regional trade integrations: a comparative study

United Nations Conference on Trade and Development Virtual Institute Research Material

REGIONAL TRADE INTEGRATIONS: A COMPARATIVE STUDY THE CASES OF GAFTA, COMESA, AND SAPTA/SAFTA Authors Parts A and B (Comparative Analysis and GAFTA) Talib Awad, Professor of International Economics, Faculty of Business, University of Jordan([email protected]) Amir Bakir, Assistant Professor, Faculty of Business, University of Jordan([email protected]) Part C (COMESA) Rojid Sawkut, Lecturer, University of Mauritius ([email protected]) Sannassee Vinesh, Senior Lecturer, University of Mauritius ([email protected]) Seetanah Boopen, Lecturer, University of Technology of Mauritius ([email protected]) Fowdar Suraj, Lecturer, University of Mauritius ([email protected] ) Part D (SAPTA/SAFTA) Meeta Keswani Mehra, Associate Professor, CITD/SIS, Jawaharlal Nehru University, India Manoj Pant, Professor of Economics, CITD/SIS, Jawaharlal Nehru University, India

With research assistance from Saptarshi Basu Roy Choudhury, M.Phil. Scholar, CITD/SIS, Jawaharlal Nehru University, India Amit Sadhukhan, M.Phil. Scholar, CITD/SIS, Jawaharlal Nehru University, India

UNCTAD Copyright © 2008 1

mailto:[email protected]




TABLE OF CONTENT A. COMPARATIVE ANALYSIS............................................................ 6

I. Introduction .......................................................................... 6 II. Type and coverage of trade agreements ............................. 6 III. Intra-regional trade development and characteristics .......... 7

III.1 Intra-regional trade development ................................................ 7 III.2 Commodity structure.................................................................. 8 III.3 Trade direction .......................................................................... 8

IV. Impact of trade agreements on intra-regional trade ............. 9 IV.1 Revealed comparative advantage analysis .................................. 9 IV.2 Partial regression analysis ......................................................... 10

V. Intra trade obstacles ............................................................. 11 VI. Conclusion ........................................................................... 12

B. REGIONAL TRADE INTEGRATION: THE CASE OF GAFTA . 14 I. Introduction .......................................................................... 14 II. Overview: Theoretical background ....................................... 15

II.1 Economic analysis of regional economic integration............................... 15 II.2 Economic effects on member countries ................................................. 15

II.2.1 Trade creation and diversion ........................................................ 15 II.2.2 Scale and competition effects ....................................................... 18

II.3 Winners and losers .............................................................................. 19 III. Regional economic integration and the Great Arab Free Trade

Area ..................................................................................... 19 III.1 Previous economic integration efforts ......................................... 19 III.2 The Greater Arab Free Trade Area ...................................................... 20

IV. Trend and characteristics of intra-Arab trade ....................... 24 IV.1 Growth of intra-Arab trade .......................................................... 24 IV.2 Intra-Arab trade commodity structure .......................................... 25 IV.3 Intra-Arab trade destination ........................................................ 26

V. Measuring the effect of GAFTA on intra-Arab trade ............. 26 V.1 Revealed comparative advantage ............................................... 28 V.2 Partial regression analysis ......................................................... 29

VI. Intra-Arab trade problems and obstacles ............................. 31 VII. Conclusion and recommendations ....................................... 34

C. THE CASE FOR COMESA .............................................................. 38 I. Introduction .......................................................................... 38 II. Regional trade blocks in Africa ............................................. 38 III. Economic structure of some COMESA Members ................ 50 IV. Measuring the effect of COMESA on intra-COMESA trade: A

preliminary analysis ............................................................. 55 V. The role of COMESA exports in COMESA economic growth

............................................................................................. 56


D. REGIONAL TRADE AGREEMENTS IN SOUTH ASIA: PROSPECTS FOR INTRA-REGIONAL TRADE COOPERATION............................................................................................................................................................. 65

I. Introduction and scope ......................................................... 65 II. Geo-political milieu of regional cooperation in South Asia ... 67 III. SAARC Preferential Trading Arrangement (SAPTA) ........... 68 IV. South Asia Free Trade Area (SAFTA) Agreement ............... 68 V. Bilateral initiatives................................................................. 70 VI. Multilateral initiatives outside the region............................... 72 VII. Broad economic and trade characteristics of South Asia ..... 74

VII.1 Economic characteristics of the SAARC region............................ 74 VII.2 Trade characteristics of the SAARC region.................................. 75 VII.3 Assessing the scope for intra-regional trade ................................ 79

VIII. Summary and conclusions ................................................... 90

FIGURES Figure B1: Intra Arab Trade, 1980-2005................................................... 24 Figure B2: Intra exports and exports with the rest of the world (US$

millions) .................................................................................................. 27 Figure B3: Intra imports and imports with ROW ($US million) ............. 28 Figure D1: SAARC’s GDP (billion US$ in PPP terms) ........................... 74 Figure D2: Share of SAARC’s GDP in world GDP (per cent) ............... 75 Figure D3: Year to year growth of intra-group exports of SAARC

(per cent)................................................................................................ 76 Figure D4: Proportion of SAARC’s intra-group trade to its aggregate

GDP (per cent) ...................................................................................... 76 Figure D5: Proportion of SAARC’s intra-regional exports to its

exports to ROW (per cent) .................................................................. 77 Figure D6: Proportion of SAARC’s intra-regional imports to its

imports from ROW (per cent) ............................................................. 77 Figure D7: Intra-group exports and SAARC’s exports to ROW

(current US$ million) ............................................................................ 82 Figure D8: Intra-group imports and SAARC’s imports from ROW

(current US$ million) ............................................................................ 82


TABLES Table A1: Type and coverage of trade agreements ............................... 7 Table A2: Share of exports by main products, 1985-2005 .................... 8 Table A3: Regression results of intra exports for GAFTA, COMESA,

and SAPTA/ SAFTA............................................................................. 10 Table A4: Regression results of intra imports for GAFTA, COMESA,

and SAPTA ............................................................................................ 11 Table B1: Growth rates of intra trade during the period 1980-2004 .... 25 Table B2: Intra-Arab trade growth rates for selected countries (per

cent) ........................................................................................................ 25 Table B3: Relative importance of intra-Arab exports according to

commodity groups ................................................................................ 26 Table B4: RCA indices for GAFTA countries by broad commodity

groups (2003, 2004)............................................................................. 29 Table B5: Estimation results for intra exports and intra imports in

logarithm form ....................................................................................... 30 Table C1: Basic economic indicators of COMESA countries and

sectors’ share of GDP (per cent), 2002 ............................................ 51 Table C2a: Share of exports (per cent) .................................................... 52 Table C2b: Share of COMESA intra exports as a percentage of

COMESA total exports ........................................................................ 52 Table C2c: Average annual growth rate of exports and imports for

the COMESA region (per cent) .......................................................... 52 Table C3: Products exported as a percentage of COMESA total

intra exports ........................................................................................... 53 Table C4: GDP (PPP) yearly growth rate (per cent, 1995 prices) ....... 53 Table C5: GDP per capita yearly growth rate (per cent, 1995

prices) ..................................................................................................... 54 Table C6: OLS estimates in logarithm ...................................................... 56 Table C7: Cross section and random effects estimates ........................ 58 Table C8: Random effects panel estimates (COMESA exports to

COMESA and US and EU) ................................................................. 60 Table C-A: RCA for selected COMESA countries in 2000 at 2 digits

SITC ........................................................................................................ 61 Table D1: Preferences under SAPTA ....................................................... 68 Table D2: Intra-group trade of the SAARC region (US$ million) .......... 75 Table D3: Aggregate commodity composition of SAARC’s trade

with the world in 2005 .......................................................................... 78 Table D4: Commodity composition of intra-group trade for the

SAARC region in 2005 ........................................................................ 79 Table D5: Trade Intensity Indices for intra SAARC trade, 1994/95 -

2004/05 .................................................................................................. 80


Table D6: Results of multivariate regression for intra-regional exports.................................................................................................... 84

Table D7: Results of multivariate regression for intra-regional imports.................................................................................................... 84

Table D8: Unit root tests.............................................................................. 86 Table D9: RCA indices for SAARC countries for broad commodity

groups in 2005 ...................................................................................... 87


A. COMPARATIVE ANALYSIS By Talib Awad and Amir Bakir

I. Introduction Following the initiative by the UNCTAD Virtual Institute that was launched during the second annual meeting of the participating member universities, mutual interest in regional trade agreements was expressed by three research groups representing the University of Jordan, the University of Mauritius and the Jawaharlal Nehru University, India. The encouragement and coordination efforts of the Virtual Institute team played an important role in realizing this joint research project. This joint research covers three regional trade agreements in three different parts of the world, the Great Arab Free Trade Area (GAFTA) in the Arab region, the Common Market for Eastern and Southern Africa (COMESA) in the African region, and the South Asia Preferential Trade Agreement (SAPTA) (followed by the recently promulgated South Asia Free Trade Area) in the South Asian region. The comparative part of the study covers the following four dimensions: type and coverage of trade agreements, characteristics and development of intra trade, impact of trade agreements on intra trade, and problems associated with implementing the RTAs.

II. Type and coverage of trade agreements As shown in Table A1 trade integration is limited to the first stage of economic integration, as free trade area, for the GAFTA and COMESA groups, while it took the form of preferential trade agreement for the SAPTA group, which recently culminated in the ratification of SAFTA. The agreement coverage was extensive as it covered most of the countries in their respective regions; seventeen countries in the Arab group, twenty countries in South Asia and seven countries in South-East Africa. In terms of the application of the agreements, the GAFTA was the most effective and comprehensive. Its actual implementation started in 1998 with a clear objective of complete elimination of non-tariff restrictions and gradual cutting of applied tariff rates by an average of 10 per cent annually. In 2002 the zero-tariff goal was achieved among GAFTA members, and most non-tariff restrictions were eliminated in almost all goods. In the case of COMESA however, although the agreement calls for elimination of all trade barriers starting 2000, it is not clear whether this objective was actually implemented. Furthermore, SAPTA was a preferential trade agreement allowing members the freedom to choose the pace, the list of commodities and the mode of trade liberalization, resulting in limited actual tariff reductions and trade preferences among members. By comparison, SAFTA is more comprehensive as well as it specifies a phased reduction in tariffs for the member countries. However, the transition to SAFTA is of recent origin and its impact on regional trade integration remains to be seen.


Table A1: Type and coverage of trade agreements

Title Arabic Group East-South African Group

South Asian Group

Main trade agreement

GAFTA COMESA SAPTA

Type of agreement

Free Trade Area Free Trade Area Preferential Trade Agreement

Coverage 17 countries: Algeria, Bahrain, Egypt, Iraq, Jordan, Kuwait, Lebanon, Morocco, Oman, Palestine, Qatar, Saudi Arabia, Sudan, Syria, Tunisia, UAE and Yemen.

20 countries: Djibouti, Egypt, Kenya, Madagascar, Malawi, Mauritius, Sudan, Zambia, Angola, Burundi Comoros, D.R. Congo, Eretria, Ethiopia, Namibia, Rwanda, Seychelles, Swaziland, Tanzania, Uganda and Zimbabwe.

7 countries: India, Pakistan, Sri Lanka, Bangladesh, Nepal, Bhutan and Maldives. Afghanistan slated to join SAFTA in January 2008.

Starting date of agreement

January 1998 October 2000 December 1995/ January 2006

Trade liberalization measures

Elimination of quotas and other quantitative restrictions. Reduction of tariff rates on traded goods by an average of 10 per cent annually. Preferential treatment is granted to Palestine and Sudan.

Elimination of quotas and other quantitative restrictions. Reduction of tariff rates on traded goods.

Gradual concession on tariff and non-tariff measures with provision for special treatment for least developed countries. Phased elimination of tariffs over 2006-16, with different rates for the least developed members. Elimination of quantitative restrictions.

Source: Based on the three regional studies in parts B-D of this document. In addition to these three major trade agreements, many bilateral and multilateral free trade agreements, either within the regions or with countries outside the region, were signed by individual countries in each region.

III. Intra-regional trade development and characteristics

III.1 Intra-regional trade development Intra-regional trade (from now on intra trade) showed a modest growing trend in all three regions. In the GAFTA group it grew at an average annual rate of 4.8 per cent during 1980-1997. However, annual average growth rates were much higher (more than 18 per cent) after 1998 as shown in Figure B2 of the GAFTA study. However, it must be noted that these growth rates are nominal and most likely due to the recent sharp increases in primary commodities especially fuel. Intra-regional exports (intra exports) of SAARC as percentage of rest of the world exports increased from around 4 per cent over the period 1984-1994, to 5 per cent over the period 1995-2005. Correspondingly, intra-regional imports (intra imports) grew from an average of 2.4 per cent to 4.1 per cent, respectively. The growth rate of SAARC’s intra trade


increased from 8 per cent to 12 per cent, between the two time periods, respectively. As far as COMESA is concerned, intra exports as percentage of total exports increased from 4 per cent in 1980 to 4.8 per cent in 2005.

III.2 Commodity structure In the GAFTA region more than 57 per cent of intra trade in 2005 was fuel and raw materials, followed by food and beverages with a share of 17 per cent, chemicals (14 per cent), and manufactured goods (about 6 per cent). Trade patterns for GAFTA shifted in favor of food and manufactured goods at the expense of fuel and raw materials over the period of the study. In the COMESA region food and beverages share in intra trade exceeded one third, followed by manufactured goods & chemicals (31 per cent), and fuel and raw materials (22 per cent). The share of manufactured goods and food in intra trade increased at the expense of minerals and raw materials over the period of the study. In the SAARC region manufactured goods dominated intra exports with a share of over 52 per cent, followed by primary commodities including fuel (47 per cent). Over the period of the study, the share of primary commodities increased (from 34 per cent to 47 per cent) at the expense of manufactured goods (less significant decline from 59 per cent to 52per cent). Intra-group trade in textile fiber, yarn, fabrics and clothing registered a significant decline. Table A2: Share of exports by main products, 1985-2005

1985 2005 GAFTA* COMESA SAARC*** GAFTA COMESA** SAARC

Fuel and Raw Materials

96.0 43.1 34.0 57.7 21.9 47.5

Food and Beverages

0.9 17.9 - 17.2 35.2 -

Manufactured Goods & Chemicals

2.4 20.0 59.0 20.1 31.2 52.5

Source: Based on the three regional studies in parts B-D of this document. Notes: * For 1980; ** for 2000; *** for 1995. The above analysis shows that while intra trade in the GAFTA is dominated by fuel and raw materials, manufactured goods play more important roles in the intra trade of the other two regions. Over the study period GAFTA and COMESA region witnessed a shift in the pattern of intra trade toward manufactured goods while in the SAARC region the shift was towards primary products.

III.3 Trade direction In the three regions intra trade was characterized by a high level of geographical concentration and in most cases it was restricted between two or three neighboring countries. For example, in the GAFTA region, Oman exported 64 per cent of intra exports to United Arab Emirates, and Libya exported 69 per cent of its intra exports to Tunisia, and 57 per cent of Bahrain’s intra exports were destined to Saudi Arabia, and Iraq’s intra exports with Morocco and Jordan amounted to 47 per cent and 21 per cent


respectively. Similarly, within the SAARC region, trade between Sri Lanka and Maldives was the most intensive in the region, followed by India and Bhutan. In the COMESA region, Djibouti, Ethiopia, Kenya and Uganda exported more than 90 per cent of total intra exports in 2001. The geographical concentration clearly indicates the importance of geographical proximity and political influence in determining the direction of intra trade.

IV. Impact of trade agreements on intra-regional trade To evaluate the impact of regional trade agreements two analytical tools are employed: revealed comparative advantage (RCA) indices and multiple regression models. The RCA index will help identify the possibilities of intra trade expansion, a higher than one value will indicate a revealed comparative advantage in that particular sector. However if revealed comparative advantages are similar among countries of the particular region then there will be limited potential for expanding intra trade within that region. The regression approach is intended to measure the partial impact of the agreement on the regional intra trade. Since the three regional agreements were signed recently (after 1995), the sample under consideration does not permit to divide the sample into two: one before and one after the signing of agreement due to lack of degrees of freedom. An alternative simple approach would be the use of a dummy variable that takes the value of zero in the absence of the agreement, and of one when the agreement is effective. Two models were employed for this purpose: the first for modeling intra exports and the other for modeling intra imports. For the purpose of identifying the substitution effect between intra trade and trade with the rest of the word (ROW), exports to ROW was introduced as a an explanatory variable in the first model, and imports from ROW was introduced as an explanatory variable in the second model. These two variables can serve the purpose of capturing the effects of trade agreements other than intra trade agreements. In addition, total GDP was used as explanatory variable in each equation to capture the economic size effect in each region. Also time was introduced in the model with the purpose of capturing the growing trend in intra trade over time.

IV.1 Revealed comparative advantage analysis Based on the RCA analysis for the three regions, the GAFTA countries showed a high concentration in revealed comparative advantage in primary goods mainly fuel and ores & metals, with almost negligible advantages in other products. This implies a very limited potential for expanding intra trade in the region. COMESA members have comparative advantage in quite similar products which suggests that complementarities as a way to stimulate trade might be limited among some COMESA members. For the SAARC region, there is also a lack of complementarities as indicated by similar RCA indices. This points toward a high degree of competition in export structures, albeit at the aggregated commodity level.


IV.2 Partial regression analysis The above specified regression model is used to explain the effects of the trade agreements on intra trade in the three regions. For the three regions a log linear specification was used. While the GAFTA and SAPTA's samples cover the period (1980-2005), the COMESA study used a slightly smaller sample covering the period (1985-2004). The results of applying the OLS for the three regions are summarized in Table A3: Table A3: Regression results of intra exports for GAFTA, COMESA, and SAPTA/ SAFTA

Variable GAFTA† COMESA SAPTA/ SAFTA

ROW exports 0.44 (4.3)**

0.85 (2.3)**

0.95 (5.3)***

Time 0.29 (5.4)**

0.15 (1.95)*

-0.22 (-3.8)***

GDP -0.007 (-0.27)

0.63 (1.92)*

0.54 (1.7)

Trade dummy -0.05 (-0.5)

0.3 (1.2)

0.06 (1.7)

R-square 0.83 0.91 0.99 Notes: † After correcting for first order autocorrelation. Figures in parenthesis are t-statistics. * Significant at 10 per cent level; ** significant at 5 per cent level; *** significant at 1 per cent level. The results of applying the OLS are satisfactory and consistent with economic theory in all three regions. The model fit is acceptable for the three regions as evident from the high R-squared values. As with regard to the substitution coefficient it turned out to be positive and statistically significant at least at the 5 per cent level. The positive sign indicates the absence of substitution effects between intra and ROW trade. The time trend variable coefficient was significant in all three regions (of varying degrees), indicating a positive trend in intra exports of the first two regions while it showed a decreasing trend in the SAPTA region over the period of study. The scale variable (GDP) was only significant (at 10 per cent level) for the COMESA region, which indicates toward a positive effect of economic size on intra exports in that region (This is also true for the SAARC region, although at a level of 11 per cent). Surprisingly, the trade dummy variable came out non-significant in all three regions indicating a negligible role of the trade agreements on intra exports of the three regions.


Table A4: Regression results of intra imports for GAFTA, COMESA, and SAPTA

Variable GAFTA COMESA SAPTA ROW imports 0.91

(6.2)** 1.53

(2.3)*** 1.9

(4.1)*** Time 0.16

(4.4)** 0.45

(1.86)* 0.26

(2.2)** GDP 0.01

(0.5) 0.43

(1.93)* -1.5

(-1.8)* Trade dummy 0.15

(1.8)* 0.8

(0.86) 0.14

(2.3)** R-square 0.92 0.95 0.97

Notes: Figures in parenthesis are t-statistics. * Significant at 10 per cent level; ** significant at 5 per cent level; *** significant at 1 per cent level. Table A4 shows the results for the intra imports variables. The substitution variable turned out to be positive and significant at the 1 per cent for COMESA and SAPTA and at the 5 per cent level in GAFTA, indicating the absence of substitution between intra and rest of the world imports. Similar to intra exports, the coefficients of the time variable were positive and significant for the three regions (at 5 and 10 per cent levels) indicating an increasing trend in intra imports during the period. Unexpectedly, the GDP coefficients were not significant at the 5 per cent level in any of the three regions. In fact, in the SAARC region, the sign of the coefficient of GDP was found to be negative, perhaps indicating less reliance of members on imports from each other on account of a larger and more diversified domestic production. The coefficient of the trade dummy was estimated to be positive and significant for the SAPTA region indicating a significant impact of the trade agreement on intra imports in this region. The difference in these results and those thrown up by the RCA analysis for SAARC, need to be resolved through a more rigorous analysis. However the coefficient for GAFTA region is also significant but only at the 10 per cent level.

V. Intra trade obstacles The three regions have somewhat similar intra trade problems related to political, economic, administrative, tariff and non-tariff, transportation and external factors. Political factors The most important factor is political mistrust and instability among selected countries of the region. The three regions suffer from political deadlocks or even unrest, either with neighboring countries or within the country itself. In addition, special-interest politics from industry and other lobby groups play a dominant role in determining the extent of trade relations or exchange of preferences. The stronger the political differences in the region the weaker the possibilities of successful trade relations. Also, mobility of labor and capital face significant degrees of friction in terms of residence permits and transfers.


Economic factors The economic factors that impede intra trade in the three regions include economic structure and policies. The countries of the respective region may or may not have similar economic structures, but evidently, limited differences in product-wise comparative advantage and absence of complementarities. The products that are similar are mainly primary including fuels and fuel products, agricultural products and raw materials, and to some extent manufactured products. Also, with the exception of some countries such as India, most of the countries are characterized by a small market and insignificant demand levels compared to the world market. Many of the countries, except those oil-producing ones, suffer from heavy foreign debt burden, poverty, low productivity, unemployment, inflation and non-competitiveness. The economic policies that impede intra trade include monetary restrictions, exchange rate controls and financial restrictions on across-the-border goods in the form of fees, stamps, etc. Administrative and technical constraints The administrative restrictions include all forms of procedures that are attached to the border crossing of goods such as customs documents and verifications, transit facilities, handling and inspection. The technical constraints include the applications of standards and specifications, health and environmental conditions, certificate of origin and value added verification. Transportation The transportation of goods depends heavily on land transport which is characterized by high costs since it relies on heavy vehicles. Conflict resolution and information The degree of trade conflict resolution and availability of information varies across the regions but it is in any case not well developed.

VI. Conclusion Intra trade in the three regions covered by this study was of relatively small importance in comparison to total trade of the regions. It did not exceed 10 per cent during the study period. Intra trade showed a modestly growing trend in the three regions over the period under study. Trade patterns within each region showed a great degree of similarity as indicated by the revealed comparative advantage analysis. Intra trade is dominated by primary products in GAFTA and COMESA while manufactured goods were more prominent in SAPTA. Fuel is more prominent than other products in GAFTA compared to the other two regions. Manufactured goods' share in intra trade increased over the period of the study in GAFTA and COMESA. In all regions intra trade was characterized by a high level of geographical concentration, mainly among neighboring countries.


The commodity and geographical concentration of intra trade in the regions was reflected in very similar RCAs. This suggests a very limited potential for expanding intra trade in the three regions. The results of the partial econometric analysis supported the above conclusion as the coefficient measures of the trade agreement effect turned out to be statistically insignificant in all three regions. Furthermore, the results showed absence of crowding-out effects between intra and total trade for all regions.


B. REGIONAL TRADE INTEGRATION: THE CASE OF GAFTA

By Talib Awad and Amir Bakir

I. Introduction During the last five decades most nations of the world moved toward trade liberalization under the umbrella of the General Agreement on Trade and Tariffs (GATT) and its later successor the World Trade Organization (WTO). WTO's membership has been ever expanding, and now covers more than 188 countries of the world. During the 1960s and 1970s there were a number of countries (mainly developing countries) who adopted an inward-looking import substitution strategy which were largely unsuccessful. By the 1980s and 1990s, most nations shifted toward a strategy of outward-looking export-led growth and dismantled all forms of trade barriers. These trade liberalizing trends were also accompanied by movements of most nations toward engaging in some form of regional economic integration. Regional integration agreements, although partial and discriminatory in nature, were seen as a first step towards more general and comprehensive trade liberalization, and hence were encouraged by GATT/WTO. An exemption from the principle of most favored nation (MFN) was granted to any group of countries that got involved in any form of regional integration agreement (RIA). Since the mid-1980s there has been a dramatic increase in regional integration activity. Of the 194 RIAs notified to GATT/WTO at the beginning of 1999, 87 were notifications since 1990. Now almost all countries are members of at least one RIA, and more than one third of world trade takes place within such agreements. New developments include the expansion and deepening of the EU; the construction of new and more open RIAs between developing countries; and the advent of RIAs in which both high-income and developing countries are equal partners. The latter development is lead by the North American Free Trade Area (NAFTA) which, in 1994, extended the Canadian-USA free trade agreement to Mexico. The following section provides a brief theoretical economic background of RIAs. The rest of the chapter is divided as follows: section two provides a theoretical background for regional economic integration, section three covers previous regional trade integration efforts among Arab countries as well as the most recent agreement namely, the Great Arab Free Trade Area (GAFTA) , section four is devoted to the analysis of intra Arab trade, section five attempts to evaluate the impact of GAFTA on intra Arab trade, section six highlights obstacles to intra Arab trade, and conclusions and recommendations are presented in the last section.


