## _wp0766

## Source details

**Canonical URL:** [_wp0766](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0766.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0766.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0766.pdf.json)

---

### I. Introduction
- Madagascar plans to start phasing out its customs tariffs on imports from the Southern African Development Community (SADC) in 2007.
- The paper uses the Global Trade Analysis Project (GTAP) model to simulate scenarios quantifying the impact of the SADC FTA on all sectors of the economy.
- Key simulation result summary:
  - Reciprocal elimination of customs tariffs between SADC countries and Madagascar would only have a limited impact on Madagascar’s real GDP.
  - The actual liberalization under the SADC FTA affects less than 6 percent of Madagascar total imports.
  - Benefits of the SADC FTA are limited under regional liberalization alone, but become significant when combined with multilateral liberalization because:
    - The liberalization then affects all Madagascar trade.
    - A multilateral liberalization reduces the cost associated with potential trade diversion.

### II. Madagascar trade patterns and trade policy
- Tariff and openness indicators:
  - Simple average MFN tariff: 16.2 percent at end-2005; 13.6 percent in 2006.
  - SSA average: 14.5 percent; excluding other duties and charges the average is 13 percent in Sub-Saharan Africa.
  - Tariff bands declined from five to four (0, 5, 10, 20); except for 19 tariff lines (out of 6155) on oil, oil products, and gases, all tariffs are ad valorem.
  - Trade openness increased from 15 percent of GDP in 1985 to about 33 percent in 2005.
- Preferential agreements and coverage:
  - Madagascar member of COMESA since 1993 and part of the COMESA FTA since 2000.
  - Madagascar’s imports from SADC account for 12.5 percent of total imports.
  - Four SADC members that also belong to COMESA already have duty-free access to Madagascar; these four account for 6.9 percent of Madagascar imports and 55 percent of Madagascar imports from SADC.
  - Net effect: the SADC FTA will eliminate customs tariffs on 5.6 percent of Madagascar imports, not 12.5 percent.
  - 88 percent of Madagascar imports that will be actually liberalized by the SADC FTA are from South Africa.

### III. Joining SADC: process and motivations
- Timetable and treatment:
  - Madagascar to start phasing out tariffs on SADC imports in 2007; by 2012 tariffs on SADC imports almost fully eliminated.
  - Tariff reduction discriminates between South Africa and rest of SADC; tariffs remain on sugar products and on oil, petroleum products, and gases.
  - SADC aims for a customs union by 2010.
- Government expectations:
  - Regional integration to increase trade openness and foster investment, notably FDI from South Africa.
  - Madagascar expects gains from elimination of tariffs to SADC members not part of COMESA (SACU members, Mozambique, Tanzania). In 2006 these countries’ average tariffs were 11.4, 12.1, and 7.2 percent respectively.
- Constraints from literature and past experience:
  - Historical limits to intra-regional liberalization due to import-substitution policies, tariff-revenue constraints, unequal distribution of gains, severe trade-regime distortions, high transaction costs, limited product complementarities, and institutional constraints.
  - Complementary policies (macroeconomic stability, appropriate exchange rate, investment policies) often not implemented.
- Sectoral expectations:
  - Investment in 2005 was 22.5 percent of GDP, 10.4 percentage points higher than saving.
  - Madagascar expects increased FDI; December 2006 protocol signed with South Africa on mutual investment promotion.
  - Agriculture: 26 percent of GDP and 31 percent when agro-industry is added.
  - Malagasy clothing industry expects reduced tariff on its export to South Africa (36.6 percent) to boost exports; South Africa’s average tariff on textiles and apparel from SADC in 2006 was 8.3 percent.
  - Rice accounts for 1/3 of agricultural production and is singled out.

### IV. The impact of the SADC FTA — model and methodology
- Model and benchmark:
  - GTAP CGE model; database benchmarked to 2001; results in 2001 U.S. dollars.
- Aggregation and focus:
  - Regions aggregated into 8 regions; industries into 6 industries.
  - SADC split into Madagascar, South Africa, and rest of SADC.
  - Industry aggregation: Agriculture and food, Rice, Textiles, Apparel, Other manufacturing, Services.
- Two-step simulation approach:
  1. Update GTAP for textiles and clothing liberalization in early 2005: elimination of bilateral quota restrictions on textiles except U.S. and E.U. quotas; “export taxes” on Chinese textiles and clothing to U.S. and E.U. halved.
  2. Simulate full SADC FTA (total reciprocal elimination of import tariffs on intra-regional trade) under two scenarios:
     - Scenario 1: all prices and wages flexible to maintain current level of employment.
     - Scenario 2: real wages of skilled and unskilled labor rigid so SADC FTA can affect level of employment.

### V. GDP, welfare and decomposition of effects
- Overall headline:
  - Under all scenarios, impact of the SADC FTA on Madagascar’s real GDP and welfare is negligible.
- Table 4 — Impact on Real GDP (in millions of 2001 U.S. dollars; percentage changes in parentheses)
  - Scenario 1 / Scenario 2
    - Madagascar: -0.1   (0) / 1.3  (0)
    - South Africa: 34.4   (0) / 595.9  (0.5)
    - Rest of SADC: -111.1  (-0.2) / 500.6  (0.9)
    - European Union: -0.5  (0) / -76.0  (0)
    - United States: 1.0  (0) / -52.0  (0)
    - Canada: -0.1  (0) / -0.8  (0)
    - China: 0.4  (0) / -10.3  (0)
    - Rest of the World: -39.0 (0) / -201.0  (0)
    - World: -115.0 (0) / 757.7  (0)
- Madagascar-specific GDP effect:
  - Scenario 1: no change in real GDP.
  - Scenario 2: real GDP increases by 0.03 percent (difference due to increased employment).
- Welfare measures and decomposition:
  - Welfare measured as equivalent variation in income using base period prices.
  - Welfare components decomposed into allocative efficiency effects, technology effects (set to zero in presented simulations), endowment effects, and terms of trade effects.
  - Appendix II selected welfare totals (millions of 2001 U.S. dollars):
    - Madagascar: Scenario 1 TOTAL -1.0; Scenario 2 TOTAL 0.5
    - South Africa: Scenario 1 TOTAL 255.5; Scenario 2 TOTAL 833.8
    - WORLD: Scenario 1 TOTAL -115.1; Scenario 2 TOTAL 756.5

