REGIONAL INTEGRATION AND TRADE IN AFRICA: AUGMENTED GRAVITY MODEL APPROACH

Transcription

1 REGIONAL INTEGRATION AND TRADE IN AFRICA: AUGMENTED GRAVITY MODEL APPROACH Edris H. Seid The Horn Economic & Social Policy Institute (HESPI) 2013 African Economic Conference Johannesburg, South Africa October 2013

2 Outline of the Presentation Introduction Composition & structure of African Trade Review of related literature (theoretical & empirical) Estimation techniques Discussion and Estimation Results Concluding Remarks

3 INTRODUCTION African countries are fragmented and the continent as whole has been marginalized in the global market. Global share of merchandise exports of SSA IN 2000 was 1.5 percent which grew to 2.3 percent compared to 6.8 percent for South Eastern Asia. There are as many countries with small population size. Around 20 countries have less than 6 million population each. Such fragmented & small economies do not support large number of firms. Hence regional integration has been viewed by many as way forward for African countries to reap the benefits of economies of scale and accelerate industrialization and promote growth.

4 The history of regional integration arrangement in Africa goes back to early 20 th century. In 1910 SACU was formed among Botswana, Lesotho, South Africa, & Swaziland). Uganda & Kenya formed custom union in 1917, which was joined by Tanzania. In 1975 ECOWAS treaty was signed; in 1981 the PTA for Eastern and Southern Africa states was established w/c became common market in 1993 and renamed as COMESA. SADCC was formed in 1980 and transformed into SADC in 1992.In 1986 formed IGADD; expanded its mandate in 1995 and renamed as IGAD. Now there are 14 RECs in the continent of w/c eight of them are recognized as pillars of African Economic Community (AEC).

5 These RECs are CEN-SAD, COMESA, EAC, ECCAS, ECOWAS, IGAD, SADC and UMA. Now every country in the continent is member of at least one REC; 25 of them belong to 2 RECs, 17 are members of 3 and 6 countries are member of 4 RECs. This reflects the fact that there is overlapping membership which hinders the integration process. Despite the proliferation of RECs, intra-bloc trade remains low compared to other trading blocs in developing Asia as well as Developing America. Intra-regional trade constitute small fraction of the continent s global exports. It was 8.5 % in 2010 which grew to 10.8 % in Though low it grew by 15% annually on average in value terms in

6 Composition & Structure of Africa s Trade Global merchandise export share of Africa is low compared to other developing regions. In 2010 the share of Developing Asia was 33 percent while that of Africa was 3.3 percent. Where as Developing America s share was 6 percent. Among the African RECs, CEN-SAD, SADC & UMA have relatively larger shares.

7 Global Share of Merchandise Exports by Region ( ) Africa Developing America Developing Asia European Union

8 Since 2000, the continent as a whole has experienced growth in exports as well as imports. But still the exports composition is mainly dominated by primary commodities with fuel being the major export item constituting 59% of the global merchandise export of Africa in While manufactured exports constituted only 17 %

9 Africa s Merchandise Exports by commodities as percentage ( ) All food items Agricultural raw materials Fuels Manufactured Goods

10 African countries are oriented towards Western industrial countries especially towards EU for their imports and exports. EU and USA had dominated as Africa s import partner and export destination; but nowadays China in particular and BRICS in general have emerged to be important trading partner for the continent.

11 Export Destinations for selected African RECs ( ) EU Middle East Rest of Africa USA Japan China India Russia Brazil South Africa Turkey COMESA ECOWAS IGAD SADC UMA

12 Intra-regional, global exports in USD million ( ) RECs ECOWAS SADC IGAD COMESA Year Intra Total Share of Intra Intra Total Share of Intra Intra Total Share of Intra Intra Total Share of Intra

13 Intra-African exports by commodities in percentage ( ) All Food Items Agricultural raw materials Fuels Manufactured Goods

14 Given the emphasis placed on regional integration as key strategy and increased level of interest and strong commitment by many African countries, intra-africa trade remains low. Hence it is a timely subject to investigate Does such proliferation of regional economic communities in Africa contribute to expansion of intra-regional trade? What are the main factors behind the low level of intra- African trade despite large number of RECs? It explores the determinants of bilateral trade in Africa with particular emphasis to selected RECs namely COMESA, SADC, ECOWAS and IGAD. It also examines the impact of multilateral trade resistance on bilateral trade.