II. Overview: Theoretical background

II.1 Economic analysis of regional economic integration Regional integration agreements are groupings of countries formed with the objective of reducing barriers to trade between members. In theory, there are several types of regional economic integration schemes. The simplest is a free trade area (FTA) that eliminates tariffs on goods among the member countries, while leaving national tariffs against non-member countries unchanged. Beyond this there is a wide range of policy options open to countries considering integration, many of which turn on the `depth' of integration sought by member countries ranging from modest trade liberalization, through full economic integration, to the formation of shared institutions. In comparison to a free trade area, a customs union, in which a common external tariff is set, involves greater sharing of sovereignty and requires establishing procedures for revenue sharing, but in return can yield much greater market integration. In a free trade area where countries set different external tariffs the free internal circulation of goods is impossible; border formalities have to be maintained to ensure that external imports do not all enter through the member with the lowest external tariff, for re-export to other member countries. Since these imports include intermediate goods that are further processed in member countries, in practice this involves enforcing complicated `rules of origin' governing trade flows within the RIA. It is increasingly recognized that tariffs and quotas alone may be just a small part of the overall barriers to trade created by an international border. Rules of origin create restrictions, and so do measures such as anti-dumping rules, duplicative customs procedures, differing national product standards, and simple border red tape.

II.2 Economic effects on member countries When analyzing the economic effects of such agreements, economists usually distinguish between effects on member countries and on the world trading system. Effects on member countries include the benefits and costs of trade creation and trade diversion, as well as gains from increased scale and competition. `Deeper' integration can be pursued by going beyond abolition of import tariffs and quotas, to further measures to remove market segmentation and promote integration. Effects on the world trading system are not clear-cut. There is little evidence that regionalism has retarded multilateral liberalization, but neither is there support for the view that continuing expansion of regional agreements will obviate the need for multilateral liberalization efforts. Our analysis will focus on the first group of effects.

II.2.1 Trade creation and diversion The modern analysis of RIAs is mainly based on Viner's (1950) distinction between trade-creating and trade-diverting effects of RIAs. The classical source of gains from trade is that global free trade allows consumers and firms to purchase from the cheapest source of supply, hence ensuring that production is located according to comparative advantage. In contrast, trade barriers discriminate against foreign supply,


inducing domestic import competing producers to expand even though they have higher costs than imports. This in turn starves domestic export sectors of resources and causes them to be smaller than they otherwise would be. Since a RIA liberalizes trade, reducing at least some of the barriers, doesn't it follow that it too will generate gains from trade? Viner's contribution was to show that the answer is "not necessarily". The gains-from-trade argument applies if all trade barriers are reduced, but need not apply to a partial and discriminatory reduction in barriers, as in a RIA. This is because discrimination between sources of supply is not eliminated, it is just shifted. If partner country production displaces higher cost domestic production then there will be gains from trade creation. But it is possible that partner country production may displace lower cost imports from the rest of the world, and this is welfare reducing trade diversion. The analysis of trade creation and trade diversion constitutes one of the first formal analyses of the more general problem of 'second-best welfare economics'. Given that distortions remain in place in some activities in the economy, it is not necessarily the case that removing just some of the distortions (e.g. eliminating trade barriers on partner countries and leaving them in place on external countries) is welfare enhancing. In the literature on regional integration the response to the fundamental ambiguity created by the second-best took three main forms. First, authors established circumstances under which there is no interaction between formation of the RIA and external trade flows, so no possibility of trade diversion. Meade (1955) pointed out that if trade barriers with non-members take the form of fixed quantitative restrictions, then a RIA must raise the total welfare of member countries since there is no possibility that imports from the rest of the world are displaced. Ohyama (1972) and Kemp and Wan (1976) showed how, when external trade barriers take the form of tariffs, it is possible to adjust these to hold external trade volumes constant, so preventing trade diversion from occurring. Second, researchers identified conditions, in terms of changes in endogenous variables, for welfare gain. For example, welfare increases if the initial tariff-weighted change in trade volume is positive (Meade 1955). If internal tariffs are close to zero, then reducing them to zero raises welfare if it increases tariff revenues earned on external trade (Ethier and Horn 1984). The third approach is to identify features of economies (in terms of their underlying exogenous characteristics) under which they are more or less likely to gain or lose from RIA membership. Lipsey (1957) argued that joining with countries that are already one's largest trading partners is unlikely to lead to diversion, since the fact that the countries were originally the largest trading partners suggests that they are the lowest cost source of supply. Similar reasoning, including transport costs in the costs of supply, leads to the 'natural trading bloc' argument (Wonnacott and Lutz 1989, Summers 1991). Venables (2000) shows that those members of an RIA with comparative advantage most different from the world average are most likely to lose from trade diversion, as their trade is diverted to partner countries with comparative costs between theirs and the world average.


Empirical work on trade creation and trade diversion has taken two main forms; econometric studies of changes in trade flows, and simulation studies of the full general equilibrium effects of RIA membership. Econometric studies seek to quantify the changes in trade flows attributable to membership of a RIA, and thereby identify trade creation and diversion. A variety of different econometric models have been developed, the most common being based on the gravity model which estimates bilateral trade between countries as a function of their GDPs, populations, the distance between them, and physical factors such as sharing a land border, and being landlocked or an island. Dummy variables capture whether or not countries are in a particular RIA, their estimated effect indicating whether countries in a RIA trade more or less than would otherwise be expected. Using this technique, Bayoumi and Eichengreen (1997) found that the formation of the European Economic Community EEC reduced the annual growth of member trade with other industrial countries by 1.7 percentage points, with the major attenuation occurring over 1959-61, just as trade preferences were phased in. Soloaga and Winters (2001) looked at a wide range of RIAs, producing a mixed picture with little evidence of widespread trade diversion. Overall, there appears to be weak evidence that external trade is smaller than it otherwise might have been in at least some of the blocs that have been researched, but the picture is sufficiently mixed that it is not possible to conclude that trade diversion has been a major problem. Furthermore, it cannot be inferred that trade diversion has been economically damaging without information on relative costs and tariff structures, variables that are not revealed in this sort of aggregate exercise. The second empirical approach is based on computable equilibrium modeling. This involves construction of a full computer model of the economies under study which simulates the effects of the policy changes associated with the RIA. Such a model typically contains a great deal of microeconomic detail, so it can be used to predict changes in production in each sector, and changes in factor prices and real incomes. In models that assume a perfectly competitive environment, the combined effects of trade diversion and trade creation typically give very small welfare gains – just a fraction of 1 per cent of GDP (see Baldwin and Venables 1997 for a survey). The strength of these models is that they have sufficient microeconomic structure for the effects of a policy change to be traced out in detail, and its real income effects to be calculated. They are also often used for prediction to estimate the likely effects of a policy change before it is implemented. But they have the major weakness that they are not usually fitted to data as carefully, or subject to the same statistical testing, as econometric models. The cost of the microeconomic detail is a complexity that makes rigorous econometric estimation impossible. Although the focus of the trade creation and diversion literature has been on the changes in trade flows induced by regional integration, two consequent effects are important. The first is that changes in trade flows may change world prices, possibly improving the terms of trade of member countries, although this gain arises at the expense of outside countries. For example, if trade diversion occurs then RIA imports from outside countries are reduced, and any reduction in import prices that this causes is a terms of trade gain. Empirical work on this issue by Winters and Chang (2000) shows that Brazil's membership in Mercosur has been accompanied by a significant decline in the relative prices of imports from non-member countries.


The second is that changes in tariffs and trade volumes will lead to loss of government tariff revenue. This can occur directly (as intra-RIA tariffs are cut) and as a consequence of trade diversion (as imports are diverted away from external, tariff inclusive, sources of supply). Its cost depends on the social cost of raising funds by alternative means, and can be severe in some developing countries. For example Cambodia derived 56 per cent of its total tax revenues from customs duties prior to its entry into the Association of South East Asian Nations (ASEAN), and Fukase and Martin (1999) argue that entry into ASEAN provided a powerful stimulus for the introduction of a value added tax. Finally, it is generally accepted that RIAs have a better chance of being net trade-creating if they meet certain basic criteria, including the following; (i) relatively high initial tariff barriers among potential members; (ii) geographic contiguity; (iii) broadly similar stages of overall economic development; but (iv) varied national production structures. According to these criteria, the Arab countries of the Middle East and North Africa should be good candidates for the formation of an effective RIA. In practice, this has not yet happened, despite a plethora of multilateral and bilateral agreements designed to reduce trade barriers.

II.2.2 Scale and competition effects A second mechanism through which member countries are affected by RIA membership derives from the fact that countries may be too small to support, separately, activities that are subject to large economies of scale. Regional cooperation offers a route to overcome the disadvantages of smallness, by pooling resources or combining markets. These scale benefits can arise in public projects (see World Bank 2000) and also at the level of the private firm, which typically interact in imperfectly competitive market structures. These considerations are absent from the trade creation and trade diversion approach outlined above, which is based on the perfect competition and constant returns to scale paradigm of traditional trade theory. It was only in the 1970s and 1980s that formal analysis of the interaction between trade, economies of scale and imperfect competition began with the 'new trade theory', and these techniques have now been applied extensively to regional integration. The basic argument is that there is a trade-off between the extent to which firms can achieve economies of scale, and the intensity of competition in the market. For a given size market, larger firms means fewer firms and hence more monopolistic outcomes. If regional integration combines markets, then it shifts this trade-off, potentially allowing firms to be bigger and markets to be more competitive (Smith and Venables 1988). For example, there might be an initial situation in which two economies each have two firms in a particular industry, and these firms exploit their 'duopoly' power, setting prices well above marginal cost. After formation of the RIA this becomes four firms in one combined RIA market. This increases the intensity of competition, and possibly induces merger (or bankruptcy), perhaps leaving only the three most efficient firms. The net effect is increased competition, increased firm scale, and lower costs. Oligopoly competition is likely to be more intense than the original duopolies, and surviving firms are larger and more efficient, so they can better exploit economies of


scale. A further source of gains comes from possible reductions in internal inefficiencies that firms are induced to make. If the RIA increases the intensity of competition, it may induce firms to eliminate internal inefficiencies (X-inefficiency) and raise productivity. Since competition raises the probability of bankruptcy and hence layoffs, it also generates stronger incentives for workers to improve productivity, and increases labor turnover across firms within sectors.

II.3 Winners and losers Another concern accompany RIAs is about the distribution of its costs and benefits between member countries. Do central regions gain at the expense of peripheral ones, and do poor countries tend to catch up or get left behind? The evidence is, broadly speaking, that RIAs composed of developed countries tend to show convergence (for example the narrowing of per capita income differentials observed in the EU, see Ben- David 1993). However, the picture for RIAs composed of developing countries is more mixed, with some examples of divergent performance (World Bank 2000). The analytical literature on these questions is quite sparse, but provides several clues why this might occur. First, as mentioned above, trade diversion is more likely for countries with 'extreme' comparative advantage, suggesting that in a RIA amongst developing countries it might be the lowest income countries that experience diversion. For example, their imports of manufactures might be diverted from non-member countries to a partner that has a comparative advantage in manufactures within the RIA, but not relative to the world at large. Second, industries might tend to cluster in locations that have relatively good market access, or that are well supplied with business services or provision of other intermediate goods. This is more likely to occur in developing countries than in developed ones, partly because of their sparser provision of business infrastructure, and partly because the small size of their manufacturing sectors means that clustering is less likely to run into congestion and other sources of diminishing returns. The clustering may lead to wages being bid up in one member country at the expense of others.

III. Regional economic integration and the Great Arab Free Trade Area

III.1 Previous economic integration efforts

Soon after most Arab countries obtained their independence from British and French colonialism, efforts were initiated at the formal level to move toward both economic and political integration in the Arab region. As early as 1950, a Treaty for Joint Defense and Economic Cooperation was signed by Egypt, Jordan, Lebanon, Saudi Arabia, Syria and Yemen at the meeting of the Arab League's Economic Council. Its main economic objectives were tariff reduction and liberalization of capital and labor flows. It was followed up in 1953 by the Convention for Facilitating Trade and Regulating Transit, which aimed at eliminating barriers to trade in agricultural goods and minerals. Efforts to lower tariffs on manufactures were largely thwarted by Iraq, Saudi Arabia and Yemen, which relied heavily for revenue on import duties (Awad 1995).


The next move towards integration, involving efforts to create an Arab Common Market, began in the late 1950s, when Egypt, Jordan, Morocco, Syria and Kuwait agreed in principle to unify economic policies and legislation. The five states ratified this proposal in 1964, but the intention of effectively abolishing duties and quantitative restrictions over a 10-year period was undermined by four rounds of negotiations on exemptions. Attempts to establish a common external tariff were finally abandoned in 1971. Another attempt to promote regional integration was the 1981 Arab League-sponsored Agreement for the Facilitation and Promotion of intra-Arab Trade. The agreement represented a declaration of intent to negotiate full exemption from tariffs and non-tariff restrictions for manufactured and semi-manufactured goods. As with previous agreements, the 1981 effort had little effect on formal trade liberalization or actual trade. It lacked binding commitment to its terms and a timetable for implementation, and featured a "positive list" approach, whereby specific products for liberalization must be listed (as opposed to the "negative list" approach, whereby liberalization covers all items other than those specifically listed for continuing protection). Meanwhile, subregional arrangements, such as the 1981 Gulf Cooperation Council - arguably the only effective Arab trade agreement to date, with its successful promotion of trade liberalization and free movement of capital and labor among member states - and the abortive 1989 Arab-Maghreb Union, along with a proliferation of strictly bilateral arrangements between trade partners have become a feature of the Arab scene. Bilateral treaties are now estimated to number more than 45. They typically comprise limited preferential arrangements, mostly confined to varying degrees of tariff exemptions for specific categories of agricultural goods and raw materials; their partial nature has had the net effect of hindering rather than stimulating wider inter-Arab trade (Hoekman 1995). While noble political aspirations helped to initiate successive integration efforts, less noble political frictions equally often helped subsequently to vitiate them. Lack of a forceful time table for practical implementation of such agreements, together with lack of coordination with the private sector, poor infrastructures, partiality in the nature of the agreements, differences in economic and political systems, spread of corruptions and red tape, and regional political instability were among the main factors behind the poor achievements of these early formal attempts.

III.2 The Greater Arab Free Trade Area In February 1997, the Arab League launched a free trade programme, known as the Greater Arab Free Trade Area, or GAFTA, in which member states were asked to come up with specific commitments regarding elimination of tariffs, non-tariff measures and rules of origin. The so called Executive Program was an effort to revive the 1981 Agreement for Facilitation and Promotion of Trade among the members. All the 22 member states of the Arab League, except Algeria, Djibouti, the Comoros Islands and Mauritania have endorsed the Agreement and have committed to the Executive Program.


The programme calls for tariff reductions over a ten year period at a rate of 10 per cent per year, meaning that tariffs would be reduced to zero by 2007. In addition, the members agreed to bind their national tariff schedules as of December 31, 1997. By September 1999, 14 countries had applied tariff reduction schemes and fulfilled the tariff commitments. For the rest of the members who did not ratify the agreement, the bound tariffs will be applied at the time they notify the Arab League of their endorsement of the programme. Regarding liberalization of industrial products, members are allowed to draw up a list of exceptions from tariff reductions during the first years of the programme. The exceptions are intended to enable local industry to restructure and improve its competitiveness before having to face competition from other GAFTA countries' imports. In the area of agriculture, the programme offers members the opportunity to suspend tariff reductions on some produce during the peak harvest seasons. Each GAFTA country is allowed to submit ten produce items for suspension, with a total exemption for all the items of 45 months. So far, 10 GAFTA countries have submitted a list of 30 fruits and vegetables. Non-tariff barriers Although the programme calls for a schedule to reduce non-tariff barriers, at this point the GAFTA countries have not tackled these barriers. A committee on non-tariff barriers has agreed on a list of goods whose imports are prohibited for religious, health, environment and national security reasons, and the list is to be reviewed on a yearly basis. The committee is also supposed to sort out other goods submitted by members and start negotiations for their elimination. The programme calls for the application of international rules regarding subsidies, countervailing measures, safeguards and anti-dumping measures. This should be possible under WTO agreements. However the programme does not explicitly refer to the WTO agreements since some Arab countries are still seeking membership. GAFTA rules of origin The programme offers rules of origin for duty-free treatment. The GAFTA value added requirement is set at 40 per cent, and there are two methods for calculating origin. The first is based on the local value added approach. The other is the net cost approach, which subtracts specified imported expenses from the transaction price to determine the base for calculating the ratio of foreign to domestic content. An important feature of the programme is the ongoing scheme for the elaboration of detailed preferential rules of origin for GAFTA-made products. This scheme adopted rules for full commutation of origin among the GAFTA countries. This means that materials obtained from, for example, Jordan, and incorporated into a product made in Egypt, may be considered as if they were obtained in Egypt. Finally, the programme also calls for the need for harmonization of preferential rules of origin to comply with the Euro-Mediterranean free trade agreements underway.


Monitoring GAFTA implementation GAFTA is managed by the Council of Ministers of Member Countries and by a permanent executive body. So far, GAFTA has a functioning Secretariat that comes under the Economics Department of the Arab League Secretariat. The programme also calls for the chambers of industry and commerce in Arab countries to monitor implementation. The involvement of the private sector is expected to enhance the transparency of the members' trade practices. GAFTA achievements prospective The GAFTA programme calls for across-the-board tariff reduction offering the advantage of transparency and ensuring that high tariffs are reduced faster than lower tariffs. However, the extent to which this approach will boost intra-regional trade flows depends on the magnitude of tariff dispersion as well as the effective rate of protection across industries in various countries. Tariff structures among the members are uneven. Low-tariff countries, namely the Gulf countries (Bahrain, Kuwait, Qatar, Oman, Saudi Arabia and the United Arab Emirates) have simple average tariffs of four to 12 per cent, while countries such Egypt, Jordan, Morocco, Syria and Tunisia have tariffs of 40 to 100 per cent. Tariffs reach 235 per cent in Syria. In addition, tariff escalation exists in many countries, particularly in textiles, clothing, leather and basic metal industries where average tariffs on finished products are multiples of the lower tariff levels on raw materials. An across-the-board tariff reduction could have a significant effect on trade creation. However, the most serious shortcomings of the programme are loopholes allowing the exclusion of certain industrial agricultural products from tariff reduction. The list of exceptions already submitted includes consumer goods competing with domestic production (e.g. textiles, clothing and leather articles). Moreover, the transition period for the excluded industrial products would allow for pressures from interest groups to resist market opening at all. The exclusion of certain agricultural products during the crop/harvest seasons for the full transitory period of ten years would also substantially limit the trade creation effect. The use of quotas has been declining in GAFTA countries partly as a result of unilateral trade reforms implemented by countries such as Morocco, Tunisia, Egypt and Jordan. In these countries, non-tariff barriers take the form of import licensing for safety and health standards. In the Gulf Cooperation Council (GCC) countries, non-tariff barriers are relatively low. However, Syria still imposes an import license requirement on virtually all imports. Customs and administrative procedures associated with importing are still important trade barriers within the region. The programme calls for some harmonization related to customs clearance procedures, but it does not address harmonization of other major regulations regarding product standards, testing and certification procedures, and environmental standards. Transportation is also a problem. The members signed an earlier agreement to simplify the transit of goods and persons, but there has been no legal enforcement for the implementation of this agreement, and there are periodic reports citing barriers to cross-border trade such as closure of roads, delays and cumbersome cross-border regulations such as no driving on weekends or the refusal of visas professional drivers of certain nationalities.


At first, the commutating of rules of origin for products made in GAFTA countries may help create backward and forward linkages between the member countries and increase the potential for trade. However the further extension of commutation of the Euro-Mediterranean Agreements and the GAFTA countries would be even more beneficial to Member Countries and would also help reduce the emerging hub-and-spoke nature of the Euro-Mediterranean Agreements. Deepening integration of GAFTA So far, GAFTA has been limited to regional liberalization of merchandise trade. This effort at regional integration alone is not sufficient to ensure a more credible path for further domestic market efficiency allowing GAFTA member countries to face greater competitiveness from world exporters. There is a need to extend regional liberalization to the areas of trade in services and labor. Linking GAFTA with the Euro-Mediterranean trade areas through incorporation of binding rules and various policy harmonization that are included in the Euro-Mediterranean Agreements would enhance market efficiency, allowing GAFTA countries to face greater competition from emerging world exporters in Asia and Eastern Europe. The absence of provisions for free movement of services and the right of establishment in GAFTA reduces the benefits of the agreement, given that services trade is generally complementary to merchandise trade. Barriers to trade in services in GAFTA countries are clearly important. A regional agreement which allows free entry into services activities should help domestic services companies to expand to regional markets before venturing into the more competitive world markets. Liberalization of transportation services should lie at the heart of intra-regional efforts to facilitate merchandise trade. Maritime transport remains by far the principal means of intra-regional transport of goods and a regional agreement on common shipping policy principles is needed. Land transportation services are an important cross-border mode of passenger and freight transport in the GAFTA countries. However, the regulatory regime for road transport is highly restrictive in the GAFTA countries. For instance, cross border freight transport returning with cargo is not allowed by most of the GAFTA countries. Internal coasting trade is also not allowed in virtually all the GAFTA countries. A regional agreement regarding land transport is needed. Financial services There is also the need to harmonization some of the banking and insurance regulations as well as a provide for the right of establishment with countries in the region. There is relatively little integration among the financial institutions in the GAFTA countries, although it is more significant in banking activities. Egypt, being the most advanced in liberalizing the financial sector among the reforming MENA (Middle East and North Africa) countries, has discretionary procedures for allowing new entry in this sector. Since domestic financial institutions in most of the GAFTA countries have comparable high operating costs and comparable technology, the real efficiency gains would come from integrating domestic financial markets into the world financial markets in order to facilitate capital and technological flows into the region. For many GAFTA countries, labor services are the mainstay of their service exports. From the viewpoint of efficiency, free movement of skilled labor enhances the


efficiency of production. Cooperation among GAFTA countries to promote professional/technical labor mobility is desirable. Some important measures for freer mobility of professional and skilled labor in the context of GAFTA include recognition of professional certification standards, visa and work permit practices and measures allowing the temporary movement of corporate personnel, engineers, architects, accountants, lawyers and computer programmers. Political and economic considerations dictate that integration at the bilateral, regional, inter-regional and multilateral levels needs to proceed in a systematic way on parallel tracks. Hopefully, building closer commercial ties in the region can play a role in supporting political initiatives to reactivate the peace process. Economically, the emergence of an integrated market in the region may yield advantages that substantially exceed the current dimension of intra-regional trade. This paper's final objective is to provide an objective evaluation of the potential success of GAFTA in expanding intra-Arab trade (IAT). But first it is necessary to review the trend and main characteristics of Arab countries' trade.

IV. Trend and characteristics of intra-Arab trade

IV.1 Growth of intra-Arab trade Trade in the region has witnessed reasonable growth during the period 1980-2005; it grew at an annual rate of 2 per cent. The growth rates of total exports and imports reached 1.8 per cent and 2.3 per cent, respectively. Similarly but more outstandingly, IAT grew by 5.9 per cent. However, the percentage of IAT in total trade did not exceed 10 per cent. The percentage of intra exports in total exports has been rising over the period while the percentage of intra imports in total imports has been declining since 1990 to reach 8 per cent in 2005, as shown in Figure B1. Figure B1: Intra Arab Trade, 1980-2005

0%

2%

4%

6%

8%

10%

12%

14%

1980 1985 1990 1995 2000 2005

Exports

Imports

Source: Calculated based on data taken from the Arab Monetary Fund website (www.amf.org.ae). It is interesting to note that intra trade growth rates are higher during the period 1998-2004 compared to the period 1980-1997. This might suggest that the GAFTA


agreement had a significant role in enhancing intra trade, as shown in Table B1 below. Table B1: Growth rates of intra trade during the period 1980-2004

1980-1997 1998-2004 1980-2004 Intra exports 3.5 14.7 5.7 Intra imports 3.5 12.6 6.0 Intra trade 3.5 13.7 5.9

Source: Calculated using exponential growth formula based on data taken from the Arab Monetary Fund website (www.amf.org.ae). Indeed, as shown in Table B2 individual countries trade growth rates vary significantly among themselves. For example, Syrian and Egyptian intra exports grew by 10.2 per cent and 9.7 per cent respectively, while the corresponding growth rates for Morocco and Jordan did not exceed 4 per cent. Similarly, Egyptian intra imports grew by 9.6 per cent while that of Saudi Arabia was less than 2 per cent, as shown in Table B2. Table B2: Intra-Arab trade growth rates for selected countries (per cent)

Country Intra exports Intra imports Jordan 4 5.1 Syria 10.2 2.6 Egypt 9.7 9.6 Morocco 2.9 6.2 Saudi Arabia 7.2 1.3 United Arab Emirates 8.7 8.1

Source: Calculated by the researchers using the exponential growth formula for the period 1980-2004. Data are taken from the Arab Monetary Fund, Annual Statistical Book, several issues.

