International trade helps in improving welfare by allowing higher levels of consumption and investment than otherwise possible. In a labour surplus country like Pakistan it also helps in generating higher employment and higher wage rates with positive implications for income distribution and poverty, thus further raising the level of social welfare.

While the promotion of multilateral trade, mandate of WTO, promises higher rate of economic growth across the globe, the spread of regional blocs denies the advantages of free trade to the countries that are not part of the group. Therefore, for enjoying benefits of higher trade, including improved resource allocation, higher level of technical and X-efficiency and wider options for consumers and exposure to new ideas, technologies and products, the South Asian countries must ensure trade facilitation, higher investment and economic cooperation in other areas of economic development for the countries of the region.

Although the South Asian Association for Regional Cooperation (SAARC) has been in existence for about 20 years, intra-regional trade is still around 4 per cent of their total trade [See IRS (2004)]. The South Asian Preferential Trade Agreement (SAPTA) signed in the mid-90s has helped little in promoting the intra-regional trade because most of the products of export interest to the regional countries were excluded from the preferential treatment. It reflects mistrust and unwillingness of the South Asian countries to increase their interdependence, that augurs well for political and economic cohesion.

The SAARC summit declaration of Islamabad promises a South Asian Free Trade Area (SAFTA). It calls for reduction in import duties to 20 per cent by 2006 and between 0-5 per cent by 2013 but allows the less developed economies to reduce the rate of duties to 0-5 per cent by the year 2016. SAFTA allows the countries to notify the negative list which will not enjoy concessional import duties. Obviously, if the negative list is quite large, the impact of the agreements will be little. Considering the significance of trade to welfare, it is hoped that the South Asian countries will keep the negative list small.

INTRA-REGIONAL TRADE AND SAPTA
For promoting intra-regional trade, a preferential treaty (SAPTA) was signed in 1994 and in the first round that came into effect in 1995. Concessions were granted on 226 products in 1995 after the first round. It took four rounds to agree on 4700 products out of 6000 by the year 2000, with India leading the table, Maldives being the last as is evidenced in Table 1.

Table 1: Preferences under SAPTA by Countries

Despite the concessions to a large number of products, there has hardly been any increase in intra-regional trade in the total international trade of the region. This is because firstly, negotiations under SAPTA have been conducted mainly on a product-by-product basis, which allows some flexibility to each member country, but takes a lot of time. Second, the tariff cuts offered under SAPTA have not been enough. For example, India has offered the preferences on many products and the margins have been maximum but its MFN rates are typically higher than those of its partners, and as such concessions had very little impact. Third, most of the products to which were given concessions are not widely traded in the region. Confining solely to the tariffs and leaving para-tariff and non-tariff measures out of the negotiations has limited the growth of intra-regional trade as has high local content criterion. The countries have comparative advantage in similar products which also tends to reduce the trade potential.

INTRA-REGIONAL TRADE FLOWS
Intra-regional trade is rather low, only four per cent of the total trade. The shares of intra-regional trade in imports and exports are quite different. The shares of each country in intra-regional imports and exports vary significantly across different countries.

Despite efforts to strengthen regional economic cooperation through SAPTA, intra-regional trade is only 4 per cent of the total trade though there have been fluctuations around this level since 1995.

While intra-regional trade has been low, its patterns vary sharply from country to country. For example, the share of intra-regional imports in total imports of Bangladesh, Nepal and Sri Lanka stood at 11.7, 33.2 and 10.1 per cent respectively in 2000. Pakistan and India met only 2.3 and 0.7 per cent respectively of their import requirements from the region in the same period. Bangladesh's share of the regional imports quadrupled and that of Sri Lanka increased by almost one half over the 1985-2000 period. The shares of India and Nepal first declined but then recovered in recent years though only a little higher than their respective shares in 1985. The share of Pakistan shows some fluctuations but it is increasing.