15 Review of Related Literature Theoretical Foundation of Gravity Model Gravity model has wide range of application and has empirical success as ex post analysis such as: migration, commuting, bilateral FDI & bilateral trade flows. But it was criticized for lack of theoretical foundations. Following the criticism, economists had formulated the theoretical justifications: Anderson(1979), Bergtrand (1985, 1989 & 1990), Deardoff(1998), Helpman(1987), Helpman & Krugman (1985), Anderson and van Wincoop (2003) among others. Anderson (1979) provides theoretical justification based on CES preferences and Armington assumption. Anderson and van Wincoop(2003) developed theoretically estimable gravity model w/c takes MTR into account.

16 Empirical Literature There has been many empirical works done on the determinants of bilateral trade & role of RECs in the world & in Africa in particular Tinbergen (1962) was the first to apply gravity model for international trade analysis. Eichengreen and Irwin (1995) analysed the impact of history on trade using gravity model. And found that omission of historical factors overstates the impact of trading blocs. Frankel (1997) applies gravity model to investigate the role of RIAs (European community, Mercosur, ASEAN) on trade.

17 Yeats(1997), Foroutan and Pritchett (1993), Cassim (2001), Alemayehu and Haile (2008). The studies done on Africa show that the performance of the RECs is disappointing compared to other RECs in other regions like Asia. Hence this study tries to uncover the main factors behind the low level of intra-regional trade in Africa despite the existence of large number of RECs applying the popular gravity model and using panel data for all African countries in

18 Estimation Techniques The gravity model has been used extensively in international trade studies since the independent pioneering works of Tinbergen (1962) and Poyhonen (1963). The traditional gravity model predicts that bilateral trade b/n countries is determined by GNP of the exporting and importing countries and the geographic distance between them.

19 Mathematically the traditional gravity model is expressed as: T ij Y i Y 2 j D Y is the GNP of countries and D represents distance Following the works of Tinbergen, Linneman (1966) augmented the gravity model by including population. ij 1 3 ij

20 Further augmented by including dummy variables that affect the trade costs b/n pair of countries such as adjacency, language, colonial relation, and RTAs (Frankel [1997], Silva and Tenreyro [2006], Cheng and Wall [2005]). Anderson and van Wincoop (2003) argue that the traditional gravity models with bilateral friction alone do not fully explain the trade flow b/n countries. Bilateral trade b/n partners is also influenced by Multilateral Trade Resistance (MRT). Hence some (Silva and Tenreyro [2006]) include atheoretical index called remoteness in the gravity to control for MRT.

21 The traditional augmented gravity model with the remoteness proxy is given as follows: T ij Y i Y 2 j D 3 ij N 4 i N 5 j 1 9 RTA 6 ij A 7 ij REM 8 it REM jt ij Traditionally it is common to log-linearize this eqn & estimate it by OLS. Anderson and van Wincoop (2003) argue MRT can be better handled in the estimation by controlling for importer and exporter time varying individual effects

22 The Anderson-van Wincoop gravity model is given as: lnt ijt 0 1dt 2dit 3d jt 4 ln Dij 5tij 6 C ijt U ijt This specification helps us to account for MRT which may change overtime due to change in the composition of trade partners and also it controls for global events. Silva and Tenreyro (2006) argue that log-linearization of the equation poses serious econometric problem and changes the property of the error term. Besides heteroscadasticity of the error, there are ZERO trade b/n countries.

23 Hence in this study PPML is applied which can address the problems and gives us unbiased and efficient estimators. Poisson regression model w/h explains volume of trade b/n countries has a Poisson distribution with a conditional mean : The conditional mean is exponentially related to the independent variables.

