East and South Asia in global trade: Regional integration and geopolitics

East and South Asia in global trade: Regional integration and geopolitics Shiro Armstrong Crawford School of Economics and Government Australian National University Preliminary Draft: please do not cite Abstract This paper measures the integration of the East Asian and South Asian economies in the global economy and their potential for regional and global economic integration through merchandise trade. East Asian economies are shown to be achieving much more of their potential trade than South Asian economies, both in terms of intra regional trade but also in trade with the rest of the world. South Asia realises more of its potential trade with trade partners outside the region than within its own region. East Asian economies in general achieve more of their trade potential among East Asian partners than globally. India is shown to be achieving more of its trade potential with East Asia, most prominently with China, than within its own region. The Sino Japan, India Pakistan and cross straits Taiwan mainland China bilateral relationships are examined to explain that integration into the global economy, starting with deep regional economic integration, has allowed economic relations to dominate and constrain difficult political relations in East Asia while in South Asia, adverse political relations have hampered development of the economic relationships. There is also evidence of trade in the less integrated South Asia being affected more by the ups and downs of political relations than in East Asia.

Contents Background... 1 Trade performance... 3 Measuring trade performance... 3 Results... 6 Trade and politics... 10 Politics in explaining trade performance... 10 Sino Japan relations... 13 Cross Straits relations... 14 India Pakistan relations... 15 Conclusions... 18 References... 19

Background Economic integration can encompass many different dimensions of opening up and increasing economic activity and interdependence across borders. It essentially involves the removal of barriers at and behind the border with the aim of increasing welfare from increased trade, investment and economic activity that comes from the ability to specialize, develop the economy and take advantage of mutually beneficial exchange with other countries. This paper focuses on trade, the most obvious measure and indicator of economic integration. Integration of East Asian and South Asian economies into the global economy show very stark differences in how entire regions approach and achieve openness. Countries in East Asia 1 have pursued policy strategies of openness to trade and investment that have led to their having high trade shares with other East Asian economies as well as globally. East Asia is one of the most economically integrated regions in the world with low barriers to trade and investment within the region but also towards the rest of the world (Armstrong and Drysdale, 2011). South Asia 2, on the other hand, is one of the least economically integrated regions in the world (ADB, 2008) and, if measured in terms of intraregional trade as a share of total trade, is the region with the lowest integration globally (Sally, 2011). Intraregional trade in South Asian was 3.5 per cent of total trade in 2009 3, up from a low of 2 per cent in 1967 but significantly lower than the 19 per cent in 1948 (Kumar and Singh, 2009). High intraregional trade is not an end goal in itself and trading partnerships in open economies should be determined by comparative advantage and market forces, but the advantages of proximity are important (Linnermann, 1966; Frankel, Stein and Wei, 1997; Disdier and Head, 2008). South Asian economies had on average trade to GDP ratio of 42.7 per cent in 2008 showing that they are still relatively less open than East Asia which had an average trade to GDP ratio of 63.2 per cent in 2008 4. Including services trade, the estimates rise to 51 per cent and 67.3 per cent for South Asia and East Asia, respectively. South Asian economies have low trade both within the region and globally despite growth over time. The rest of the paper focuses on merchandise trade and not services 1 East Asia is defined in this study as integrating East Asia as the ADB (2008) study defined which is most of the economies encompassing geographic East Asia except North Korea, Burma/Myanmar and the Pacific Island economies. 2 For the purposes of this study South Asia follows the World Bank s definition and is comprised of SAARC members less Afghanistan: Bangladesh, Bhutan, India, the Maldives, Nepal, Pakistan and Sri Lanka. 3 2009 and 2010 data are not complete for South Asian intraregional flows in the United Nations Comtrade database or the World Bank's World Development Indicators. 4 Source: World Bank's World Development Indicators online. East Asia is 'East Asia and the Pacific' as defined by the World Bank. 1

