Will free trade agreement between India and China reduce India’s trade deficit?

Trade between India and China has been rising exponentially with a widening trade deficit for India, which has raised alarm by businesses, and some Indian parliamentarians have started accusing China of unfair trade practices. Nevertheless, both countries intend to negotiate for free trade arrangements between them based on complementarity. This study examines how much reduction in trade deficit due to different preferential trading arrangements is feasible under hypothetical full export potential scenarios using a stochastic frontier gravity model. The empirical analysis shows that India’s potential gain is high when the influence of India’s existing ‘behind the border’ constraints is eliminated.


Introduction
Trade balance of a country is the difference between the value of its exports of goods and services and the value of its imports of goods and services. 1 When this balance is negative, it means that the concerned country is importing more than it is exporting. Initially, a trade deficit raises the standard of living of a country's residents, since they now have access to a wider variety of goods and services for a more competitive price. It can reduce the threat of inflation, since the products are priced lower. A trade deficit can also indicate that the country's residents are feeling wealthy enough to buy more than that the country produces. On the other hand, if the trade deficit persists over a long period of time, it can cause jobs outsourcing because, as a country imports certain goods rather than buying domestically, the local companies may have to close down their businesses. As a consequence, fewer jobs in that industry are created in the home country. Drawing on these arguments, many business corporations and trade unions propose reducing the trade deficit to increase employment. They often blame trade agreements between countries for causing such deficits. On the other hand, in this context, Hamanaka (2012) has argued that few countries, which have experienced trade interdependence with a particular country, have always opted to sign a free trade agreement (FTA) to solve issues associated with high levels of trade interdependence. However, it should be noted that these arguments are different in contents from the arguments levelled in favour of multilateralism by Bhagwati (1993).
In the literature, there are mixed empirical evidences across the continents with respect to arguments for both in support of and against FTAs between two countries. For example, in the case of free trade agreements of the United States, there are two views: one group argues that US FTAs have never exerted any significant influence on the US trade deficit, and over the past four years until 2011 FTA has resulted in a US-manufactured goods surplus of about $50 billion with its FTA member-countries (US Chamber of Commerce 2013). Criticising the US Chamber of Commerce's assertion that FTAs are solutions to US trade deficit and not problems, the 'Public Citizen', which is a national, non-profit consumer advocacy organization founded in 1971 in Washington DC, shows that US trade deficit with FTA member countries surged over the years with eliminating several thousands of jobs in the USA, while trade deficit decreased with non-FTA trading partners (Public Citizen 2013). Another example concerns the FTA between EU and Korea that came into force on July 1, 2011. Das (2012) has empirically proved that the FTA between EU and Korea has produced positive impact in terms of welfare gains to both countries, though of different magnitudes. Thus, it is an interesting empirical question: whether free trade agreements will widen the trade deficit between member countries.
Currently, the trade between India and China, which are the dynamic emerging Asian economies (Tseng and Cowen 2005;Rahman and Andreu 2006), has been growing exponentially with a widening trade deficit, which has raised alarms among businesses and trade unions in India. The trade deficit has been persistent in spite of the fact that India's average and peak tariffs rates are higher than that of China. During 2010, China's total merchandised exports to the world were US $ 1,550 billion against US $ 218 billion of India. China's uninterrupted growth is characterized by higher productivity, lower wages and exploitation of economies of scale among other parameters. India lagged behind China in almost all macroeconomic parameters including trade. For example, while China's total trade was 2720 billion in 2010, India's figure was hovering around 564 billion during the same period. China's share in the world trade was only 1.5 per cent during 1950, increased to 10 per cent in 2010, whereas India's share has dropped from 2.2 per cent during 1950 to 2.05 per cent in 2010. It is evident from the volume of trade of both countries, at least in the near future, India is not at all a competitor to China, but obviously China is the fierce competitor for India in the world market. This is because of the fact that China has started liberalizing its economy much before and faster than India, as a result of which China's FDI flow is ten times larger than that of India (Garnaut 2004).