Table 3: Percentage Shares of Intra-Regional Imports in Total Imports

INTRA-REGIONAL EXPORTS
In intra-regional exports, Bangladesh's share has gone down from 7.7 per cent in 1985 to 1.6 per cent in 2000, of Nepal from 38.3 per cent to 30.0 per cent, of Sri Lanka from 3.8 to 1.8 per cent and of Pakistan from 5.3 per cent to 2.9 per cent. However, India increased its share from 3.3 per cent in 1985 to 4.4 per cent in 2000. [For details see Kemal et al (2003)]. Smaller countries have the pro-regional bias in their trade structure while larger countries, both Pakistan and India, have an anti-regional bias in their trade structure. Unless all the trade partners benefit from trade liberalisation, trade expansion in SAARC can expand little.

Table 4: Percentage Shares of Intra-Regional Exports in Total Exports

India's trade has not only an anti-region bias, the index of trade balance for India falls short of unity; its exports to the region have invariably been higher than its imports. On an average, Indian imports from SAARC countries have been less than its exports to the region. Pakistan, on an average, has TB less than 0.5, while other countries, in general, have TB greater than one [see Kemal et al, (2003)].

REVEALED COMPARATIVE ADVANTAGE
revealed comparative advantage ratios, a concept developed by Balassa (1965), is simply a ratio of the share of a given product in a country's exports to its share in world exports. A country is said to have a revealed comparative advantage (disadvantage) in product if the ratio exceeds or falls short of unity. However, it may give misleading results amid distortions in the market. Therefore the pattern of "true" comparative advantage may differ from the one suggested by the revealed comparative advantage ratios.

The finer the disaggregation, the more useful would be the revealed comparative advantage ratios. Here we report results at three digit classification and the reader is referred to [Kemal et al (2002)] for one and second digit classification. The revealed comparative advantages of various countries are examined below.

Bangladesh has comparative advantage in fish, vegetables, jute, tea, leather, textile yarn, made-up articles of textile material, clothing, and woven cotton fabrics. India has comparative advantage in food, beverages and tobacco products including meat, fish, crustaceans, rice, fruits and nuts, tea and coffee, spices, feeding stuff for animals, tobacco and tobacco products; a wide range of 'crude materials' including oilseeds, cotton, stone, sand and gravel, iron ore, ores and concentrates of basic metals, and crude animal, vegetable materials, petroleum, oils and preparations, fixed vegetable oils; in chemicals and related products including nitrogen-function compounds, other organic chemicals, synthetic organic coloring material, medicinal and pharmaceutical products, perfumery, cosmetic and soaps, and insecticides and herbicides; leather; articles of textile and clothing; machine tools, household equipment, and steel products; and motor vehicles, motor cycles, and bicycles.

Nepal has comparative advantage in men and women's clothing, knitted or crocheted, floor coverings, textile clothing accessories, and essential oils and perfumes etc. Pakistan's revealed comparative advantage is in fish and crustaceans, rice, fresh and dried fruits, sugar, molasses and honey, and spices, vegetables, roots and tubers; crude materials including cotton, besides oilseeds and oleaginous fruits, worn clothing, stone, sand and gravel and crude animal and vegetable materials; textile and clothing; leather; floor coverings; medical instruments; baby carriages; and toys and cutlery.

Sri Lanka has comparative advantage in fish, crustaceans, other cereal meals and flour, fruits and nuts, tea, and spices; crude materials such as synthetic rubber, fuel wood, oilseeds and oleaginous fruits, paper and paperboard, vegetable textile fibers, and crude vegetable materials; rubber tyres and articles; wood manufactures; made-up articles of textile materials;, pottery, pearls and precious stones; materials of rubber; textile yarn, and woven fabrics of textile materials; and electric power machinery. The profile of revealed comparative advantage suggests that the pattern of revealed comparative advantage is quite similar across the South Asian countries.