trade despite its importance in South Asian trade due to data availability 5. A region that has low economic integration can face problems beyond losing out on benefits to trade from economic proximity. Political interactions tend to occur at a much higher frequency between neighbours and higher economic interdependence can ameliorate and dampen adverse effects of political tensions (Mansfield and Pollins, 2003). Political tensions can act as a large barrier to economic integration and lack of economic integration and interdependence can act as a constraint on the improvement of political relations. Such an equilibrium might be thought to characterize the India Pakistan relationship, for example. The East Asian case is very different with bilateral relationships that have unresolved historical issues and political mistrust enjoying high levels of trade and economic integration between them. Some of these relationships, such as that between Japan and China and across the Taiwan Straits between China and Taiwan, show trends of improving political relations over time as economic relations deepen. But just how integrated are individual economies in South Asia and East Asia, within their region and with the global economy, and how much trade can we expect within and beyond these regions? Are the low intraregional trade shares in South Asia expected given the economic structures of South Asia economies and the other determinants of these countries' trade patterns? Are the high intraregional trade shares in East Asia due to how far East Asia has already proceeded with integration into the global economy or for other reasons? This paper estimates potential trade between countries and answers those questions through analysis of the achievement of that potential. It is not clear from trade shares whether the small intraregional trade and trade between certain partners, such as India China trade, is above or below what might be expected given the underlying determinants and characteristics of these countries' trade. Without benchmarking trade performance, it is not clear whether India's trade with China and East Asia, seemingly at the expense of its trade with South Asian partners, is due to policy strategy or because the East Asian economies are larger and more dynamic. The paper is organized as follows. The next section introduces the model to measure trade performance. Then the results of achievement of trade potential for East Asia, South Asia and other regions, as well as bilateral relations within each region, are presented. Political distance and other measures are then used to explain some of the differences in trade performance. The following section looks at the trade and political relationships of some of the key bilateral relationships in both regions before drawing some conclusions on the 5 Services trade will have to be included in the analysis in a different manner as different modes of services are affected by distance to varying degrees. 2

interaction between economic and political relations. Trade performance Measuring trade performance Trade performance is defined as the achievement of potential trade with potential trade being estimated as the maximum amount of trade achievable given the characteristics of the trading relationship which are benchmarked by the characteristics of all other trade relationships (Armstrong, 2007). Trade potential is measured using stochastic frontier analysis applied to a gravity model of trade and was pioneered by Drysdale, Huang and Kalirajan (2000) 6. The frontier, or potential trade, is defined by the characteristics of the most liberal trading relationships globally. The trade frontier is based on the determinants of trade and is defined by the best trade 'technologies worldwide The standard gravity model estimates the mean effects of all determinants of trade with variations from that mean all summing to zero. The stochastic frontier approach of the gravity model allows a composite residual that separates out impediments to trade (inefficiencies in the stochastic frontier analysis literature related to production) that reduce trade from the frontier, from random shocks and measurement error, and mostly from unobservable factors. The performance of trade relationships, or their achievement of potential trade, can also be thought of as a measure of economic distance where, since geographic distance is already controlled for, high performance shows low resistance to goods or capital flows between countries. The most liberal and free flowing trade and investment relationships are characterised by low economic distance Figure 1 is a stylized diagram of the difference between a gravity model with and without a frontier. With a gravity model, the comparison of where a trade flow performs is relative to the mean, or what the model would predict trade is, whereas it is an upper bound, or potential, with the frontier approach. 6 For other studies that use this approach see Kalirajan and Findlay (2005), Kalirajan (2008), Kang and Fratianni (2006) and Armstrong and Drysdale (2011), 3

Figure 1 Frontier and conventional gravity model comparison potential X Frontier Gravity Conventional Gravity actual X Y Trade determinants Source: Kumbhakar and Lovell (2000). The vertical axis shows the amount of trade and the horizontal axis is a vector of the determinants. There are many variants of the gravity model in use with the choice of variables and specifications varying considerably. Not all are consistent with derivations from micro foundations and not all have theoretic underpinnings. This paper uses a variation of the reduced form version of the Anderson and van Wincoop (2003) model that is widely used but adds a complementarity variable to reflect endowment structures between economies. The model to be estimated is ln x ijt ln 0 1 ln y it 2 ln y jt 3 ln Dist ij 4 Border ij 5 COMP ijt ijt v ijt u ijt where x ijt is the trade from i to j at time t and is from the IMF s Direction of Trade Statistics and the United Nations' Comtrade database (various years). Trade flows are calculated from imports instead of exports for accuracy 7. 7 Importers have less incentive to under report and imports are a more accurate reflection of trade flow values than reported exports. The exception is European trade where there is a tax incentive to underreport imports due to the value added tax structure but import flows were used for consistency. This is common practice. 4