IV.2 Intra-Arab trade commodity structure The analysis of commodity groups uses the SITC classification and for this purpose the following commodity groups were chosen: Food and Beverages, Raw Materials and Fuel, Chemicals, Machines and Transport Equipment, and Manufactured Goods (Table B3). Fuel and raw materials share in intra trade exceeds 50 per cent, followed by food and beverages with a share of 17 per cent, then chemicals with a share of 14 per cent-15 per cent, followed by manufactured goods with a share of 8 per cent-9 per cent, and finally machines and transport equipment with a share of 5 per cent-6 per cent. Over the period under investigation in this study, the development of the relative importance of the above mentioned commodity groups, as shown in Table B3, reveals that food and beverages and manufactured goods have realized a dramatic gain in relative importance on the expense of raw materials and fuel.


Table B3: Relative importance of intra-Arab exports according to commodity groups

Intra exports Group/year 1980 1998 2002 2005 Food and Beverages 0.9 13.0 18.2 17.2 Raw Materials and Fuel 96.0 55.0 52.2 57.7 Chemicals 1.0 16.0 16.2 14.1 Machines and Transport Equipment 0.7 5.0 5.5 5.0 Manufactured Goods 1.4 11.0 7.9 6.0

Source: Arab Monetary Fund, Annual Statistical Book, several issues.

IV.3 Intra-Arab trade destination Intra-Arab trade is characterized by a high level of geographical concentration, and in most cases, it is restricted mainly among two partners or three. For example, in 2005, Oman exported 64 per cent of intra exports to the United Arab Emirates, Libya exported 69 per cent of its intra exports to Tunisia, 57 per cent of Bahrain’s intra exports were destined to Saudi Arabia, and Iraq’s intra exports with Morocco and Jordan amounted to 47 per cent and 21 per cent, respectively. In contrast, the case for intra imports is very similar. For instance, Oman imported 91 per cent of its intra imports from the United Arab Emirates, and 70 per cent of Jordan’s intra imports came from Saudi Arabia, 56 per cent of Iraq’s intra imports came from Jordan, about one half of Libya’s intra imports came from Tunisia, and intra imports of Morocco, Bahrain and the United Arab Emirates from Saudi Arabia amounted to 58 per cent, 54 per cent and 52 per cent, respectively (Arab Monetary Fund 2005).

V. Measuring the effect of GAFTA on intra-Arab trade Intra-Arab trade over the past twenty five years grew steadily and substantially, while Arab trade with the rest of the world (total trade less intra trade) declined in the 1980s but gained momentum in the 1990s and continued to grow steadily thereon. Figures B.2 and B.3 show the development of exports and imports, respectively. The growth in trade is due to several factors, but mainly trade agreements among Arab countries themselves and with the rest of the world, as well as the economic growth in the world are the main factors behind those trade developments.


Figure B2: Intra exports and exports with the rest of the world (US$ millions)

0

40

80

120

160

200

240

280

320

360

400

440

480

520

560

600

1980 1985 1990 1995 2000 2005

Intra

Rest of World

Source: Arab Monetary Fund, Annual Statistical Book, several issues. Although it is possible to consider some degree of substitution between intra trade and trade with the rest of world but this element is suspected to be relatively weak. Also, it is possible to consider some effects resulting from elements of past experience and previous trade trends on what is happening in the present and what is going to happen in the future. To evaluate the impact of regional trade agreements two analytical tools are employed: the revealed comparative advantage (RCA) analysis and a multiple regression model. The RCA index will help identify the possibilities of intra trade expansion, a higher than one value indicating a revealed comparative advantage in that particular sector. However if RCAs are similar among countries of the particular region then there will be limited potential for expanding intra trade within that region.


Figure B3: Intra imports and imports with ROW ($US million)

0

50

100

150

200

250

300

350

1980 1985 1990 1995 2000 2005

Rest of World

Intra

Source: Arab Monetary Fund, Annual Statistical Book, several issues.

V.1 Revealed comparative advantage Revealed comparative advantage (RCA) indices for a group of countries help identify as to whether there exists high or low potential of intra-group trade in a particular product category. The RCA for the jth country for the ith product can be defined as follows: RCAij = (Xij / Xiw) / (∑Xij / ∑Xiw), Where, commodity i = {1, 2, 3,…, n.}, Xij = exports of the i-th product from country j, ∑Xij = total exports from country j, Xiw = exports of the i-th product from the world, ∑Xiw = total exports from the world. Clearly, if the RCAs for a commodity for all the countries are close to each other or 1, there is less scope for intra-regional trade in that commodity for the group of countries. Correspondingly, if a country’s RCA is greater than 1 for a commodity, while for the rest of countries the indices are less than 1, it is indicative of a high scope for intra-group trade in that commodity. Table B4 shows the RCA's for GAFTA countries. It reveals that for food products six countries had value of RCA greater than one, with Morocco being the highest (2.8), followed by Jordan, Lebanon, Syria, Sudan and Egypt. In agricultural products only three countries showed values greater than one; i.e. Egypt (3.8), Sudan (3.3) and Syria (1.7). Fuel products showed the strongest RCA for most Arab countries ranging between 6-8.9, the relevant countries are Algeria (8.9), Kuwait, Yemen, Oman, Qatar, Saudi Arabia, Sudan, Syria, Bahrain and the United Arab Emirates (6.1). Five countries showed higher than one values of RCA in ores and metal products including Bahrain (6.2), Jordan (3.8), Lebanon (2.7), Morocco (2.6) and Egypt (1.1). For manufactured products none of GAFTA countries showed a revealed comparative advantage except Tunisia (1.1). Similarly only Jordan had an RCA greater than one


(1.8) for chemicals. For machinery and transport products no country in the region showed a relative comparative advantage. Based on the above RCA analysis it can be concluded that the GAFTA countries showed a high concentration in revealed comparative advantage in primary goods, mainly fuel and ores and metals, with almost negligible advantages in the other products. This implies a very limited potential for expanding intra trade in the region. Table B4: RCA indices for GAFTA countries by broad commodity groups (2003, 2004)

Country Total Food Agriculture Fuel

Ores &

Metal Manufactured

Goods Chemical

Machinery &

Transport Algeria 0.0 0.0 8.9 0.1 0.0 0.1 0.0 Bahrain 0.1 0.1 7.1 6.2 0.2 0.3 0.1 Egypt 1.2 3.8 4.3 1.1 0.4 0.7 0.0 Jordan 2.1 0.2 0.1 3.8 1.0 1.8 0.3 Lebanon 2.0 0.8 0.0 2.7 0.7 0.8 0.3 Morocco 2.8 1.0 0.3 2.6 0.9 1.0 0.4 Oman 0.7 0.0 8.0 0.3 0.2 0.1 0.2 Qatar 0.0 0.0 7.9 0.0 0.2 0.9 0.0 Saudi Arabia 0.1 0.1 8.0 0.1 0.2 0.9 0.0 Sudan 1.4 3.3 7.9 0.1 0.0 0.0 0.0 Syria 1.9 1.7 7.1 0.3 0.1 0.1 0.0 Tunisia 1.0 0.5 0.9 0.6 1.1 0.9 0.4 UAE 0.5 0.1 6.1 1.0 0.4 0.3 0.0 Yemen 0.7 0.2 8.4 0.1 0.0 0.0 0.0 Kuwait 0.0 0.1 8.5 0.2 0.1 0.4 0.0

Source: Calculated based on data from UNCTAD Handbook of Statistics 2005. Calculations are based on either 2003 or 2004 depending upon data availability.

V.2 Partial regression analysis As GAFTA became effective starting 1 January 1998 and data on intra trade at the time of writing is only available up to 2004, it is not possible to perform a subset regression analysis for the two periods: before and after the agreement. Hence, a multiple regression model is fitted for the whole sample. The effect of GAFTA on intra trade is simply introduced using a “Dummy” variable that takes the value of zero in the absence of the effect of the agreement, and one when the agreement was effective (1998-2004). Two models were fitted: the first represents intra exports and the latter represents intra imports. For the purpose of identifying the substitution effect between intra trade and trade with the rest of the world (ROW), exports with ROW was introduced as a an explanatory variable in the first model, and imports from ROW was introduced as an explanatory variable in the second model. These two variables can serve the purpose of capturing the effects of trade agreements other than intra trade agreements. GDP was introduced as an extra explanatory variable to capture the economic size effect of the region. In addition, a time trend was introduced with the purpose of capturing the growth trend in intra trade that may be attributed to other than GAFTA factors. Hence the two models to be estimated take the following form:


143210 uGaftadBTimeBGDPBNetxBBIntrax +++++= (1)

243210 uGaftadCTimeCGDPCNetmCCIntram +++++= (2)

Where: Intrax = intra exports, Netx = exports with the rest of the world, GDP = Gross domestic product, Gaftad = Dummy variable takes the value one for the period 1998-2004, and zero otherwise, Time is a simple time trend, Intram is intra imports, Netm refers to imports with the rest of the world, and u1 and u2 are white disturbance terms. BBi and Ci are unknown parameters to be estimated. For the purpose of data stationarity and to obtain the long run elasticity directly the log-linear form is chosen. The results of applying ordinary least squares (OLS) on the available sample (1980-2004) are shown in Table B5. Table B5: Estimation results for intra exports and intra imports in logarithm form

Parameters Estimate t-statistics constant 3.5 1.25 lnetx 0.44 4.3** lgdp -0.007 -.27 ltime 0.29 5.35** gaftad -0.04 -0.48 R-square 0.84 constant -1.77 -1.02 lnetm 0.91 6.2** lgdp 0.01 0.5 ltime 0.16 4.4** gaftad 0.15 1.82*** R-square 0.92

Notes: * Significant at 10 per cent level; ** significant at 5 per cent level; *** significant at 1 per cent level. The import equation was estimated using Coherene–Orcutt procedures to correct for serial correlation. The high values of R-square and Fisher-statistic (not shown here) imply that both estimated models can be considered of high significance, i.e. explaining the variation in the data to a large extent. More than 80 per cent of the variations in intra exports, and more than 90 per cent of the variations of intra imports were explained by the independent variables. The variables representing trade with ROW (lnetx and lnetm) took a positive sign and were significant at the 1 per cent level. This implies that there is no evidence of any significant substitution effects between intra trade and trade with ROW. It can be concluded that the two types of trade are indeed complements. The coefficients of the time variables were positive and significant at the 5 per cent level for both equations, indicating some positive growth of intra trade that cannot be clearly attributed to GAFTA. The GDP variable failed to affect significantly either intra exports or intra imports. This can be an indication of the limited effect of the region's economic size on intra trade, and can be explained by lack of production diversification and limited production capacities in these countries. In addition, most traded goods between Arab countries are mostly essential and primary commodities that have low income elasticities.


Furthermore, the coefficient of the GAFTA dummy is statistically insignificant at the 5 per cent level in both equations, indicating to limited success of the agreement in affecting intra-regional trade. In summary, the estimation results lend support to the analysis of revealed competitive advantage above, indicating a lack of effectiveness of the GAFTA agreement during the short period of its application.

VI. Intra-Arab trade problems and obstacles The relatively insignificant contribution of IAT to total Arab trade can be attributed to several factors and can be identified within six main groups; namely, political, economic, structural, geographical, administrative and technical factors (see Alafouri et al. 2004). It is worth mentioning here that since oil figures high in total Arab trade, intra trade registered low significance as a per cent in total Arab trade. Hence, if oil trade is excluded, intra trade's significance increases from about 10 per cent to about 20 per cent. It is imperative that in the absence of trade problems and obstacles intra trade could be improved to a great extent. Having said that, some problems are akin to the Arab region in particular and perhaps this caused the low share of intra trade in total trade. First: Political factors The Arab countries got engaged on an individual basis, with different international alliances and established strong ties with them which reflected negatively on intra-Arab relationships, thus, leading to weaknesses in the economic and trade relations among Arab countries. The effects of those factors can be summarized as follows:

a) The strong political ties in addition to the trade and economic agreements of Arab countries with other countries of the world, especially cooperation agreements, loans and foreign aid, engineered stronger trade and economic relations in favor of the rest of the world rather than among the Arab countries themselves.

b) The variety of economic systems in the Arab countries in terms of the degree of involvements of the public and the private sectors in economic activity poses unfruitful and unproductive difficulties of trade transactions and cooperation across borders. In addition, most economic and trade decisions of Arab countries are directed heavily by their governments, with very limited degree of coordination with the private sector.

c) The Arab countries are characterized by being small economies where the private sector is largely dominated by small interest groups, and since the openness to trade may result in conflicting interests, this has caused large setbacks and retardation in the development of trade.

d) The Arab countries face major constraints and difficulties regarding the movement of people and capital. For example, it is not easy to obtain residence permits or undertake capital transfers.


Second: Economic factors The economic factors that affect the growth of intra-Arab trade are related to monetary and financial constraints, the countries' economic structures and the respective economic policies, which can be clarified as follows:

a) The monetary constraints include currency transfer restrictions, foreign currency conditions, multiple exchange rates, import licenses and insurance certificates.

b) The financial restrictions consist of additional and unnecessary fees with the purpose of collecting revenues regardless of neither the nature of fee nor its purpose. For example, stamp (authentic) fees on certificate of origin charged by the respective consulates (embassies) as a percentage of the value of the certificate rather than per certificate. Also, other stamp fees charged on other merchandise-related documents such as bills.

c) The economic structures of the Arab countries are very similar. On the one hand, the economic structures of some Arab countries consist mainly of oil production and its derivatives, while on the other; the economic structures of other Arab countries depend on the agricultural sector and raw materials. This similarity in economic structures impedes intra trade to a great extent. This issue is exacerbated when we consider further that since the economy is small in the respective countries and hence, income is relatively low, the purchasing power is weak to overcome this similarity problem and extend demand across country borders to enhance trade efforts.

d) Several Arab countries depend on oil production to a large extent and since oil prices fluctuate considerably, demand fluctuates accordingly. Hence, when oil prices are high trade increases and vice versa when oil prices are low.

e) Several Arab countries adopt conservative economic policies and are under the pressure of economic reform, tight spending and debt control which result in conservative actions towards openness and trade extension.

f) The spread of globalization issues worldwide and the joining of WTO opened the Arab markets to severe competition from Asian products which hindered the domestic efforts to enhance intra trade and exacerbated sales problems even for domestic products in their respective domestic markets.

g) The products of the Arabic markets lag behind world market products in terms of competition, quality and production costs. For example, Asian products are cheaper and western products are better in quality and have cost advantage especially in terms of transportation costs.

h) The consumer demand and its growth rate in the Arab countries are relatively small compared to the rest of the world which makes selling in the world market more attractive due to its higher potential.

i) Intra-Arab trade is dominated by consumer goods and raw materials while world trade consists largely of durable and capital goods. As the value of the former and its growth are considerably lower than the latter this creates a widening gap between them.

Third: Tariffs, fees and taxes Tariffs were considered the main obstacle to intra trade development during the past two decades. However, as a result of the signing by all Arab countries of the GAFTA agreement which became effective, it is expected that all the negative effects of tariffs on intra trade will be eliminated substantially.


Nevertheless, the existence of taxes and fees with similar effects as tariffs that are imposed on intra trade commodities among the Arab countries still deter intra trade to a great extent. For example, all imported commodities are subject to stamp fees, consulate fees, statistics fees, custom services fees, transit fees and other additional taxes. Usually those fees and taxes are imposed as a percentage on the value of the merchandise rather than as per document/license which increases costs unnecessarily and lessens the effectiveness of tariff reductions. Although all Arab countries signed and agreed on GAFTA but some individual countries, especially those that were granted exemptions, still chose to impose tariffs on intra trade commodities. Fourth: Administrative and technical restrictions

a) The administrative restrictions consist of procedures related to the release of merchandise from customs at the border point, which includes document presentations, verifications, transit services, waiting, handling, inspection, valuation, etc. Those procedures suffer from highly unorganized impotent bureaucratic procedures and administrative inefficiencies. In addition, some Arab countries still adopt quantitative restrictions on intra trade such as limiting trade within specific institutions in the public sector or restricting import licenses within a specific group of individuals and companies, as well as in cases were bilateral agreements dictate certain trade restrictions.

b) The technical restrictions are related to standards and specifications, health and environmental conditions, certificate of origin and value added. The issue here is not whether to have or not to have standards and specifications and health and environmental conditions; on the contrary, the existence of such controls is necessary. The problem under consideration involves the variety and multiplicity of tests and other requirements that keep changing from time to time between countries, thus incurring higher costs and requiring extensive time. As far as the certificate of origin is concerned, the products entering trade are subject to a 40 per cent value added requirement of Arabic origin in order to get a preferential treatment under the GAFTA. Although some progress has been made in this respect between Arab countries this item is considered one of the main obstacles to the development of intra-Arab trade.

Fifth: Transportation The transportation sector in Arab countries is characterized by excessive costs since it depends heavily on land transport across the borders in the case of adjacent countries, mainly through the use of heavy vehicles. In the case of relatively long-distance transport (between eastern and western Arab countries), marine transportation is used. In contrast world trade depends to a large extent on railways. Further more, Arab-Europe marine transportation is much more efficient than Arab-Arab transportation. In general, the transportation sector in the Arab countries cannot be considered developed to the extent that it can facilitate trade transactions. On the contrary it poses a major obstacle. Perhaps the geography of the region implies trade concentration among two to three trade partners in most cases.


Sixth: External factors: E.g. conflict resolution and information Trade transactions between individuals, companies and institutions involve a commitments and responsibilities that need to be enacted and respected to the utmost intent. However, in some cases, disagreements occur on any item under implementation in any trade transaction which implies the need for a well identified mechanism that is agreed upon by all partners in order to be able to resolve any disagreement or conflict that might arise in this context. It is worth mentioning here that this item applies to trade in general and not to intra trade in particular. In this respect a conflict resolution mechanism is being implemented by the Arab countries but it is not identified as a very efficient one and is not fully accepted by all partners. In addition, the trade-related information services lack efficiency and suffer weaknesses in the provision of the necessary information. For example, a lot of information could be provided on markets, commodities, consumption, standards and specifications, quality, laboratory test requirements, fees, and other trade services such as storage facilities, transportation, transit, banks, insurance, etc.

VII. Conclusion and recommendations Most economic integration efforts in the region were initiated at the political level, without parallel support from the private sector leading to a limited success. Throughout the period of the study, regional trade grew at a modest rate but it did not exceed 10 per cent of total trade. However, intra trade grew more rapidly after the signing of GAFTA compared to the period before. In addition, most intra-Arab trade was concentrated in fuel and raw materials, and to a lesser extent in food and manufactured goods. Over the period under investigation, food and manufactured goods had realized a substantial gain in relative importance at the expense of raw materials and fuel. Also, intra-Arab trade was characterized by a high level of geographical concentration, and in most cases it was concentrated mainly among two or three neighboring countries. The impact of GAFTA on intra-Arab trade was evaluated using two analytical tools: The revealed comparative advantage analysis and multiple regression technique. The RCA analysis showed similar competitive advantages in Arab countries, concentrated in raw materials and primary products including fuel, which indicates a limited potential for expanding intra-Arab trade. In addition, the regression estimation results supported the RCA analysis indicating lack of effectiveness of the GAFTA agreement during its first period of application. Expansion of trade within the region still faces many obstacles including poor infrastructure, political instability in the region, spread of corruption and red tapes in some countries, inefficient transportation network and bureaucratic custom procedures at the border. It should be noted that the study findings are contingent upon the available limited data. Deeper and richer analysis including using more advanced econometric technique can be used in the future as larger sample become available.


Finally the recommendations of the study are as follows: 1) Trade integration among Arab countries taking the form of an FTA is not enough;

it needs to be deepened to higher levels of integration, and has to be accompanied by parallel efforts on production diversification and capacity building.

2) If trade integration is to be successful, it has to deal with other non tariff barriers, including rules of origin, duplicative customs procedures, differing national product standards, bureaucracy and corruptions at border, and poor infrastructures particularly transportation.

3) Policy makers should coordinate more with the private sector in terms of designing trade policies and its implementation. The private sector should be encouraged to play a more active economic role in Arab economies, and to participate more effectively in the decision making process.

4) Finally to achieve the needed political and social stability needed for any successful economic integration, economic reform and democratization must be pursued by all countries of the regions.


References Alafoury A., M. Murad and M. Abdelelrasheed Ali (2004). "The Greater Arab Free

Trade Area (GAFTA)", Published in the proceedings of the conference "Intra Arab Trade and Economic Integration", held at Amman, Jordan, 20-22 September 2004.

Arab Monetary Fund (2005). The Annual Arab Economic Report, 2005, P. 148.

Awad T. (1995). 'International trade: theory and applications'. Institute of Monetary Studies, Amman, Jordan.

Baldwin, R.E. and A.J. Venables (1997). `International Economic Integration,' in G. Grossman and K. Rogoff (eds.) Handbook of International Economics, vol. 3.Amsterdam: North Holland.

Bayoumi, T. and B. Eichengreen (1997). "Is Regionalism Simply a Diversion: Evidence from the Evolution of the EC and EFTA", in Ito, T. and Krueger, A. (eds) Regionalism versus Multilateral Trade Arrangements, Chicago, The University of Chicago Press.

Ben-David, D. (1993). "Equalizing Exchange: Trade Liberalization and Income Convergence", The Quarterly Journal of Economics, 108, 653-679.

Ethier, W. and H. Horn (1984). "A new look at economic integration", in H. Kierszkowski (ed) Monopolistic competition and international trade, Oxford.

Fukase, E. and W. Martin (1999). "Evaluating the Implications of Cambodia's Accession to the ASEAN Free Trade Area: A General Equilibrium Model (CGE) Approach", processed, The World Bank. Washington, D.C.

Hoekman, Bernard (1995). "The World Trade Organization, the European Union and the Arab World." World Bank Policy Research Paper No. 1513, World Bank, Washington, D.C.

Kemp, M. and H.Y. Wan (1976). "An elementary proposition concerning the formation of customs unions", Journal of International Economics, 6, 95-97.

Lipsey, R.G. (1957). "The theory of customs unions; trade diversion and welfare", Economica 24, 40-46.

Meade, J.E. (1955). The theory of customs unions, North-Holland, Amsterdam.

Ohyama, M. (1972). "Trade and welfare in general equilibrium", Keio Economic Studies, 9, 37-73.

Smith, A. and A.J. Venables (1988). "Completing the Internal Market in the European Community: Some Industry Simulations", European Economic Review, 32, 1501-25.

Summers, L. (1991). "Regionalism and the World Trading System", in Policy Implications of Trade and Currency Zones, Federal Reserve Bank of Kansas.

UNCTAD (2005). Handbook of Statistics 2005, Geneva.

Venables, A.J. (2000). "Winners and losers from regional integration agreements", CEPR discussion paper, London.


Viner, J. (1950). The Customs Union Issue, Carnegie Endowment for International Peace.

Winters, L.A. (1998). "Regionalism versus Multilateralism" in Baldwin R., Cohen D., Sapir A., and Venables A. (eds.) Market Integration, regionalism and the global economy, CEPR and Cambridge University Press, Cambridge.

Winters, L.A. and W. Chang (2000). `Regional Integration and the Prices of Imports: An Empirical Investigation", Journal of International Economics, vol. 51, No. 2.

Soloaga, Isidro and Alan Winters (2001). "Preferential Trade Agreements in Asia and the Pacific", Asia Development Outlook, 2002.

Wonnacott, R.I. and M. Lutz (1989). "Is there a Case for Free Trade Areas", in J. Schott(ed.), Free Trade Areas and US trade Policy, Institute for International Economics, Washington, D.C.

World Bank (2000). Trade blocs, Policy Research Report, Washington, D.C.

World Trade Organization (WTO) (1995). Regionalism and the World Trading System, by WTO Secretariat.