With the exception of India and Sri Lanka, the South Asian countries enjoy comparative advantage in a relatively narrow range of products. Bangladesh, Nepal and Pakistan out of 71 commodity groups have revealed comparative advantage in only 7, 5 and 12 commodity groups while India and Sri Lanka have comparative advantage in 26 and 21 product categories; and none of the countries has comparative advantage in capital intensive and high value-added products. Despite the similar comparative advantage, there is sill some scope for increasing intra-regional trade. South Asian countries could import veneers, plywood, particle boards and other textile fabrics woven from Bangladesh; 43 products ranging from various food items to machinery and transport equipment from India; oilseeds and oleaginous fruits from Nepal; molasses and honey, cotton, clothing, crude animal and vegetable materials, fabrics, cutlery, live animals, and surgical instruments from Pakistan; and synthetic rubber, fuel wood, raw or processed textile fibers, residual petroleum products, tobacco, rubber articles, and electric power machinery and parts from Sri Lanka.

TRADE COMPLEMENTARITIES
Regional trading arrangements are likely to succeed in strengthening intra-regional trade if the trade structures of member countries exhibit strong complementarities. The low values of trade complementarity indices highlight the absence of strong complementarity of almost all the countries with other South Asian countries. The trade complementarity between Bangladesh and India has increased and is highest in the region. The pattern of complementarity between India's imports and its trading partners' exports shows lack of trade complementarity in exports of South Asian countries to India. Except for Sri Lanka, the index is around 10 per cent. The structure of Nepal's imports exhibits some compatibility with the exports of Bangladesh, India, and Pakistan. While trade complementarity between Nepal and Bangladesh has improved, it has weakened in Nepal's trade with India. On average, Nepal's import structure exhibits the lowest complementarity with exports of Sri Lanka and the complementarity is higher for trade between Pakistan and India. Exports of Bangladesh, Nepal and Sri Lanka depict weak compatibility with imports of Pakistan. While Sri Lanka and India are mostly compatible, exports of Nepal and Bangladesh do not match imports of Sri Lanka. The trade complementarity between Sri Lanka and Pakistan, though not substantial, has strengthened.

The South Asian region has an almost identical pattern of comparative advantage in some products, and that there is no strong complementarity in the bilateral trade structures of South Asian countries. Similarities in the trade structures and absence of comparative advantage in capital intensive and high value-added products i.e. the products that are normally imported by countries in the region may have limited the growth of intra regional trade in South Asia.

INTRA-INDUSTRY TRADE IN SOUTH ASIA
While trade would take place only if there are differences in factor endowments, Grubel-Lloyd (1975) argue that differences in technology and human capital can lead to intra-industry trade even in products with identical factor input requirements. Krugman (1981) argues that industries in which increasing returns are achieved at a fairly low level of output can accommodate many producers, with each producing differentiated products. Under these circumstances, each country will specialise in different varieties of the product and engage in intra-industry trade. The growth of regional integration schemes involving cross-country production sharing arrangements also increases intra-industry trade. Yeats (1998) points out that production sharing has become a major factor in regional trading arrangements, approximately 30 per cent of the world trade in manufactured goods, largely of intra-industry variety.

On the basis of two-way trade in similar products the Grubel-Lloyd has provided the index of intra-industry trade. The intra-industry trade index ranges between 0 and 1 with larger values indicating a greater degree of intra-industry trade. In the chemicals and related products category, the bilateral intra-industry trade between Bangladesh and India largely consisted of inorganic chemical elements, oxides and halogen salts, fertilizers, and insecticides and herbicides. Significant intra-industry trade took place in basic manufactures, such as made-up articles of textile material, floor coverings, and nails and screws. Intra-industry trade in machines and transport equipment is hardly noticeable, except for some trade in ships, boats and floating structures in the year 1995. Intra-industry trade has strengthened over time in such miscellaneous manufactured goods as clothing, knitted or crocheted women's clothing, articles of apparel and textile fabrics and clothing accessories of textile fabrics.

No intra-industry trade took place between Bangladesh and Nepal during this period. A low level of intra-industry trade occurred in chemical and related products consisting of dyeing, tanning extracts and synthetic tanning materials, medicinal and pharmaceutical products, and monofilament rods between Bangladesh and Pakistan,. A moderate to high degree of intra-industry trade was indicated in textile yarn, woven textile fabrics, special yarns, made-up articles of textile materials, and manufactures of base metals. In 1998, Bangladesh and Pakistan engaged in significant intra-industry trade in several industrial products; machinery and transport equipment, prominent among them being rotating electrical plants and parts, agricultural machinery excluding tractors, and pumps for liquids. In miscellaneous manufactured goods, intra-industry trade was confined to knitted and crocheted women's clothing, and articles of plastic.