y it is country i s size (GDP) at time t is from the World Development Indicators (WDI) and at current prices. rdist ij the relative distance from i to j. The (great circle) distance is betweeen capital cities and relative to the square root of the sum of all distances for i and j between their trading partners. The data are from the Chemical Ecology of Insects website 8 and the CEPII geodesic Distances database 9. Border ij is a variable that takes on the value of one if i and j share a common land border, zero otherwise. COMP ijt is an index of complementarity of i s trade with j at time t and is from Drysdale (1967) and Drysdale and Garnaut (1982): C ij k k X i M w M i k k X i M w M i M M k j j where X is exports, M is imports, subscripts denote country (i, j and world) and superscript k implies commodity k. The index is calculated at the three digit level from the Australian National University s International Economic Databank 10 for all combinations of countries and years. The index captures the complementarity of trade structures between countries and the higher the index the higher degree of complementarity. Multilateral (or third country) effects are controlled for using time invariant country pair fixed effects (µ ij) and time varying country pair fixed effects (µ ijt) as is commonly done to avoid bias that Anderson and van Wincoop (2003) identify (see Baldwin and Taglioni, 2006). The disturbance term v ijt accounts for random variation in trade similar to the disturbance term in the standard ordinary least squares model. The nonnegative (or one sided) disturbance term, u ijt, measures the difference between potential trade and actual trade. More precisely, it is the amount of trade that falls short of the frontier for trade from country i to j at time t. Other variables commonly included in gravity models, such as trade barrier measures (tariff levels), measures of risk or economic freedom, regional groupings and language similarity, are not included here but are instead included in explaining the trade performance in a separate model later in the paper. The core determinants as per Anderson and van Wincoop (2003) are used to estimate the potential and then all trade impediments, policy variables and other institutional quality indicators are used first to explain the distance from the frontier. 8 http://www.chemical ecology.net/ 9 http://www.cepii.fr/anglaisgraph/bdd/distances.htm 10 http://iedb.anu.edu.au/ 5

The model is estimated for a dataset that is a representative sample of world trade comprised of bilateral flows between the top 65 trading nations for the period 1990 to 2006. This period is used because it includes the Indian reforms of the early 1990s, the period when Chinese imports were almost completely banned by Taiwan, China's accession to the WTO and other significant events is before the global financial crisis. Also, the availability of the political distance data, discussed later, is limited to the period 1990 2004. The model can be estimated using the statistical programs GAUSS, STATA, LIMDEP, and FRONTIER 4.1. The results of the estimation using FRONTIER 4.1 are shown in Table 1. Results Table 1 Maximum likelihood estimation frontier gravity estimation results (1) (2) (3) Dependent variable ln(trade ijt) OLS Frontier country pair Frontier w country pair & time dummies Constant 24.28 *** 20.52 *** 21.05 *** (0.1207) (0.1214) (0.127) lngdp it 0.85 *** 0.77 *** 0.78 *** (0.0033) (0.0034) (0.0033) lngdp jt 0.79 *** 0.76 *** 0.76 *** (0.0031) (0.0031) (0.0031) rdist ij 2.42 *** 2.27 *** 2.24 *** (0.0243) (0.0237) (0.0229) Comp ijt 1.11 *** 1.33 *** 1.089 *** (0.0329) (0.0307) (0.0302) Border ij 1.73 *** 1.97 *** 1.98 *** (0.0137) (0.0155) (0.0145) sigma squared 1.77 52.04 *** 7.908 *** (0.1529) (0.173) Gamma 0.89 *** 0.88 *** Mu 5.37 5.29 log likelihood function 103285 101457 101324 Number of observations 60565 60565 60565 Note: Standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001. Coefficient estimates for time dummies are not included to save space. 6

