consumption correlates with a 0.9 percent increase in imports.
- A one percent increase in investment correlates with a 0.3 percent increase in imports.
- A one percent increase in exports correlates with a 0.5 percent increase in imports.
- Translating to import intensity measures: about 20-40 percent of consumption and 60 percent of investment are linked to imports.
- Imports demand in oil-exporting countries is markedly more responsive to exports and investment compared to other countries.
- SSA imports demonstrate greater sensitivity to fluctuations in consumption rather than to variations in investment.
- Consumption goods constitute about 40 percent of the import basket in SSA, versus 30 percent in other emerging economies.
- The analysis focuses on components of final demand and does not separate final versus intermediate input demand.

### Exchange rate effects on imports and exports
- A one percent appreciation of the real effective exchange rate (REER) is associated with bolstering import growth by 0.1 percent.
- The REER–imports relationship primarily reflects trends in East African Community (EAC) or non-resource-intensive countries; it is less pronounced for resource-intensive countries.
- The prevalence of fixed exchange rate regimes, especially among resource-intensive economies and monetary unions such as CEMAC, may hinder exchange rates as an adjustment mechanism.
- A one percent depreciation of the REER is associated with boosting exports in the region by 0.2 percent (quantitatively small).
- Exchange rate pass-through to exports is significant only for non-resource-intensive (mostly in the EAC) and other resource-intensive countries.
- For oil exporters, exchange rate depreciation does not seem to influence exports, likely because oil prices are globally determined and priced in U.S. dollars.

### Exports and global environment sensitivity
- A one percent surge in global growth correlates with a 2.6 percent growth in the region’s exports.
- The sensitivity is strongest for non-resource-intensive countries: a one percent rise in global growth correlates with a 3.7 percent growth in exports.
- For other country groupings (including oil exporters and other resource-intensive countries), the correlation is less than 2 percent.

### Heterogeneity across country groups and regional blocs
- The importance and correlation of factors with exports and imports vary across countries, contingent on economic structure.
- Oil exporters: imports demand more responsive to exports and investment; exchange rate depreciation ineffective for exports.
- Non-resource-intensive countries (notably in the EAC): exchange rate flexibility has more influence on trade outcomes; exports more responsive to global growth.
- Monetary unions and fixed-rate regimes (e.g., CEMAC) show limited exchange rate adjustment channels.

### Policy implications and recommendations
- Policy design must account for country-specific economic structure and heterogeneity in trade dynamics.
- For non-resource-intensive countries:
  - Policies to promote exchange rate flexibility could be instrumental in restoring trade competitiveness.
  - Exchange rate depreciation can be an effective leverage to restore trade balance.
- For commodity/commodity-exporting countries (oil exporters):
  - Currency depreciation may be less effective in restoring trade balance; alternative adjustment mechanisms are needed.
- Countries with exports vulnerable to global fluctuations:
  - Implement robust countercyclical macroeconomic policies to mitigate the impact of global shocks.
- Given the strong correlation between imports and domestic demand:
  - Fiscal consolidation can be a policy instrument to restore external balance by moderating import demand.
  - Maintaining adequate foreign exchange reserve levels is important to guard against unforeseen spikes in import demand.
- The limited role of exchange rates in some groups suggests complementary policies (fiscal, reserve management, structural) are required for external sustainability.

### Research and data considerations for future work
- Future studies could refine analysis by distinguishing final from intermediate components of demand to better capture input-intensity and global value chain linkages.
- Deeper investigation into the role of intermediate goods demand is warranted, given current analysis concentrates on final demand components and African countries’ lower integration into global value chains.

### Final synthesis
- The complex trade dynamics in SSA require a nuanced, country-specific policy approach.
- Findings support tailoring exchange rate, fiscal, reserve management, and countercyclical policies to country economic structure to foster long-term growth, external sustainability, and resilience.

*Source: IMF Working Paper — 7. Conclusion and Policy Implications*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2025/english/wpiea2025045-print-pdf.pdf_