### VI. Employment, wages and factor demand
- Scenario 2 employment and wages:
  - Skilled labor employment: +0.1 percent.
  - Unskilled employment: +0.05 percent.
  - Real wages remain constant; nominal wages slightly decline by -0.05 percent.
- Table 5 — Change in Real Wages in the SADC (in percent)
  - Madagascar: Unskilled labor 0.02; Skilled labor 0.05
  - South Africa: Unskilled labor 0.26; Skilled labor 0.25
  - Rest of SADC: Unskilled labor 0.98; Skilled labor 1.11
  - Note: Scenario 2 assumes no change in real wages of both skilled and unskilled labor.
- Table 11 — Demand for Factor of Production in Madagascar (percentage change; first figure scenario 1, second scenario 2)
  - Textiles: Unskilled Labor 1.0, 0.9; Skilled Labor 1.0, 0.9; Capital 1.0
  - Clothing: Unskilled Labor 0.3, 0.4; Skilled Labor 0.3, 0.4; Capital 0.3
  - Food: Unskilled Labor -0.1, 0; Skilled Labor -0.1, 0; Capital -0.1
  - Rice: Unskilled Labor -0.1, 0; Skilled Labor -0.1, 0; Capital -0.1
  - Other manufacturing: Unskilled Labor 0, -0.1; Skilled Labor 0, -0.1; Capital 0
  - Services: Unskilled Labor 0, 0.1; Skilled Labor 0, 0.1; Capital 0
- Employment shifts favor textiles and clothing; under Scenario 2 increases in textiles and clothing employment do not occur at the expense of other sectors.

### VII. Trade creation versus trade diversion; geography of trade changes
- Net effects:
  - South Africa benefits from net trade creation; allocative efficiency gains and replacement of inefficient domestic production by imports (rice and apparel).
  - Madagascar and rest of SADC experience trade diversion tendencies.
- Madagascar trade flows:
  - Imports from South Africa increase by about 9 percent; imports from rest of SADC increase by 2 percent; imports from non-SADC decline by 0.5 percent.
  - Total trade barely increases: Madagascar total trade 0.3 (Scenario 1), 0.4 (Scenario 2).
  - Madagascar trade with South Africa increases by +12.5 percent; exports to South Africa expected to double.
  - Madagascar’s trade with rest of SADC increases by less than 1 percent; exports to rest of SADC drop by 8 percent.
- Table 7 — Change in Volume of Trade of SADC members (in percent)
  - Madagascar: Exports 0.4, 0.5; Imports 0.3, 0.4
  - South Africa: Exports 0.5, 0.4; Imports 2.1, 2.8
  - Rest of SADC: Exports 3.8, 4.4; Imports 4.9, 5.9
- Table 9 — Change in the Geographical Structure of Trade (in percent)
  - SADC: Total trade 2.7, 3.2; Intra-SADC 25.8, 27.1; Trade with third countries -1.8, -1.5
  - Madagascar: Total trade 0.3, 0.4; Trade with SADC 4.8, 5.1; Trade with third countries -0.2, -0.1
  - South Africa: Total trade 1.6, 1.9; Trade with SADC 22.6, 23.7; Trade with third countries -1.4, -1.2
  - Rest of SADC: Total trade 3.6, 5.2; Trade with SADC 29.0, 30.3; Trade with third countries -2.7, -2.1

### VIII. Sectoral impacts; textiles and apparel focus
- Table 10 — Change in Madagascar Trade Balance (in millions of 2001 US dollar)
  - Food -1.4, -1.5
  - Rice 0.0, 0.0
  - Textiles 1.9, 2.1
  - Clothing 0.9, 1.0
  - Other manufacturing -2.0, -2.3
  - Services 0.3, 0.1
  - TOTAL -0.3, -0.6
- Textiles and clothing outcomes:
  - Exports of textiles to South Africa jump by more than 500 percent; exports of clothing by 200 percent.
  - Imports from South Africa increase by about 57.8 percent (textiles) and 116.0 percent (clothing) in Madagascar’s structure of trade.
  - Madagascar total exports of textiles limited to about 1 percent of exports; apparel about 0.5 percent.
- AGOA relevance:
  - AGOA third-country provision originally scheduled to expire in September 2007 but recently extended until 2012; elimination would restrict eligibility to apparel made with African or U.S. fabrics.
  - Madagascar’s apparel industry imports inputs mostly from Asia; the extension gives time to diversify sourcing and markets.
- Caveats limiting expansion:
  - Textiles and clothing treated as sensitive under SADC Trade Protocol; liberalized more slowly.
  - SADC rules of origin and South Africa’s willingness to protect its textiles and clothing may restrain expansion.

### IX. Investment, capital, dynamic considerations and scenario variants
- Base assumption: capital endowment fixed in Scenarios 1 and 2.
- Relaxing capital constraint (scenario 1b and 2b) effects (Table 12 — contributions to growth in percent):
  - GDP: Scenario 1 0.00; Scenario 1b 0.02; Scenario 2 0.03; Scenario 2b 0.43
  - Consumption: Scenario 1 -0.02; Scenario 1b -0.01; Scenario 2 0.01; Scenario 2b 0.38
  - Investment: Scenario 1 0.01; Scenario 1b 0.01; Scenario 2 0.02; Scenario 2b 0.08
  - Exports: Scenario 1 0.10; Scenario 1b 0.09; Scenario 2 0.10; Scenario 2b 0.22
  - Imports: Scenario 1 -0.08; Scenario 1b -0.07; Scenario 2 -0.10; Scenario 2b -0.25
- When both labor and capital endowments can rise (Scenario 2b):
  - GDP increase: 0.4 percent
  - Investment increase: 0.08 percent
  - Bulk of GDP contribution comes from increased consumption.
- Welfare and endowment effects (U.S. dollars):
  - Endowment-related employment effect: $10 million US
  - Larger endowment in capital effect: about $8 million US
  - Madagascar welfare: Scenario 1 loss of $1 million US; Scenario 2b gain of $17.5 million US
  - Per capita welfare: Scenario 1 about $0.06 per capita; Scenario 2b $1 US per capita
- Conclusion: allowing both labor and capital to expand removes binding constraints and magnifies gains.