There are only a few products in which the Grubel-Lloyd indices show a reasonable intensity of intra-industry trade between Bangladesh and Sri Lanka. These products include textile yarn, woven fabrics, special yarns, and printed matter. Some intra-industry trade is discernible in soap and cleansing preparations, monofilament rods, cotton fabrics, and made-up articles of textile materials.

In the chemicals and related products group, the Grubel Lloyd index indicated some intra-industry trade between India and Nepal in nitrogen compounds, inorganic chemical elements, perfumery and cosmetics, and a moderate degree of intra-industry trade in monofilament rods, and miscellaneous chemical products. In some years, intra-industry trade was significant in some basic manufactures such as articles of textile and clothing, leather, rubber tyres, plywood, floor coverings, mineral manufactures, rails and railway track construction materials, copper, aluminium, metal containers, wire products, and equipment of base metal. In machinery and transport equipment, no significant intra-industry trade occurred except in heating and cooling equipment. Other miscellaneous manufactured goods in which intra-industry trade was indicated were mainly women's clothing, footwear, road motor vehicles, articles of plastic, and works of art.

Intra-industry trade took place between India and Pakistan in nine items in chemicals and related products in which, the prominent among them being medicinal and pharmaceutical products and soap and cleansing preparations. The Grubel-Lloyd indices show some intra-industry trade in basic manufactures, such as leather, articles of paper and paperboard, embroidery, made-up articles of textile materials, floor coverings, lime, cement and fabricated construction materials, nails and screws, and manufactures of base metal. There are many products in the category of machinery and transport equipment in which intra-industry trade occurred between the two countries. These products range from textile and leather machinery and parts to heating and cooling equipment, and from data-processing machines, to medical apparatus. In miscellaneous manufactured goods, intra-industry trade mainly consisted of medical and measuring instruments, photographic supplies and musical instruments.

Intra-industry trade between India and Sri Lanka in chemicals and related products was confined mainly to carboxylic acids and nitrated derivatives, other organic chemicals, dyeing, tanning extracts and synthetic tanning materials, and essential oils. In the category of basic manufactures, leather and leather products, pottery, pearls and precious stones, construction material of rails, and metal containers figured prominently in intra-industry trade between the two countries. In some years, there was significant intra-industry trade in non-electric engines and motors, paper and pulp machinery, and electric power machinery. Other major products in which the GL indices showed a somewhat high degree of intra-industry trade were articles of apparel, motor cars and other motor vehicles, and photographic supplies.

Intra-industry trade between Nepal and Pakistan was confined only to metal salts, while intra-industry trade between Nepal and Sri Lanka consisted of trade in rubber tyres and textile yarn. Pakistan and Sri Lanka traded a variety of products intra-industry. In chemicals and related products, the GL index indicated some intra-industry trade in inorganic chemical elements, metal salts, and medicinal and pharmaceutical products. In some years, intra-industry trade appeared significant in a number of basic manufactures, such as leather, articles of textile and clothing, rubber tyres, floor coverings, wire of iron or steel, and manufactures of base metal. In machinery and transport equipment, intra-industry trade was indicated mainly in textile and leather machinery and parts, printing and book binding machinery, pumps, non-electrical parts, equipment for electricity distribution, electrical machinery and apparatus, and motor cycles and cycles. Other miscellaneous manufactured goods in which intra-industry trade occurred were furniture, bedding and mattresses, articles of apparel, medical instruments, printed matter, and articles of plastic [For details, see PIDE (2003)].

The following observations can be made regarding the nature and extent of intra-industry trade in the South Asian region. First, the historical pattern of intra-industry trade amongst the South Asian countries is highly erratic, and there are only a few products in which intra-industry trade has occurred on a sustained basis. Second, with a few exceptions, leather products, textiles and clothing, and some basic machinery and tools dominate the intra-industry trade profiles of the South Asian countries. Third, the proportion of intra-industry trade in total trade has been very low for most of the products. This is also reflected in the average bilateral Grubel-Lloyd indices of intra-industry trade. In 1995, intra-industry trade of Bangladesh was only 1.0 per cent, 0.08 per cent, and 0.01 per cent of its bilateral trade with India, Pakistan, and Sri Lanka, respectively.