The results are consistent with the findings of the vast literature on gravity models and show the expected sign and magnitude for the coefficient estimates of GDP, distance and border variables. A higher degree of complementarity in trade explains higher trade. Log likelihood ratio tests determine that the specification in column 3 of country pair and time dummy variables is the best fit for the data. Log likelihood ratio tests also determined that a truncated normal distribution for the one sided error term fits the data better than a half normal distribution. Trade performance, or realisation of potential is given by actual trade as a ratio of potential trade. Table 2 shows that the world average trade performance is 50 per cent over time and is fairly steady. Although results for selected years only are shown in Table 2, the world trade performance is either 49 per cent or 50 per cent in each year for the whole period. Table 2 Trade performance results, selected regions and years Region 1990 1995 2000 2006 Average World average 0.49 0.50 0.50 0.49 0.50 East Asian Exports 0.54 0.57 0.59 0.60 0.58 East Asian Imports 0.54 0.55 0.54 0.54 0.54 East Asian Intraregional 0.63 0.64 0.66 0.67 0.65 South Asian Exports 0.39 0.43 0.41 0.41 0.41 South Asian Imports 0.45 0.45 0.41 0.44 0.43 South Asian Intraregional 0.31 0.35 0.24 0.25 0.27 Europe intraregional 0.53 0.54 0.57 0.54 0.54 North America Intraregional 0.47 0.58 0.59 0.58 0.57 Source: Author's calculations Note: Europe is EU 15. At 60 per cent for exports and 54 per cent for imports in 2006, East Asian trade on average realises more of its trade potential than the world on average; East Asia averages close to those numbers over the whole sample period. East Asian intraregional trade is at 67 per cent of potential in 2006 which is the highest in the world with North America at 58 per cent and Europe at 54 per cent. The realisation of South Asian regional trade potential is the lowest among these regions, at 25 per cent in 2006. South Asia realises more of its potential trade with trade partners outside the region than within its own region. South Asia realises over 40 per cent of both export and import potential with extra regional trading partners, on average, compared to around 27 per cent of its intraregional 7

trade potential. South Asia's intraregional trade performance has also been falling over time. The results are consistent with those obtained in Armstrong, Drysdale and Kalirajan (2008) which also used a stochastic frontier applied to a gravity model but with different specifications. In that study ASEAN had the highest intraregional trade performance, followed by East Asia more broadly, then North America, Europe and with South Asia the lowest. Table 3 shows selected bilateral and country average results for the same years as Table 2. The results are selected and averaged from the 60,565 bilateral results (the same as the number of observations in model see Table 1). Pakistani trade is closer to potential, on average, than Pakistani trade with India. India is achieving higher trade potential, on average, than in the bilateral relationship with Pakistan as well. Average export trade performance is around 40 per cent of its potential over time with average import trade performance falling over time from 46 per cent to 41 per cent of potential in 2006. By comparison, Pakistan's trade with India is lower, achieving an average of 21 per cent of trade from Pakistan to India over the period 1990 2006 and 32 per cent of potential trade in the other direction over the same period (Table 3). India's trade with East Asia is closer to potential than its trade with the rest of South Asia. Both India's exports to, and imports from, East Asia are rising over time with exports at 53 per cent of potential and imports at 59 per cent of potential in 2006, both higher than the world average. India's trade is showing bias, or stronger performance, in trade with East Asia than with Indian trade overall, a successful result of global openness as well, perhaps, of the look East policy strategy. Some bilateral relationships in East Asia are worth special note. Trade between Japan and China realised more of its potential than Japan's trade on average and China's imports, while on par with China's average export performance. At an average of 59 per cent over the entire period for Chinese exports to Japan, and 57 per cent for imports from Japan, the relationship is realising more of its potential than the average of its global trading relationships, at 50 per cent of potential. While the global average is steady over time, the Japan China relationship registers improvement in realised potential over the entire period. 8

Table 3 Country and bilateral trade performance, selected years Exporter Importer 1990 1995 2000 2006 Average World average 0.49 0.50 0.50 0.49 0.50 East Asia China World 0.50 0.54 0.58 0.65 0.57 World China 0.46 0.48 0.47 0.54 0.48 China East Asia 0.58 0.62 0.64 0.68 0.63 East Asia China 0.49 0.55 0.60 0.65 0.58 China Japan 0.55 0.57 0.60 0.64 0.59 Japan China 0.48 0.56 0.57 0.62 0.57 Japan World 0.53 0.50 0.52 0.54 0.52 World Japan 0.46 0.42 0.43 0.45 0.43 China Taiwan 0 0.18 0.45 0.67 0.36 Taiwan China 0.42 0.64 0.65 0.71 0.63 Taiwan East Asia 0.68 0.63 0.64 0.67 0.64 East Asia Taiwan 0.67 0.67 0.67 0.67 0.67 Taiwan World 0.61 0.60 0.61 0.62 0.60 World Taiwan 0.54 0.58 0.58 0.58 0.57 South Asia India World 0.37 0.45 0.45 0.47 0.44 World India 0.39 0.38 0.41 0.50 0.42 India South Asia 0.31 0.23 0.18 0.29 0.26 South Asia India 0.26 0.37 0.38 0.42 0.35 India East Asia 0.43 0.50 0.51 0.53 0.50 East Asia India 0.44 0.47 0.50 0.59 0.50 India Pakistan 0.30 0.28 0.24 0.24 0.21 Pakistan India 0.32 0.27 0.27 0.30 0.32 Pakistan World 0.38 0.42 0.37 0.37 0.39 World Pakistan 0.46 0.46 0.36 0.41 0.41 Sri Lanka World 0.45 0.46 0.47 0.43 0.46 World Sri Lanka 0.53 0.53 0.47 0.45 0.50 Source: Author's calculations Notes: data for the year 2000 was missing for some intra South Asian trade so the data displayed is the average of data from 1999 and 2001. The data are relatively stable over time. Cross straits trade performance between Taiwan and China is also very high in 2006 at 71 per cent from Taiwan to China and 67 per cent the other way. And this achievement can be measured in the context of their having been virtually no direct trade between the two economies in the early 1990s with Taiwan pursuing a policy of unilateral bans on imports from China that were only 9