India-China relationship has changed significantly in recent years because of its economic imperatives. Growing economic interactions are evident from the fact that during 2004-2005, China for the first time became the third largest importer of Indian goods next to the USA and the United Arab Emirates. The USA's share in India's exports was 16.74 per cent, which was 8.96 per cent in the case of UAE and 5.79 per cent for China during [2004][2005]. Concerning imports, the performance is even more spectacular. From the state of almost non-entity compared to few years back, China has made incredible progress in its market penetration to India surpassing all its past predictions. China now has emerged as the largest exporter to Indian market surpassing the traditional allies of the USA and the UK. During 2004-2005, China's share in India's total imports was 6.30 per cent, i.e. the highest of all trading partners, followed by the USA, whose share was 5.88 per cent. Ever since India-China bilateral trade has been growing at a faster rate, which stood at $ 18 billion in 2005 increased to $ 73.9 billion in 2011 and is expected to reach $ 100 billion in 2015. China's exports to India continued to surge, which left the trade balance in China's favour at US $ 27.07 billion in 2011 in spite of India's higher tariffs rates. In comparison, India's exports to China increased to US $ 23.4 billion registering a growth of 12.26 per cent in 2011 compared to the previous year, though China's average and peak tariffs rates are much lower. In order to enhance bilateral trade and promote economic cooperation, both China and India have started negotiations for establishing an FTA based on complementarity and comparative advantage. However, the widening India's trade deficit with China has attracted the attention of Indian businesses, trade unions, and politicians, who have started questioning the usefulness of FTA with China. Their concern is that FTA with China will destroy the small manufacturing units and will increase unemployment rate, mainly due to the cheap imports from China, which enjoy hidden subsidies in China.
Though the Government of India has been vigorously trying to conclude increasing number of bilateral/regional trading arrangements with different countries in the past few years, however, the ongoing discussion in the public arena about the surging India's trade deficit with China has slowed down the progress of bilateral free trade agreement with China (Government of India 2005), which holds membership in important regional groupings such as the Asia Pacific Economic Cooperation (APEC) (Whalley and Antkiwicz 2005). It is in this context, the objective of this paper is to examine whether India's FTA with China will further widen the trade deficit between them, as discussed in the literature. The analysis is done with a hypothetical preferential trading arrangement (PTA) of 50% reduction in existing tariffs and if tariffs are withdrawn completely by both the countries through FTA. This paper is divided into five sections. Section 2 deals with the methodology adopted and the nature of data used for empirical analysis in this paper. The next section discusses the trading patterns of India and China between 1995 and 2010. Section 4 shows the likely impact of PTA and FTA on bilateral trade between India and China. In a final section, the concluding remarks of this paper have been presented.

Theoretical modelling and data
It is rational to argue that trade deficit between any two countries arises mainly due to trade policies of both exporting and importing countries. Hence, the literature argues that countries tend to blame each other for their restrictive trade policies. What countries always overlook is to examine whether there are any specific characteristics besides their trade policy that contribute to trade deficit. In the trade deficit issue, most of the time, countries, as in the case of India, complain that they are not able to increase their exports due to the restrictive trade policy of the partner country. Bilateral export growth is determined by many factors: 'behind the border' constraints in exporting and importing countries such as infrastructure bottlenecks and inefficient institutions; trade policy related 'explicit beyond the border' constraints such as tariffs and exchange rate; and 'natural' factors such as partner country's population, gross domestic product (GDP), and distance (Kalirajan 2007). Of these, the exporting country can exert influence only on its own 'behind the border' constraints, which have direct bearing on the exporting country's export potential. The 'behind the border' constraints in the exporting country are the negative influences on exports of the existing infrastructure that deal with trade, such as connectivity between production point and exporting point, and of the inefficiency of institutions like port authority. Thus, even if there are no constraining factors in the partner country, the existing 'behind the border' constraints in the exporting country can reduce export flow. Therefore, it is imperative that exporting countries first eliminate their 'behind the border' constraints, which will facilitate reducing trade deficit. Hence, gauging the impact of 'behind the border' constraints on export flow is crucial in reducing trade deficit. As there are many 'behind the border' constraints in the exporting country, it is very difficult for the researchers to have full knowledge about these constraints. Nevertheless, drawing on Kalirajan (2007), the combined effect of all 'behind the border' constraints on export flow can be modelled and measured in a stochastic frontier gravity model framework as follows: wherethe term X ij represents the actual exports from country i to country j. Logarithm is represented as 'ln'. The term f (Z i ;b) is a function of the determinants of potential bilateral trade (Z i ), and b is a vector of unknown parameters. The single-sided error term, u i , is country-specific and represents the combined influence of all 'behind the border' constraints in the exporting country with respect to the concerned trading partner. It creates the difference between actual and potential exports of the exporting country to the relevant partner country. u i takes values between 0 and 1. When u takes the value 0, this means that the country-specific 'behind the border' constraints are not important and the actual exports and potential exports are the same. When u takes the value other than 0 (but less than or equal to 1), this means that the country-specific 'behind the border' constraints are important and they constrain actual export from reaching the potential export. Thus, the term -u i represents the difference between potential and actual exports in logarithmic values that is a function of the inefficiencies that are within the exporting country's control. It is assumed to be following a truncated normal distribution with mean µ and variance ó 2 , truncated above zero. The double-sided error term v i captures the influence on trade flows of other variables, including 'behind the border' constraints of the partner country, and conventional statistical errors. v i is assumed to be randomly distributed following a normal distribution.