Table 5: Average Bilateral Grubel Lloyd Indices of Intra Industry Trade: 1995

Intra-industry trade can play a pivotal role in promoting regional integration in South Asia because it can flourish even in situations where the trade and production structures of the trading partners lack strong complementarities. This leads to trade expansion and dynamics scales of economy. The South Asian countries can strengthen their trade linkages by devising mechanisms to promote intra-industry trade within the region. One way to accomplish this is through regional production sharing arrangements that involve the initiation of part of a manufacturing process for a specific good in one country and the transfer of the activity to another for processing. The South Asian countries can achieve greater economic cooperation and integration by evolving a vertically integrated regional production structure in sectors that can boost regional economy. This would allow the South Asian economies to specialise in different lines of production within a particular industry and thus achieve benefits of specialisation and economies of scale. It must, however, be pointed out here that the regional production sharing arrangements generally emerge in response to a combination of factors including low tariffs, wage differentials, low transportation costs, and favourable government policies.

IV. CONSTRAINTS TO INTRA-REGIONAL TRADE
Identical comparative advantage, lack of communication link, restrictive trade policies, lack of finances and political problems cause weak trade linkages in South Asia.

IDENTICAL COMPARATIVE ADVANTAGE
The South Asian countries have an almost identical pattern of comparative advantage in a relatively narrow range of products. Similarly, their bilateral trade structures hardly show any complementarities in the trade structure. Together with absence of comparative advantage in capital intensive and high value-added products, which are normally imported by countries in the region, they act as structural constraints on expanding intra-regional trade.

Though South Asian countries have undergone major structural reforms and share of industrial sector has increased sharply, their industry is not diversified. With the exception of India and, to some extent Pakistan, these resource constraints have prevented the South Asian countries to invest in high value-added exportable products, and have made these countries dependent on industrialised countries for their capital goods and technology. The regional exports largely consist of raw materials and traditional products, such as textiles and garments, and some regional countries are direct competitors in the world export market for these products. The import requirements of the region mainly consist of capital goods and high-tech products. In this way, the trade pattern of the South Asian countries is tilted towards the developed countries.

Despite the demand for the South Asian products in the region, there is a rather limited capacity to generate exportable surpluses of the product in accordance with the specifications. It has marred the growth of intra-regional trade. The specifications of products imported and exported are different.

LACK OF COMMUNICATION LINKS
There are no communication links between the South Asian countries and as such the production, consumption and trade patterns of potential trading partners within the region may not be known. There are hardly any ships which call on specifically for the export of South Asia to other regional countries. Similarly, inadequate trade facilitation mechanisms contribute to the unrealised potential of intra-regional trade in certain areas. For example, Nepal's trade with other countries in the region depends on transit facilities provided by India. These facilities often involve high handling and transportation charges and delays in delivery, thus hampering the flow of trade between Nepal and its trading partners in the region.

RESTRICTIVE TRADE POLICIES
Restrictive trade policies also cause the low level of intra-regional trade. The restrictions have been more severe on the export interest of South Asia. However, the South Asian countries have substantially liberalised their economies in the past decade or so. Some trade liberalisation has also occurred under the SAPTA regime, according to which almost 5000 products from all SAARC member countries are entitled to preferential duty treatment. The trade regimes of South Asia are still quite restrictive. There is, however, a general perception that the trade liberalisation episodes including SAPTA have not made any significant impact on intra-regional trade in South Asia.

POLITICAL PROBLEMS
Political differences have also undermined efforts to foster regional economic cooperation in South Asia. India and Pakistan, the two largest economies of the region, have not been able to realise the full potential of their bilateral trade owing to various political compulsions. The small South Asian countries have been skeptic towards regional economic cooperation initiatives, fearing that a large trading partner like India will dominate the region economically to the detriment of their domestic industries. The political conflicts as well as differences in economic outlooks have hindered intra-regional trade in South Asia. The prospects of trade cooperation have enhanced with improved Pak-India relationships and signing of SAFTA.