C. THE CASE FOR COMESA By Rojid Sawkut, Sannassee Vinesh, Seetanah Boopen and Fowdar Suraj

I. Introduction RTAs in Africa Africa is home to some 30 regional trade arrangements (RTAs), many of which are part of deeper regional integration schemes. African RTAs have largely been motivated by the continent’s desire to promote growth through regional cooperation. Many African countries are landlocked small economies with inadequate infrastructure. Although Africa has 12 per cent of the world’s population, it produces just 2 per cent of the world’s output because its productivity is low. In 2003, sub-Saharan Africa’s GDP was 17 per cent lower than Australia’s; when South Africa is excluded, Africa’s GDP is about the same as Austria’s. RTAs, by creating larger markets, are thought to enable African countries to exploit economies of scale and enhance domestic competition as well as to raise returns on investment and, hence, attract more foreign direct investment (FDI). On average, each African country belongs to four RTAs (World Bank 2004) and on top of the list are many Eastern and Southern African countries. As for Mauritius, it is a member of South African Development Community (SADC), Common Market for Eastern and Southern Africa (COMESA) and the Indian Ocean Commission (IOC). Overlapping membership There is overlap of membership among Regional Economic Communities (RECs) in the Eastern and Southern African region to an extent unparalleled anywhere else in the world.1 For example, almost half of COMESA members are also members of SADC, whose membership is smaller than COMESA's. This may tend to weaken the integration process. It leads to costly competition (even for attention and resources); conflict; inconsistencies in policy formulation and implementation; unnecessary duplication of functions and efforts; fragmentation of markets and restriction in the growth potential of the sub-region. Yet, as most RECs in the Eastern and Southern African region wish to move to a Customs Union (CU), member states with multiple memberships at present will have to strike the balance of the costs and benefits of belonging to one or another CU grouping.2

II. Regional trade blocks in Africa IOC The Indian Ocean Commission (IOC) is a regional organization regrouping four ACP states (Comoros, Madagascar, Mauritius and Seychelles) plus one ultra-peripheral region of the EU (Reunion, an overseas department of France). Set up in 1984, the IOC is one of the first formal experiences of regional cooperation in this part of the

1 See Jakobeit et al. (2005). 2 From a legal as well as from a technical point of view a country cannot apply two different common external tariffs (CET) and therefore cannot be a member of more than one CU. Hence, the current pattern of overlapping membership becomes impossible to maintain once COMESA and SADC also become CUs in addition to SACU and EAC.


vast region constituted by the Indian Ocean. The approach, essentially political in nature, was at the time part of the drive to reinforce cooperation within the Southern hemisphere. The objectives and missions established by the founders of the IOC were primarily to strengthen links between the peoples of its member states and improve their standard of living, promoting cooperation in a number of areas: diplomacy, economy, trade, agriculture, fishing, the conservation of resources and ecosystems, culture, science and education. In terms of development, however, these islands are not all on an equal footing and are in fact at various levels, which are often, very far apart. Reunion is part of the developed world; Comoros and Madagascar are members of the group of least developed countries (LDCs); while Mauritius is classified as a newly industrialized country and the Seychelles as a middle-income country.3

ECOWAS The Economic Community for Western Africa States (ECOWAS) was founded in 1975 with the idea of a community embracing all of Western Africa. The idea was particularly promoted by Nigeria out of the conviction that a broader community would reduce her dependence on oil and increase her influence in French dominated regions. The ECOWAS Treaty of 1975 envisaged the creation of a common market among member countries with a phased reduction of tariffs and non-tariff barriers on products of community origin until their complete elimination for all categories of goods and all countries by 1989; the establishment of a common external tariff by 1994; fiscal and monetary harmonization; and close cooperation in all areas of economic activity. ECOWAS is composed of 15 countries, i.e. all the countries in Western Africa (Benin, Burkina Faso, Cape Verde, Côte d'Ivoire, Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Niger, Nigeria, Senegal, Sierra Leone and Togo) and had an estimated total GNP of $US64 billion.4 It also has a wealth of mineral resources and a huge variety of agricultural products. ECOWAS is the largest and most diversified economic community in sub-Saharan Africa (SSA). The GNI per capita in 1989 was only $US326 which increased to $US370in 2003. The structure of the ECOWAS economies is shown below.

Source: World Bank. http://siteresources.worldbank.org/EXTAFRREGINICOO/Resources/EWS-aag.pdf

3 See http://ec.europa.eu/development/body/publications/courier/courier201/pdf/en_018.pdf. 4 Currently there are 15 members in ECOWAS. One new country joined in 1977 (Cape Verde) and one withdrew in 2002 (Mauritania).


It should be noted that the economic activity of ECOWAS is concentrated heavily on extractive industry and agriculture for exports. Fourteen out of the sixteen ECOWAS countries derive over 60 per cent of their export revenues from just one or two commodities. The population in ECOWAS is more than 250 million, but its combined economy is only $US125 billion. Over the 1974-2004 period, the ECOWAS economy grew 3.2 per cent annually (in nominal dollar terms), just slightly ahead of the 2.8 per cent annual increase in population. One should bear in mind that Nigerians constitute half of ECOWAS’ combined population and that ECOWAS-wide averages are dominated by developments in Nigeria. Against the background of stagnant oil production, the Nigerian economy has trailed economic developments elsewhere in the region and the share of the Nigerian economy in the total ECOWAS economy fell from two-thirds in 1974 to one half in 2004. In contrast, rigorous economic growth in Benin, Burkina Faso, Cape Verde, Ghana, Mali and Senegal was significantly higher than in other ECOWAS countries. Progress towards regional integration was initially slow and uneven, but renewed momentum was provided by enactment of a revised ECOWAS treaty in July 1993. The revised treaty clarified the community’s economic objectives (establishment of a common market and a single currency) and facilitated the establishment of community-wide bodies such as a parliament, a social council and a court of justice. In 1994 a subgroup of seven member countries - Benin, Burkina Faso, Cote d’Ivoire, Mali, Niger, Senegal and Togo - established the West African Economic and Monetary Union (WAEMU), which Guinea-Bissau joined in 1997. Already sharing a common currency - the CFA franc - with a regional central bank located in Dakar, Senegal these countries recognized that they could pursue subregional integration at a somewhat faster clip than what could be expected for the broader ECOWAS membership. By 1999 the WAEMU countries had adopted a regional pact of convergence, stability, growth and solidarity aimed at strengthening economic performance through deeper regional integration and in 2000, a customs union was created with the elimination of tariffs on intra-WAEMU trade and establishment of a common external tariff (CET) on imports from outside the WAEMU area. With WAEMU countries already sharing a common currency, monetary cooperation among non-WAEMU countries (with the exception of Cape Verde and Liberia) takes place within the so-called Second Monetary Zone where the aim is to establish a common non-WAEMU currency that is then subsequently merged with WAEMU’s CFA franc. As monetary unification requires macroeconomic stability, members of the Second Monetary Zone have committed to comply, inter alia, with various fiscal rules and limits on price and exchange rate volatility. In this area compliance has also remained spotty, and the timeline for introducing the common currency has been pushed back on several occasions (the last postponement shifted introduction of the common currency back from 2005 to December 2009).5

5 See Simplice and Lynge (2007).


CEAO The Communauté Economique de l’Afrique de l’Ouest (CEAO) was founded in 1973 by the Treaty of Abidjan and comprises seven members: Burkina Faso, Côte d’Ivoire, Mali, Mauritania, Niger and Senegal were founding members. Benin became a member in 1984. The CEAO operates as a free trade area for agricultural products and raw materials and as a preferential trading area for approved industrial products, with a regional cooperation tax replacing import duties and encouraging trade among members. CEAO envisions eventual creation of a customs union and coordination of fiscal policies. Community headquarters are in Ouagadougou, Burkina Faso. SACU SACU was established through the Customs Union Agreement of 1910 between the Union of South Africa and the so-called BLS states, Botswana, Lesotho and Swaziland. The 1969 Customs Union Agreement between South Africa, Botswana, Lesotho and Swaziland replaced that agreement. After independence, Namibia also formally joined the Union in 1990. The Union was created in 1910, soon after the Republic of South Africa was formally created as an independent state. The new union replaced an older one which had been in existence since 1889. The main reasons for the establishment of SACU were that of regional integration and the facilitation of trade between the members of the agreement in order to improve economic development of the whole area, in particular the less advanced members. The main feature of SACU is the overwhelming dominance of South Africa over the other three members. The economic environment of SACU The Southern African Customs Union (SACU) consists of five member states, Botswana, Lesotho, Namibia, South Africa and Swaziland. It was established through the Customs Union Agreement of 1910. The 1910 Agreement was re-negotiated leading to the 1969 Agreement. The Agreement was further re-negotiated in 1994 leading to the current SACU 2002 Agreement that came into force on 15 July 2004. The main objective for the establishment of SACU was that of Regional Integration, Trade Facilitation between the Members States and Trade Negotiations between SACU and third parties, in order to improve the economic development of the Member States. The aim of the Southern African Customs Union (SACU) is to maintain the free interchange of goods between member countries. It provides for a common external tariff and a common excise tariff to this common customs area. All customs and excise collected in the common customs area are paid into South Africa’s national Revenue Fund. Revenue is shared among members according to a revenue-sharing formula as described in the agreement. South Africa is the custodian of this pool. Only the member states shares are calculated with South Africa receiving the residual. SACU revenue constitutes a substantial share of the state revenue of the BLNS countries (Botswana, Lesotho, Namibia and Swaziland). SACU countries have very different levels of economic scale, structure and development. Botswana and South Africa are upper middle-income countries (although very different in economic structure), Namibia and Swaziland are lower middle-income countries, and Lesotho is a least developed country. However, all


SACU countries face common challenges such as eradicating poverty; combating HIV/AIDS; promoting sustainable economic growth and development, and a more equitable income distribution; reducing high unemployment rates; and being more fully integrated into the world economy. Monetary and exchange rate policies within SACU are largely led by South Africa. Indeed, Lesotho, Namibia, South Africa and Swaziland have a Common Monetary Area (CMA) Agreement, where the loti (Lesotho's currency), the Namibian dollar and the lilangeni (Swaziland's currency) are pegged to, and freely convertible into, the South African rand at par. Although a non-member of the CMA, Botswana has pegged its currency (pula) to a basket of currencies, including the rand, and has generally contained its fluctuations against the rand. This monetary stance and the constraint on customs revenue due to SACU membership somewhat disciplined fiscal policy within the customs union. This relative harmonization of policies has resulted in comparable levels of inflation (around 7 per cent per year) throughout SACU. Nevertheless, the lack of formal harmonization of fiscal policies leaves substantial leeway for expenditure-induced public deficits. SACU countries have generally been also slow in implementing structural reforms. SACU members (except South Africa) depend on a limited number of products. The share of agriculture in SACU's GDP decreased somewhat during the last few years, but it remains an important sector with linkages to other activities. Food insecurity in Lesotho, Namibia and Swaziland has increased recently due to several factors, including unfavorable weather conditions (drought and floods), making emergency food aid necessary. The food security objective has driven the move towards tariffs as the main trade policy instrument in the sector, although some non-tariff measures still exist. Mining and quarrying provides a key anchor for growth and development, particularly in Botswana, Namibia and South Africa, where mineral taxation, for example, has enabled the improvement of basic infrastructure and social services. Furthermore, the sector is one of the most important foreign exchange sources and a magnet for foreign investment in these countries. Large mineral endowment has meant low tariff protection, and private investment continues to be actively encouraged in the sector. Certain labor-intensive manufacturing activities, particularly textiles and clothing, have expanded strongly in recent years throughout SACU, largely as a result of new incentives provided under non-reciprocal preferential arrangements. The general development of the manufacturing sector, however, has largely been hampered by supply-side constraints, such as high production costs, limited access to financing, low capacity utilization, and low quality of products. Moreover, subject to concessions, the tariff structure (mainly the relatively high rates on semi-finished goods) is not conducive to investment in certain manufacturing activities. Export opportunities in the services sector remain largely unexploited, other than in South Africa and to some extent in Botswana and Namibia. In tourism, for example, where SACU’s attractions are among the best in Africa, inadequacies in infrastructure and marketing/promotion, financial constraints and lack of skilled labor have constrained the development of the subsector. Further liberalization and investment in services, in general, should improve the efficiency of other economic activities and


the competitiveness of SACU’s exports, especially by reducing costs related to, inter alia, telecommunications, transport and energy. South African Development Community The Southern African Development Coordination Conference (SADCC) was established in 1980 in terms of the ‘Lusaka Declaration: Southern Africa: Towards Economic Liberation’ by the governments of the nine Southern African countries of Angola, Botswana, Lesotho, Malawi, Mozambique, Swaziland, Tanzania, Zambia and Zimbabwe. The positive experiences gained in working together in the group of Front Line States, to advance the political struggle, had to be translated into broader cooperation in pursuit of economic and social development.6 From 1977, active consultations were undertaken by representatives of the Front Line States, culminating in a meeting of Foreign Ministries of the Front Line States in Gaborone, in May 1979, which called for a meeting of ministers responsible for economic development.7

That meeting was subsequently convened in Arusha, Tanzania, in July 1979. The Arusha meeting led to the birth to the Southern African Development Co-ordination Conference a year later. This transformation occurred in August 1992, when the Heads of State and Government of the Southern African Development Co-ordination Conference met in Windhoek, Namibia, to sign a Declaration and Treaty establishing the new SADC - the Southern African Development Community. The SADCC leaders had come to realize that although the coordination conference had served them well and had demonstrated the crucial need to cooperate in their development efforts, time had come to give the organization a legal and more formal status. There was also a need to shift the focus of the organization from coordination of development projects to a more complex task of integrating the economies of member states. Hence the Treaty, which is the blueprint for building a Community of Southern African states. The Southern African Development Community (SADC) was established on 17 August 1992 by Angola, Botswana, Lesotho, Malawi, Mozambique, Namibia, Swaziland, Tanzania and Zimbabwe. Originally known as the Southern African Development Coordination Conference (SADCC), the organization was formed in Lusaka, Zambia on 1 April 1980, following the adoption of the Lusaka Declaration. SADC is an intergovernmental organization aimed at promoting economic development. The objective of SADC is to foster harmonized regional development through the concerted undertaking of economic activities on a sector-by-sector basis. Current member states are Angola, Botswana, Democratic Republic of Congo (DRC), Lesotho, Malawi, Mauritius, Mozambique, Namibia, Seychelles, South Africa, 6 As the countries of the Southern African region gradually became independent, starting with Zambia in 1964, the newly independent governments believed that it was their responsibility to give as much support as they could to liberation groups fighting for their independence in neighboring states. So important was this mission that once Angola and Mozambique won their independence in 1975, they joined Zambia, Botswana and Tanzania to form a political coalition known as the Front Line States. They selected this name because they formed the front line between the independent states of Africa and the colonies still under the control of European settlers within the southern Africa region. The Front Line States provided invaluable assistance for the liberation struggle in neighboring states through direct material assistance and by providing bases for the armies of the liberation movements. Each time one of the nations of the region won their independence, they joined the Front Line States and lent their support for remaining liberations struggles http://exploringafrica.matrix.msu.edu/students/curriculum/m20/activity4.php. 7 SADCC grew Out of a political grouping, the Frontline States, whose objective was to brine about independence under majority rule in Zimbabwe, Namibia and South Africa.


Swaziland, Tanzania, Zambia and Zimbabwe. SADC headquarters are in Gaborone, Botswana. All SADC counties are contracting parties to the General Agreement on Tariffs and Trade (GATT), extending MFN treatment to each other. They are members of the World Trade Organization (WTO). SADC and its member states are expected to act according to the following principles:

• Sovereign equality of all member states; • Solidarity, peace and security; • Human rights, democracy and the rule of law; • Equity, balance and mutual benefit; • Peaceful settlement of disputes.

SADC objectives

• Achieve development and economic growth, alleviate poverty, enhance the standard and quality of life of the people of Southern Africa and support the socially disadvantaged through regional integration;

• Evolve common political values, systems and institutions; • Promote and defend peace and security; • Promote self-sustaining development on the basis of collective self-reliance,

and the interdependence of member states; • Achieve complementarities between national and regional strategies and

programmes; • Promote and maximize productive employment and utilization of resources of

the region; • Achieve sustainable utilization of natural resources and effective protection of

the environment; • Strengthen and consolidate the long-standing historical, social and cultural

affinities and links among the people of the region. The ultimate objective of SADC as a community is, therefore, to build a region in which there will be a high degree of harmonization and rationalization to enable the pooling of resources to achieve collective self-reliance in order to improve the living standards of the people of the region. COMESA The Common Market for Eastern and Southern Africa (COMESA) evolved out of the former Preferential Trade Area (PTA) which had existed since the early days of 1981. The establishment of COMESA followed a treaty signed in November 1993 and was ratified in December 1994. COMESA was established, as defined by its treaty “as an organization of free independent sovereign states which have agreed to co-operate in developing their natural and human resource for the good of all their people”.8 COMESA strategy of economic prosperity through regional integration has as main focus the formation of a large economic and trading unit that is capable of overcoming some of the barriers that are faced by individual states.

8 See the COMESA website at http://www.comesa.int/about/Overview/view.


COMESA, the 209-nation African trade bloc, launched its free trade area (FTA) on 31 October 2000 with nine10 of its members initially participating in the project, which dismantled trade barriers and guaranteed free movement of goods and services in the region. COMESA was pushing ahead with its plan to adopt a common external tariff and a customs union in December 2004 and to adopt a common currency for regional members by 2025. The target for the CU was initially 2004 but this target has already been missed and the new target is 2012. Article 45 of the COMESA Treaty provides that “within the CU, custom duties and other charges of equivalent effect imposed on imports shall be eliminated”. Non-tariff barriers shall also be eliminated. Furthermore, a CET with respect to all goods imported into the member states from third countries shall be established and maintained. At its Eighteenth Meeting in Lusaka, Zambia in December 2004, the Council of Ministers decided that member states should work towards harmonizing their external tariffs as a transition strategy towards realizing the COMESA CET, as follows:11

Category Range for tariff harmonization CET target rate Raw materials 0% to 5% 0% Capital goods 0% to 5% 0% Intermediate goods 10 to 15% Not agreed Final goods 25 to 40% Not agreed Whilst there is agreement on the CET target rates for capital and raw materials, work is ongoing on rates for intermediate and finished products. With regard to the rates for intermediate products, many of which are direct inputs for industry, there is a need to define a rate that will ensure industrial competitiveness. In that regard, an exercise is ongoing that will provide for exceptions, exclusions and derogations. This exercise will also take into account goods used as inputs in production, goods under a rebate system, merit goods, goods of social significance and goods covered by international agreements. With regard to the rates for finished products, there is agreement to harmonize tariffs in the range of 25-40 per cent.12 The origins of the COMESA can be traced as far back as the mid-sixties. Before the Lagos Plan of Action and the Final Act of Lagos were adopted, the countries of Eastern and Southern Africa had already initiated the process towards creating an Eastern and Southern African cooperation arrangement. The African integration agenda was put forward in 1958, but it was not until 1960 and 1963 that the OAU took practical steps. In 1965 the United Nations Economic Commission for Africa (ECA) convened a ministerial meeting of the then politically independent states of Eastern and Southern Africa to consider proposals for the establishment of a mechanism for the promotion of subregional economic integration. The meeting,

9 The twenty members are Angola, Burundi, Comoros, DR Congo, Djibouti, Egypt, Eritrea, Ethiopia, Kenya, Madagascar, Malawi, Mauritius, Namibia, Rwanda, Seychelles, Sudan, Swaziland, Uganda, Zambia and Zimbabwe. 10 The nine members which initially joined the FTA are Djibouti, Egypt, Kenya, Madagascar, Malawi, Mauritius, Sudan, Zambia and Zimbabwe. 11 For an assessment of the impact of COMESA forming its CU, the readers are directed to “The impact of COMESA forming its custom union – A CGE based analysis” by Rojid Sawkut presented at the CSAE conference, St Catherine College, University of Oxford, March 2007. 12 See the COMESA website, Custom Union page. http://www.comesa.int/trade/Folder.2004-04-19.2132/MS-Office-Document.2005-05-15.2735/view.


which was held in Lusaka, Zambia, recommended the creation of an Economic Community of Eastern and Southern Africa States. An Interim Council of ministers, assisted by an Interim Economic Committee of officials, was subsequently set up to negotiate the treaty and initiate programmes on economic cooperation, pending the completion of negotiations on the treaty. After the preparatory work had been completed, a meeting of the Heads of State and Government was convened in Lusaka on 21 December 1981 at which the Treaty establishing the PTA was signed. The Treaty came into force on 30 September 1982 after being ratified by more than seven signatory states as provided for in Article 50 of the Treaty. COMESA replaced the former PTA and established itself as an organization of free independent sovereign states, which have agreed to cooperate in developing their natural and human resources for the good of all their people. The Treaty establishing COMESA was signed on 5 November 1993 in Kampala, Uganda and was ratified a year later in Lilongwe, Malawi on 8 December 1994. Principles of COMESA The Treaty establishing COMESA binds together free independent sovereign states which have agreed to cooperate in exploiting their natural and human resources for the common good of all their peoples. In attaining that goal, COMESA recognizes that peace, security and stability are basic factors in providing investment, development, trade and regional economic integration. Experience has shown that civil strives, political instabilities and cross-border disputes in the region have seriously affected the ability of the countries to develop their individual economies as well as their capacity to participate and take full advantage of the regional integration arrangement under COMESA. It has now been fully accepted that without peace, security and stability there cannot be a satisfactory level of investment even by local entrepreneurs. Therefore, in pursuit of the aims and objectives stated in Article 3 of the COMESA Treaty, the member states of COMESA have agreed to adhere to the following principles:

• Equality and inter-independence of the member states; • Solidarity and collective self-reliance among the member states; • Inter-State cooperation, harmonization of policies and integration of

programmes among the member states; • Non-aggression between the member states; • Recognition, promotion and protection of human and people's rights in

accordance with the provisions of the African Charter on Human and People's Rights;

• Accountability, economic justice and popular participation in development; • The recognition and observance of the rule of law; • The promotion and sustenance of a democratic system of governance in each

member state; • The maintenance of regional peace and stability through the promotion and

strengthening of good neighborliness; and • The peaceful settlement of disputes among the member states and the

promotion of a peaceful environment as a pre-requisite for their economic development.


Aims and objectives of COMESA The aims and objectives of COMESA are defined in the Treaty and its Protocols. They have been designed so as to facilitate the removal of the structural and institutional weaknesses in the member states and the promotion of peace; security and stability so as to enable them attain sustained development individually and collectively as a regional bloc. These are as follows:

• To attain sustainable growth and development of the member states by promoting a more balanced and harmonious development of its production and marketing structures;

• To promote joint development in all fields of economic activity and the joint adoption of macro-economic policies and programmes; to raise the standard of living of its peoples, and to foster closer relations among its member states;

• To cooperate in the creation of an enabling environment for foreign, cross-border and domestic investment, including the joint promotion of research and adaptation of science and technology for development;

• To cooperate in the promotion of peace, security and stability among the member states in order to enhance economic development in the region;

• To cooperate in strengthening the relations between the Common Market and the rest of the world and the adoption of common positions in international fora; and

• To contribute towards the establishment, progress and the realization of the objectives of the African Economic Community.

The COMESA agenda is to deepen and broaden the integration process among member states through the adoption of more comprehensive trade liberation measures such as the complete elimination of tariff and non-tariff barriers to trade and elimination of customs duties; through the free movement of capital, labor, goods and the right of establishment; by promoting standardized technical specifications, standardization and quality control; through the elimination of controls on the movement of goods and individuals; by standardizing taxation rates (including value added tax and excise duties), and conditions regarding industrial cooperation, particularly on company laws, intellectual property rights and investment laws; through the promotion of the adoption of a single currency and the establishment of a Monetary Union; and through the adoption of a CET. By agreeing to the above, member states have agreed on the need to create and maintain:

• A full free trade area guaranteeing the free movement of goods and services produced within COMESA and the removal of all tariffs and non-tariff barriers;

• A customs union under which goods and services imported from non-COMESA countries will attract an agreed single tariff all COMESA States;

• Free movement of capital and investment supported by the adoption of common investment practices so as to create a more favorable investment climate for the entire COMESA region:

• A gradual establishment of a payments union based on the COMESA Clearing House and the eventual establishment of a common monetary union with a common currency;


• The adoption of a common visa arrangement, including the right of establishment leading eventually to free movement of bona fide persons.

COMESA achievements COMESA, as well as is predecessor the PTA, has achieved a lot in the area of trade, customs, transport, development finance and technical cooperation. Impressive progress has also been made in the productive sectors of industry and agriculture. Trade facilitation and trade liberalization measures are bearing fruit.

• Intra-COMESA trade, valued at about $US4.2 billion, is growing at the rate of 20 per cent per annum. Trade with third countries is growing at about 7 per cent per annum.

• Transport transit facilitation measures have resulted in a reduction of costs by 25 per cent.

• Inter-state movement of persons, goods and means of transport has been facilitated.

• In the sector of telecommunications, special emphasis has been placed on network development to enable direct telecommunication links through more reliable infrastructure.

• Establishment of several important institutions was achieved, including the PTA Trade and Development Bank, the COMESA Clearing House, the COMESA Re-insurance Company and the COMESA Leather and Leather Products Institute.

• The PTA Bank has, over the years, been very active in promoting investments and providing trade financing facilities.

• The Re-Insurance Company has, since its establishment in 1992, been able to carve out a reasonable share of the regional insurance business and is now transacting business in some nineteen countries.

• Investment in the region is promoted and this issue is addressed through facilitation of bilateral agreements; promoting export drives by individual member states, and identifying specific projects, which have the potential to act as growth poles between two or more member states.

