ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

How Promising Is BIMSTEC?

In this article we attempt to determine to what extent the Bay of Bengal Initiative for Multisectoral Technical and Economic Cooperation economies are ready to form a free trade area. The present analysis suggests that the BIMSTEC region has many characteristics that would be desirable for an FTA. We base our analysis on a few criteria such as price, income, geographical characteristics and trade. In general, we find there are favourable indications for the BIMSTEC economies to enter into a successful regional trade agreement. Forming an FTA would be expected to create relative advantage for the member countries.

How Promising Is BIMSTEC?

In this article we attempt to determine to what extent the Bay of Bengal Initiative for Multisectoral Technical and Economic Cooperation economies are ready to form a free trade area. The present analysis suggests that the BIMSTEC region has many characteristics that would be desirable for an FTA. We base our analysis on a few criteria such as price, income, geographical characteristics and trade. In general, we find there are favourable indications for the BIMSTEC economies to enter into a successful regional trade agreement. Forming an FTA would be expected to create relative advantage for the member countries.


newly formed economic cooperation region between Bangladesh, India, Myanmar, Sri Lanka, Thailand, Bhutan and Nepal, is gathering attention. Better known as the Bay of Bengal Initiative for Multisectoral Technical and Economic Cooperation (BIMSTEC), the initiative for this economic cooperation, was undertaken during 1997. BIMSTEC was formed with the idea of imparting greater socio-economic cooperation among the member nations in the areas of technology, transport and communication, energy, tourism, agriculture, fisheries and human resources development. In addition to sectoral cooperation, BIMSTEC also wanted to strengthen cooperation in the areas of trade and investment. All seven BIMSTEC members have openly agreed to set up a free trade area (FTA) by July 2007. Negotiations on FTA regarding services and investment have begun during January 2006.

Importance of BIMSTEC

During the last two decades, regional trading agreements (RTAs) have gained increased prominence. Repeated failures of multilateral negotiations, especially at various ministerial meets of the World Trade Organisation (WTO), has lead to an increase in the number of RTAs. Also, increased internationalisation of markets (i e, globalisation) and the fear of losing out to other inefficient producers have put pressure on individual countries to become part of any RTA. Around 200 RTAs notified under the General Agreement on Tariffs and Trade (GATT) and the WTO are in force today. BIMSTEC is one of them. Most of the developing countries are now a member of one or more RTAs.

Compared to south Asian free trade area (SAFTA), the BIMSTEC FTA seems to be more promising. A deeper economic integration process within South Asian Association of Regional Cooperation (SAARC) sometime suffers because of political tension between India and Pakistan. Such things are less likely to happen in case of BIMSTEC. It is believed that negotiation under the BIMSTEC umbrella will be easier than under SAFTA because all the BIMSTEC members are purely guided by economic interests rather than by political interests.

The purpose of this paper is to analyse whether BIMSTEC is likely to emerge as a success story in terms of greater flow of goods and services in the south-east Asian region. We base our analysis on the basis of the following factors: prices, income, trade barriers (tariff and non-tariff barriers), economic structure, and geographical characteristics of the BIMSTEC member countries. These factors are essential in shaping up the formation of a successful RTA.

Why More Trade?

Success of a RTA is measured in terms of increased flow of goods and services. The more the economies trade among themselves the greater will be tendencies for further economic integration.1 Since trade affects growth, a greater flow of goods and services are likely to see less opposition in the way of economic integration.

Trade affects growth in three primary ways. First, trade encourages a flow of resources from low productive sectors to the high productive sectors, leading to an overall increase in output. Export growth may affect total productivity growth through dynamic spillover effects on the rest of the economy [Feder 1983]. The possible sources of this positive dynamic spillover include more efficient management styles, better forms of organisation, labour training, and knowledge about technology and international markets [Chuang 1998]. Second, with unemployed resources, an increase in export sales leads to an overall expansion in production and a fall in unemployment rate. As production increases, firms because of increase in the scale of operations (economies of scale) become more efficient [Helpman and Krugman 1985]. Third, international trade also allows for the purchase of capital goods from foreign countries and exposes an economy to technological advances of the developed countries. Recent theoretical work suggests that capital goods import from technologically advanced countries may increase productivity and thereby growth, since knowledge and technology are embodied in equipment and machinery and therefore transferred through international trade [Chuang 1998].