### X. Terms of trade, price changes and net exports
- Table 6 — Change in the Terms of Trade (In percent)
  - Madagascar: Terms of Trade -0.07 (Scenario 1), -0.06 (Scenario 2); Px -0.05, -0.05; Pm 0.02, 0.01
  - South Africa: Terms of Trade 0.68, 0.72; Px 0.67, 0.71; Pm -0.01, -0.02
  - Rest of SADC: Terms of Trade -0.31, -0.41; Px -0.11, -0.21; Pm 0.20, 0.21
- Net exports:
  - Do not contribute to increasing Madagascar GDP.
  - For South Africa and rest of SADC net exports tend to have a negative impact.

### XI. Fiscal implications
- Madagascar fiscal context:
  - Tax revenue-to-GDP ratio about 10 percent of GDP in 2005.
  - About half of tax revenue is from taxes on international trade.
- Revenue implications of phasing out tariffs on South African imports:
  - South Africa trade-weighted average tariff faced by Madagascar: 10.2 percent.
  - Rest of world trade-weighted average tariff faced: 7.2 percent.
  - South Africa share: 4.8 percent of Madagascar total imports but 6.2 percent of Madagascar dutiable imports.
  - Estimated reductions:
    - Reduction in Madagascar revenues from customs tariff: about 8.0 percent.
    - Reduction in taxes on international trade (customs duties, tax on petroleum products, VAT, and excise taxes): about 2.6 percent.
  - Second-round effect: simulated decline in imports from non-SADC implies further negative impact on customs revenue.
  - SADC has no general fiscal compensation provision except privileged access to the SADC regional development fund.

### XII. Limits of FTA-only gains and role of multilateral liberalization
- Reasons gains are limited under SADC-only liberalization:
  - (i) it is a small trade liberalization;
  - (ii) regional (discriminatory) liberalization involves trade diversion costs.
- Illustrative combined scenario (full intra-SADC liberalization + uniform nondiscriminatory 10 percent multilateral reduction of applied customs tariff):
  - Scenario 1 (SADC only): Real GDP 0.0; Welfare -1.0 (millions US$); Welfare per capita -0.1 (dollars)
  - Scenario 1 (SADC + multilateral 10% cut): Real GDP 0.12; Welfare 7.9; Welfare per capita 1.6
  - Scenario 2 (SADC only): Real GDP 0.0; Welfare 0.5; Welfare per capita 0.0
  - Scenario 2 (SADC + multilateral 10% cut): Real GDP 1.59; Welfare 94.7; Welfare per capita 5.3
- Conclusion: nondiscriminatory multilateral liberalization both increases coverage and eliminates trade diversion costs, producing much larger gains than regional-only liberalization.

### XIII. Policy implications and recommendations to maximize gains
- Domestic policies to relax resource rigidities (especially labor) can increase welfare gains (Scenario 2 shows larger positive impacts).
- Multilateral liberalization and measures to attract investment or relax capital restrictions could raise the limited investment-driven GDP contribution.
- Address supply-side constraints to better meet increased SADC demand (South Africa faces supply-side constraints limiting production response).
- For Madagascar specifically:
  - Policies to diversify fabric sourcing, develop domestic fabric production, and exploit expanded access to South Africa could help maintain AGOA benefits and amplify textile/apparel gains.
- Fiscal policy consideration: anticipate and plan for reductions in customs tariff revenue (estimated customs tariff revenue reduction about 8.0 percent; taxes on international trade reduction about 2.6 percent).

*Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0766.pdf*

### References..............................................................................................................

### _wp0766 - References..............................................................................................................

### I. Introduction
- In 2007, Madagascar plans to start phasing out its customs tariffs on imports from the Southern African Development Community (SADC) and gain better market access to the bloc.
- The paper uses the Global Trade Analysis Project (GTAP) model to simulate several scenarios to quantify the impact of the SADC FTA on all sectors of the economy.
- Key simulation result summary:
  - Reciprocal elimination of customs tariffs between SADC countries and Madagascar would only have a limited impact on Madagascar’s real GDP.
  - The actual liberalization under the SADC FTA affects less than 6 percent of Madagascar total imports.
  - Benefits of the SADC FTA are limited under regional liberalization alone, but become significant when combined with multilateral liberalization because:
    - The liberalization then affects all Madagascar trade.
    - A multilateral liberalization reduces the cost associated with potential trade diversion.

### II. Madagascar trade patterns and trade policy
- Unilateral tariff liberalization:
  - Madagascar reduced its simple average MFN tariff from 16.2 percent at the end of 2005 to 13.6 percent in 2006, slightly below the SSA average of 14.5 percent.
  - Excluding other duties and charges, the average is 13 percent in Sub-Saharan Africa.
  - Tariff bands declined from five to four (0, 5, 10, 20); except for 19 tariff lines (out of 6155) on oil, oil products, and gases, all tariffs are ad valorem.
  - Authorities intend to “further streamline the tariff structure,” “eliminate tariff distortions by reviewing the categorization of merchandise,” and conclude a “study of the advisability and fiscal and economic impact of a uniform nonzero customs tariff and of other options” (IMF, 2006b).
- Preferential trade agreements and overlapping memberships:
  - Madagascar has been a member of COMESA since 1993 and part of the COMESA FTA since 2000.
  - Madagascar’s imports from SADC account for 12.5 percent of its total imports.
  - Four SADC members already have duty-free access to Madagascar because they also belong to the COMESA FTA; these four countries account for 6.9 percent of Madagascar imports and 55 percent of Madagascar imports from SADC countries.
  - Therefore, the SADC FTA will lead to elimination of customs tariffs on 5.6 percent of Madagascar imports, not 12.5 percent.
  - 88 percent of Madagascar imports that will be actually liberalized by the SADC FTA are from South Africa.
- Trade openness trend:
  - Madagascar trade openness (share of exports and imports in GDP) increased from 15 percent of GDP in 1985 to about 33 percent in 2005.
- Footnotes and data sources:
  - SADC members listed in Table 1; IMF’s Direction of Trade database used for import shares.