Major concerns about COMESA The magnitude of COMESA’s external indebtedness is a source of serious concern. The external debt of the COMESA region has increased twenty-fold since 1970 and debt service ratios, which in 1970 were insignificant, averaged 45 per cent of export earnings in 1989-90, making the region one of the most heavily indebted in the world. While member states borrowed heavily to maintain incomes and investments, the collapse of their export earnings undermined attempts to reduce their debts. Debt relief to the COMESA region, and SSA as a whole, has been limited in relation to the magnitude of the problem and inflows of Official Development Assistance continue to decline. On the production side, both the agricultural and industrial sectors have been in decline. For many COMESA countries, agriculture constitutes between 50 and 76 per cent of GDP but the growth of agricultural output, at an average of 2 per cent per year over the last three decades, has barely matched that of population growth, so has not contributed effectively to sustainable growth and development. Agricultural exports have declined, budgetary allocations to agriculture have remained small and


inadequate and an anti-poor bias in agricultural policy across much of the region, notably through over-taxation of crops, inadequate spending on market infrastructure for small-holder producers, and insufficient investment in research of local foods have combined to adversely affect the region’s trade share of exports in the world market, which has dropped by 50 per cent since 1970. Food imports are increasing at about 8 per cent a year and COMESA’s current bill for cereals is over $US2 billion. This heavy and chronic dependence on food imports is particularly dangerous for COMESA, not only because it’s debt and trade problems impose serious limits on its ability to purchase food in world markets, but also because there is no guarantee that food aid and/or commercial imports will be available when needed in the required quantities and quality. Although industry grew roughly three times as fast as agriculture in the first decade of independence, the past few years have seen an alarming reversal in many states where de-industrialization, as a short-term effect of structural adjustment, has set in. Progress in the manufacturing sector has fallen far short of the target growth rate of 8 per cent per annum projected in the second Industrial Development Decade for Africa (IDDA II) as a result of entrenched structural rigidities, weak inter-industry and inter-sectoral linkages, lack of access to advanced technologies and poor institutional and physical infrastructure. The African continent’s share of world manufacturing value added (MVA) rose from 0.7 per cent in 1970 to 1 per cent in 1982 and fell to 0.8 per cent in 1994. Most African industries have a very low capacity utilization rate and current structural adjustment programmes have as yet to have a positive impact on the industrial sector. Population is expanding at a rate of around 3.2 per cent, outstripping agricultural and food production and COMESA now has twice the population it had in 1965 and more than five times the population it had at the beginning of the century. The region has also experienced, over the last few years, unprecedented droughts, leading to widespread food shortages and famine. There is growing and widespread poverty in the COMESA region, especially among the rural communities, aggravated by the decline in expenditures on social services, including health, education and public utilities, nutrition has worsened and mortality continues to increase. There is a major crisis in employment in all countries, especially among the youth in cities and towns. Unemployment in most countries is as much as 30 per cent or more of the active labor force and under-employment is just as serious. The majority of the region’s population still dwells in the villages and earns their living cultivating between one and fifteen hectares. The COMESA region has also had to contend with civil strife, ethnic wars and political instability, which have also contributed to the decline in economic growth. In summary, the economic performance of the COMESA region has been rather disappointing over the last decades, with overall economic growth of the COMESA region having averaged 3.2 per cent a year since 1960 and only marginally above the level of the region’s population growth. Nevertheless, performance of the region has been remarkable in the very recent past: In 2005, the COMESA region recorded an overall GDP growth rate of 5.8 per cent, buoyed by high prices for petroleum and metal products. During the same period, the average growth rate for the African continent was 4.9 per cent. Total COMESA exports increased by 54 per cent in 2005,


while imports surged by 46 per cent; primarily due to higher fuel prices. The oil-exporting economies saw the fastest growth with Sudan recording 13.4 per cent growth. Non-oil economies, that export mineral and metal commodities, also saw strong increases. The DRC grew at 6.2 per cent and Uganda at 5.8 per cent.13

III. Economic structure of some COMESA Members14 COMESA members vary largely in terms of size in many respects. For example, as seen from Table C1, they differ in geographic, demographic and economic terms. With a 2002 combined population of about 257 million (Table C1), these economies are, for example, potentially rich in human resources. Ethiopia and DRC are the two largest countries in the region with population size of 67.2 and 51.5 million respectively. Comoros and Seychelles are the two smallest countries in population size. Land area varies considerably ranging from 450 sq km in Seychelles to 2'267'050 sq km in DRC. Likewise, GDP per capita also vary largely across countries. In 2002, GDP per capita at constant 1995 prices ranges from $US515 in Malawi to more than $US9'500 in Mauritius. Heterogeneity exists to a large extent among COMESA members as far as economic structures are concerned. Some countries (like Burundi, DRC and Rwanda) are still highly relying on the agricultural sector (more than 40 per cent of total economic activity) while some others (like Seychelles, Mauritius and Kenya) are already heading towards the services sector (more than 60 per cent of total economic activity). Angola on the other hand is the only member state having more than 50 per cent of its resources (68.1 per cent) employed in the industry sector. Table C1 also shows that the prevailing level of inflation varies from country to country. For example, while Zimbabwe and Angola are plagued by rampant hyperinflation (140.1 per cent and 118.8 per cent respectively), Burundi and Uganda are affected by declining inflation rates (-1.4 per cent and 0.3 per cent). These very large increases in price level experienced are mainly attributed to the civil strife in Angola and the severe currency shortage in Zimbabwe. The general negative inflation rate prevailing in countries like Burundi and Uganda is due to deflation in the commodities market.

13 See http://www.iss.co.za/dynamic/administration/file_manager/file_links/STATE%20OF%20INTEGRATION%20REPORT %20-%2006TH%20NOV%202006.PDF?link_id=3893&slink_id=3873&link_type=12&slink_type=13&tmpl_id=3. 14 For a more elaborate study on the economic structure of COMESA states, readers are referred to the PDH thesis (not yet published as of 0ctober 4th 2007) of one of the authors, namely Sawkut Rojid.


Table C1: Basic economic indicators of COMESA countries and sectors’ share of GDP (per cent), 2002

Inflation, consumer

prices (annual %)

Population, total

GDP per capita, PPP

(constant 1995 $US)

Agriculture, value added (% of GDP)

Industry, value added (% of GDP)

Services, value added (% of GDP)

Angola 118.8 13'121'000 1890.6 7.8 68.1 24.1 Burundi -1.4 7'071'000 561.3 49.3 19.4 31.3 Comoros 586'000 1499.3 35.4 10.6 54.0 DR Congo 31.5 51'580'000 578.4 56.3 18.8 24.9 Ethiopia 1.6 67'218'000 693.1 39.9 12.4 47.6 Kenya 2.0 31'345'000 902.1 16.4 19.0 64.6 Madagascar 15.9 16'437'000 659.3 32.1 13.3 54.7 Malawi 14.7 10'743'000 514.9 36.5 14.8 48.7 Mauritius 6.7 1'212'000 9577.2 7.0 31.0 62.0 Rwanda 2.5 8'163'000 1126.0 41.4 21.3 37.3 Seychelles 0.2 840'000 .. 2.9 30.0 67.1 Uganda -0.3 24'600'000 1228.9 31.6 22.0 46.4 Zambia 22.2 10'244'000 742.7 22.2 26.1 51.7 Zimbabwe 140.1 13'001'000 .. 17.4 23.8 58.8

Source: WDI 2004 database, The World Bank. COMESA patterns of trade15

As shown in Table C2b, the shares of COMESA members’ exports to the COMESA block has increased in 2005 compared to 1980, with a peak around the year 2004. Table C2a shows that COMESA as a destination of exports is particularly important for Djibouti, Ethiopia, Kenya and Uganda with their share of exports to COMESA in 2001 being 24 per cent, 20 per cent, 23 per cent and 25 per cent respectively. The EU is the main destination for COMESA members’ products. These countries products are exported to the EU under several initiatives, namely the Everything But Arms (EBA)16 initiative and the Cotonou Agreement17 (following the Lomé Convention). Under both these agreements some of these countries have preferential market access in the EU. The Cotonou agreement has already phased out and the ACP countries together with the EU are in the process of negotiating new trade treaties called Economic Partnership Agreements. Some countries have already initiated and interim agreement towards this at the end of December 2007. The EPA should be signed by end 2008.

15 For a detailed insight in Intra Industry Trade (IIT) analysis for COMESA, the readers are directed to the PHD thesis of Rojid Sawkut (not published as of 4 October 2007). 16 EBA is an initiative of the European Union under which all imports to the EU from the least developed countries are duty free and quota free, with the exception of armaments. It entered into force on 5 March 2001. 17 The Cotonou Agreement is a treaty between the European Union and the group of African, Caribbean and Pacific states (ACP countries). It is aimed at the reduction and eventual eradication of poverty while contributing to sustainable development and to the gradual integration of ACP countries into the world economy.


Table C2a: Share of exports18 (per cent) EUROPE USA COMESA 1990 2001 1990 2001 1990 2001 Angola 36 26 56 47 <1 <1 Burundi 86 72 7 0.7 1 19 Comoros 75 47 18 24 <1 <1 DR Congo -- -- 17 13 <1 7 Djibouti 53 2 <1 <1 4 24 Egypt 45 37 8 9 <1 2 Ethiopia 41 36 11 5 12 20 Kenya 40 33 2 5 34 23 Madagascar 54 42 17 20 3 5 Malawi 51 40 17 14 2 12 Mauritius 78 68 12 19 1 7 Rwanda 59 31 21 4 <1 -- Seychelles 19 48 1 3 <1 7 Sudan 38 8 3 <1 6 2 Uganda 82 45 8 2 1 25 Zambia 37 44 2 -- 1 9 Zimbabwe 44 43 6 5 13 17

Source: Calculated from WTA. Table C2b: Share of COMESA intra exports as a percentage of COMESA total

exports 1980 1990 2000 2004 2005

COMESA intra export 555 889 1'328 2'294 2'716

COMESA total exports 13'851 18'268 24'777 40'106 56'311

Share of intra exports to total exports 4.0 4.9 5.4 5.7 4.8

Source: UNCTAD 2006/07. Table C2c: Average annual growth rate of exports and imports for the COMESA region (per cent) 1980-1990 1990-2000 2000-2005Exports 4.1 4.1 18.1 Imports 6.4 5.2 12 Source: UNCTAD 2006/07. As can be seen from Table C3, the range of products traded within the COMESA has been submitted to significant changes throughout the period of study. Exports of ‘food and live animals’, which in the 1980’s were ranked third in priority, were the main intra-COMESA exports in 2000. ‘Manufactured goods classified chiefly by material’ have remained the second most important area of exports within the bloc. Alternatively, ‘mineral fuels, lubricants and related materials’, which were the main intra exports in 1980 was ranked third in 2000. Likewise, machinery and transport equipment which made up 6 per cent of the COMESA total intra exports in 1980 was

18 The authors also calculated the shares for two more years, 1995 and 2000 and for the region Asia, and the data can be made available on request.


reduced to 2.7 per cent in 2000. In contrast, exports in ‘beverages and tobacco’ and ‘crude materials, inedible except fuels’ experienced steady increases from 1980 to 2000. ‘Animal and vegetable oils, fats and waxes’ and ‘miscellaneous manufactured articles’ exported as a percentage of COMESA total intra exports have also increased, but with minor fluctuations within the past two decades. Rojid (on-going PHD thesis) notes that: COMESA member states exports on similar product categories are becoming more intense (given by an increased IIT index value). The number of COMESA members with a Grubel-Lloyd index higher within COMESA than within the world increased from 5 to 11 during the 1980-2002. This comes to confirm the hypothesis that trade within developing countries is more of IIT. Table C3: Products exported as a percentage of COMESA total intra exports19

1980 1985 1990 1995 2000 Food and live animals 15.4 42.0 22.8 33.5 30.1 Beverages and tobacco 2.5 4.1 4.3 4.8 5.1 Crude materials, inedible, except fuels 4.1 4.1 7.6 9.8 10.2 Mineral fuels, lubricants and related materials 39.0 24.5 10.2 5.4 11.7 Animal and vegetable oils, fats and waxes 0.2 0.1 1.7 0.4 1.4 Chemicals and related products, n.e.s. 8.9 6.3 7.5 8.1 8.7 Manufactured goods classified chiefly by material 19.3 13.7 22.7 25.1 22.5 Machinery and transport equipment 6.5 3.0 17.7 6.6 2.7 Miscellaneous manufactured articles 4.2 2.0 5.2 5.9 7.4 Commodities and transactions not classified elsewhere 0.1 0.1 0.3 0.3 0.3 TOTAL 100 100 100 100 100

Source: Calculated from WTA. Trade performance Over the last 2 decades, as seen from Table C4, growth performance in SSA has been relatively poor compared to other developing regions. During the period 1986 to 1994 the performance of COMESA bloc was even worse than average SSA countries (1.24 per cent compared to 1.42 per cent). However, since 1995, the performance of the SSA region improved and more interestingly, the performance of COMESA improved more than average SSA. Table C4: GDP (PPP) yearly growth rate (per cent, 1995 prices)

1986-1994 1995-2002 Sub-Saharan Africa 1.42 3.06 East Asia & Pacific 11.03 6.95 South Asia 5.79 5.38 Low & middle income 3.12 4.42 COMESA 1.24 3.62

Source: Author’s calculation based on WDI 2004 database, The World Bank. According to Table C5, which takes demographic growth trends into consideration, the GDP growth rate of COMESA since the mid-1980s has been sufficient to provide an increase in the standard of living. However, it is important to note the high variation in the growth performance of COMESA members reflecting the diverse characteristics of these economies. During the period 1986 to 1994, average growth

19 Results at 2 digit and 3 digit SITC level are available on request from the author.


rates of GDP per capita ranges from -5.5 per cent in Congo to 5.5 per cent in Mauritius. Between 1995 and 2002, GDP per capita growth was 4.16 per cent in Uganda in comparison to -3.6 per cent in Madagascar. COMESA as a bloc has done much better than SSA countries taken as a whole. Table C5: GDP per capita yearly growth rate (per cent, 1995 prices)

1986-1994 1995-2002 Angola -3.76 0.63 Burundi -1.09 3.15 Cameroon -4.67 1.32 DR Congo -5.54 2.84 Ethiopia -0.50 -1.50 Kenya 0.05 2.24 Madagascar -1.39 -3.63 Malawi -0.88 -0.34 Mauritius 5.53 2.64 Rwanda -4.77 -0.62 Seychelles 4.40 -1.20 Sudan 2.19 0.15 Uganda 2.39 4.16 Zambia -2.00 3.72 Zimbabwe 0.35 2.39 Sub-Saharan Africa -0.98 0.63 South Asia 3.19 3.15 Middle East & North Africa 0.38 1.32 Lower middle income 0.56 2.84 COMESA 1.26 2.42

Source: Author’s calculation based on WDI 2004 database, The World Bank. In the context of regional trade arrangements, members of the same bloc exporting similar products are less likely to benefit from the arrangement as trading within them will be limited. The essence of trade creation within a regional bloc lies on diversified export potentials among members of the region. Economists have used imports and exports values to generate different indicators as a means to find out whether there exist potential trade benefits between trading partners. One of the ways to further analyze intra-COMESA trade flows is to look at the revealed comparative advantages (RCA). In the context of regional trade arrangements, the presumption is that country groupings that have a narrower range of RCA indices (and in similar products) are less likely to find grounds for sustained exporting within the RTA. The technique of RCA ratios is used to measure the importance of a country’s exports of a particular product in its total exports relative to world exports of that product to the world market. In order to measure the RCA of a country in a certain product category, we compare the share of that product category in that country’s exports to the share of that product category in overall world exports. Thus this measure captures the extent to which a country exports more of a product than the average country. The RCA is defined as follows:


RCA = ( X ij / X j ) / ( X iw / X w ),

Where X ij = exports of i-th product from country j

X j = total exports from country j

X iw = exports of i-th product from the world X w = total exports from the world. An RCA index above unity indicates relative advantage while that below unity denotes relative disadvantage. According to our results in the Appendix the range of comparative advantages for Zimbabwe, Zambia and Kenya are less concentrated than other COMESA countries. Angola, Djibouti, Comoros, Mauritius and Seychelles have the most concentrated range of comparative advantages. None of the COMESA countries has comparative advantages in machinery and equipment products (including telecommunication and electrical machinery and road vehicles). Except from that, COMESA members have comparative advantages in almost all product lines. In a logical sense therefore, there is potential for more trade within COMESA. However, from the results obtained we should also note that some countries have comparative advantage in quite similar products. This tends to suggest that to some extent complementarities as a way to stimulate trade might be limited among some COMESA members.

IV. Measuring the effect of COMESA on intra-COMESA trade: A preliminary analysis20

Two preliminary models were employed in an attempt to measure the effect of COMESA on intra-COMESA trade over the period 1985-2004 using a time series analysis. The first represents intra exports and the latter represents intra imports and we take the aggregate value of the countries in the COMESA. The purpose of these models is to identify the substitution effect between intra-COMESA trade and COMESA trade with the rest of the world. Therefore imports and exports with the rest of the world have been introduced as explanatory variables in these models. These two variables can serve the purpose of capturing the effects of trade agreements other than intra trade agreements. In addition, time (t) was introduced in the model with the purpose of capturing the growing trend in intra trade over time. Consequently, the following summarizes the model:

143210 Re uCOMESAtgdpstxIntrax +++++= βββββ (1)

2423210 Re uCOMESAtgdpstmIntram +++++= βαββαα (2) Where: Intrax = intra exports, Restx = exports with the rest of the world, COMESA = Dummy variable takes the value one for the periods 2000-2005, and zero otherwise, t is the time trend (taking values 1, 2…etc), Intram is intra imports, and Restm refers to imports from ROW. We also include GDP due to the predominance of the scale effect. The trade flows and GDP are measured in current million dollars and has been

20 For a detailed econometric analysis of COMESA trade using a gravity approach, see Rojid (2006).


obtained from the World Trade Analyzer and the Penn World Table respectively. u1, u2 are disturbance terms and are assumed to have zero mean and constant variance. αi and βi are unknown parameters to be estimated. Table C6: OLS estimates in logarithm

Parameter Estimate t-statistics Dep: log(Intrax) Constant -3.233 2.11* Log Restx β1 0.85 2.25** Log GDP β2 0.63 1.92* Time β3 0.154 1.95* Dummy β4 0.03 1.22 R-squared 0.91 DW 1.97 Dep: log(Intram) Constant 1.75 2.16** Log Restm α1 1.53 2.34*** Log GDP α2 0.43 1.93* Time α3 0.45 1.86* Dummy α4 0.08 0.86 No of Obs. 20 R-squared 0.95

Notes: * Significant at 10 per cent level; ** significant at 5 per cent level; *** significant at 1 per cent level. Of interest to us, the effect of COMESA on intra trade, as revealed by the insignificance t-values of the dummy variables in both models shows that the COMESA agreements had no significant effects on intra trade. It should be noted that this might be due to its early stage and that the agreement may hopefully matter in the future. The variable representing trade with the rest of the world obtained a positive sign and was significant at the 5 per cent level for both specifications. It thus appears that there might be no significant substitution effects between intra trade and trade with ROW for the sample of countries. In other words, for instance on the exports side, the positive and significant coefficient of Restx would imply that, trade agreements at least like COMESA, do not necessarily lead to an increase in exports within COMESA at the cost of exports to ROW. The time variable in both models turned out to be significant with a positive sign and validates the fact that there has been an overall growth trend in both intra exports and intra imports. Further, although to a lesser extent, GDP is also seen to have a positive effect on both intra-exports and intra imports. Thus increases in aggregate regional GDP leads to higher increase in intra trade. In both regressions, the value of R–squared is quite high (above 0.9) implying that the independent variables have effective predicting power for the dependant variables.

V. The role of COMESA exports in COMESA economic growth21

To model the growth effects of trade in the COMESA region, a classical economic growth function was extended with standard control variables such as investment, 21 The readers are guided to Rojid (2006) for a detailed explanation on the trade potential of COMESA.


education, labor, financial development22 and trade. This is consistent with works from the literature (Barro 1991; Mankiw, Romer and Weil 1992; Benhabib and Spiegel 1994; Levine and Renelt 1992; Levine et al. 2000; among others). The key attraction of growth regressions is that they provide a way of testing directly for productivity effects of various determinants. More importantly for the sake of this study, the variable interesting for us, namely exports, has been decomposed into exports of COMESA countries to other member states and into exports to non-COMESA countries. This will permit us to gauge the effect of COMESA trade on economic growth. The implied theoretical model is thus

),,,,,( iztijtititititit NONCOMESACOMESAFDLABEDUIVTfOUTPUT = (

1)

here i denotes COMESA member countries (in our analysis they are exporting

UTPUTit denotes the respective COMESA members economy’s output (measured

he econometric model and preliminary tests Recall equation 1 above and taking logs on both sides of the equation24 and denoting the lowercase variables as the natural log of the respective uppercase variable results in the following:

wcountries), j denotes COMESA importing countries, Z denotes non-COMESA importing countries and t is the time period (that is 1985-2000). Therefore COMESAijt would imply exports of one COMESA member to another COMESA member at time t. Oas the real GDP at constant price in $US), IVTit (see Delong and Summers 1990, 1994; Reinhart 1989 and more recently Sala-i-Martin 1997 and Arin 2004) is the level of investment in the country (measured as the investment ratio). EDUit is the level of literacy and quality of labor (measured by the secondary enrolment ratio). It is included following the arguments and empirical evidences of Mankiw, Romer and Weil (1992), Barro (1998) and more recently Temple (2001). Human capital can be thought of as affecting economic growth in the sense that workers with higher levels of education or skills should, ceteris paribus, be more productive and more inventive and innovative. Higher levels of human capital may also encourage capital accumulation, or may raise the rate of technological catch-up for follower countries (Temple 2001). LABit is the labor force (measured as the number of people in employment), FDit is the level of financial development (measured by the ratio of liquid assets to GDP (see Levine et al. 2000; Levine 1997). Export23 has been decomposed into COMESAijt that is the amount of exports to COMESA member states(j) by COMESA countries(i) and NONCOMESAijt which is the amount of exports to non-COMESA countries (z) by COMESA countries(i), both measured as a percentage of GDP. The data series for OUTPUT, IVT, FD were generated from the International Financial Statistic (IFS) various yearbooks, the secondary enrolment ratio from the World Development Report (various issues) and the two sets of export measures have been obtained from the World Trade Analyzer. T

22 Seetanah (2008) discusses the link between financial development and economic growth.

ee work from Dollar (1992), Sachs

untries

ge change or output elasticity.

23 The role of exports in economic development has been well documented in the literature, sand Warner (1995) and Edwards (1998). The authors supported the idea that increased trade openness raised economic growth through access for a country to the advances of technological knowledge of its trade partners, access bigger markets and encouraging the development of R&D through increasing returns to innovation and also through providing developing cowith access to investment and intermediate goods that are vital to their development processes. 24 This is for ease of interpretation whereby the coefficient can be expressed in terms of percenta


itititititititit noncomesacomesafdlabeduivtoutput εβββββββ +++++++= 6543210

(2) where β0 is the constant term, β1 , β2 , β3, β4, β5 and β6 represent the elasticity of output relative to investment , education, labor, financial development, exports to COMESA

e sample and t the time period, that is 1985-2000. Thus the number of

85-2000). The results reveal that exports to COMESA embers have not had a significant impact on the economic development of the

n the block. Rather, on the other hand, exports to non-COMESA countries

(COMESA exports to COMESA and non-COMESA)

(COMESA exports to COMESA and non-COMESA)

countries and exports to non-COMESA countries. i denotes the respective countries in 25th

observations comes to 192. Analysis We start by running the cross section regressions as a preliminary exercise (averaged over the sample period 19mcountries ihad a positive and significant link to the economic development for the COMESA bloc. As expected investment level, education, employment and financial development proved to be statistically significant at 5 per cent and also have the expected signs. Table C7: Cross section and random effects estimates

Variable Cross section estimates Random effects estimates

Constant

8.13 (3.22)***

ivt

esa

comesa

***

***

***

*

10.22

**

*

***

**

edu labt fd com non

0.31 (2.49)*** 0.41 (4.16)0.59 (2.96)0.32 (3.26)0.19 (0.83)0.68 (1.85)

(1.85)* 0.45 (1.87)* 0.73 (4.13)*0.45 (2.07)0.17 (6.91)0.11 (1.34)0.36 (2.30)

R2

Number of observatHausman Te

ions. st 2=0.411

0.61 12

0.56 192 Prob>Chi

Notes: Dependent variable output = log of gdp, 1985-2000. *significant at 10 per cent, ** significant at 5 per cent, ***significant at 1 cent. The small letters denotes variables in natural logarithmic and t eses.

26

se shortcomings. With panel data, the issue is whether to se a random effects or a fixed effects estimation approach. Accordingly, to determine

per values are in parenth

The limitations of using a single-equation OLS cross sectional regression model and pooled OLS are known (see Kennedy 2003). In the next section we use panel data techniques to overcome theu

25 These are Angola, Burundi, Comoros, Ethiopia, Kenya, Malawi, Mauritius, Rwanda, Egypt, Uganda, Zambia and Zimbabwe. 26The most serious limitations being that simple cross section may produce biased and inconsistent estimates since they may not

t take into consideration the endogeneity of some of the regressors. It ignores dynamics and throws away information (Attanasio eal. 2000) and may suffer from omitted variable bias.


which of these estimators are more appropriate to use in the present case, both a fixed effects (FE) and a random effects (RE) estimator were initially used to estimate the equation and the Hausman specification test was performed to evaluate the assumption in the random effects model. In fact the Hausman test tests the null hypothesis that the coefficients estimated by the efficient random effects estimator are the same as the ones estimated by the consistent fixed effects estimator.27 The p-value, of the Hausman test reported in Table C7, shows that the test favors the random effects approach as the null hypothesis was rejected. Note that it has also been argued that since panel data techniques are employed, the issue of non-stationarity of the variables is less serious (Garcia-Mila et al. 1996). From the random effect findings,28 it is interesting to note that the coefficient of COMESA, that is exports to COMESA member states, is positive but not significant, suggesting that it has not impacted significantly on the economic performance of the

lock. This confirms the results from the cross section estimates. This can be

tries except COMESA, US and US (xpother) for further insights and articularly to capture the effect to exports to EU and US and their importance to

bexplained by the fact that, as we mentioned above, COMESA internal trade is very minimal as a share of COMESA trade to the world. Exports to non-COMESA countries (NONCOMESA) is observed to have been very influential to the economic development of the countries in the sample and the coefficient of 0.36 suggests that a 1 per cent increase in trade to non-COMESA countries brings a 0.36 per cent increase in the output level of the country. This is principally due to the large export volume from COMESA to EU and USA under the Lomé Convention/ Cotonou Agreement and the AGOA preferential regime. Below we analyze growth due to exports to EU and US only. Investment, education, labor and financial development report signs and magnitudes of elasticity as generally predicted by the literature. Interestingly the growth of COMESA countries are principally determined by their level of investment and particularly education as judged by the high output elasticities reported. For instance a 1 per cent increase in the investment and education level of the country is associated with a 0.45 per cent and 0.73 per cent respective increase in level of output (see Table C7). As a further analysis we segregated the exports of COMESA members into exports to other COMESA countries (xpcom), exports to the US and EU (xpUSEU) and exports to the other counpCOMESA economic growth. Table C8 below shows the results from the panel estimates.