It is important to note that a potential export is not the level of export without any restrictions (free trade). That is, a potential export is defined here as the maximum level of export for given levels of the determinants ('behind the border' constraints in partner country, explicit 'beyond the border' constraints, and 'natural' factors) of export and no restrictions on export, which is the absence of 'behind the border' constraints, within the exporting country. Using the joint density functions of u i and v i , the potential export of a country can be estimated by the maximum likelihood methods.
The computer software STATA can be used for empirical estimation. Both tariffs and trade data used in this analysis are taken from the World Bank (2013a). In the case of both India and China, we have taken simply mean tariffs rather than weighted tariffs. The distance and other gravity data are taken from CEPII (2012). The gross domestic products and the variables needed to form the real effective exchange rates are taken from the World Bank (2013b). The data for empirical analysis cover the period of 1995-2010.

India's trade with China: trends and patterns
India's total exports to China, its total imports from China, its growth rates and trade balances between 1996 and 2010 are shown in Table 1. India-China trade has been growing very rapidly since the mid-nineties. In 1995-1996, India's exports to China were $ 333.2 million, which grew to$5.6 billion during 2004-2005 and to $15.5 billion in 2010-2011. The trend of India's imports from China has shown a similar trend. India's imports from China were $ 813.2 million in 1995-1996, which increased to $ 43.48 billion during 2010-2011. Total trade between India and China touched $ 17 billion in 2005, which increased to $ 59 billion in 2010-2011. Total trade between two countries has been growing at an annual exponential rate of 28.13 per cent between 1994 and 1995 and 2004-2005, and then to 49% during 2005-2006 to 2010-2011, which is much higher than the rate of growth of India's overall trade during this period.
A significant feature of the India-China bilateral trade is that trade balance in favour of China has also been increasing over the years, which stood at $ 1.5 billion during 2004-2005 compared to $ 0.6 billion in 2000-2001. The deficit further increased, and it stood at $ 28 billion during 2010-2011. This has raised concern among Indian policymakers and business corporations. If the growth rate of India's exports to China is not maintained at the current level, it is expected that this gap will keep on widening. In terms of the commodity composition of India's exports, its principal commodities exports to China from 1996 to 2010 are shown in Table 2. Among other items, iron ore/concentrates are the most important items in India's exports basket to China over the years. This group is followed by copper, which is the second largest item that India exports to China. Cotton is another important item in India's exports basket to China. Other major product categories of India's exports basket to China are shown in the ascending order of importance in 2010 in Table 2. It is worth noting that India's exports basket to China has not changed significantly over the years. Table 3 shows India's imports of principal commodities from China from 1996 to 2010. India's major import items from China have been the telecommunications equipment. Imports of these items have been consistently increasing over the years. The second largest imported items are computer equipment. India imports substantial amount of manufactured fertilizers from China, which are the third largest import items from China in last several years. However, these imports suddenly shot up during 2007, though it was modest in the early years. Other important commodities, which cover more than 80 per cent of India's imports from China, are given in Table 3 in the ascending order in value in 2010.
Trade as one of the important channels facilitating the technology spillover among the trading partners is well established in the literature (Grossman and Helpman 1991). When countries import high value-added technologies, this means that they are also importing embodied technological progress, which contribute to improvement in productivity of labour and capital (Romer 1990). However, such technology spillovers to domestic firms largely depend on the absorptive capacity of developing countries (Cohen and Levinthal 1989). The technology absorptive capacity depends on the levels of both entrepreneurial and labour skills, which are determined by various government policies such as the trade policy, education policy, and research and development (R&D) policy (Teixeira and Fortuna 2010). In this context, it is worth examining whether India's imports from the world in general and from China, in particular, resemble the imports of embodied technological progress. Tables 4 and 5 show India's top 20 exports to and imports from the world, respectively. India's imports from China shown in Table 3 when compared with imports from the world shown in Table 5 imply that the feasibility of technological spillover from the former and the latter sources, such as special industrial machineries (SITC 728) and telecommunication equipment (SITC 764), cannot be ruled out. Therefore, imports from China have the potential to contribute to improvement in productivity and thereby to growth in India. Whether Indian domestic producers have effectively handled such technology spillover is another empirical question, which is beyond the scope of this paper. Nevertheless, some conjectures can be made: lack of industrial growth, which has been hovering around 6.5% for the past two years, has reflected in the low share of India's exports to the world, which has been stagnant over the years; infrastructure including power supply is very weak, as the government is spending only about 5% of GDP on infrastructure, which also discourages the inflow of foreign direct investment; it is necessary to revive the long-term finance for industrial development; and by providing easier credit access to small and medium enterprises, they can tap into foreign technological spillover smoothly.