### III. Joining SADC: process and motivations
- Timetable and asymmetric treatment:
  - Madagascar plans to start phasing out its tariff on imports from SADC in 2007; by 2012, tariff on SADC imports will be almost fully eliminated (Table 2).
  - Madagascar’s tariff reduction on SADC imports discriminates between South Africa and the rest of SADC, reflecting SADC provisions of asymmetry of treatment between member states at different levels of development.
  - Tariffs will remain on sugar products as well as oil, petroleum products, and gases.
  - SADC plans to achieve a customs union by 2010.
- Expected benefits highlighted by Malagasy authorities:
  - Regional integration expected to increase trade openness and foster investment (foreign direct investment, especially from South Africa).
  - Government statements: joining SADC seen as promoting strong economic growth by attracting new investments and opening Madagascar to the world economy.
  - Madagascar expects to benefit from elimination of tariffs on its exports to SADC members not part of COMESA (members of SACU, Mozambique, and Tanzania). In 2006, these countries’ average tariff was 11.4, 12.1, and 7.2 percent respectively.
  - Preferential access by Democratic Republic of Congo and Angola to Madagascar exports will be increased.
- Rationale and constraints from literature and past experience:
  - Potential trade from African RTAs may be low; gravity-model studies suggest actual intra-African trade is not below expectations (Foroutan and Pritchett (1993); Chauvin and Gaulier (2002)).
  - Historical limits to intra-regional liberalization in Africa due to: (i) import-substitution policies, (ii) tariff-revenue constraints, (iii) unequal distribution of costs and benefits, (iv) severe trade-regime distortions, (v) high transaction costs from inadequate infrastructures, (vi) limited product complementarities, and (vii) institutional constraints and conflicting goals from multiple memberships.
  - Complementary policies (macroeconomic stability, appropriate exchange rate, investment policies) often not implemented, limiting growth effects from liberalization.
- Investment and sectoral expectations:
  - Investment is highlighted as crucial; in 2005, investment was 22.5 percent of GDP, 10.4 percentage points higher than saving.
  - Madagascar expects increased FDI, notably from South Africa; in December 2006 Madagascar and South Africa signed a protocol on mutual investment promotion.
  - Agriculture accounts for 26 percent of GDP and 31 percent when agro-industry is added (IMF, 2005b).
  - The Malagasy clothing industry expects that reduction of a high customs tariff on its export to South Africa (36.6 percent) will boost exports and help diversify away from U.S. and EU markets.
  - In 2006, South Africa’s average tariff on textiles and apparel from SADC was 8.3 percent.
  - Rice sector singled out because rice accounts for 1/3 of agricultural production.

### IV. The impact of the SADC FTA
#### A. Model and methodology
- Model: Global Trade Analysis Project (GTAP), a Computable General Equilibrium (CGE) model based on neoclassical theory; GTAP database benchmarked to 2001 and results presented in 2001 U.S. dollars.
- Aggregation:
  - Regions aggregated into 8 regions; industries aggregated into 6 industries (Table 3).
  - SADC split into Madagascar, South Africa, and rest of SADC; rationale: Malagasy expectations about South Africa, evidence in literature on South Africa as exception for intra-African trade potential, and 88 percent of Madagascar imports liberalized under SADC FTA being from South Africa.
  - Industry aggregation highlights: Agriculture and food, Rice, Textiles, Apparel, Other manufacturing, Services.
- Two-step approach:
  1. Update GTAP database for liberalization of trade in textiles and clothing in early 2005:
     - All bilateral quota restrictions on textiles are eliminated except U.S. and E.U. quotas on imports.
     - The “export taxes” on Chinese textiles and clothing to the U.S. and E.U. markets were halved.
  2. Simulate implementation of a full SADC FTA (total and reciprocal elimination by all SADC countries of import tariffs on intra-regional trade) under two alternative scenarios:
     - Scenario 1: All prices and wages flexible to maintain current level of employment.
     - Scenario 2: Real wages of skilled and unskilled labor in all countries and regions assumed rigid so the SADC FTA can affect the level of employment.

#### B. A marginal impact on GDP and welfare
- Objective: Assess impact on real GDP and welfare (welfare measured as equivalent variation in income using base period prices).
- Summary finding: Under all scenarios, the impact of SADC FTA on Madagascar’s real GDP and welfare is negligible.
- Table 4: Impact on Real GDP (in millions of 2001 U.S. dollars, percentage changes in parentheses)
  - Scenario 1 / Scenario 2
    - Madagascar: -0.1   (0) / 1.3  (0)
    - South Africa: 34.4   (0) / 595.9  (0.5)
    - Rest of SADC: -111.1  (-0.2) / 500.6  (0.9)
    - European Union: -0.5  (0) / -76.0  (0)
    - United States: 1.0  (0) / -52.0  (0)
    - Canada: -0.1  (0) / -0.8  (0)
    - China: 0.4  (0) / -10.3  (0)
    - Rest of the World: -39.0 (0) / -201.0  (0)
    - World: -115.0 (0) / 757.7  (0)
- Scenario-specific GDP effect for Madagascar:
  - In scenario 2, Madagascar’s real GDP would increase by 0.03 percent, while in scenario 1 it would not be affected.
  - Difference arises from the impact of increased employment under scenario 2.
- Welfare decomposition note:
  - Changes in welfare can be decomposed into allocative efficiency effects, technology effects, endowment effects, and terms of trade effects; GTAP welfare measures use equivalent variation in income.

*Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0766.pdf*

### Appendix I does not report the technology effect which is set to zero in the simulations presented in this paper.

### _wp0766 - Appendix I does not report the technology effect which is set to zero in the simulations presented in this paper.

### Employment and wages
- Scenario 2: SADC FTA increases skilled labor employment by +0.1 percent and unskilled employment by +0.05 percent.
- Real wages are assumed to remain constant in Scenario 2; nominal wages slightly decline by -0.05 percent (Table 5).
- Table 5 — Change in Real Wages in the SADC (in percent):
  - Madagascar
    - Unskilled labor 0.02
    - Skilled labor 0.05
  - South Africa
    - Unskilled labor 0.26
    - Skilled labor 0.25
  - Rest of SADC
    - Unskilled labor 0.98
    - Skilled labor 1.11
  - Note: "1/ Scenario 2 is based on the assumption that there is no change in real wages of both skilled and unskilled labor."

### Welfare impacts and distribution
- Madagascar: impact is marginal and mostly due to a deterioration of the terms of trade; only Scenario 2 (employment increase) yields a welfare gain via an endowment effect.
- SADC as a whole: FTA would improve welfare by $74 million US to $1.2 billion US, but distribution ambiguous — South Africa gains under both scenarios; welfare impact on the rest of SADC and Madagascar ambiguous.
- World welfare:
  - Scenario 1: decline by 115 millions US dollars.
  - Scenario 2: increase by 757 millions US dollars.
- In both scenarios, SADC area trade with the rest of the world declines.

### Terms of trade and price changes
- Table 6 — Change in the Terms of Trade (In percent):
  - Madagascar
    - Terms of Trade -0.07 (Scenario 1), -0.06 (Scenario 2)
    - Px -0.05, -0.05
    - Pm 0.02, 0.01
  - South Africa
    - Terms of Trade 0.68, 0.72
    - Px 0.67, 0.71
    - Pm -0.01, -0.02
  - Rest of SADC
    - Terms of Trade -0.31, -0.41
    - Px -0.11, -0.21
    - Pm 0.20, 0.21
  - Note: "1/ Changes in terms of trade on non-SADC countries (as well as import and export prices) are close to zero."

### Trade creation versus trade diversion
- South Africa: appears to benefit from net trade creation.
  - Allocative efficiency gains and replacement of some inefficient domestic production by imports (rice and apparel).
  - Example sectoral changes:
    - Rice value added decreases by about 2 percent; rice imports increase by 3 percent.
    - Apparel value added declines by 0.4 to 1.3 percent while apparel imports increase by about 8 percent; exports increase by 4 percent.
  - Exports to SADC increase by 25 percent (Appendix III).
  - South Africa real GDP simulated to be higher by 0.8 to 1.4 percent.
  - Import prices for South Africa do not increase; imports from third countries would increase.
  - SADC exports to South Africa account for only 2.1 percent of South Africa’s imports (excluding trade with other SACU countries).
- Madagascar and rest of SADC: the FTA tends to be trade diverting.
  - Madagascar: imports from South Africa increase by about 9 percent; imports from rest of SADC increase by 2 percent; imports from non-SADC decline by 0.5 percent.
  - Trade diversion not offset by trade creation for Madagascar because:
    - (i) efficiency gains negligible;
    - (ii) total trade barely increases (+0.4 percent at most, Appendix II);
    - (iii) domestic production not significantly affected by imports (Appendix IV).

### Impact on GDP and its components
- Table 7 — Change in Volume of Trade of SADC members (in percent):
  - Madagascar
    - Exports 0.4 (Scenario 1), 0.5 (Scenario 2)
    - Imports 0.3, 0.4
  - South Africa
    - Exports 0.5, 0.4
    - Imports 2.1, 2.8
  - Rest of SADC
    - Exports 3.8, 4.4
    - Imports 4.9, 5.9
- Table 8 — Decomposition of the Change in Real GDP (in percent):
  - Madagascar
    - GDP 0.00, 0.03
    - Consumption -0.02, 0.01
    - Investment 0.01, 0.02
    - Exports 0.10, 0.10
    - Imports -0.09, -0.10
  - South Africa
    - GDP 0.0, 0.5
    - Consumption 0.2, 0.6
    - Investment 0.2, 0.5
    - Exports 0.2, 0.2
    - Imports -0.6, -0.7
  - Rest of SADC
    - GDP -0.2, 0.9
    - Consumption -0.3, 0.6
    - Investment 0.5, 0.8
    - Exports 1.6, 1.8
    - Imports -1.9, -2.4
- Net exports do not contribute to increase Madagascar GDP; for South Africa and rest of SADC net exports tend to have a negative impact.

### Investment, capital and dynamic considerations
- Investment positively contributes to real GDP but impact limited in Madagascar and smaller than in South Africa or rest of SADC.
- Capital account would improve marginally by 0.1 to 0.2 percent owing to limited increase in inflows of foreign capital.
- Model assumption: scenarios 1 and 2 assume endowment in capital is fixed. Relaxing this assumption (section G) leaves Scenario 1 contribution unchanged but increases Scenario 2 investment contribution from 0.02 to 0.08 percent (Table 12).
- Model does not allow for dynamic gains affecting investment and FDI inflows; historical note: following trade reform in the late 1980s, Madagascar experienced a significant increase in FDI inflows (Harrison and Revenga, 1995), but that may be related to export processing zones (EPZ) introduced in 1989 rather than tariff cuts.