27 For a detailed treatment of the fixed and random effects model see among other Green (1997). 28 We have also tried to use 5 and 10 year average observations for the panel as it is believed that annual growth can be influenced by short term effects. The results obtained were overall similar to the ones presented in the text. However since 5 and 10 years average dramatically decreased the number of observations (we could not obtained longer period data or more countries in the sample due to data limitations), thus making the estimates less reliable, it was judged not appropriate to be included as an integral part of the discussion.


Table C8: Random effects panel estimates (COMESA exports to COMESA and US and EU)

Variable Random effects Estimates (COMESA exports to COMESA and US and EU)

Constant ivt edu lab fd xpcom xpUSEU xpother

3.34 (1.77)* 0.48 (1.99)* 0.56 (2.56)*** 0.33 (2.37)** 0.13 (2.31)* 0.07 (1.21) 0.43 (2.61)*** 0.18 (1.89)*

R2

Number of obs. Hausman Test

0.66 192 Prob>Chi2=0.643

Notes: Dependent variable gdp= log of gdp, 1985-2000. *significant at 10 per cent, ** significant at 5 per cent, ***significant at 1 per cent. The small letters denotes variables in natural logarithmic and t values are in parentheses. It is observed that COMESA exports to the EU and US have had a significant and interestingly a larger impact (with an output elasticity of 0.43) than COMESA exports to the other countries (that is excluding exports to COMESA members, US and EU on the COMESA bloc economic performance. The above results yet again validate the fact that exports to other COMESA countries have had an insignificant effect on economic progress in member states and that growth is mainly due to exports to the EU and US.


Appendix Table C-A: RCA for selected COMESA countries in 2000 at 2 digits SITC Angola Comoros Djibouti Kenya Mauritius Zambia Zimbabwe00 live animals other than in div 03 0.0 0.0 2.5 0.1 0.1 0.2 2.7 01 meat and meat prep 0.0 0.0 0.0 0.3 0.2 0.0 1.1 02 Dairy products and birds egg 0.0 0.0 0.0 0.1 0.0 2.2 2.7 03 fish 0.5 0.0 0.1 2.2 4.1 0.1 0.2 04 cereals and cereals prep 0.0 0.0 0.0 1.5 1.5 1.8 3.8 05 veg and fruits 0.0 0.0 0.1 10.8 0.1 1.7 2.0 06 sugar and sugar prep 0.0 0.0 0.0 9.0 45.3 12.3 16.9 07 coffee, tea 0.0 536.1 0.3 190.3 0.3 4.6 9.9 08 feeding stuffs for animals 0.0 0.0 0.0 0.0 1.1 2.6 1.0 09 misc. edible prod 0.0 0.0 0.3 4.8 0.0 0.0 5.9 11 beverages 0.0 0.0 0.0 0.2 0.5 0.0 0.3 12 tobacco 0.0 0.0 0.0 5.3 0.0 8.4 179.0 21 hide, skins raw 0.0 0.0 3.6 2.9 0.0 1.2 3.7 22 oil-seeds 0.0 0.0 1.4 0.3 0.0 1.6 5.3 23 crude rubber 0.0 0.0 0.0 0.0 0.0 0.0 0.0 24 cork and wood 0.1 0.0 0.0 0.0 0.0 0.3 2.4 25 pulp and waste paper 0.0 0.0 0.0 0.0 0.0 0.0 0.0 26 textile fibres 0.0 0.0 0.1 7.6 0.3 3.3 14.0 27 crude fertilizers 0.2 0.0 1.8 4.6 0.5 0.4 8.5 28 metalliferous ores 0.0 0.0 0.3 0.1 0.3 1.4 1.3 29 crude animal 0.0 0.0 0.8 16.8 0.8 4.5 3.9 32 coal, coke 0.0 0.0 0.0 0.0 0.0 0.3 7.4 33 petroleum 5.2 0.0 0.1 1.4 0.0 0.0 0.3 34 gas 0.4 0.0 0.0 0.0 0.0 0.0 0.0 35 electric current 0.0 0.0 0.0 0.0 0.0 7.9 0.1 41 animal oils 0.1 0.0 0.0 0.1 0.0 0.0 6.2 42 fixed veg fats and oils, crude 0.0 0.0 0.1 2.7 0.1 0.1 0.2 43 animal or veg fats and oils processed 0.0 0.0 0.0 4.3 0.0 0.4 0.3 51 organic chemicals 0.0 0.0 0.1 0.0 0.0 0.0 0.2 52 inorganic chemicals 0.0 0.0 0.1 0.3 0.1 0.2 0.6 53 dyeing, tanning 0.0 0.0 0.0 0.8 0.1 0.0 0.5 54 medicinal products 0.0 0.1 0.0 1.1 0.1 0.0 0.2 55 essential oils 0.0 14.5 0.0 2.7 0.1 0.2 0.5 56 fertilizers other than 272 0.0 0.0 0.0 0.0 2.1 0.5 2.8 58 plastic in primary forms 0.0 0.0 0.0 0.1 0.0 0.0 0.3 59 chemical materials n.e.s 0.0 0.0 0.0 1.1 0.0 0.1 0.6 61 leather 0.0 0.0 1.1 0.7 0.2 0.4 8.8 62 rubber manufactures 0.0 0.0 0.0 0.3 0.0 0.0 0.3 63 cork and wood manufactures 0.0 0.0 0.1 1.4 0.1 0.4 1.0 64 paper, paperboard articles 0.0 0.0 0.1 1.5 0.2 0.0 0.5 65 textile yarn 0.0 0.0 0.0 0.5 1.9 1.2 0.5 66 non-metallic minerals 1.4 0.0 0.0 1.6 0.6 2.5 0.8 67 iron and steel 0.0 0.0 0.1 0.5 0.0 0.1 6.4 68 non-ferrous metal 0.0 0.0 0.0 0.1 0.0 21.4 5.2


Angola Comoros Djibouti Kenya Mauritius Zambia Zimbabwe69 manufactures of metal 0.0 0.0 0.0 0.6 0.2 2.2 1.2 71 power generating machinery and equipment 0.0 0.0 0.0 0.0 0.0 0.2 0.1 72 machinery specialized for particular industries 0.0 0.1 0.0 0.1 0.2 0.4 0.3 73 metal working machinery 0.0 0.0 0.0 0.0 0.0 0.0 0.1 74 general industrial machinery 0.0 0.0 0.1 0.0 0.1 0.3 0.1 75 office machines 0.0 0.0 0.0 0.0 0.0 0.0 0.0 76 telecommunications equipment 0.0 0.0 0.0 0.0 0.0 0.0 0.0 77 electrical machinery 0.0 0.0 0.0 0.0 0.0 0.2 0.1 78 road vehicles 0.0 0.1 0.0 0.1 0.0 0.2 0.2 79 other transport equipment 0.0 0.0 0.0 0.0 0.0 0.1 0.1 81 prefabricated buildings 0.0 0.0 0.0 0.2 0.0 0.0 0.1 82 furniture 0.0 0.9 0.2 0.3 0.1 0.0 1.5 83 travel goods, hand bags 0.0 0.0 0.0 0.0 0.7 0.0 0.4 84 articles of apparel and clothing accessories 0.0 0.0 0.0 0.2 16.4 1.2 0.6 85 footwear 0.0 0.1 0.0 1.3 0.1 0.9 0.4 87 professional and scientific apparatus 0.0 0.0 0.0 0.0 0.2 0.0 0.0 88 photographic apparatus 0.0 0.0 0.0 0.0 1.5 0.0 0.0 89 miscellaneous manufactured articles 0.0 0.0 0.0 0.7 1.0 0.5 0.9 93 - Special transactions & commod., not class.t 0.0 1.1 0.1 0.0 0.0 0.4 0.0 94 - Animals, live, n.e.s., incl. zoo-animals 0.0 17.8 0.0 0.5 110.1 0.8 0.3 95 - Armored fighting vehicles, arms of war & ammunition 0.0 0.0 0.0 0.0 0.0 0.0 0.1 97 - Gold, non-monetary 0.0 0.0 2.4 1.3 0.1 1.9 0.9 99 - Non-identified products 9.1 0.0 316.0 0.0 0.0 0.0 0.0 Number of product lines with revealed comparative advantage (99 not counted) 2 4 4 18 7 19 25 Source: Authors calculations.


References Attanasio, O.P., L. Picci and A.E. Scorcu (2000). "Saving, Growth, and Investment: A

Macroeconomic Analysis Using a Panel of Counties", Review of Economics and Statistics, 82:182-211.

Arin, K. (2004). Fiscal Policy, Private Investment and Economic Growth: Evidence from G-7 Countries. Available at SSRN: http://ssrn.com/abstract=438785.

Barro, Robert J. (1991). "Economic growth in a cross-section of countries". The Quarterly Journal of Economics, CVI, 2, 407-443.

Barro, Robert J. (1998). Notes on Growth Accounting. NBER Working Paper No. 6654, Cambridge, MA.

Benhabib, J. and M.M. Spiegel (1994). ‘The role of human capital in economic development: evidence from aggregate cross-country data’. Journal of Monetary Economics, 34, 143-173.

DeLong, J. and L. Summers (1990). Equipment Investment and Economic Growth. NBER working paper number W3515, available at SSRN: http://ssrn.com/abstract=226830.

Delong, J. and L. Summers (1994). Equipment Investment and Economic Growth: Reply. Quarterly Journal of Economics, 109(3), 803-807.

Dollar, D. (1992). Outward-Oriented Developing Economies Really Do Grow More Rapidly: Evidence from 95 LDCs, 1976-1985. Economic Development and Cultural Change, 40(3), 523-44.

Garcia-Mila, Teresa, Therese McGuire and Rob Porter (1996). "The Effect of Public Capital in State-Level Production Functions Reconsidered", The Review of Economics and Statistics.

Green, W.H. (1997), Econometric Analysis, Prentice-Hall.

Edwards, S. (1998). Openness, productivity and growth: what do we really know? Economic Journal 108, 383– 398.

Jakobeit, Cord, Trudi Hartzenberg and Nick Charalambides (2005). Overlapping membership in COMESA, EAC, SACU and SADC. Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) GmbH. http://tanzania.fes-international.de/info-service/docs/overlapping-membership-in-comesa-eac-sacu-and-sadc.pdf.

Kennedy, P. (2003). A guide to econometrics, 5th edition, Oxford, Blackwell.

Levine, R. and D. Renelt (1992). A sensitivity analysis of cross-country growth regressions. American Economic Review, 82(4), 942-963.

Levine, R. (1997). Financial development and economic growth: views and agenda, Journal of Economic Literature, 25, 688–726.

Levine, R., N. Loayza and T. Beck (2000). “Finance and the sources of growth”, Journal of Financial Economics, 58, 261-300.

Mankiw, N.G., D. Romer and D.N. Weil (1992). A contribution to the empirics of economic growth. The Quarterly Journal of Economics, CVI, 2, 407-437.


http://ssrn.com/abstract=226830

Reinhart, C. (1989). Private Investment and Economic Growth in Developing Countries. IMF Working Paper No. 89/60, available at SSRN: http://ssrn.com/abstract=884880.

Rojid, S. (2007). “Impact of COMESA Forming a Custom Union – A CGE Analysis”- presented at the Centre for the Study of African Economies (CSEA) Conference 2007 on Economic Development in Africa, University of Oxford, March 2007.

Rojid, S. (2006). “COMESA Trade Potential: A Gravity Approach”, Applied Economic Letters, 13, 947-951.

Sachs, J. and A. Warner (1995). Natural Resource Abundance and Economic Growth. Available at SSRN: http://ssrn.com/abstract=225459.

Sala-i-Martin, X. (1997). I just ran two million regressions. American Economic Review 87 2 (1997), pp. 178–183.

Seetanah, B. (2006). ‘Financial development and economic growth: An ARDL approach for the case the small island state of Mauritius’, forthcoming in Applied Economics.

Seetanah, B. and J. Khadaroo (2008). ‘Transport and economic performance: The case of Mauritius’, forthcoming in Journal of Transport Economics and Policy.

Simplice, G. and Lynge, N. (2007). ECOWAS – Fiscal revenue implications of the prospective economic partnership, World Bank, Washington, D.C.; http://www.worldbank.org/afr/wps/wp103.pdf.

Temple, J.R.W. (2001). Generalizations that aren’t evidence on education and growth. European Economic Review, 45(4-6), 905-918.

World Bank (2004). Global Economic Prospects 2005: Trade, Regionalism, and Development. Washington, D.C.




http://www.worldbank.org/afr/wps/wp103.pdf

D. REGIONAL TRADE AGREEMENTS IN SOUTH ASIA: PROSPECTS FOR INTRA-REGIONAL TRADE COOPERATION

By Meeta Keswani Mehra, Manoj Pant, Saptarshi Basu Roy Choudhury and Amit Sadhukhan

I. Introduction and scope The report attempts an analysis of trade liberalization efforts in contributing to regional integration in the South Asian region. Most recently, these efforts have culminated in the ratification of the South Asia Free Trade Area (SAFTA) Agreement that was signed on 6th January 2004 by the South Asian Association for Regional Cooperation (SAARC) countries, namely, India, Pakistan, Sri Lanka, Bangladesh, Nepal, Bhutan, and Maldives and came into force on 1st January 2006. Afghanistan is slated to join the SAFTA in January 2008. While the Free Trade Agreement (FTA) within South Asia is of recent origin, subregional cooperation in trade and development has been in existence for over three decades. This has taken the form of instituting the SAARC, followed by the formation of the SAARC Preferential Trading Agreement (SAPTA) as well as several bilateral and preferential trading arrangements in the region. The important ones in the latter category are the India-Sri Lanka Free Trade Area, the Indo-Nepal Trade and Transit Treaty and its extensions, the India-Bhutan trade relationship, the BIMSTEC, the India-ASEAN Free Trade Area and the India-Singapore Comprehensive Economic Cooperation, all of which have contributed, in some measure, to promote intra-regional trade. The report aims to briefly trace these developments towards regional trade cooperation, identifies constraints posed in the past and the scope of intra-regional trade under the aegis of the recently implemented SAFTA in contributing towards regional integration and economic cooperation in the South Asia. The South Asian region comprises eight SAARC nations, namely India, Pakistan, Sri Lanka, Bangladesh, Nepal, Bhutan, Maldives and, recently Afghanistan as members, The region accounts for around 23 per cent of the world population but produces a mere 2.3 per cent of global GDP. In 2006 the region’s share in world trade was 1.6 per cent, although aggregate trade (exports plus import) flows accounted for a little over 30 per cent of their GDP. The striking economic performance of the South Asian region since the mid 1990s has made it emerge as one of the fastest growing regions of the world. This has led many to believe that greater economic integration and freer trade will play an ever critical role in furthering potential for economic growth and improve other development indicators such as literacy, health, access to energy and infrastructure, and such like. This part of the report begins by providing the geo-political context in which these attempts toward regional cooperation were forged and helps to understand as to why there remains vast unexploited potential for greater regional cooperation and gains from trade to be reaped. This is followed by an account of the two agreements, the SAARC Preferential Trading Arrangement (SAPTA) and the South Asia Free Trade Area (SAFTA), as well as a brief discussion of the key bilateral/ multilateral


arrangements forged within and with countries outside the region. Thereafter, the report examines the crucial economic indicators/ characteristics that help in bringing out the potential for economic integration, in particular, expansion of intra-regional trade, among the SAARC countries. The first of the quantitative analyses focuses on the estimates of trade intensity between pairs of countries in the region. This is accompanied by an aggregated regional time series analysis of the extent to which intra-regional trade flows have been determined by regional trade trends with the rest of the world, the regional GDP levels and the formation of preferential or free trade areas. An estimation of revealed comparative advantage (RCA) indices by broadly defined sectors forms the groundwork for some complementarity analysis of prospective commodity trade flows within the region. The report concludes by noting that:

1. Although there was a faster increase in intra-SAARC trade in the post-SAPTA period, there is less conclusive evidence as to whether this uptrend can be ascribed to the formation of the SAPTA. By comparison, SAFTA is of very recent origin and its impact on trade flows is yet to be realized.

2. Despite the growth in intra-regional trade over 1995-2005, the region as a whole had 95 per cent of its exports directed to the rest of the world and imported close to 96 per cent from the rest of the world. Thus, its trade ties continued to be primarily with countries outside the region.

3. The bilateral trade intensity indices and changes in it did not depict a steady trend over 1995 through 2005. The results were mixed, depicting a decline in trade intensity between some pairs of countries, while an increase in case of some others.

4. The trend in intra-SAARC exports and intra-SAARC imports was found to be positively related to SAARC’s exports and imports to the rest of the world, respectively. The PTA/ FTA dummy was also found to have a significant positive coefficient, implying that SAPTA and other bilateral trade agreements within the region have contributed to the expansion of intra-group trade at the aggregate regional level. The regression results are contrasting with those of the trade intensity analysis and need to be reconciled with more detailed country wise as well as product wise analysis, including the trade analysis of newer service-oriented sectors such as banking, insurance, tourism, telecommunications and information technology.

5. The RCA analysis of some broad product categories pointed toward an absence of intra-regional trade complementarity. In fact, it showed a high degree of competition in the export composition, thus pointing toward limited scope for regional trade integration. Interestingly, however, despite limited regional trade potential, regional integration did have a significant positive impact on intra-regional trade in South Asia.

6. On the qualitative front, realization of the full potential of the SAFTA Agreement has been circumscribed by both domestic protectionist interests and political deadlocks between countries of the region. Owing to its large membership, in terms of trade on a genuine Most-Favored-Nation (MFN) basis, liberal rules of origin and streamlined trade facilitation measures, the SAFTA Agreement has substantial scope for improvement.


II. Geo-political milieu of regional cooperation in South Asia At the time of their independence and for a few years thereafter, the intra-regional trade among three of the large economies of SAARC, namely India, Pakistan and Sri Lanka, varied between 12-19 per cent of their total trade. This was a result of the dual effect of trade protectionism in the developed countries and low trade barriers in the South Asian subcontinent (Baysan et al. 2006). In the immediate period that followed, countries of the region embarked upon a strategy of import-substitution industrialization, anti-export discrimination, export controls and taxes, massive public sector participation in production activity and tight regulation of the private sector. The political problems between Indian and Pakistan at the time of partition in 1947 had a profound impact in inducing an inward-looking trade regime. The other countries of the region also adopted a similar path of protectionist trade. Emerging as an independent nation in 1971, Bangladesh also followed an import-substitution strategy. Nepal and Bhutan, two small land-locked countries in the region, had stronger trade ties with India at that time, but had erected barriers to trade with the rest of the world. Maldives, a small island state in the Indian Ocean was the only exception in that it had a small production base and continued to rely on trade within and outside the South Asian region. With the exception of Sri Lanka, the regime of protectionism lasted for over four decades. While the developed countries were adopting outward-oriented trade policies through the 1950s and 60s and East Asia in the 1970s, the South Asian region closed its doors to international trade. The strategy of import-substitution industrialization was effective in limiting trade. In fact, the anti-trade bias impinged upon intra-region trade more than the trade with the rest of the world: official statistics point toward intra-regional trade as a share of total trade reducing from 19 per cent in 1948 to 4 per cent by the end of 1950s, and 2 per cent by 1967. It rose marginally in the early 1970s and then declined again to reach a low level of 2 per cent in 1990 (World Bank 2004). Barring Sri Lanka that had taken the initiative to liberalize its trade regime as early as in the 1970s, protectionist regimes prevailed in the region until the 1990s. India and Bangladesh embarked upon liberalizing trade policies by reducing tariffs in the early 1990s. Freer trade regimes were ushered in steadily during the 1990s and the early 2000s. Nepal and Pakistan began to open up by reducing its tariffs in 1997. Until then, Bangladesh had among the highest tariff levels in the region. Thus, most of the 1990s witnessed a change away from stringent protectionism and quasi-autarkic economic structures to a steady integration of the South Asian countries with the world economy. However, the pace of reforms was cautious, selective, and far slower than that adopted in the other Asian countries, including China. In fact, in response to the East Asian financial crisis, India and Bangladesh backed off on tariff reforms in 1997/98, only to revive it in the early 2000s. Concomitant with the general opening up to the world economy, South Asian countries have been working toward broader multilateral regional and bilateral cooperation (both within and outside the region). The formation of the SAARC in 1985 was a critical milestone covered to allow for discussions on better economic cooperation and political harmony in the region, as well as to collaborate in international forums on the issues of common interest. The SAPTA Agreement emerged under the aegis of the SAARC as a means toward promoting trade and economic integration in the region.


III. SAARC Preferential Trading Arrangement (SAPTA) The formulation of SAPTA in 1993 and its implementation in December 1995 marked the first step toward the formation of an economic union among the SAARC countries. The agreement (which terminated on 31 December 2003) had exclusive coverage of trade in goods and provided for gradual concessions on tariffs and non-tariff measures in various stages. SAPTA included provisions for special treatment of LDCs, including projects to identify export potential in specific sectors of their economy. It also permitted countries to pull out of the agreement in the face of balance of payments difficulties. However, it was somewhat open-ended, in that the member nations had the freedom to choose the pace, the list of items and the mode of trade liberalization. Thus, despite its provisions and the four rounds of negotiations for the exchange of preferences, the actual exchange under SAPTA remained rather limited. At the end of the first round of negotiations, in 1995, concessions were granted on 226 products. It took four rounds to agree upon 4'700 products (out of 6000) by the year 2000, with India leading the tally (Table D1) (Kemal 2004). Table D1: Preferences under SAPTA

Number of items Bangladesh 521 Bhutan 233 India 2554 Maldives 178 Nepal 491 Pakistan 491 Sri Lanka 199 SAARC 4667

Source: Weerakon and Wijayasiri 2003 as quoted in Kemal 2004. The persistence of high levels of protectionism, absence of meaningful offers and exclusion of several large tradable sectors from tariff reductions, lack of trust among the member nations and domestic political disturbances impeded the success of SAPTA. Moreover, the agreement was institutionally weak in handling trade-related disputes. The dispute settlement body of the agreement failed to impose time-bound or legally binding decisions on the member countries involved in the dispute. To make the transition from a preferential trade agreement to a full-fledged free trade agreement, the Inter-Governmental Expert Group was established in 1996 to chalk out a plan of action toward formation of an FTA. This transition from SAPTA to SAFTA was envisaged to be completed by 2001. However, it took ten years of deliberations to sign the SAFTA Agreement in January 2004 and another two years to activate it on 1 January 2006.

IV. South Asia Free Trade Area (SAFTA) Agreement The objectives of SAFTA indicate that the Agreement is more comprehensive in scope and coverage (USAID 2005). It aims to “strengthen intra-SAARC economic


cooperation to maximize the realization of the region’s potential for trade and the development of their people.” The Agreement encompasses not just the removal of tariff barriers but also non-tariff barriers, harmonization of standards and certifications, trade facilitation, creating fairer competition and creating institutional mechanisms for greater economic cooperation at the regional level (including communication and transport infrastructure, removal of investment restrictions, reducing exchange barriers and such like). The institutional arrangements to manage the implementation of the agreement comprise the SAFTA Ministerial Council (SMC) and the Committee of Experts (CoE). The SMC is envisaged as the highest decision-making body of SAFTA that will hold responsibility for the administration and implementation of the Agreement within its legal framework. The SMC has the Ministers of Commerce/Trade of the Contracting States as its members with the Chairperson changing every year on a rotational basis. The SMC is backed by a CoE comprising senior economists with expertise in trade matters. The CoE also acts as the dispute settlement body. SAFTA specifies a phased reduction in tariffs for each of its member countries, spanning the ten year period, 2006-2016. These reductions are envisaged to proceed in two stages: 2006-08 and 2008-16. Moreover, tariff reductions will take place at different rates for the least developed members (LDMs) namely Nepal, Bhutan and Maldives as against the non-least developed members (NLDMs). NLMDs such as India, Sri Lanka and Pakistan will reduce tariffs from the existing levels to a maximum of 20 per cent and those tariff levels which are already below this will be cut by 10 per cent per annum over 2006-08. Correspondingly, the LMDs will reduce tariffs to a maximum of 30 per cent by 2008, and those levels that are below 30 per cent will be reduced by 5 per cent annually until 2008. In the NLDMs, the subsequent 8-year period will have to attain tariff levels of 0-5 per cent for products from LDMs by the third year of this phase and, thereafter over the remaining 5 years, at a rate no less than 15 per cent per annum. The LDMs will cut down tariffs to 0-5 per cent over the period 2008-2016, at a rate no less than 10 per cent annually. To be eligible for tariff preferences, additional requirements in the form of rules of origin, sensitive lists, safeguards and balance of payments, will have to be satisfied. The rules of origin would specify how a good is being classified (as a proportion of value added) in terms of it being produced in a member country. Here, more stringent rules would tend to constrict scope for trade expansion. The agreement permits exclusion of certain goods from the grant of preferences on grounds of national security, public morals, and health or heritage value. Countries experiencing balance of payments concerns or imports posing potential threat to competitive products at home may also temporarily shelve the grant of preferences under the agreement. As regards non-tariff barriers, the most significant component addressed is the elimination of quantitative restrictions, which are not consistent with the World Trade Organization (WTO) provisions and are not on the sensitive lists. Moreover, member nations have to regularly inform on all non-tariff barriers to the SAARC Secretariat. Unlike tariff reductions, however, no specific time frame or procedure has been laid down for the removal of non-tariff barriers.