gradually lifted over the period under study. Actual to potential Taiwanese trade in East Asia was very high (at around 65 per cent) throughout the entire period under study despite its severely limiting imports from China. Taiwan's trade in East Asia realised much more of its potential than Taiwan's trade with the rest of the world. I return to discussion of this relationship below. Trade and politics What are the implications of South Asia only realizing 27 per cent of its potential intraregional trade compared to 65 per cent in East Asia, 54 per cent in Europe and 57 per cent in North America? What is inhibiting India and Pakistan from realising more of their trade potential and having their potential economic interdependence improve their political relationship? The rest of this paper attempts to address these questions. Politics in explaining trade performance The trade performance can be explained by variables that reflect trade policy, domestic and partner economic and political conditions, as well as institutional factors. This section indentifies a number of variables that proxy these determinants of trade performance and allow a measure of whether certain variables such as political distance between countries affect the trade performance as measured in detail above. There is a vast literature on the interaction between trade and interstate conflict (or cooperation) that reveals a complicated relationship with causality between trade and politics depending on the countries, the symmetry in the relationship, the type of goods traded and the time period (Reuveny and Kang, 1996; Mansfield and Pollins, 2003). Each bilateral relationship faces different circumstances at different points in time and a few of the key relationships in East Asia and South Asia that have resulted in different outcomes in the interaction between politics and economics, are worth reviewing. Political distance between countries causes uncertainty to increase and acts as an impediment to trade or investment. Significant and prolonged low intensity conflicts, as well as positive political developments fostering cooperation in the relationship, can affect a trade or economic relationship (Armstrong, 2009). The political distance variable used in this study is a measure of sentiment towards another country, separated into both positive and negative elements, based on event coding from newspaper articles. The data are from King (2003) and are updated to 2004 data in the new IDEA dataset. Events or actions 10

between countries are given weights on a scale of severity that has 200 categories and, to reduce reporting bias, is all coded from Reuters Business Briefs. There is an established literature which employs, tests and develops such event data (Mansfield and Pollins, 2003). The data are events or actions between countries and is high frequency (monthly) which are aggregated to an annual measure for this study and the cooperation net of conflict is used here as in Armstrong (2009) and Armstrong and Drysdale (2011). In order to estimate whether political distance between countries has an effect on trade, or in this case whether it helps explain the realisation or lack of realisation of trade potential, it is included as an explanatory variable alongside other variables with the actual to potential trade ratio as the dependent variable. The other variables include indexes of economic freedom for both exporter (i) and importer (j) which are from the Fraser Institute (Gwartney et al., 2008). The average tariff rate for each exporter and importer is also extracted from the index of economic freedom. There is assumed to be very little correlation between the economic freedom index and the tariff rate, as the economic freedom index is calculated from many variables of which the tariff rate is only one. A language index is also included an is based on Boisso and Ferrantino (1997) which is an index that takes a value of 0 if none of the population of country i speaks the same language as in country j and a value of 10,000 if all of the population in both countries speak the same language. Due to the lack of data availability, the language similarity index does not change over the 27 year period. It is preferable to other language similarity variables such as in the CEPII database that are often binary, taking the value of one if two countries share the same language and zero otherwise. Finally, regional grouping variables are included to see whether being a member of a particular regional grouping helps explain some of the realisation of trade potential. The results for the model are presented in Table 4. 11