### Changes in trade geography and sectoral composition
- Table 9 — Change in the Geographical Structure of Trade (in percent):
  - SADC: Total trade 2.7 (Scenario 1), 3.2 (Scenario 2); Intra-SADC 25.8, 27.1; Trade with third countries -1.8, -1.5
  - Madagascar: Total trade 0.3, 0.4; Trade with SADC 4.8, 5.1; Trade with third countries -0.2, -0.1
  - South Africa: Total trade 1.6, 1.9; Trade with SADC 22.6, 23.7; Trade with third countries -1.4, -1.2
  - Rest of SADC: Total trade 3.6, 5.2; Trade with SADC 29.0, 30.3; Trade with third countries -2.7, -2.1
  - Note: "1/ Total trade is the sum of imports and exports measured at world price."
- Madagascar:
  - Total trade barely increases (0.4 percent at most) while trade with SADC increases by about 5 percent and trade with the rest of the world declines.
  - Madagascar’s trade with South Africa increases by +12.5 percent; exports to South Africa expected to double (Appendix III).
  - Madagascar trade with rest of SADC increases by less than 1 percent; Madagascar’s exports to rest of SADC drop by 8 percent.

### Sectoral winners and losers; textiles and apparel focus
- Table 10 — Change in Madagascar Trade Balance (in millions of 2001 US dollar):
  - Food -1.4, -1.5
  - Rice 0.0, 0.0
  - Textiles 1.9, 2.1
  - Clothing 0.9, 1.0
  - Other manufacturing -2.0, -2.3
  - Services 0.3, 0.1
  - TOTAL -0.3, -0.6
- Textiles and clothing are major beneficiaries:
  - Exports of textiles to South Africa jump by more than 500 percent; exports of clothing by 200 percent.
  - Imports from South Africa increase by about 58 percent (textiles) and 116 percent (clothing).
  - Madagascar total exports of textiles limited to about 1 percent of exports; exports of apparel to about 0.5 percent.
- AGOA relevance:
  - AGOA’s third-country provision (duty- and quota-free access for African apparels made from non-African and non-U.S. fabrics) was initially scheduled to expire in September 2007 but recently extended until 2012; elimination would restrict eligibility to apparel made with African or U.S. fabrics.
  - Madagascar’s apparel industry imports inputs mostly from Asia; the extension gives time to diversify sourcing and export markets.
- Caveats limiting textile/apparel expansion:
  - SADC Trade Protocol treats textiles and clothing as sensitive; liberalized more slowly.
  - SADC’s strict rules of origin and South Africa’s willingness to protect its textiles and clothing might prevent strong expansion.

### Resource allocation and factor demand in Madagascar
- Production composition shifts toward textiles and clothing:
  - Real value added of textiles expands by about 0.9 percent; clothing by about 0.3 percent.
  - Other goods’ value added declines slightly.
  - South Africa: clothing value added declines by -0.4 to -1.3; textiles value added increases up to 0.9 percent.
  - Food and processed food winners: South Africa +1.4 to +1.9 percent; Madagascar -0.1 percent.
- Table 11 — Demand for Factor of Production in Madagascar (percentage change) — "1/ First figure refers to scenario 1 and the second to scenario 2."
  - Unskilled Labor / Skilled Labor / Capital:
    - Food -0.1 / 0, -0.1 / 0, -0.1
    - Rice -0.1 / 0, -0.1 / 0, -0.1
    - Textiles 1.0, 0.9 / 1.0, 0.9 / 1.0
    - Clothing 0.3 / 0.4, 0.3 / 0.4, 0.3
    - Other manufacturing 0, -0.1 / 0, -0.1 / 0
    - Services 0 / 0.1, 0 / 0.1, 0
- Employment shifts to textiles and clothing for both skilled and unskilled labor; when total employment grows (Scenario 2), increases in textiles and clothing employment no longer occur at expense of other sectors — employment in other manufacturing does not decrease and may increase slightly.

### Policy implications and maximizing gains
- Scenarios interpretation:
  - Scenario 1: quantity of each endowment fixed — reflects full utilization or rigidities preventing full utilization; effect mainly resource reallocation across sectors.
  - Scenario 2: labor endowment flexible — allows changes in employment; realistic given Madagascar’s labor market structure.
- To maximize gains from the SADC FTA:
  - Domestic policies that relax resource rigidities (especially labor) can increase welfare gains (Scenario 2 shows larger positive impacts).
  - Multilateral liberalization and measures to attract investment or relax capital restrictions could raise the limited investment-driven GDP contribution.
  - Address supply-side constraints to better meet increased SADC demand (South Africa faces supply-side constraints limiting production response).
  - For Madagascar, policies to diversify fabric sourcing, develop domestic fabric production, and exploit expanded access to South Africa could help maintain AGOA benefits and amplify textile/apparel gains.

*Italic: Source — _wp0766 - Appendix I does not report the technology effect which is set to zero in the simulations presented in this paper.*

### 8.3 million on a population of 18.1 million (MIGA, 2006), 82 percent of the labor force was

### _wp0766 - 8.3 million on a population of 18.1 million (MIGA, 2006), 82 percent of the labor force was

### Impact of the decline in the price of capital (Table 12)
- Table 12 (In percent) contributions to growth:
  - GDP: Scenario 1 0.00, Scenario 1b 0.02, Scenario 2 0.03, Scenario 2b 0.43
  - Consumption: Scenario 1 -0.02, Scenario 1b -0.01, Scenario 2 0.01, Scenario 2b 0.38
  - Investment: Scenario 1 0.01, Scenario 1b 0.01, Scenario 2 0.02, Scenario 2b 0.08
  - Exports: Scenario 1 0.10, Scenario 1b 0.09, Scenario 2 0.10, Scenario 2b 0.22
  - Imports: Scenario 1 -0.08, Scenario 1b -0.07, Scenario 2 -0.10, Scenario 2b -0.25
- Under scenario 1b (relax fixed capital endowment), results are similar to relaxing fixed employment (scenario 2): SADC FTA increases real GDP by 0.02 percent of GDP and investment increase remains negligible.
- When both labor and capital endowments can rise (scenario 2b), effects are larger:
  - GDP increase: 0.4 percent
  - Investment increase: 0.08 percent
  - Bulk of GDP contribution comes from increased consumption.