Besides more relaxed tariff reductions and quantitative restrictions, the LDMs have been extended favorable treatment, that is, the agreement provides for revenue compensation for these countries for the loss of customs revenue. It deemed necessary to establish the compensation mechanism for this before the tariff reductions were to take-off in January 2006. The implications of SAFTA to integrate the economies of the region remain to be seen. Most recently, the Declaration of the Fourteenth SAARC Summit held in New Delhi on 3-4 April 2007 stressed the need for ensuring effective market access under SAFTA through steady trade liberalization as well as catalyze cooperation in other areas, namely, trade in services and investment promotion, the inclusion of which would be critical to SAFTA’s success. The subsequent analysis reveals that intra-regional trade has grown at a faster pace over the last decade as compared to the period prior to that. However, it is hard to discern if this growth is ascribable to stronger bilateral and multilateral trade ties among region’s countries (both within and outside South Asia) or to the regional multilateral trade liberalization initiatives, such as the SAPTA or the SAFTA.

V. Bilateral initiatives The countries of the region have perceived the economic gains that could accrue from trade and hence, have continued to undertake initiatives for freer trade. In order to overcome the political deadlocks associated with the multilateral approach, countries have been forging bilateral links with countries within and outside the South Asian region. There exist several trade-related bilateral treaties between the countries of the region; a discussion on these ensues. Indo-Sri Lanka Free Trade Area An immediate implication of liberalization embarked upon by Sri Lanka in the 1970s and lasting through the 1980s and the 1990s was the economic gains from trade accruing to exporters from India. The Indo-Sri Lanka FTA was forged to address the concerns of growing trade deficit in Sri Lanka vis-à-vis India. In view of the political deadlocks impeding the progress of SAPTA, an FTA with India was perceived by the Sri Lankan government as an effective option. The agreement came into force in January 2000 and included provisions for phased tariff elimination, identification of negative lists, the rules of origin and safeguards provisions, including anti-dumping and anti-subsidies policies. While the provision of negative lists did constrain exports from one country to another in case of some commodity groups, in general, the agreement has been instrumental in boosting trade between the two nations. Studies point toward the fact that this has been in an entirely new range of products, not previously exported by Sri Lanka to India or vice versa (Baysan et al. 2005; World Bank 2004). Batra (2005) states that, consequent to this free trade accord, the Sri Lankan exports to India rose to 3.6 per cent of its aggregate exports from only 1 per cent in 1999. In terms of relative ranking, India emerged as the fifth-largest destination for Sri Lankan exports in 2002, compared to its rank in the 20s in the mid-1990s. At the same time, India is the largest


source of imports for Sri Lanka, with its share at 14 per cent of total imports into the latter. India-Nepal Treaties of Trade and Transit The trade links between India and Nepal are of historical origin. The first official trade agreement between the two countries, the Indo-Nepal Treaty of Trade was formalized in 1950. It was revamped in 1961 and subsequently in 1971 to incorporate the specific provisions on transit facilities extended by India to Nepal for the latter’s trade with third countries, as well as to seek cooperation in controlling unauthorized trade across the borders. The Treaty had to be suspended during March 1989 to May 1991 on account of a trade and transit crisis, but again revived in December 1991 following the change of governance in Nepal. In December 1996, a new Treaty of Trade was forged with the proviso for automatic renewal in every five years. It permitted more relaxed rules of origin requirements for Nepalese manufactures. Subsequently, a Treaty of Transit was signed to free up transit routed between Nepal and Calcutta (World Bank 2004). In 2002, India signed an FTA with Nepal that enabled several Indian joint ventures to be operational in Nepal, as also to expand the volume of trade and investment flows. As it stands, almost 200 Indo-Nepalese joint ventures are active in Nepal and India’s trade with Nepal accounts for as much as 40 per cent of Nepal’s aggregate trade, also entailing a positive trade balance for it (Batra 2005). Trade Agreement between India and Bhutan Traditionally, there has been an inherent recognition of free trade arrangements between India and Bhutan as per Article V of the Treaty of Perpetual Peace and Friendship signed by the two, way back in 1949. Being landlocked, Bhutan has historically had strong trade links with India. The trade relations were made explicit and formalized in 1972 with the promulgation of the Agreement on Trade and Commerce, which has been renewed periodically in 1983, 1990 and 1995. This bunch of agreements has extended its domain to include technical and financial assistance from India for the economic development and diversification of the Bhutanese economy. In the interest of industry protection, the agreement allows Bhutan to impose non-tariff barriers on commodities from India, with the proviso that these will be no stricter in impact than those applied to goods from third countries. Currently, Bhutanese trade with India is dominated by electricity exports to India from the Chukka hydroelectric project, financed and built by Indian companies. By far, this economic activity is larger than production and trade in any other agricultural or manufactured good (World Bank 2004). In 1998, Bhutanese exports to India accounted for as high as 94 per cent of its aggregate merchandise exports (UNCTAD 2006/07). Trade Agreement between Bangladesh and Bhutan On account of geographical proximity, Bangladesh is the second largest destination (after India) for Bhutanese exports. In May 2003, Bangladesh and Bhutan renewed the earlier (1980) Agreement on trade for another five years to offer tariff concessions to exploit each other’s market potentials. Besides tariff preferences, the accord covers aspects relating to trade and transit facilitation, controlling illegal cross-border trade, and cooperation in transport infrastructure. Despite the preferential agreement, Bangladesh’s import duty rates on exportable goods from Bhutan (namely apples and oranges) continue to be excessive. Consequently, although the two-way trade volume


is quite insignificant, accounting for only 4 per cent of Bhutan’s total exports in 1998 and a negligible fraction of Bangladesh’ aggregate export value over 2000-04, Bhutan enjoys a trade surplus with Bangladesh (UNCTAD 2006/07). Indo-Bangladesh Trade and Transit Agreement Traditionally, India and Bangladesh have shared an intensive trade relationship with each other, wherein, Bangladesh was the largest export market for Indian commodities in the region. After its independence, in 1972, Bangladesh forged a trade and transit agreement with India that provided for the use of waterways, railways and roadways for commerce between the two countries. This agreement, which was renewed on a yearly basis, was revoked during 1977-80. But subsequently, it culminated in a new agreement being signed in 1980, which has been renewed periodically. While facilitating existing trade flows, these agreements remained limited in scope and were beset with constraints on cross-border movement of cargo by road, illegal movement of goods, and lack of meaningful preferential trade concessions (World Bank 2004). While the share of Bangladesh’s exports to India in its total exports rose marginally from 0.72 per cent in 1995 to 1.26 per cent in 2004, the share of India’ exports to Bangladesh in the former’s aggregate exports fell steeply from 3.4 per cent to 1.7 per cent during 1995-2005 (UNCTAD 2006/07). India-Thailand Free Trade Area India and Thailand forged an FTA in 2003 for establishing a free trade area that would cover goods, services and investment over the next ten years. The Indo-Thai FTA includes in its domain as many as 84 items and is quite comprehensive in approach in that, in the first phase itself, it includes services, investment, economic cooperation and goods like food items, tourism, auto parts, and electronic goods. As per the agreement, negotiations on goods were to commence in January 2004 and concluded in March 2005. The FTA for zero duty imports is envisaged to be effective by 2010. India-Singapore Comprehensive Economic Cooperation A comprehensive agreement for economic cooperation between India and Singapore was signed in April 2003, which came into force in August 2005. The aim of this comprehensive agreement is to expand the individual partner country’s domestic market, through economic integration, and achieve growth and development through the enhancement of trade and investment flows. Since then, bilateral trade and investments between India and Singapore have grown steadily to reach unprecedented levels. In 2006, India became Singapore’s twelfth largest trading partner with bilateral trade between them valued at almost Singapore $20 billion. This amounted to an increase of around 20 per cent over 2005, making India one of the Singapore’s fastest-growing trade partners. The agreement also envisages to promote regional economic integration and with other countries in the South East Asian region, in particular the ASEAN.

VI. Multilateral initiatives outside the region Bangkok Agreement Instituted in 1975, the Bangkok Agreement is the oldest preferential trade agreement among the Asian developing countries. Formally called the Agreement on Trade


Negotiations among Developing Member Countries of the Economic and Social Commission for Asia and the Pacific (ESCAP), the Bangkok Agreement was originally signed by Bangladesh, India, the Lao People’s Democratic Republic, the Republic of Korea and Sri Lanka. The agreement covers concessions on tariffs, reduction of non-tariff barriers, according MFN status to member countries, rules of origin and preparation of national lists for exchange of preferential concessions. Despite some expansion in scope and coverage, there has been no appreciable impact on trade flows among the member countries under the Bangkok Agreement. This is because, first, the number and level of concessions allowed by countries were not sufficient to boost the trade flows markedly, and second, over the 1990s, both unilateral liberalization efforts as well as multilateral initiatives such as the SAPTA had begun to play a more important role. However, China’s accession to the agreement in 2001 was perceived as a significant milestone covered in terms of revitalizing it, but in view of the minimal grant of concessions by China therein, it seems that the agreement would largely maintain a symbolic or political stature (World Bank 2004). BIMSTEC The Bay of Bengal Initiative for Multi-Sectoral Technical and Economic Cooperation (BIMSTEC) was launched in December 1997 and has membership of Bangladesh, India, Myanmar, Sri Lanka, Thailand, Bhutan, and Nepal. At a ministerial meeting held in February 2004, the BIMSTEC member countries concluded an FTA Framework Agreement to encourage goods trade and investment among themselves, as well as to attract outsiders to trade with and invest in BIMSTEC. All the members, except Bangladesh (due to its domestic procedures), became signatories to the framework agreement in the 6th Ministerial Meeting. Bangladesh joined the framework agreement later, in June 2004. The Trade Negotiating Committee (TNC) was setup and had its first meeting in Bangkok in September 2004. As stated in the adopted Terms of Reference, Thailand would be the permanent chair of the TNC although the host country is proposed to be rotated. TNC’s negotiation areas cover trade in goods and services, investment, economic cooperation, trade facilitations and technical assistance for LDCs in the domain of BIMSTEC. It was agreed that once the negotiations on trade in goods are completed, the TNC would proceed with concessions on trade in services and investment. India-ASEAN Framework Agreement on Comprehensive Economic Cooperation India had entered into an FTA Framework Agreement with the ASEAN in October 2003. As per the Indo-ASEAN pact, the negotiations on commodities trade were scheduled to conclude by June 2005 and that for services and investments to commence at the beginning of 2005 and get completed in 2007. A TNC was constituted to frame the rules of origin and work out the modalities for tariff reduction to make the FTA effective. India has agreed to grant a special and differential treatment to the ASEAN group and align its peak tariff levels accordingly. Through January 2006 to December 2011, India will reduce its tariff levels for Brunei, Cambodia, Laos, Indonesia, Malaysia, Myanmar, Singapore, Thailand and Vietnam, in reciprocity to tariff concessions from Brunei, Indonesia, Malaysia, Singapore and Thailand. The new ASEAN members like Cambodia, Laos, Myanmar and Vietnam (CLMV) will do so by December 2016. Philippines, which has expressed its reservations to the FTA has agreed to eliminate


its tariffs on a reciprocal basis for India by 2016. India will unilaterally extend concessions on 11 tariff lines to CLMV. The volume of trade and FDI flows between ASEAN and India has started expanding recently, particularly with Malaysia and Singapore.

VII. Broad economic and trade characteristics of South Asia A broad discussion of the economic features and trade characteristics of the countries of the SAARC region provides the context for the analysis of scope for future regional trade integration.

VII.1 Economic characteristics of the SAARC region The aggregate GDP in PPP US$ of the SAARC region has increased continuously over the period 1984-2007, registering a compound annual rate of growth of 8.4 per cent over this period (Figure D1). This growth performance implies that it is one of the fastest growing regions of the world. Interestingly, the compound growth rate over the period 1984-1994 was 8.3 per cent, whereas the same variable, over the period 1995-2005, grew at 8 per cent annually. So clearly, in the post-SAPTA period, there was no spurt in growth as far as aggregate regional GDP is concerned. Figure D1: SAARC’s GDP (billion US$ in PPP terms)

0100020003000400050006000

YEAR19

8519

8719

8919

9119

9319

9519

9719

9920

0120

0320

05

Source: IMF 2007. Again, SAARC’s share in aggregate world GDP did not increase significantly. In the year 1984, SAARC’s GDP (in PPP US$) was 4.8 per cent of the aggregate world GDP; it rose to 6.1 per cent in the year 1995, and in 2007 it stood at 7.8 per cent of the world GDP (Figure D2).


Figure D2: Share of SAARC’s GDP in world GDP (per cent)

0123456789

YE

AR

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

Source: IMF 2007.

VII.2 Trade characteristics of the SAARC region A look at the trend in trade flows within the SAARC region helps capture the effectiveness of SAPTA/ SAFTA or other bilateral arrangements as the trade liberalization efforts have unfolded at the regional level. Clearly, as indicated in Table D2, the SAARC did seem to witness an upswing of intra-group trade in the post-SAPTA period. What remains to be shown is whether this uptrend could be ascribed to the formation of multilateral preferential/ free trade areas or due to overall trade liberalization embarked upon by the countries of the region in the 1990s. Table D2: Intra-group trade of the SAARC region (US$ million)

Year 1984 1989 1994 1999 2004 2005 SAARC exports 614.95 862.25 1'433.54 2'180.00 5'705.84 7'062.04

Source: UNCTAD 2006/07. At the same time, what is interesting is to note that the growth of intra-group exports was quite fluctuating over the period, even becoming negative in some years, during the period of analysis (Figure D3).


Figure D3: Year to year growth of intra-group exports of SAARC (per cent)

-0,20

-0,10

0,00

0,10

0,20

0,30

0,40

0,50

YEAR19

8519

8719

8919

9119

9319

9519

9719

9920

0120

03

Source: UNCTAD 2006/07. In quantitative terms, intra-regional trade depicted a steadily growing trend in recent years. The compound growth rate of intra-SAARC group trade was close to 8 per cent over the period 1984-1994, whereas the same variable grew around 12 per cent over the period 1995-2005. So, we can observe a faster pace of growth of intra- group trade in the post-SAPTA time period. However, a look at the ratio of SAARC’s intra-group trade to its aggregate GDP is more revealing. It shows that over the entire period of analysis, namely 1984-2005, this share remained below 1 per cent, implying large unexploited potential for intra-regional trade. In the pre-integration period (1984-1994) the average intra-regional trade of SAARC was 0.24 per cent of its GDP, whereas the same variable rose to a mere 0.50 per cent in the post integration (1995-2005) period. Thus, preferential trading agreements made only a slight contribution to increase the share of intra-group trade in aggregate regional GDP (Figure D4). Figure D4: Proportion of SAARC’s intra-group trade to its aggregate GDP (per cent)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

YEAR19

8519

8719

8919

9119

9319

9519

9719

9920

0120

03

Source: UNCTAD 2006/07. Next, if one analyzes the proportion of SAARC’s intra-regional exports to its exports to the rest of the world (ROW), it is evident that it rose from an average of around 4 per cent over the period 1984-1994 and around 5 per cent over the period 1995-2005.


So, it is clear that for the region as a whole approximately 95 per cent of aggregate exports are directed towards ROW (Figure D5). Figure D5: Proportion of SAARC’s intra-regional exports to its exports to ROW (per cent)

0

1

2

3

4

5

6

7

YEAR198

519

87198

919

9119

9319

95199

719

9920

0120

03

Source: UNCTAD 2006/07. Correspondingly, the proportion of intra-group imports in total imports was an average of 2.4 per cent during 1984-1994 and rose to 4.1 per cent over the period 1995-2005. This implies that a bulk of imports, around 96 per cent, was sourced from outside the region (Figure D6). Figure D6: Proportion of SAARC’s intra-regional imports to its imports from ROW (per cent)

0

1

2

3

4

5

6

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

year

Source: UNCTAD 2006/07. A look at the commodity composition of SAARC’s exports to and imports from ROW as in 2005 shows that manufactured goods exports (especially in other manufactured goods category) are most important (at 72 per cent), followed by exports of textile fibers, yarn, fabrics and clothing (at 29 per cent) and primary commodities (around 27 per cent). Across countries of SAARC, manufactured goods exports predominate in all countries, except Maldives, while Bangladesh, Pakistan, Nepal and Sri Lanka have large shares of textile exports. As far as imports are concerned, the product group that again leads the tally is manufactured goods (at 52 per cent) of which a bulk is


contributed by machinery and transport equipment and other manufactures. This is followed by primary products and fuel imports, at shares of 42 per cent and 30 per cent respectively. Among the SAARC countries, the relatively larger importers of manufactures are Bhutan, followed by Sri Lanka and Bangladesh and India leads in fuel import dependence (Table D3). Table D3: Aggregate commodity composition of SAARC’s trade with the world in 2005 Product Export

share Import share

Aggregate trade share

Primary commodities, including fuels (SITC 0 + 1 + 2 + 3 + 4 + 68)

27% 42% 36%

All food items (SITC 0 + 1 + 22 + 4) 10% 5% 7% Agricultural raw materials (SITC 2 - 22 - 27 - 28) 2% 2% 2% Ores and metal (SITC 27 + 28 + 68) 6% 4% 5% Fuels (SITC 3) 9% 30% 21% Non-ferrous metals (SITC 68) 2% 2% 2% Manufactured goods (SITC 5 to 8 less 68) 72% 52% 60% Chemical products (SITC 5) 10% 10% 10% Machinery and transport equipment (SITC 7) 9% 23% 18% Other manufactured goods (SITC 6 + 8 less 68) 54% 18% 32% Iron and steel (SITC 67) 4% 3% 4% Textile fibers, yarn, fabrics and clothing (SITC 26 + 65 + 84)

29% 4% 14%

Source: Estimated from data from UNCTAD 2006/07. Further, a look at SAARC’s Intra-group trade by broad products categories reveals that food, fuels, chemical products, other manufactured goods and textiles fibers, yarn, fabrics and clothing are dominant, with their shares in intra group exports as in 2005 at 21 per cent, 18 per cent, 12 per cent, 29 per cent and 17 per cent respectively (Table D4). Their dominance prevails even in case of intra-group imports. Over the ten year period, 1995-2005, the share of primary commodities increased substantially from 32-35 per cent to 47-48 per cent while that of manufactured goods declined (but less significantly) from around 59 per cent to 52-53 per cent. The share of textile fiber, yarn, fabrics and clothing also depicted a decline in case of both intra-exports and imports.


Table D4: Commodity composition of intra-group trade for the SAARC region in 2005 Product category Exports Imports 1995 2005 1995 2005 Primary commodities, including fuels (SITC 0 + 1 + 2 + 3 + 4 + 68)

39% 48% 32% 47%

All food items (SITC 0 + 1 + 22 + 4) 30% 21% 23% 25% Agricultural raw materials (SITC 2 - 22 - 27 - 28) 4% 4% 5% 4% Ores and metal (SITC 27 + 28 + 68) 2% 6% 3% 8% Fuels (SITC 3) 2% 18% 2% 10% Non-ferrous metals (SITC 68) 1% 5% 1% 5% Manufactured goods (SITC 5 to 8 less 68) 59% 52% 59% 53% Chemical products (SITC 5) 10% 12% 14% 14% Machinery and transport equipment (SITC 7) 14% 11% 13% 13% Other manufactured goods (SITC 6 + 8 less 68) 35% 29% 32% 26% Iron and steel (SITC 67) 5% 6% 4% 5% Textile fibers, yarn, fabrics and clothing (SITC 26 + 65 + 84) 23% 17% 22% 14%

Source: Estimated from data from UNCTAD 2006/07.

VII.3 Assessing the scope for intra-regional trade

a) Trade Intensity Index (TII) The analysis of trade intensity aims to bring out whether a country, given the relative size of its own export market and the relative size of its partner’s import market, does export to that partner as much as is expected. The formula for TII is given as: Iij= (Xij / Xi) / [Mj / (Mw-Mi)], where Iij, the intensity of country i’s exports to country j, is defined as the share of country j in country i’s total exports (Xij / Xi) as a proportion of the share of j’s imports, Mj, in total world imports, net of i’s imports from the rest of the world (Mw – Mi). Intuitively, the TII denotes the export intensity of a country vis-à-vis the partner country, normalized by the aggregate import intensity of the partner country with the rest of the world. In general, trading partners with TIIs greater than unity are said to have an ‘intensive’ trade relationship, namely, that the countries trade more than would be expected, given the relative size of the market for imports. The trade intensity analysis for the countries of the SAARC region helps bring the degree of trade intensity and the trend in it over the years 1994/95 and 2004/2005. The data were taken from the merchandise trade statistics of UNCTAD (UNCTAD 2006/07) and indices have been estimated based on trade flows expressed in current $US.


Table D5: Trade Intensity Indices for intra SAARC trade, 1994/95 - 2004/05 Partner Bangladesh India Nepal Pakistan Sri Lanka

Reporter/ Exporter ↓

1994 or

1995

2004 or

2005

1994 or

1995

2004 or

2005

1994 or

1995

2004 or

2005

1994 or

1995

2004 or

2005

1994 or

1995

2004 or

2005

Bangladesh 1.1 1.2 1.2 0.4 4.3 2.8 3.8 1.4

Bhutan 58.2 35.3* 152.1 122.9* 1.6 13.2* India 27.3 12.8 20.3 49.3 1.1 2.9 13.6 24.9 Maldives 0.2 0.5 132.3 141.6 Nepal 1.1 7.4** 19.8 55.7** Pakistan 15.5 11.2 0.7 1.6 1.5 1.3 7.1 11.6

Sri Lanka 2.8 1.9 1.2 6.8 0.2 0.3 6.6 2.9 Source: Estimated from data from UNCTAD 2006/07. Notes: * Figures refer to year 1998. ** Figures pertain to 2003. From the indices compiled in Table D5, it is clear that the intensity of exports within the region generally exceeded unity, with the exception of Maldivian exports to India and Sri Lankan exports to Nepal. In relative terms, however, Bangladesh’s export intensity with Nepal and India remains low. In fact, even with its other trading partners, namely Pakistan and Sri Lanka, with whom Bangladesh’s exports depict higher trade intensity, there has been a distinct decline from the year 1994/95 to 2004/05. Pakistan’s trade intensity with India and Nepal are also relatively low. It shows a marginal increase with respect to India and a decline with Nepal over the period under consideration. Bhutan’s trade intensity with both Bangladesh and India has been very high but showed a decline over 1994/95 and 2004/05. Moreover, India’s trading intensity with Bangladesh, Pakistan’s intensity with Bangladesh and Sri Lanka’s intensity with both Bangladesh and Pakistan depicted also registered a fall over the concerned period. Thus, the formation of SAPTA did not contribute to raising the tendency of trade amongst the SAARC countries. The interesting exceptions were strong and growing trade links between India and Sri Lanka and vice versa, Bhutan’s with Nepal, Nepal’s with Bangladesh and India, and Pakistan’s with Sri Lanka, where greater export integration was observable in the post-SAPTA period. Being landlocked, both Bhutan and Nepal have traditionally had strong trade ties, reflective in high trade intensities with respect to India. On the aggregate, the trade intensity indices and changes therein over time from the pre- to the post-SAPTA period did not depict a consistent positive trend over time (Pitigala 2005). In fact, Pitigala (2005) goes on to show that the intensity indices, when corrected for geographical proximity, fall below what would be expected for some of the large economies of the region, namely, Pakistan, Bangladesh and Sri Lanka’s trade with India. The evidence of uneven trade intensity changes implies that neither the volume of trade nor forging of multilateral preferential trading arrangements could be important determinants of growth in intra-regional trade. Pitigala (2005) also demonstrates that the SAARC countries have shown an increasing tendency to trade with partners elsewhere in the world, on account of cultural, ethnic and religious affinity. For example, Pakistan demonstrated the most robust evidence in this regard, with its growing trade with geographically far placed countries such as the Islamic countries of Africa and the Middle East as compared to the countries of the SAARC region. The same is true for India’s growing trade ties with selected countries of the ASEAN region.


Since the examination of bilateral trade intensities provides a mixed picture as regards trends in regional integration in the post-SAPTA period, we carry out a more aggregated investigation of regional trade flows over the past two and a half decades. This is based on a simple regression analysis.

b) Partial regression analysis for determinants of intra-regional trade

The partial regression analysis aims to assess the extent to which the formation of SAPTA and other bilateral trade agreements have impacted intra-SAARC trade at the aggregate level. As discussed earlier, the region, in general, embarked on trade liberalization in the early 1990s. SAPTA came into force in December 1995 and the subsequent transition to a free trade agreement happened in January 2006 in the move to SAFTA. To study the effects of these trading agreements on intra-regional trade, we carry a regression analysis, separately for both aggregate export and import flows within the region, through the 25-year period, 1980-2005. Thus, we have a more or less commensurate time period for both before as well as after the SAFTA agreement. If we look at the time series of intra-group exports and intra-group imports for the region over this period, it depicts a somewhat fluctuating tendency in the initial years, but grows steadily thereafter, especially in the 1990s with substantially high rate of growth in the last few years of our analysis. The exports from the SAARC region to ROW (total exports minus intra-group exports) show an even steeper rate of growth over the period than that of the SAARC. A similar trend for ROW imports (total imports minus intra imports) is observable (see Figures D7 and D8 below). The growth in intra exports and intra imports can be ascribed to various factors such as trade agreements among the SAARC countries themselves as well as those forged with ROW, growth in the size of the economies of these countries and that of the world economy as a whole.