Table 4 Explaining trade performance (1) (2) (3) Freedom index Exporter 0.0248 *** 0.0238 *** 0.0196 *** (0.0008) (0.00084) (0.000835) Freedom index Importer 0.0247 *** 0.0229 *** 0.0207 *** (0.000673) (0.000698) (0.000691) Exporter tariff 0.000795 ** 0.000280 0.000300 (0.000292) (0.000302) (0.000297) Importer tariff -0.00232 *** -0.00274 *** -0.00256 *** (0.000296) (0.000307) (0.000301) Language similarity 0.0000104 *** 0.0000103 *** 0.00000929 *** (0.000000418) (0.000000411) (0.000000414) Political distance -0.0000129 (0.0000221) EA* Political dist 0.000394 *** (0.0000959) SA* Political dist -0.00423 *** (0.000742) Lagged political dist -0.0000131-0.00000945 (0.0000217) (0.0000212) EA* Lagged political dist 0.000377 *** -0.0000132 (0.0000935) (0.0000920) SA* Lagged political dist -0.00542 *** -0.00240 ** (0.000821) (0.000850) APEC 0.104 *** (0.00328) ASEAN 0.0919 *** (0.00921) NAFTA 0.00267 (0.0259) EU 0.0491 *** (0.00386) SAARC -0.177 *** (0.0168) ANDEAN 0.0637 *** (0.0123) MERCOSUR 0.103 *** (0.0116) constant 0.179 *** 0.204 *** 0.238 *** (0.00661) (0.00705) (0.00708) N 40181 38984 38984 Note: Standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001. Three different models are estimated with different specifications for political distance and with the last model (in column 3) including a variable for regional groupings. Higher scores of economic freedom for both the exporter and importer are associated with higher realisation of trade potential. Higher tariffs by the exporting country reduces the distance to the frontier but higher tariffs by the importer is associated with lower realisation of trade performance, as would be 12

expected. Language similarity is associated with increased trade performance. Being a member of APEC, ANDEAN, ASEAN, MERCOSER and the EU are associated with higher realised trade potential within that group. Being a member of SAARC is associated with lower trade performance among SAARC members a result that is the same as that discussed above. The results suggest NAFTA does not have a statistically significant effect on trade performance. a result which is unexpected. All three NAFTA members are part of APEC so the NAFTA effect may be picked up in the APEC variable. The political distance variables that are included are for the whole sample (Political distance), as well as that variable interacted with a dummy variable for East Asia and South Asia. They are also included with a one year lag to reflect the lagged effect which often varies but is generally accepted to be roughly one year. (Reuveny and Kang, 1998). The results show that trade performance in South Asia is more affected by political distance than it is in East Asia. This is an important result. There is no strong evidence that politics affects trade performance at the global level consistent with the idea that most trade is conducted under the World Trade Organization system and is arms length and so would not be so sensitive to politics, unlike investment that requires more commitment to economic engagement in another country (Armstrong and Drysdale, 2011). The lack of a global result for politics affecting trade is not surprising given that the direction of causality between political distance and trade is not clear (Polachek, 1980; Reuveny and Kang, 1998; Mansfield and Pollins, 2003) and that the direction of effect is also unclear. While improved political relations (a narrowing of political distance) is usually associated with an increase in trade, trade has been found to increase as well as decrease political amity and enmity between countries (Barbieri and Schneider, 1999; Mansfield and Pollins, 2001). Given that the effect of political relations on trade differ by country pair and over time, the next section takes a closer look at some key relations in East Asia and South Asia that can further the understanding of the findings thus far. Sino Japan relations The economic relationship between China and Japan is one of the largest in the world. In terms of trade, it is behind only China US and US Canada in aggregate size. Given that they are the second and third largest economies globally, and they are in close proximity, a high level of trade is expected. But high trade shares and economic interdependence have occurred with the backdrop of unresolved historical issues that flare up from time to time and strain the 13