SAFTA AGREEMENTS AND POSSIBILITIES OF INTRA-REGIONAL TRADE
SAARC's Islamabad declaration in 1994 promises South Asian Free Trade Area SAFTA) which would come into force on January 1, 2006, to be fully implemented by the end of 2015. Under the agreement, tariff reductions, rules of origin, safeguards, institutional structures, and dispute settlement will be sorted out by various committees. It also calls for harmonisation of standards and customs procedures, mutual recognition of test results and transport infrastructure cooperation. These measures would hopefully help in promoting the intra-regional trade.

SAFTA's tariff reduction program calls upon India, Pakistan, and Sri Lanka, more developed countries, to reduce tariffs to 20 per cent by 2006. While Nepal, Bhutan, Bangladesh and Maldives, less developed countries, are required to reduce tariffs to 30 per cent by 2006. Following that Pakistan and India in five years, Sri Lanka in six years and other SAARC countries in 8 years shall have to reduce their tariffs to 0-5 per cent. Moreover, India, Pakistan, and Sri Lanka will reduce their tariffs on imports from the relatively less developed countries to 0-5 per cent by January 1, 2009. The agreement calls for elimination of all quantitative restrictions for products on the tariff liberalisation list. The agreement allows a negative list, but it provides for reviewing the number of products on the sensitive list at four-year intervals to reduce the list and expand the free trade coverage of the Agreement.

Regional groupings have proliferated around the globe. While some have been successful, others have not. World Bank (2004) reviews the regional experiences around the world and points out six broad conclusions for the success of such groupings of free trade. Firstly, a regional trade agreement does not automatically result in increased trade and growth. Whereas the intention at the time of formation of the group is always to promote intra-regional trade and economic cooperation in all the fields, a large number of interest groups emerge who on the grounds of injury to their industry, call for exemption of reduction in the import duties. The experience with SAPTA has been disappointing for this reason. SAFTA allows a sensitive list and if the list is large, then it may not see higher intra-regional trade. Whenever trade would be promoted, the industries in which the country does not have comparative advantage will close down. Moreover, agreements that kept in place high external border barriers, protect inefficient activities and undermine the competitiveness of all countries.

Second, the trading arrangements with unilateral efforts among members to reduce external protection have been more successful. Reducing trade barriers vis-à-vis the rest of the world creates an incentive for all members to export. It augments competition that drives domestic productivity [see Muendler (2002)]. When external protection is generally low, trade creation usually dominates trade diversion, and so the risk that regional agreements will be a drag on growth is substantially reduced. Indeed regional agreements where members have had low external protection have enjoyed greatest success [see Baldwin and Venables (1995) and Burfisher, et al, (2003)].

Third, the agreements between the countries with different factor endowments have shown more consistent success because of the opportunities to exploit different comparative wage rates, capital availability, technological levels that give rise to differing factor proportions in production [Schiff and Winters (2003) and Lederman et al, (2003)]. However, this conclusion runs contrary to the success of EU. The promotion of intra-industry trade would result in higher growth even if factor endowments are similar. Fourth, a regional integration framework that helps in trade creation and competition amongst regional countries would help in lowering domestic prices and providing new technology. It is impossible to have the benefits of a regional agreement without exposing the member economies to new competition [Hoekman and Schiff (2002)].
Fifth, competition in services also results in successful integration. Lowering the cost of telecommunications, finance, business services, and retail and wholesale commerce would result in productivity gains. Finally, there is a need to streamline borders transactions through trade facilitation. Increase in efficiency within the region often spills over into trade outside the region as well, because improving customs or improving efficiency of ports helps both intraregional trade and international trade.