Figure D7: Intra-group exports and SAARC’s exports to ROW (current US$ million)

0

20000

40000

60000

80000

100000

120000

140000

1980

1983

1986

1989

1992

1995

1998

2001

2004

Rest of the WorldExportIntra Export

Source: UNCTAD 2006/07. Figure D8: Intra-group imports and SAARC’s imports from ROW (current US$ million)

0

20000

40000

60000

80000

100000

120000

140000

160000

180000

200000

1980

1983

1986

1989

1992

1995

1998

2001

2004

Rest of the WorldImportIntra Import

Source: UNCTAD 2006/07. We try to probe into the determinants of intra-group trade. One aspect that needs a closer scrutiny is whether there exists a degree of substitution between intra-group trade and the region’s trade with ROW. The other is the extent to which growth in intra-group trade could be explained by the size of the economy or the underlying time trend. Finally, the effect of the institutional change is captured by a dummy variable, which is a proxy for the formation of SAPTA and strengthening of bilateral trade ties amongst member countries of the SAARC. We specify two models in the partial analysis, one representing intra-regional exports of the SAARC countries and the second representing intra-regional imports. To explain these two variables through the last 25 years, we introduce aggregate exports of the SAARC countries to ROW as


an explanatory variable in the first model and aggregate imports from ROW in the second model. These two explanatory variables capture trade agreements other than intra-regional trade agreements and help identify the substitution effect between intra-group trade and trade with ROW. We also introduce aggregate GDP of the SAARC countries as another explanatory variable in both the export and import equations. This is because the growth of GDP reflects the expansion in the size of economic activity and is postulated to have an important effect on intra-regional trade. Time is included as the third explanatory variable in the model, and it captures the growing trend in intra-group trade. Finally, we introduce a dummy variable in the model which assumes a value zero from 1980-1994 and one from 1995 onwards. This binary variable intends to capture the effect of the various regional trade agreements (particularly SAPTA and SAFTA) on intra-SAARC trade.

c) The model The analysis spans the period 1980-2005. The estimation relies on time series data, wherein for each year under consideration we capture the regional scenario by taking aggregate values of the variables for the eight SAARC countries put together. The data are obtained from UNCTAD’s Handbook of Statistics (UNCTAD 2006/07). The following two equations summarize the two models:

0 1 2 3 4Re 1Intrax stx gdp t SAPTA uβ β β β β= + + + + + (1)

0 1 2 3 4Re 2Intram stm gdp t SAPTA uα α α α α= + + + + + (2) where Intrax denotes SAARC’s intra exports, Restx refers to SAARC’s exports to ROW, t is the time trend, and SAPTA is the dummy variable that takes a value of 1 for the period 1995-2005 and 0 otherwise. Intram is intra-regional imports and Restm denotes imports of SAARC from ROW. The aggregate trade flows and GDP are measured in current million dollar and the time trend is scaled with 1980 taking a value of 1, 1981 taking a value of 2 and so on. iα and iβ are unknown parameters to be estimated. and are the error terms, assumed to be distributed with zero mean and a constant variance. This completes the specification of the model.

1u 2u

From the data plotted in Figures D7 and D8, the signs of 1β and 1α are expected to be positive. What is also predictable is that GDP will be positively related to intra-group exports due to the predominance of the scale (of economic activity) effect, while its impact on the intra-group imports may be positive or negative. It may be positive as a higher GDP implies higher consumption and imports. On the other hand, it may be negative since a large scale of production within the region may, by itself, reduce the need to rely on imports from other countries in the region. Intuitively, the time trend and the SAPTA dummy are expected to impact intra-group trade in a positive fashion. Intuitively, these are the probable findings of our estimation exercise.


d) Estimation results We first estimated the two models with the absolute values of the variables. Subsequently, we also attempted a log linear transformation of the dependent variable and select independent variables, namely, the rest of the world exports and imports and GDP. The estimation involving log values gives us much better results. We, therefore, present here the results of the second estimation exercise. Equation (1) is estimated by regressing log(Intrax) on log(Restx), log(gdp), time and the dummy, and similarly equation (2) is estimated by regressing log(Intram) on log(Restm), log(gdp), time and the dummy variable. Tables 6 and 7 summarize the results of the estimation using the OLS method. Table D6: Results of multivariate regression for intra-regional exports Dependent variable: log(Intrax) Method: Least Squares Sample: Time series from 1980-2005 Included observations: 26 Variable Coefficient Std. Error t-Statistic Prob. Constant -4.012579 1.199617 -3.344883 0.0031 Log(Restx) 0.952823 0.179324 5.313410 0.0000 Log(gdp) 0.543024 0.327965 1.655740 0.1126 Time -0.220178 0.057933 -3.800573 0.0010 Dummy variable 0.062638 0.037047 1.690785 0.1057 R-squared 0.985399 Mean dependent var 3.135289Adjusted R-squared 0.982618 S.D. dependent var 0.351733S.E. of regression 0.046372 Akaike info criterion -3.133185Sum squared resid 0.045158 Schwarz criterion -2.891243Log likelihood 45.73140 F-statistic 354.3245Durbin-Watson stat 1.609263 Prob(F-statistic) 0.000000 Table D7: Results of multivariate regression for intra-regional imports Dependent variable: log(Intram) Method: Least Squares Sample: Time series from 1980-2005 Included observations: 26 Variable Coefficient Std. Error t-Statistic Prob. Constant 2.294030 2.537075 0.904203 0.3761 Log(Restm) 1.908277 0.467199 4.084501 0.0005 Log(gdp) -1.483981 0.816548 -1.817382 0.0835 Time 0.257076 0.114829 2.238764 0.0361 Dummy variable 0.140803 0.061886 2.275186 0.0335 R-squared 0.970692 Mean dependent var 3.127460Adjusted R-squared 0.965109 S.D. dependent var 0.386653S.E. of regression 0.072223 Akaike info criterion -2.247065Sum squared resid 0.109540 Schwarz criterion -2.005124Log likelihood 34.21185 F-statistic 173.8796Durbin-Watson stat 1.510671 Prob(F-statistic) 0.000000


As can be seen from Table D6, the estimation yields a high value of R2 (of 0.98) and a significant F-statistic. All the explanatory variables have got significant coefficients except GDP and the dummy which are significant only at around 10 per cent level of significance. Restx is highly significant and got a positive sign, conforming to our expectation that there is no significant substitution between intra-group exports and exports to ROW. On the aggregate, the two grow together in a complementary fashion. Contrastingly, the time variable depicts a negative relationship to intra-trade. The dummy coefficient is also positive implying that the SAPTA and other bilateral trade agreements have contributed to intra-group exports, at least at the aggregate regional level. We might need to reconcile this with the results of the trade intensity analysis (presented above) that throws up mixed results for pairs of countries as regards changes in bilateral trade intensities in the pre- and post-SAPTA period. What is clear from the regression analysis is that, at least for the SAARC countries as a group, SAPTA and other trade arrangements have had a positive effect on intra-regional export. Further, GDP has also got a positive effect on intra-exports, albeit it is a less significant determinant of it. Interestingly, the time variable has a counter-intuitive sign, although the size of the coefficient is smaller relative to Restx and GDP. Though the volume of intra exports has increased over the years, the time variable does not seem to capture this. To verify this, we dropped the time trend variable and estimated the model again. This neither altered the signs of the other three independent variables, nor did Restx or the SAPTA dummy lose their significance. Table D7 compiles the estimation results for intra-group imports and indicates that the results are highly significant here as well. The value of R2 is 0.96, and the F-value is significant. The effect of Restm is significantly positive and so are the dummy variable and time. This clearly explains that, similar to intra-group exports, the trade agreements have made a positive contribution to intra-group imports. The time variable shows a positive sign implying an overall growth trend in intra imports. In terms of the size of the coefficients, Restm explains most of the direct (or positive) variation in Intram, followed by time and the SAPTA dummy. Interestingly, the result also implies that as the aggregate regional GDP increases, the member countries rely less on imports from each other, presumably because now their own production base is larger and (maybe) more diversified. The size of the coefficient of log(gdp) is also large and significant. We also check the univariate properties of the data series by running the Augmented Dickey-Fuller (ADF) test and the Phillips-Peron (PP) test. Both of these show that selected variables, namely, log(Intrax), log(Intram), log(Restx), log(Restm) and log(GDP) are integrated of order 1, that is, all these series are stationary in the first difference. Table D8 summarizes the results of these unit root tests.


Table D8: Unit root tests Level 1st Difference Variables (in natural log) ADF PP Critical

Value (1%)

ADF PP Critical Value (1%)

Intrax 1.6186 2.4343 -3.7241 -4.4845 -4.4768 -3.7378 Intram 0.7364 0.9076 -3.7241 -5.2452 -5.2368 -3.7378 Restx 1.8766 1.4972 -3.7241 -3.5771 -3.5771 -3.7378 Restm 2.5418 2.5418 -3.7241 -3.8302 -3.8040 -3.7378 GDP 1.4442 1.1706 -3.7241 -2.9006 -3.5527 -3.7378

e) Revealed comparative advantage (RCA) So far the analysis has focused on either bilateral or regional trade analysis for all the commodities put together. As against this, the RCA indices for a group of countries helps to identify whether there exists a high or low scope of intra-group trade in a particular product category. The RCA for the jth country for the ith product can be defined as follows: RCAij = (Xij / Xiw) / (∑Xij / ∑Xiw), where, commodity i = {1, 2, 3,…, n.}, Xij = exports of ith product from country j, ∑Xij = total exports from country j, Xiw = exports of ith product from the world, ∑Xiw = total exports from the world. Clearly, if the RCAs for a commodity for all the countries are close to each other or 1, there is less scope for intra-regional trade in that commodity for the group of countries. Correspondingly, if a country’s RCA is greater than 1 and deviates from other countries’ RCA for a commodity, it is indicative of a high scope for intra-group trade in that commodity. A look at the RCA indices for the year 2005 for the countries of SAARC as compiled in Table D9, we find that for primary commodities, Bhutan (at 1.28), India (at 1.18), and Sri Lanka (at 1.13) depict comparative advantages. Within this category, the RCA indices for all food items point to an interesting result that all the indices exceed 1, with those for Nepal, Sri Lanka and Bhutan relatively higher than others. Thus, these three countries demonstrate greater comparative advantage in food items relative to the other SAARC countries. In case of agricultural raw material, Bangladesh (at 1.23) and Sri Lanka (at 1.31) have comparative advantages. In case of ores and metals, Bhutan (at 5.24), India (at 2.18), and Nepal (with 1.29) have comparative advantages, and for non-ferrous metals, Bhutan (with 7.9), Sri Lanka (with 1.5) and Nepal (at 1.6) have comparative advantages. Bhutan leads the tally in both of these commodities. Given their high import dependence, none of the countries of the region depict a comparative advantage in fuels. In case of manufactured goods, the RCA values for all the countries are close to unity, which implies less potential for intra-regional trade. Within manufacturing, in case of chemical products, Bhutan (at 1.5) and India (at 1.09) reveal a comparative advantage. None of the countries demonstrate a comparative advantage in machinery and transport equipment, since the RCAs are low or even close to zero. For other


manufactured goods, for almost all the countries there is evidence of comparative advantage, again implying a low potential for intra-regional trade. As regards textile fibres, yarn, fabrics and clothing, Bangladesh (with 16.53), Bhutan (with 2.0), India (with 3.47), Nepal (with 9.93), Sri Lanka (at 9.64), and Pakistan (at 13.16) have comparative advantages vis-à-vis the rest of the world. Table D9: RCA indices for SAARC countries for broad commodity groups in 2005

Product Bangladesh Bhutan India Nepal Pakistan Sri Lanka Primary commodities, including fuels 0.422 1.288 1.187 1.056 0.738 1.135 All food items 1.228 2.036 1.379 3.196 1.869 3.450 Agricultural raw materials 1.239 0.404 0.985 0.688 0.950 1.311 Ores and metal 0.026 5.242 2.182 1.296 0.123 1.092 Fuels 0.033 0.042 0.869 0.000 0.317 0.001 Non-ferrous metals 0.007 7.900 0.963 1.607 0.025 1.596 Manufactured goods 1.250 0.953 0.974 1.032 1.141 0.978 Chemical products 0.030 1.567 1.098 0.638 0.286 0.126 Machinery and transport equipment 0.010 0.033 0.287 0.011 0.048 0.118 Other manufactured goods 3.817 2.168 2.031 2.869 3.303 2.762 Iron and steel 0.124 9.821 1.574 1.644 0.137 0.021 Textile fibers, yarn, fabrics and clothing 16.534 2.017 3.472 9.934 13.169 9.641

Source: Estimated based on data from UNCTAD, 2006/07. Theoretically, the higher the divergence in factor endowments, demonstrated in divergent comparative advantage, the greater is the prospect for regional trade integration. Evidently, for the member countries of SAARC, there is lack of complementarity as indicated by similar RCA indices. This points toward a high degree of competition in export structures, albeit at the aggregated commodity level. This is also corroborated by Pitigala (2005) for a much finer commodity classification. VIII. Problems and obstacles of SAPTA/ SAFTA The key problems envisaged in the implementation of SAFTA could be ascribed to various factors in the economic, political, administrative, logistical and external domains. Economic factors The moot question that arises out of an agreement is whether an FTA in the region provides sufficient economic gains to its members or not (Baysan et al. 2006). As for SAFTA, the following comments are in order:

• Although India, Pakistan, Bangladesh and Sri Lanka are largely populated countries, they are relatively small compared to other developed economies in the world in terms of GDP and trade flows. The economic size of the region is less than one twentieth of the world in terms of GDP and, after dropping the values for India, this is further reduced to 0.4 per cent. This involves the risk


of trade diversion since there is less possibility of the most efficient suppliers belonging to the region itself (Baysan et al. 2006).

• As brought out earlier, the SAARC economies’ export structures are highly competitive. Pitigala (2005) in his study, found low complementarity and high degree of competition in export structures of the SAARC countries. Our report also corroborates this through the RCA analysis. According to theory, the larger the difference in factor endowments, the higher will be the probability of success of an RTA. However, the absence of diversity in the export structure of the SAARC economies poses a serious question on this issue.

• More recently, prospects of trade in services yielding efficiency gains under the SAFTA agreement are being discussed. According to Pohit (2004), since the services sector requires simultaneous presence of the factors of production, trade in services facilitates the movement of these factors between countries. Thus, trade in services like tourism, investment, insurance, energy, infrastructure development, information technology and communication can prove very important for the SAARC nations to gain from SAFTA, which otherwise can be limited due to similarity of export structures. In the long run, the scope of SAFTA needs to be broadened to augment the gains from it while, in the short run, the agreement needs to have a clearer and shorter schedule for implementation (Batra 2005).

• Services trade liberalization could be beneficial for all the countries providing two-way gains, or else, countries except India might be viewed to lobby against SAFTA due to the fear of being dominated by Indian goods (Pohit 2004).

Political factors The success of the agreement is contingent on political commitment and harmony among all the member countries.

• The political tensions between India and Pakistan are viewed as a major obstacle to the fulfillment of SAFTA. As Kemal (2004) points out, the two largest economies of the region, namely India and Pakistan, have not been able to realize the full potential of the bilateral trade between them because of persistent political problems.

• The domestic conflict in Sri Lanka and Pakistan, and the international tensions on Pakistan’s borders generate instability and reduce confidence which impede growth, especially cross-border investment flows. These perturbations or political conflicts elsewhere in the region can lead to a fall in output, which in turn, can have serious consequences on the vulnerable groups in individual countries (Global Economic Prospects 2007).

Protectionism (tariff and non-tariff)

• Historically, the combination of politics and protectionism has been responsible for the small share of intra trade in total SAARC trade. According to Kemal (2004), trade liberalization under SAPTA was not effective due to protectionist tendencies culminating in the absence of meaningful exchange of preferences amongst the member countries.

• According to Baysan et al. (2006), except Sri Lanka, the SAARC counties suffer from high applied duties on non-agricultural and agricultural goods. On an average, the level of applied duties on non-agricultural goods ranges from 10.7 per cent in Sri Lanka to 25.4 per cent in Bangladesh. The level of


protection is also high in case of agricultural goods: it ranges from 19.6 per cent in Pakistan to 40 per cent in India. Bangladesh has recently increased the use of para tariffs apart from custom duties and also enhanced nominal protection on several import-competing industries.

• While all the member countries are also members of WTO, if the MFN tariffs are close to preferential tariffs then intra-regional trade may not grow as rapidly as expected otherwise.

Administrative factors

• The SAFTA agreement expresses political commitment for trade facilitation in the region. Therefore, it needs a creative and effective framework for its accomplishment (USAID 2005).

• The sub-regional economies have engaged in large informal and unrecorded trade for a long period. For example, illegal trade and smuggling via long and porous Indo-Pak border have considerably increased the transactions cost of sub-regional trade (Das 2007).

• The enforcement of trade facilitation norms is critical. SAFTA covers institutional structures and dispute settlement/ resolution mechanism and, accordingly, the policymakers need to employ many departments and agencies to look after the issue of compliance. Unlike tariff reductions, which are administered by a single government agency, trade facilitation covers many procedures that take place along many steps of a cross-border commercial transaction and, thus, requires measures more than the standard uniform bureaucratic approach (USAID 2005).

• SAARC security discourses must be expanded to include political, social and environmental perspectives in order to achieve sustainability (Thakur and Newman 2004). To check violence and disorder, joint law enforcement, intelligence and linkages with international organizations need to be implemented (Khan et al. 2007).

Transport-related factors

• According to Pohit (2004), removal of barriers to transportation services can be very important for trade between India and Bangladesh. According to him, a transit route through Bangladesh will benefit the north-eastern states of India by a more direct access to the other regions of India. Moreover, Bangladesh can export power to these states and reap the benefits from economies of scale.

• A large volume of trade between India and Pakistan takes place via Dubai and Singapore. This third country channel increases the transactions cost considerably and the SAFTA agreement need to address the decline in transaction cost and increase the volume of intra trade (Das 2007).

Others

• To maximize the potential benefits from the cooperation, right policy measures need to be taken. According to Das (2007), these can include reduction in tariff and non tariff barriers, eliminating sectoral and product exclusion and having clear rules of tariff-rate quotas.

• A continuing process of unilateral or multilateral trade liberalization would be helpful for the South Asian countries to diversify more and evolve new comparative advantages and complementarities (Pitigala 2005).


VIII. Summary and conclusions

The South Asia subregion is one of the most populous and fast growing regions of the world. While the aggregate GDP of the region has been growing in the range of 7-8 per cent per annum, the performance in terms of per capita GDP has been lower, implying the region has a long way to go to enhance the average incomes and quality of life of its large populace. In 2006, region’s share in world trade was a mere 1.6 per cent, although aggregate trade flows accounted for around 30 per cent of the regional GDP. Since the 1990s, individual countries of the region have been consistently striving for trade liberalization, through forging trade agreements both within and outside the region. These have taken the form of unilateral efforts to reduce protectionism as well as broader multilateral regional and bilateral cooperation efforts. The implementation of the SAPTA (in 1995) and then SAFTA (since 2006) agreements have been key attempts at promoting regional trade, and larger economic cooperation among the countries of the SAARC region. The key bilateral initiatives have taken the form of the Indo-Nepal and Indo-Bhutan Trade Agreements, India-Sri Lanka Free Trade Area, India-Singapore Comprehensive Economic Cooperation, BIMSTEC and the India-ASEAN Framework Agreement on Economic Cooperation. The analysis of data indicates that there was a faster increase in intra-SAARC trade in the post-SAPTA period, but there is less conclusive evidence as to whether this uptrend can be ascribed to the formation of the PTA or to the overall (unilateral or bilateral) trade liberalization programmes adopted by the countries of the region in the early 1990s. Moreover, through 1985-2005, the growth rate of intra-regional trade depicted an uneven trend. Despite the growth in intra-regional trade volumes, over 1995-2005, the region as a whole had 95 per cent of its exports directed toward the rest of the world and imported close to 96 per cent from it. Thus, a bulk of the commodity trade exchange has been with the countries outside the region. On the aggregate, the bilateral trade intensity indices and changes therein did not depict a steady trend over 1995 through 2005. The results were mixed, depicting a decline in trade intensity between some pairs of countries, while an increase in case of some others. Thus, neither the growth in overall trade nor the forging of multilateral PTA/ FTAs were significant determinants of intra-regional trade. For some of the large economies of the region, there was in fact a decline in trade intensity in 2005 as compared to 1995. India’s trading intensity Bangladesh, Pakistan’s with Bangladesh and Sri Lanka’s with both Bangladesh and Pakistan depicted a fall during 1995-2005. For the region as a whole, regression of intra-group trade on extra-group trade (or trade with ROW), GDP, time and PTA/FTA dummy variables points to a strong fit. Changes in intra-SAARC exports and intra-SAARC imports are significantly and positively explained by variations in SAARC’s exports and imports to ROW, respectively. The dummy variable also has a significant positive coefficient, implying that SAPTA and other bilateral trade agreements within the region have contributed to intra-group trade at the aggregate regional level. The regression results are contrasting


with those of the trade intensity analysis and need to be reconciled with more detailed country-wise and product-wise analysis, including analysis of trade patterns in services, such as insurance, banking, tourism and hotels, telecommunications and information technology. The RCA analysis points toward absence of complementarity and, in fact, shows a high degree of competition in the export composition, thus pointing toward limited scope for regional trade integration. RCA analysis carried out for finer category of products would be more revealing. Interestingly, however, despite limited regional trade potential, regional integration did have significant positive impact on intra-regional trade in South Asia. Having run through the quantitative indicators, at this point, it is worthwhile to delve into those problems and obstacles of the SAFTA Agreement that are qualitative in nature and hence circumscribe the future intra-group trade opportunities. Some of these, in fact, are inherent flaws envisaged at the inception of the agreement itself. Although India, Pakistan, Bangladesh and Sri Lanka are largely populated countries, they are relatively small compared to other developed economies in the world in terms of GDP and trade flows. This involves a risk of trade diversion. In addition, the success of the agreement is contingent on political commitment and harmony among all the member countries. Thus far, a combination of politics and protectionism has constrained the growth in intra-regional trade within the SAARC countries. Under the auspices of SAFTA, potential benefits from liberalizing services trade are now being discussed. What remains to be ascertained is whether trade on an MFN basis or a bilateral agreement by just opening up the border between the countries or a multilateral endeavor such as the SAFTA, will be the most effective in achieving greater regional integration. As Rodriguez-Delgado (2007) points out, RTAs are viewed as an important tool for trade liberalization and, in most of the empirical studies, they have been found to create trade more than to divert it. On account of its large and diverse membership, in terms of trade on a genuine MFN basis, liberal rules of origin and streamlined trade facilitation measures, the SAFTA Agreement has substantial scope for improvement.


References Batra, Amita (2005). India Country Chapter (Chapter 3) in South Asian Free Trade

Area: Opportunities and Challenges, USAID, October 2005.

Baysan, Tercan, Arvind Panagariya and Nihal Pitigala (2006). Preferential Trading in South Asia. World Bank Policy Research Working Paper 3813, January 2006.

Das, Dilip K. (2007). South Asian Free Trade Agreement: Prospects of Shallow Regional Integration. Centre for the Study of Globalization and Regionalization (CSGR) Working Paper Number 218/07, February 2007.

Global Economic Prospects (2007). Managing the Next Wave of Globalization: South Asia regional Prospects, Washington D.C.

IMF (2007). World Economic Outlook Database, April 2007 (Online version).

Kemal, A.R. (2004). SAFTA and Economic Cooperation. Paper presented at the SFMA Regional Conference. August 2004, Dhaka, Bangladesh.

Khan, Shaheen Rafi, Faisal Haq Shaheen, Moeed Yusuf and Aska Tanveer (2007). Regional Integration, Trade and Conflict in South Asia. International Institute for Sustainable Development Publication, January 2007.

Pitigala, Nihal (2005). What does Regional Trade in South Asia Reveal about Future Trade Integration? Some Empirical Evidence. World Bank Policy Research Working Paper 3497, February 2005.

Pohit, Sanjib (2004). SAFTA: Much Effort for Little Gains? Financial Daily from The Hindu, January 03, 2004.

Rodriguez-Delgado, Jose Daniel (2007). SAFTA: Living in a World of Regional Trade Agreements. IMF Working Paper Number WP/07/23, February 2007.

Thakur, R. and E. Newman (2004). Broadening Asia’s Security Discourse and Agenda: Political, Social and Environmental Perspectives, United Nations University Press, Tokyo, Japan.

UNCTAD (2006/07). Handbook of Statistics, 2006/07 (Online version).

USAID (2005). South Asian Free Trade Area: Opportunities and Challenges. October 2005.

Weerakoon, Dushni and Janaka Wijayasiri (2003). Implications to Member States of Progression from SAPTA to SAFTA, Institute of Policy Studies of Sri Lanka, 2003.

World Bank (2004). Trade Policies in South Asia. Volume II: An Overview. World Bank Report Number 29949. September 2004.


regional trade integrations: a comparative study

Documents