Criteria for Successful RTA

Despite these positive aspects, free trade is opposed mainly because workers and producers associated with the inefficient industries stand to lose out. There is a considerable amount of lobbying pressure by the inefficient producers demanding more protection. As raising tariff barriers is not allowed under the WTO framework, individual governments try to protect their respective economies by imposing non-tariff barriers (NTBs) like, antidumping measures, import licence, sanitary standards, etc. The answer to a successful RTA therefore, lies in controlling the factors that act against RTA, and nurturing the factors which help

Economic and Political Weekly December 23, 2006

forming and sustaining a RTA. Some of the factors that affect formation of a RTA are considered below. Intra-industry trade: RTA is more likely to happen when trade happens in similar commodities, that is, intra-industry trade. The likelihood that industry association will demand more protection is less in the case of intra-industry trade. In the presence of intraindustry trade (for example, India exporting Tata Indica cars to the US and at the same time importing Ford cars from the US), adjustment cost associated with removing trade barriers are lower. In this case jobs lost due to customers shifting to more efficient foreign suppliers may to a large extent be offset by the job enhancing expansion in foreign demand for similar, differentiated goods produced domestically. The political opposition to liberalising and expanding intra-industry trade tend to be far less when compared to trade involving dissimilar items, that is, inter-industry trade. Country characteristics: Economies that are similar in terms of size are better candidates for forming a RTA. Similarities are measured in terms of economic development and geographical proximities. The more similar are the economies, the more is the likelihood of intra-industry trade. This is because geographically near economies with similar levels of economic development have access to similar kind of technology. Consequently they tend to produce more or less similar items and tend to trade in similar commodities (closely differentiated products as in the monopolistic competition type market structure). As the literature on the gravity model of trade demonstrates, similarities in economic structure and geographical distance between respective economies are powerful determinant of trade [Linneman 1966; Frankel et al 1995; Frankel 1997]. Trade increases with economic size and proximity of the trading partners. Prices: Low technology intensive items like, leather footwear, garments, gems and jewellery, textile products, etc, which are typical of any developing country’s exports profile are very much sensitive to movement in prices, i e, are price elastic. When it comes to form a RTA, countries analyse whether such an arrangement will enable them to realise a greater demand for their exports. From the demand-side perspective, it can be argued that sustained demand growth cannot be maintained in a small domestic market, since any economic impulse based on expansion of domestic demand is bound to be exhausted. However, export markets do not exhaust quickly. RTA not only provides a platform for a greater market share but also enable countries to produce efficiently. As the literatures on monopolistic competition suggests, a way to produce exports competitively is to take advantage of economies of scale in production which can be realised from a greater market share resulting from a RTA [Helpman and Krugman 1985; Leamer 1984]. Government policies: More liberal government policies are likely to be beneficial for a RTA. There is a general consensus in the literature that trade volume, both exports and imports, increase following external sector liberalisation [Agosin 1991; Bertola et al 1991; Kohli 1991; Clarke et al 1992; Joshi et al 1996].2 Both the imports and exports of a country tend to increase with external sector liberalisation. Under the small country assumption a fall in tariff barriers reduces the price of imports and causes imports to rise. Exports also increase and this is true whether the economy has a fixed or flexible exchange rate regime. Under flexible exchange rate regimes when the economy opens up, first its imports rise. An increase in imports causes a relative increase in the supply of domestic currencies vis-à-vis the foreign currency. This happens because foreign currency are used to finance imports. With flexible exchange rates the value of domestic currency is market determined; an excess supply causes it values to depreciate. This means the price of exports for this economy falls; causing exports to rise. Under fixed exchange rate regimes, increase in exports happens in a different way. First, because of liberalisation imports increase. However, market price of domestic currency does not fall as it is fixed now. An increase in imports releases resources from the import competing sectors. A considerable portion of these resources find their use in export sectors. As a result, production of exports increases. Export prices fall, partly because of increased production and partly because input prices are cheaper with more coming from the import competing sectors. Exports increase. Higher trade volume, resulting from external sector liberalisation, is expected to increase the likelihood of a RTA formation. Geographical characteristics: This factor acts a catalyst and can have an accelerating effect on any country’s trade. Like trade affects growth, geographical characteristics of a region can also affect growth. Although a country’s geographical characteristics are not influenced by government policies, they can have an important effect on a country’s income by their influence on trade. Thus, countries’ geographical characteristics can be used as