### Welfare and endowment effects
- Welfare impacts (US$) and per capita changes:
  - Endowment-related employment effect: $10 million US
  - Larger endowment in capital effect: about $8 million dollars
  - Welfare impact of SADC FTA for Madagascar:
    - Scenario 1: loss of $1 million US
    - Scenario 2b: gain of $17.5 million US
  - Per capita welfare:
    - Scenario 1: about $0.06 per capita
    - Scenario 2b: $1 US per capita
- Mechanism: labor and capital are complementary; allowing both to increase removes constraints that limit gains when only one factor can vary.

### Scenarios and interpretation of initial endowment utilization
- Scenario interpretations:
  - Scenario 1: full utilization of both capital and employment or rigidities preventing movement to full employment.
  - Scenario 2: relaxes assumption for employment.
  - Scenario 1b: relaxes assumption for capital.
  - Scenario 2b: relaxes both employment and capital utilization.
- Policy implication: if rigidities affect utilization of endowments, complementary policies to remove these rigidities will increase gains from the SADC FTA.
- Evidence on constraints from World Bank investment climate assessment (Shah and al., 2005):
  - 67 percent of firms: cost of finance as a major or severe constraint
  - 59 percent: access to finance
  - 31 percent: skills and education of workers
  - 15 percent: labor regulations

### Limits of the SADC FTA gains and role of multilateral liberalization
- Under all scenarios, gains from the SADC FTA are limited due to:
  - (i) it is a small trade liberalization;
  - (ii) regional (discriminatory) liberalization involves costs (trade diversion).
- Illustrative combined scenario (Table 13): full intra-SADC liberalization + uniform nondiscriminatory 10 percent multilateral reduction of applied customs tariff (cut limited for SADC to trade with non-SADC countries while intra-SADC trade is fully liberalized).
  - Scenario 1:
    - SADC only: Real GDP (change in percent) 0.0, Welfare (millions of U.S. dollars) -1.0, Welfare (dollars per capita) -0.1
    - SADC and a multilateral tariff cut: Real GDP 0.12, Welfare 7.9, Welfare per capita 1.6
  - Scenario 2:
    - SADC only: Real GDP 0.0, Welfare 0.5, Welfare per capita 0.0
    - SADC and a multilateral tariff cut: Real GDP 1.59, Welfare 94.7, Welfare per capita 5.3
- Conclusion: a nondiscriminatory multilateral liberalization both increases coverage and eliminates trade diversion costs, producing much larger gains.

### Fiscal implications and trade revenue effects
- Madagascar has one of the lowest tax revenue-to-GDP ratios in the world: about 10 percent of GDP in 2005.
- About half of tax revenue is from taxes on international trade.
- Phasing out tariffs on South African imports would reduce Madagascar customs tariff revenue and taxes on international trade:
  - South Africa trade-weighted average tariff faced: 10.2 percent
  - Rest of world trade-weighted average tariff faced: 7.2 percent
  - South Africa share: 4.8 percent of Madagascar total imports but 6.2 percent of Madagascar dutiable imports
  - Resulting reductions:
    - Reduction in Madagascar revenues from customs tariff: about 8.0 percent
    - Reduction in taxes on international trade (customs duties, tax on petroleum products, VAT, and excise taxes): about 2.6 percent
- Second-round effect: simulations indicate imports from non-SADC members decline, implying further negative impact on customs revenue since only imports from non-SADC countries continue to face customs duty.
- Institutional note: SADC has no provision for compensating fiscal losses, except privileged access to the SADC regional development fund.

### Trade and sectoral impacts
- Madagascar’s trade with South Africa would increase substantially under the SADC FTA; impact on trade with rest of SADC is limited; trade with third countries would decline.
- Most change in trade driven by textiles and clothing benefiting most from SADC FTA.
- Tariff elimination would affect less than 6 percent of Madagascar imports, limiting GDP and welfare impact.

### Caveats and dynamic considerations
- Simulations are benchmark estimates; they:
  - Make no allowance for potential dynamic gains or costs (Wacziarg, 2001).
  - Do not include other gains such as increased credibility of policies or infrastructure improvements.
- SADC emphasis on infrastructure constraints and two protocols signed in December 2006 between Madagascar and South Africa to promote investment and develop economic and physical infrastructures could matter for dynamic gains.
- Further analysis is needed on fiscal impacts and dynamic channels.

### Appendix highlights
- Origin of Madagascar imports:
  - Imports from Africa represent only 15 percent of Madagascar total imports, with the share increasing since 2001.
  - Virtually all Madagascar imports from Africa are from COMESA and SADC; exceptions: Côte d’Ivoire (0.1 percent) and Maghreb countries (0.2 percent).
  - In 2005, 89 percent of COMESA imports are from countries that are members of both SADC and COMESA (90 percent of which is imports from Mauritius).
  - In 2005, only 35 percent of SADC imports were from countries belonging both to SADC and COMESA.
  - South Africa accounts for 40 percent of Madagascar imports from SADC.
  - Mauritius import value was 12 times higher in 2005 than in 2000.
- Appendix II: Decomposition of the Welfare Effect (In millions of 2001 U.S. dollars) — selected entries:
  - Madagascar:
    - Scenario 1 TOTAL -1.0
    - Scenario 2 TOTAL 0.5
  - South Africa:
    - Scenario 1 TOTAL 255.5
    - Scenario 2 TOTAL 833.8
  - WORLD:
    - Scenario 1 TOTAL -115.1
    - Scenario 2 TOTAL 756.5

*Content derived from the provided IMF PDF text.*

### 1. Total trade

### 1. Total trade

### 1.1 Aggregate trade outcomes
- Imports: 0.3 / 0.4 2.1 / 2.8 5.1 / 6.1
- Exports: 0.4 1.1 / 1.2 3.8 / 4.3
- Total: 0.3 / 0.4 1.6 / 1.9 4.4 / 5.2