Assessing the likely impact of PTA and FTA between India and China
Trade cooperation must start with the preferential tariff arrangements (PTAs) with the objective of completely eliminating tariffs to facilitate free trade arrangement between these two countries, which is the ultimate objective of bilateral cooperation. In the context of the proposed FTA, the impact of eliminating the 'behind the border' (BTB) constraints in India on its exports to China under the preferential trade agreement (PTA) and FTA arrangements is estimated in terms of percentage of exports that would be foregone by Indian businesses. The analytical tool for measuring the impact of the BTB constraints on India's export flow to China and also the impact of PTA and FTA on the bilateral trade between India and China under the assumption of no BTB constraints is the augmented gravity model suggested by Panagariya (1993) that is different from the framework adopted by Frankel and Wei (1995). While the latter approach concerns the pooling of data for different countries to estimate the gravity model imposing identical determinants of trade across countries, the former approach concerns only trade between any two concerned countries on a bilateral basis, which is more relevant in the present study. This paper considers the following hypothetical scenarios of PTA and FTA between India and China. These are as follows: 1. 50 per cent across the board tariff cuts by these two countries; and 2. 100 per cent tariff cuts, i.e. free trade between India and China.
The following augmented stochastic frontier gravity model using trade data from 2000 to 2010 is estimated separately for India and China for product-level data using the software STATA 12: where X ijt Bilateral trade between the exporting country (India or China) and trading partner at time 't'. Subscripts i and j refer to trading country and partner country, respectively. GDP = Gross National Product of the exporting country (India or China) at time t. POP = Population at time t. DIS = Distance between the exporting country (India or China) and trading partner in kilometres. It is measured as the distance between major ports of the exporting and importing countries. TRF = Average tariff rate imposed by the partner country at time t. REXCH = Real Effective Exchange Rate between the exporting country (India or China) and the partner country at time t. CONT = whether the exporting country and the partner country have common border? It is a dummy with equal to 1, if there is common borders; otherwise 0. LANG = Common official language. It is a dummy with equal to 1, if the exporting country and the partner country have the same official language; otherwise 0. t = time trend. u it = combined influence of country-specific 'behind the border' constraints in the exporting country at time t. v it = combined influence of 'behind the border' constraints of partner country, and conventional statistical errors.
The estimates of the above stochastic frontier gravity models for India and China are given in Table 6. First, the gamma coefficients for India and China are significant at the 1 per cent level, which indicates that the selected frontier model framework is appropriate to examine the trade flows of India and China. The highly significant Wald test statistics implies that the structure of the frontier gravity model with the selected independent variables has explained the variations in trade flows well. The structure of the model is further supported by the significant coefficient of 'mu', which confirms a truncated normal distribution with mean 'mu' and constant variance. The signs of the GDP of the partner country and GDP of India, which are significant at the 1 per cent level, are positive indicating that export flows would increase with the increasing size of both economies. Similar is the case with China and its trading partner GDP. The significant signs of population of India and its partner country are positive implying that an increased population increases the production and thereby the exports of India and also the increased population of partner countries increases the demand for India's exports in the importing country, respectively. Similar is the impact of increased population in China and its trading partners on China's exports. The significant time trend indicates that exports from China and India to their respective trading partners have been increasing over time.