Table 1: Intra-BIMSTEC Trade during 2003

(In US dollar million)

Bangladesh Bhutan India J a p a n Myanmar Nepal Sri Lanka Thailand World
Exports to: Bangladesh Bhutan India J a p a n Myanmar Nepal Sri Lanka Thailand Imports from: Bangladesh Bhutan India J a p a n Myanmar Nepal Sri Lanka Thailand 1358.00 428.00 30.21 4.42 11.17 273.00 61.00 131.00 2.68 3.28 5.64 30.00 2.38 4.00 10.00 3.84 29.00 55.34 2396.00 247.01 328.76 245.05 641.00 1494.22 2174.00 76.49 228.29 1076.16 879.00 51.49 1976.00 126.89 6.52 160.98 11435.00 566.70 2636.00 136.96 14.43 448.13 18266.00 2.44 73.00 125.00 0.24 439.00 33.23 259.00 140.00 2.29 915.00 2.98 217.00 13.00 1.66 28.00 4.86 345.00 7.00 0.19 1.00 5.80 957.00 375.00 1.36 0.22 161.00 9.24 227.00 193.00 0.37 1.18 8.00 9.45 799.00 16043.00 831.65 1.24 11.54 176.56 706.00 11890.00 483.39 30.57 145.89 6229.40 60641.00 473911.00 2641.70 649.40 5133.30 80521.00 9672.30 85294.00 383025.00 3204.90 996.60 6671.90 75809.00
Source: CSIRD, India.
Economic and Political Weekly December 23, 2006 5265

another variable to measure the impact of trade on income. For instance, one can argue one of the reasons for Nepal to trade less and hence poorly (per capita-wise) relative to Thailand is because the former is mountain locked and has no coastline in comparison to the latter. Thus, countries with more favourable geographical characteristics are more likely candidates for a RTA.

How Well Do BIMSTEC Members Fit These Criteria?

Given the discussion about the aforementioned criteria necessary to form a RTA in general, it is of interest to examine future prospect of the BIMSTEC area against these criteria?

Economic characteristics: When compared in terms of their economic structure, namely, value addition of services, industry, and agriculture sector, to gross domestic product (GDP), BIMSTEC nations have many similarities. Except in case of Thailand, the industrial sector constitutes roughly a fourth of GDP in all countries. All these economies are predominantly associated with service related activities. Although majority of the population still lives in rural areas, all of these nations are becoming increasingly urbanised. Geographical proximity along with similar economic profile indicates similarity in consumption, production, and trading pattern. Going by the argument of monopolistic competition (intra-industry trade and economies of scale) all these economies stand to gain the more they trade among themselves. Trade: Presently trade in the BIMSTEC region is low. One of the reasons for lower trade has to do with the closed nature of the BIMSTEC economies. Most of the BIMSTEC member countries have lower trade-GDP ratio and have initiated external sector liberalisation (that is, bringing down tariff barriers) only starting in the early 1990s.3 Presently there also exist a large number of NTBs in the region. The NTBs include procedural requirements, sanitary standards, certification and technical standards [Kelegama 2001; Mukherji 1997]. The encouraging point is that most of these economies have started to open up and have also registered healthy growth. During 2003-04, all BIMSTEC countries, except Nepal, witnessed strong economic growth in the range of 5-13 per cent as well as 4-5 per cent per capita GDP growth. As McCombie and Thirlwall (1997), Paulino (2002) and Paulino and Thirlwall (2004), have pointed out, a robust economic growth encourages a more liberalised trade regime. With similar exports’ profile, trading partners are better-off by placing less restrictions. Because countries in BIMSTEC share a similar exports’ profile they also face same types of NTBs; and hence share a similar negotiating stance for removing these barriers. Recent trends in trade data reflect this. India’s trade with other developing countries like Brazil, Sri Lanka and Thailand are on the rise.