### 2 Trade with SADC
- Imports: 4.5 / 4.8 14.1 / 15.3 29.3 / 30.6
- Exports: 6.9 / 7.7 25.0 / 26.1 28.0 / 29.5
- Total: 4.8 / 5.1 22.6 / 23.7 29.0 / 30.3

### 2.1 Trade with South Africa
- Imports: 8.5 / 8.7- 25.2 / 26.4
- Exports: 104.2 / 105.6- 13.9 / 15.1
- Total: 12.3 / 12.5- 22.9 / 34.1

### 2.2 Trade with Rest of SADC
- Imports: 1.8 / 2.4 13.9 / 15.1 65.7 / 68.0
- Exports: -7.5 / -8.2 25.2 / 26.4 64.5 / 66.8
- Total: 0.3 / 0.9 22.6 / 23.8 65.1 / 67.4

### 3 Trade with the rest of the World
- Imports: -0.5 1.3 / 2.0 -6.4 / -7.3
- Exports: 0.2 -3.6 / -3.8 0.7 / 1.1
- Total: -0.1 / -0.2 -1.2 / -1.4 -2.1 / -2.7

- Note: 1/ Imports and exports are measured at world price. Values are the results for scenarios 1 and 2.

### Appendix IV — Change in Real Value Added by Sector (selected entries)
- Madagascar
  - Scenario 1: Agr. -0.1 -0.1; Food 0.9 0.3; Rice -0.0; Textiles 0.0
  - Scenario 2: Agr. -0.0 -0.0; Food 1.0 0.4; Rice -0.0; Textiles -0.0
- South Africa
  - Scenario 1: 1.4 -2.0 -0.1 -1.3 -0.4 0.0
  - Scenario 2: 1.7 -2.5 0.4 -0.9 -0.1 0.6
- Rest of SADC
  - Scenario 1: -0.5 -0.4 3.5 6.4 -0.2 0.1
  - Scenario 2: 0.4 0.4 4.9 7.9 0.6 1.4

- Note: 1/ Changes for non-SADC regions are below 0.05 percent for all sectors in all scenarios and are thus not reported here.

### Appendix V — Change in Madagascar Structure of Trade (changes in percent)
- Change in imports at world prices (%) from / Change in exports at world prices (%) to

- To South Africa
  - Food: Imports Scenario 1 43.9; Scenario 2 43.9 — Exports Scenario 1 16.0; Scenario 2 16.5
  - Rice: Imports -; Scenario 2 - — Exports -; -
  - Textiles: Imports 57.8; 57.8 — Exports 535.9; 537.9
  - Clothing: Imports 116.0; 116.0 — Exports 204.8; 205.8
  - Other manufacturing: Imports 6.1 6.1 — Exports 27.2 28.0
  - Services: Imports -2.2 -2.2 — Exports 1.7 2.2

- To Rest of SADC
  - Food: Imports 3.0 3.0 — Exports -14.3 -13.8
  - Rice: Imports -1.4 -1.4 — Exports - -
  - Textiles: Imports 0.7 0.7 — Exports -3.5 -2.6
  - Clothing: Imports 1.3 1.3 — Exports -2.9 -2.1
  - Other manufacturing: Imports 5.7 5.7 — Exports -7.3 -6.1
  - Services: Imports -0.1 -0.1 — Exports - 0.9

- To Rest of the World
  - Food: Imports -1.2 -1.2 — Exports 0.1 0.1
  - Rice: Imports -0.2 -0.2 — Exports - -
  - Textiles: Imports -0.2 -0.2 — Exports 0.3 0.4
  - Clothing: Imports -0.5 -0.5 — Exports 0.3 0.3
  - Other manufacturing: Imports -0.7 -0.7 — Exports 0.2 0.1
  - Services: Imports -0.1 -0.1 — Exports 0.1 0.1

- Total imports / Total exports
  - Food: Imports 0.8 0.8 — Exports -0.1 -0.1
  - Rice: Imports -0.3 -0.3 — Exports - -
  - Textiles: Imports 0.5 0.5 — Exports 1.2 1.3
  - Clothing: Imports 0.1 0.1 — Exports 0.4 0.5
  - Other manufacturing: Imports 0.4 0.4 — Exports 0.2 0.2
  - Services: Imports -0.1 -0.1 — Exports 0.1 0.1

### Appendix VI — The GTAP Model (summary of model structure and limitations)
- Model type and assumptions:
  - Comparative static, global general equilibrium model based on neoclassical theory.
  - Firms maximize profits; consumers maximize utility.
  - Markets are perfectly competitive; constant returns to scale prevail.
- Production structure:
  - Firms use a composite of primary factors and a composite of intermediates with Leontief production technology.
  - Primary factor composite: CES function of labor, capital, land and natural resources.
  - Intermediate composite: Leontief of material inputs; material inputs are CES blends of domestic goods and imports (Armington approach).
- Demand structure:
  - Each region has a “super” household allocating income to private consumption, government expenditure and savings in fixed proportions.
  - Household consumption: constant difference in elasticities function of consumer goods.
  - Government expenditure: CES function of commodities.
  - Consumption and government demand are CES blends of domestic goods and imports sourced from all regions (Armington approach).
- Closure and capital/savings treatment:
  - Regional savings are homogenous and contribute to a global pool allocated among regions for investment in response to changes in expected rates of return.
  - Changes in expected rates of return are equalized across regions, implying capital (savings) mobility across regions.
  - Capital stocks are immobile across regions but perfectly mobile within a region; labor is perfectly mobile within a region.
  - Land and natural resources are industry-specific with limited transformation possibilities.
- Limitations:
  - Simplicity aids interpretation but limits treatment of adjustment paths over time and long-term effects related to investment accumulation, technology and productivity change.
  - Adjustment costs associated with trade liberalization are absent.

*Source: _wp0766 - 1. Total trade*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0766.pdf_