24 Discussion & Estimation Results Intuitive Gravity Model Result The estimation result from the intuitive gravity model result confirm that traditional variables are imp t factors behind the low level of intra-regional trade in Africa. Measures of bilateral trade cost have the expected signs. SADC and ECOWAS have positive and statistically significant coefficients while the IGAD has insignificant positive coefficient; The COMESA dummy has negative sign. SADC member states trade 38 percent more than non-member states whereas ECOWAS member trade around 6 times more compared to non-members.

25 Estimation Technique PPML Variables Xijt (Coeff.) Strd Error Ln (GDP)- expo 1.807*** (0.0254) Ln (GDP)- impo 0.214*** (0.0243) Landlocked-impo. dum *** (0.172) Landlocked-expo. dum *** (0.153) Border dum *** (0.252) Common Lang. dum (0.139) Common Colo.dum 1.033*** (0.139) Ln (Distance) *** (0.131) IGAD 0.33 (0.582) COMESA *** (0.0165) SADC 0.324*** (0.0298) ECOWAS 2.003*** (0.224) Ln(Per capita diff) ( ) Ln (area)-impo *** (0.041) Ln (area)-expo *** (0.0454) Bilateral exchange rate *** ( ) Ln (Population)-expo (0.0464) Ln (Population)-impo 0.986*** (0.0441) Ln (Remoteness)-impo *** (0.0118) Ln (Remoteness)-expo *** (0.0124) Ln (Distance)*t Constant *** Observations 30,503 Number of Paired 1,793

26 Anderson-van Wincoop Gravity Model Result Anderson-van Wincoop gravity model introduces year, importer and exporter fixed effects to control for MTR properly. It shows that adjacency, language, common facilitate trade while distance deters. colonizer The estimated effect of distance from AvW model is higher than that of the traditional gravity model. 10 % increase in distance deters trade by 20 percent suggesting that distance is one of the important components of trade cost. REC dummies have different sign. COMESA s is insignificant

27 Estimation Technique PPML Standard Variables Xijt Errors Common Border dummy 1.940*** (0.194) Common Language dummy 0.400*** (0.136) Common Colonizer dummy 0.506*** (0.151) Ln(Distance) *** (0.108) IGAD 1.651*** (0.523) COMESA (0.0266) SADC 0.193*** (0.0525) ECOWAS (0.241) Ln(Per capita difference) *** ( ) Bilateral Exchange rate 0.776*** (0.0303) Exporter Fixed Effects Yes --- Importer Fixed Effect Yes --- Year Fixed Effects Yes --- Constant (29.03) Observations 31,477

28 CONCLUDING REMARKS Despite the existence of many RECs in Africa, the continent has low level of intra-regional trade. Excluding the informal cross border trade, the intra-regional trade lingers around percent. This study finds that traditional gravity model variables are found to be the important determinants of bilateral trade flows in Africa. Besides real exchange rate between partners and difference in preference & taste among countries do affect the bilateral trade. Regional integration is thought to promote bilateral trade and then growth by bringing fragmented economies into a single large market. But the empirical researches carried out have mixed result on the contribution of regional economic communities (RECs) on intra-regional trade in African.

29 The result from the traditional gravity model shows that two of the RECs (SADC and ECOWAS) have created trade among the members. On the other hand IGAD and COMESA contrary to expectation have not contributed to the promotion of trade. A theoretical gravity model of Anderson-van Wincoop is estimated & the result gives almost the same result. But the IGAD dummy is found to be significant and positive implying that IGAD promotes intra-regional trade. The result from this paper implies that African countries should invest much on physical infrastructure to link neighboring countries, harmonize trade policies, and simplify custom procedures so that the existing regional economic communities would promote intra-regional trade.

30 In addition to this, these countries should adopt and implement coherent and coordinated trade policies to promote intra-regional policies. Most of African countries face trade facilitation constraints w/c include: Complex customs and administrative procedures and regulations, Inefficient and costly transit systems as evident by numerous informal roadblocks and checkpoints along many trade corridors in Africa. Moreover, there exist differences in rules of origin, trade documentation, and standards.

31 THANK YOU!!