To ensure the success of SAFTA, the member countries have to take a number of initiatives. Firstly, all countries must have very small negative list. If the list is large, the SAFTA would become redundant. There should be a firm basis to exclude products and ground rules should be laid down and be transparent. Second, keeping rules of origin simple and transparent and they must not become devices of protection and impediments to trade. The SAPTA rules, which are quite stringent, need to be revised. Third, ambitious agreements to establish detailed investors' protections and separate dispute panel resolution systems should be left for the future, because it could delay the progress unnecessarily. Fourth, since some of the SAARC members may dump and some may provide subsidies to exports and hence unfair competition, the transparent anti-dumping and countervailing methods may be necessary. However, anti-dumping actions and countervailing duties against regional partners can bring back the protectionist tendencies; anti-dumping mechanisms can stifle the benefits for SAFTA arrangements.

What would be the impact of SAFTA? RIS (2004) reports results of studies conducted in the framework of gravity model. It suggests that complete elimination of tariffs under SAFTA may increase the intra-regional trade by 1.6 times. It further suggests that in the dynamic framework the gains from liberalisation are at least 25 per cent higher than the static gains. However, these gains are grossly in view of SAARC's large trade potential; it exists both in trade diversion from traditional sources towards SAARC countries by removing the constraints and trade creation and expansion by easing import restrictions on products which SAARC countries are not trading in but are their major exports. While more than half the exports of manufactured goods from South Asia consist of textiles and leather products, they are subject to very high rates of import duties and/or quantitative restrictions and even outright bans in South Asia.

Similarly, rather limited trade in engineering goods is owing to a number of factors including reliance on foreign aid to finance the import of capital goods, poor quality of goods and heavy import duties on capital goods even by the countries who are themselves exporters of capital goods.

Intra-industry trade can play an important role in bolstering economic and trade relations within the region. This is because intra-industry trade can take place even in situations where the trade and production structures of the trading partners lack strong complementarities, as observed in South Asian countries. Whereas the intensity of intra-industry trade is low, the potential for widening the scope of this type of trade within the region is rather large. However, that can only be realised if the import duties are low. Intra-industry trade is largely driven by product differentiation and increasing returns to scale. Therefore, an increased level of intra-industry trade in the region can only be achieved if the regional countries develop the technological capacity to produce different product varieties at declining average cost.

Lack of trade complementarities causes weak trade linkages in South Asia. However, trade complementarities can be developed within the region if countries achieve vertical specialisation through production sharing arrangements. The manufacturing of components for automobile and engineering industries would be quite helpful. Similarly, in chemical industries the possibilities exist. Vertical specialisation would not only allow the regional trading partners to strengthen their trade ties, but also enable them to reap economies of scale by concentrating on a specific production process in the value-addition chain. Therefore, as in the case of various regional trading groups around the world, the South Asian countries can boost economic cooperation by developing vertically integrated production structures, thereby attaining vertical specialisation.

Joint ventures can pool regional resources to promote industrialisation and economic growth in South Asia. Since the South Asian countries have collectively gained substantial experience in agro-based industries, textiles and clothing, paper and pulp, and light engineering, there seems to be scope for joint ventures in these areas. The establishment of joint ventures will particularly benefit the small South Asian countries because they lack the resources to undertake industrial investment efficiently.

CONCLUSION
The signing of SAFTA has created euphoria in the South Asian countries. However, there are, at least, two possibilities which may make the agreement redundant. Firstly, all the countries are members of WTO and would reduce the tariff levels. If their MFN tariffs are close to preferential tariffs under SAFTA, intra-regional trade may not grow rapidly. Though one could argue that at lower rate of import duty, with or without SAFTA, the intra-regional and trade outside the region would flourish. Second and more importantly, if the negative list is large and includes most of the products of export interest of South Asian countries, trade would not flourish. The SAARC countries must make an effort to make SAFTA a success to expedite economic development, mainly because of the problems in global market access and the higher transaction costs of producing for the world market. The expansion of regional trade yields gains in production specialisation, efficiency and improved quality of exports, which benefit the countries participating in the regional co-operation effort.

The SAFTA has great potential and South Asian countries should accept the short-term costs for long-term benefits.

Dr Kemal is a senior economist and the Director of Pakistan Institute of
Development Economics (PIDE) based in Islamabad