Tariffs levied by partner countries exerted negative influence on exports from India and China as expected, though in small amounts. The coefficients of distance are negative for both India and China, though the estimate is larger for India than for China. This implies that China has been efficient in transporting goods and services than India to their trading partners. The contiguity dummy for India is not significant, which means India needs to improve its exporting strategy to other South Asian countries, particularly (1) Ln X ijt =á 0 +á 1 ln GDP i,t +á 2 ln GDP j,t +á 3 ln POP j,t +á 4 ln POP i,t +á 5 ln DIS i,j +á 6 ln TRF i,j,t +á 7 ln REXCH i,j,t +á 8 CONT i,j +á 9 LANG i,j, +á 10 t−u it + v it with Pakistan. On the other hand, the contiguity dummy for China is positive and significant, which is expected due to its strong connection in the East Asian production network and its FTAs with Pakistan and Sri Lanka in South Asia. Having a common official language has positive influence on exports flows from India and China, with larger influence for the latter. With the existing tariff structure and exchange rate, India could achieve on average only about 68% of its export potential to China, while China could achieve on average about 86% of its export potential to India. This means that India could increase its exports to China by 32% by simply eliminating the 'behind the border' constraints, which would certainly reduce the trade deficit between India and China significantly.
The present study is a comparative-static analysis showing the tentative increase in potential bilateral trade due to PTA and FTA between India and China. One can simulate the likely increase in potential bilateral exports due to PTA and FTA based on current levels of tariffs, which will make policy makers aware of the consequences of liberalization of tariffs. In the simulation exercise, we have taken two hypothetical scenarios: one with the 50% reduction in tariff (PTA) and the other is zero tariff (FTA). In our exercise, China's simple average mean tariff was 7.7 per cent during 2010 and India's simple average mean tariff was 11.5 per cent during the same period, which was about 50% higher than China's mean tariffs. If both India and China go for 50% reduction in their simple average mean tariff rates in 2010, the likely increase in potential bilateral exports from India to China will increase by 12%, while the percentage increase for exports from China to India works out to be 18%. It is interesting to know that under the FTA arrangement, India's potential bilateral exports to China will increase by 20%, while China's potential exports to India will increase by 28%. Thus, it is imperative that India should realize its export potential with China before it embarks on any PTA and FTA arrangements. These results corroborate the earlier findings of Erzan and Yeats (1992) in the context of FTA between the United States and the Latin American countries, which emphasized that the Latin American countries would gain significant long-term export benefits from their reduced trade barriers.
An important feature of trade between India and China has been the composition of items imported by both countries. While India imports basically finished and manufactured products and less of primary and intermediary products, China imports basically primary and intermediary products from India with least of finished products. Due to China's nature of products imported from India, tariff levels are very low. Therefore, imports are unlikely to be very high due to PTAs and FTA. On the other hand, most of India's imported items from China are manufactured goods and finished products, whose tariff levels are very high compared to primary goods. Therefore, any reduction in tariff will increase its imports substantially. It is concluded that through any kind of bilateral preferential or free trade arrangement, India will not gain much at least in the short run unless there is substantial change in eliminating the 'behind the border' constraints first to improve India's export environment.

Conclusions
India and China are two largest and fastest growing economies in Asia. One of the issues recently attracting the attention of policymakers in both countries is the widening trade deficit that is in favour of China over the years. The interesting question is whether such trade deficits can be eliminated through trade policies such as the free trade arrangement between India and China. Very recently, Dreze and Sen (2013) have argued that India's growth potential in all sectors has not been achieved and the existing gap between India's potential performance and actual achievements stems mainly from the fact that it overlooked to learn from the successful examples of the East and Southeast Asian countries. One of the lessons that India can learn from East and Southeast Asia is to provide efficient infrastructure and institutions that are conducive to growth. In the context of trade deficit, India's exporting environment is negatively influenced by the 'behind the border' constraints such as poor infrastructure and weak institutions supporting restrictive trade policies. As a consequence, if FTA between China and India is implemented without eliminating the 'behind the border' constraints, then certainly it will go in favour of China and will not reduce the trade deficit. Because, tariffs in India are very high, Chinese exporting firms have more to gain from the duty-free access to India and Indian exporting firms will gain much less from the duty-free access to China due to the prevailing low tariffs in China. Thus, it is necessary to bring the tariff structure of India, which is influenced by the inefficient institutional framework, to be in line with that of China; if not, the danger of potential losses from the transfer of tariff revenue to the Chinese firms in the form of higher profits will remain. On the other hand, an increase in duty-free imports from China might translate at least partially reduction in consumer prices in India, which will be substantially high compared to reduction in prices to the Chinese consumers because of the existence of their lower tariffs. Therefore, the possibility of welfare gains of the Indian consumers being higher than the welfare gains of the Chinese consumers due to FTA between China and India will be high.
Funding Open Access funding enabled and organized by CAUL and its Member Institutions. There is no funding provided for this article.

Conflict of interest There is no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.