Table 2: Economic Structure of the BIMSTEC Countries

Country GDP Sectoral Composition of Output (GDP)

Average Agriculture Industry Services Annual Value Added Value Added Value Added

Percentage Growth 2000-04 1995 2004 1995 2004 1995 2004

Bangladesh 5.1 31 21 18 27 52 53 India 6.2 29 22 29 26 41 52 Nepal 2.6 42 40 22 23 36 37 Sri Lanka 3.8 23 17 25 25 52 58 Thailand 5.3 11 10 40 44 49 46

Source: World Bank (2005), World Development Report, Oxford University Press, New York.

Exports in the BIMSTEC region have increased from US$ 104.9 billion in 1999-2000 to US$ 143.2 billion in 2003-04; whereas imports grew firmly from US$ 103.4 billion in 1999-2000 to US$

152.4 billion in 2003-04. Since early 1990s, Indian industries have started enjoying economies of scale [Barua and Chakraborty 2004]. Therefore there are indications that the present low intra-BIMSTEC trade is likely to flourish in the future.

Prices, income and geographical characteristics: To ascertain the importance of these factors in fostering a RTA we did an econometric analysis in the context of the BIMSTEC region (reported in the Appendix). In general, we found statistically significant coefficients associated with price, income and geographical characteristics. A statistically significant price coefficient implies that trade in the region will flourish provided the products are price competitive and there is no market access problem related to NTBs. A statistically significant income coefficient ascertains that higher GDP growth will encourage a more liberalised trade regime in the region. Lastly, statistically significant geographical characteristics coefficients’ imply that to facilitate trade there is a need to develop infrastructure in the BIMSTEC region. A country with miles of coastline without any port facilities will not be very different from countries without any coastlines. Here, infrastructures are seen as a factor complementing trade flow in the region. There is a need for the BIMSTEC member countries to invest resources for development of infrastructure. Since some of the countries are resource poor – during 2003, annual per capita GDP (measured in constant US$ 2,000) for Bangladesh and Nepal were US$ 395 and US$ 241, respectively – there is a need for foreign capital. In order to attract increased intra-regional foreign direct investment and portfolio investment flows, member countries should further strengthen macroeconomic conditions and liberalise and harmonise investment regimes.


As evident from the above discussion the BIMSTEC region has many characteristics that would be desirable to form a FTA. We based our analysis on the basis of few criteria such as price, income, economic and geographical characteristics, and trade, as an indication for forming a FTA. In general, we found there are favourable indications for the BIMSTEC economies to flourish into a successful RTA. Forming a FTA would be expected to create relative advantage for the member countries. Greater economic cooperation among BIMSTEC member nations has important implications in the form of larger market economies of scale in production, and improved resource allocation.



In the following, we estimate a simultaneous-equations model involving demand and supply functions of imports and exports of the BIMSTEC member countries. The literature generally agrees about the empirical specifications of the demand and supply functions for imports and exports [Leamer and Stern 1970; Magee 1975]. The demand for imports (M) is a function of domestic real income (GNP), the price of imports in domestic currency (PM) relative to the domestic prices (P), and the ratio of reserves (R) to imports lagged one period. There is considerable

Economic and Political Weekly December 23, 2006

evidence available that many developing countries’ capacity Export Supply: to import is constrained by the stock of real international reserves and hence this is the idea behind including reserves as an explanatory variable [Khan and Knight 1988]. It is expected



to have a positive coefficient as higher international reserves

i = 1, ..., N; t = 1,..., T. (3)

increase the ability of the country to import more. The relative price variable is expected to have a negative sign; a higher price where i denotes countries; t denotes time periods; z' i α1, z' i β1 implies a lower amount of imports demanded. The variable and z' i γ1 represents individual country specific effects for each domestic real income is expected to have a positive coefficient; of the three functions; ujit denotes the general equation specific demand for imports are expected to increase with an increase errors. in domestic real income. If zi is observed for all countries, then the entire model can

Under the assumption that world supply of imports is infinitely be treated as an ordinary linear model and can be fit by least elastic, we need not have to specify the supply function of imports squares. For the purpose of estimation we consider the classic [Khan and Knight 1988]. The foreign demand for exports is pool model and the within transformation model. If zi contains determined by the world real income (GNPW) and the ratio of only a constant term, then the ordinary least squares estimation exports price (PX) to the price of foreign substitute (PW). The provides consistent and efficient estimates of the common incoefficient on world real income is expected to have a positive tercept terms and the slope vectors. This is a classic pool model sign; demand for exports is expected to increase with a stronger (also known in the literature as least square dummy variable world real income. Similarly, the coefficient on the price variable model). Another variant of the fixed effects model is the within is expected to have a negative sign; foreign demand for exports

will fall when the price of exports increase. Table 3: Import and Export Growth Incorporating GeographicalCharacteristics

The supply of exports will depend on price of foreign substitute (PW) relative to domestic price (P), stock of fixed capital (K) 2SLS 3SLS Within 2SLS Within 3SLS and a term representing the role of imports in exports supply

Import demand (M)(M.PM/Pχ). Exporters are willing to supply more when exports Intercept -0.12241 -0.09221 _ _

price increase. Accordingly, the price variable in the export (-1.10) (-0.91) PM 0.004609 0.005326 -0.03415 -0.04155**

supply equation is expected to have a positive coefficient.

Log [—– ]

Similarly, more capital stock, and more importable inputs P (0.27) (0.32) (-1.36) (-1.76) LogGNPit 0.045657** 0.045517** 0.789564* 0.760677*

used for exports mean a higher supply of exports. Hence, the

(1.80) (1.86) (4.93) (5.34)coefficients of these two variables are expected to have positive LogMit–1 0.969939* 0.968645* 0.591720* 0.565527*

coefficients. (46.38) (47.69) (10.60) (11.45) R 0.051200* 0.049496* 0.054299* 0.063339*

We assume that the adjustment of import demand, export

Log (—)


+ α Log( demand and export supply to changes in prices and income are) + α Log(C) M (3.73) (3.72) (2.09) (2.87)

4 it −15 1i

M Log (C) 0.008768** 0.006747 _ _

not instantaneous, so we included lagged endogenous variables

(1.76) (1.37)

for the dynamic specification of the system. The other modifi-

Log (L) 4.846605** 4.485214** _ _cation of this basic model is the introduction of our measures of (1.84) (1.72)

geographical characteristics in the import demand and the export AdjR2 0.98938 0.82433

Export demand (X)

supply equations. Two basic measures of geographical charac-

Intercept -4.65571 -2.83225 _ _

teristics are used for this study. The first measure is the miles of (-1.33) (-1.44) coastline. The second measure is the area of land as a percentage PX -1.72556* -2.61554* -2.75512* -2.78969*

Log —–

of total surface area that a country shares border with other [P W ] (-1.99) (-2.16) (-2.05) (-4.27)countries. Our assumption is that the geographical characteristics LogGNPW 1.93037 1.76217** 0.277090 0.295826*

(1.52) (1.74) (1.64) (4.17)

variables, namely, miles of coastline (C) and proportion of border

LogX–1 0.999639* 0.998686* 0.940630* 0.907469*

lands (L), will facilitate trade and hence expected to have positive

(122.64) (123.00) (15.76) (20.02)signs. We do not introduce these variables in the export demand AdjR2 0.98905

equation, as the foreign demands for any country’s exports Export supply (X) Intercept 0.085843 0.086899 _ _

depends upon relative price competitiveness and not on the

(0.50) (0.88)

country’s geographical characteristics. The model is log linear, PW -0.02381** -0.00369 -0.033357 -0.00065 Log —–

with price and income coefficient reflecting the respective elas-[ ]

P (-1.93) (-0.51) (-1.31) (-0.06)ticities. Thus the equations can be written as: LogK 0.017314* 0.006845* 0.084963 0.043583 Import Demand: (2.70) (2.07) (1.04) (1.08) LogX–1 0.846173* 0.936902* 0.766956* 0.904406* ⎡ PM ⎤ (28.68) (48.34) (12.11) (20.30)

Log(M) it = α 2Log + α 3LogGNP it + α 4LogM it −1

⎢ P ⎥ Log(M.PM/PX) 0.158352* 0.062520* 0.208463* 0.085001*

⎣⎦ it

(5.29) (3.35) (3.72) (2.64) ' Log(C) 0.015179* 0.004225* _ _

+α6Log(L) + z1iα1 + u1it (1)


(3.37) (1.81) Log(L) 4.535949* 2.395974* _ _

Export Demand:

(2.11) (1.98)

⎡ P ⎤

X AdjR2 0.99273 0.79913

Log(X) it = β 2Log ⎢⎥ + β 3LogGNP W + β 4LogX it −1


⎣ W ⎦ it Notes: Figures in parentheses are t-ratios.

** Indicates that a coefficient is significant at the 10 per cent level and


+z β + u (2)

2i 1 2it * significant at 5 per cent level.

Economic and Political Weekly December 23, 2006 transformation model. Here the pooled regression is re-formulated in terms of deviation from the series means leading to the disappearance of the intercept terms and the dummies. This model is more efficient than models with dummy variables as it gives n degrees of freedom (corresponding to relevant dummies and the intercept term) back with same parameter estimates.

For estimation, we have data for seven countries in the BIMSTEC region covering the period between 1990 and 2002. However, for the entire time period (1990-2002) not all the variables could be found for all the sample countries, leading to an unbalanced panel data set. In total, we have 73 data points.

To avoid possible heteroscedasticity in errors all the quantitative variables are expressed in per capita terms. Precise definitions of variables are given below. There are three endogenous variables in the system, which are Log(X), Log(M) and Log(PX). Table 3 summarises estimates of all of the parameters using classic pooled 2SLS, classic pooled 3SLS, within transformed 2SLS and within transformed 3SLS.

Results and Analysis

All the estimates have theoretically estimated signs, except in ⎡ PW⎤one case where the coefficient of Log ⎢⎥ has come out with


P negative sign. Importantly, the geographical characteristics variables have statistically significant coefficient in three out of four cases, suggesting that they have important role in facilitating trade. Based on the within estimates, we find the income elasticity for the demand for exports are 1.93 (2SLS) and 1.76 (3SLS) respectively. Similarly, long run price elasticity demand for exports is –1.72 (Classic Pooled 2SLS) and –2.61 (Classic Pooled 3SLS). Hence, both income and price competitiveness of exports are important factors in determining exports performance of this group of countries.

Data Definitions and Sources

The sources of data are: (a) IMF, International Financial Statistics and (b) World Bank, World Development Indicators.

X: Per capita nominal exports in (constant 1995 US$); source (b).

M: Per capita nominal imports in (constant 1995 US$); source (b). R

(—): Official foreign reserves (constant 1995 US$) divided by

M nominal imports per capita; source (b). GNP: Per capita gross domestic product (constant 1995 US$); source (b).

K: Per capita gross fixed capital (constant 1995 US$); source (b). GNPW: Per capita real GNP for the world; source (b). PX: Unit value of exports (US$), 1995 = 100; source (a). PM: Unit value of imports (US$), 1995 = 100; source (a). PW: Unit value of exports of the continent of the originating country (US$), 1995 = 100; source (a).

P: Domestic consumer price index, 1995 = 100; source (a). POP: Population; source (b).




1 There are four different forms of regional trading agreements, namely,

FTA, custom unions (CU), common markets (CM) and economic unions

(EU). In forming as FTA, members remove trade barriers among themselves but keep their separate national barriers against trade with outside nations. In a CU, members not only remove trade barriers among themselves but also adopt a common set of external barriers. In a CM, members allow full freedom of factor flows (migration of labour and capital) among themselves in addition to having a CU. In an EU, members unify all their economic policies, including monetary, fiscal and welfare policies, while retaining features of a CM. A deepening of economic integration means member countries graduating from FTA to CU; thereafter from CU to CM; and finally from CM to EU.

2 In trade literature, external sector liberalisation is also known as trade liberalisation. It means a reduction in tariff barriers, phasing out of NTBs like quotas, import licences, etc, export promotion and a move towards market determined exchange rates.

3 The only exception being Thailand undertaking external sector liberalisation during early 1970s and Sri Lanka initiating liberalisation starting 1977.


Agosin, M (1991): ‘Trade Policy Reform and Economic Performance: A Review of the Issues and Some Preliminary Evidence’, UNCTAD Discussion Papers, No 41, Geneva, Switzerland.

Barua, A and D Chakraborty (2004): ‘Liberalisation, Trade and Industrial Performance in India: An Empirical Study’, paper presented in the seminar entitled ‘WTO Negotiations: India’s Post-Cancun Concerns’ jointly organised by Planning Commission and international trade and development division, JNU.

Bertola, G and R Faini (1991): ‘Import Demand and Non-Tariff Barriers: The Impact of Trade Liberalisation’, Journal of Development Economics, Vol 34, pp 269-86.

Chuang, C (1998): ‘Learning by Doing, the Technology Gap, and Growth’, International Economic Review, Vol 39, 697-721.

Clarke, R and C Kirkpatrick (1992): ‘Trade Policy Reform and Economic Performance in Developing Countries: Assessing the Empirical Evidence’ in R Adhikari, C Kirkpatrick and J Weiss (eds), Industrial and Trade Policy Reform in Developing Countries, Manchester University Press, Manchester.

Feder, G (1983): ‘On Exports and Economic Growth’,Journal of Development Economics, Vol 12, pp 59-73.

Frankel, Jeffrey A, E Stein and S Wei (1995): ‘Trading Blocs and the Americas: The Natural, the Unnatural, and the Supernatural’, Journal of Development Economics, Vol 47, pp 61-95.

Frankel, Jeffrey A (1997): Regional Trading Blocs in the World Trading System, Institute of International Economics, Washington DC.

Helpman, E and P Krugman (1985): Increasing Returns, Imperfect Competition, and International Trade, MIT Press, Cambridge.

Joshi, V and I M D Little (1996): India’s Economic Reforms 1991-2001, Oxford University Press, Oxford.

Kelegama, S (2001): ‘Bangkok Agreement and BIMSTEC: Crawling Regional Economic Groupings in Asia’, Journal of Asian Economics, Vol 12, pp 105-121.

Khan, Mohsin S and D M Knight (1988): ‘Import Compression and Export Performance in Developing Countries’, The Review of Economics and Statistics, Vol 20. pp 315-21.

Kohli, U (1991): ‘Technology Duality and Foreign Trade’, University of Michigan Press, Ann Arbor.

Leamer, E (1984): Sources of Comparative Advantage: Theory and Evidence, MIT Press, Cambridge.

Leamer, Edward E and Robert M Stern (1970): ‘Quantitative International Economics’, Allyn and Bacon, Boston, MA.

Linneman, H (1966): ‘An Econometric Study of International Trade Flows’, Amsterdam, North Holland.

Magee, S P (1975): ‘Prices, Income and Foreign Trade: A Survey of Recent Economic Studies’ in Peter B Kenen (ed),International Trade and Finance: Frontiers for Research, Cambridge University Press, Cambridge, England.

McCombie, J and A P Thirwall (1997): ‘The Dynamic Foreign Trade Multiplier and the Demand Oriented Approach to Economic Growth: An Evaluation’, International Review of Applied Economics, Vol 11, 5-26.

Mukherji, N I (1997): ‘Strengthening the Bangkok Agreement as a Regional Mechanism for Promoting Cooperation in Trade: Country Report – Thailand and Myanmar’, ESCAP.

Paulino, A (2002): ‘The Effects of Trade Liberalisation on Imports in Selected Developing Countries’, World Development, Vol 30, No 6, pp 959-74.

Paulino, A and A P Thirlwall (2004): ‘The Impact of Trade Liberalisation on Exports, Imports and the Balance of Payments of Developing Countries’, The Economic Journal, Vol 114, pp 50-72.

Economic and Political Weekly December 23, 2006

To read the full text Login

Get instant access

New 3 Month Subscription
to Digital Archives at

₹826for India

$50for overseas users


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top