political relationship. The period covered in this paper, 1990 to 2006, includes the period 2001 to 2006 where political tensions rose to a level where leadership visits were suspended between the two countries and there were large protests in China against Japan and boycotts aimed at Japanese goods. Yet these political tensions did not derail the economic relations. The trade performance results show that in fact 62 per cent of trade from Japan to China and 64 per cent of trade from China to Japan is realised, significantly higher than the global average, the average of Japan's exports, the average of Japan's imports, the average of China's imports and on par with the average of China's exports in 2006 (see Table 2). This performance against potential was not affected significantly at all by disturbance in the political relationship between these two partners. So the Japan China trade relationship realised more of its potential trade than other trade relationships, on average. That happened despite the political tensions and conflicts that rose fairly high from time to time. The period of heightened political tension, represented by the rising negative actions or events starting in 2001, coincide with a period of positive globally directed actions or events to do with China's accession to the World Trade Organization (WTO) in December 2001. It was not simply the positive news from China joining the WTO that offset the conflict and tension. Low intensity conflict existed prior to WTO accession and indeed the contemporary Japan China relationship since engagement in 1978 featured periods of high intensity conflict. But China's commitment to the global trading system since 1986 with rapid trade liberalisation and economic reform is what gave Japanese firms confidence in dealing with Chinese counterparts (Armstrong, 2009). China steadily and increasingly constrained itself to global trading rules and norms through its 15 year accession march towards membership of the GATT and later the WTO. Cross Straits relations Political relations between Taiwan and China are even more difficult and complicated than the relations between China and Japan. In addition to, and because of, the political difficulties, Taiwan had earlier banned imports from China. These bans were only gradually lifted during the 1990s and then more rapidly after both China and Taiwan s accession to the WTO (Drysdale and Xu, 2004; Armstrong, 2010). 14

As a result of the gradual lifting of bans (a considerable number of which still persist in 2011) realisation of potential trade from China to Taiwan has risen rapidly and in 2006 was at 67 per cent, significantly higher than the world average and Taiwan s average imports globally, and on par with Taiwan s imports from other East Asian countries. Political and economic relations have improved further since 2006 with a significant economic framework agreement that aims to reduce the trade and other barriers over time signed in 2010. There are Taiwanese firms in China that were not able to export products back to Taiwan and many still cannot easily repatriate profits. Much trade and capital had to pass through Hong Kong which was costly given the extra distance and transactions costs in a highly competitive region with small profit margins across the production chain. Taiwanese trade with the rest of East Asia was performing at an average of close to 60 per cent for 1990 to 2006. Taiwan was deeply integrated into the regional economy and part of the complex production networks that produces parts and components where they are cheapest to produce and then traded them to be assembled. Given that both China and Taiwan are so integrated into the regional economy, the lost potential in trade was too large and the opportunity cost of Taiwan s discriminatory protectionist policies became too apparent and acute. It was the deep integration into the regional economy by both China and Taiwan that has carried the bilateral economic relationship beyond being simply bilateral in its nature. Economic interdependence between China and Taiwan grew as they increasingly became more interdependent on the same regional economy. India Pakistan relations While the relationship between India and Pakistan is unique, and the political tensions and historical circumstances seriously differ from the Japan China, cross Straits and other relationships, there are useful inferences that can be drawn from those relationships that highlight the reasons why the negative politics has hampered the development of any significant trade between India and Pakistan. Trade from Pakistan to India was achieving 30 per cent of potential in 2006 and trade from India to Pakistan was at 24 per cent of potential, significantly below average Indian trade performance, average Pakistan trade performance and the global average. Those results are consistent other estimates that used the gravity model without a frontier suggesting that potential trade could be between 5 and 15

10 times higher than actual trade (Taneja, 2006; Batra, 2004; Kemal, Abbas, and Qadir, 2002; Khan, 2009). Positive news and events have increased in either frequency or intensity since the beginning of the last decade (see Figure 2) but they have been completely swamped by very high intensity negative conflicts. Figures 2 and 3 show the King (2003) data used above for India and Pakistan. The variable reported as action, event or news from India towards Pakistan is highly correlated with the reporting in the other direction so Figure 3 shows the net of positive and negative, which is what was used in the analysis above 11. Figure 2 Positive and negative news and events as reported from India towards Pakistan 120 100 80 60 40 20 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Positive Negative Source: data from King (2003) 11 The negative events are subtracted from the positive events to obtain a measure of net political closeness. As in utility theory, the assumption here is that positive events cancel out negative events to a certain extent. Therefore, a positive value for the political variable indicates political closeness and a negative value indicates widening political distance. A movement in a positive direction implies a narrowing political distance. 16

Figure 3 Net cooperation as reported from Pakistan towards India 80 40 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 40 80 120 160 Source: data from King (2003). Net cooperation is positive news and events less negative news and events. Both Figure 2 and Figure 3 show that the negatives outweigh the positives in the India Pakistan relationship. While that could be the result of the weighting system applied to the events data, the net cooperation variable for Japan China and Taiwan China is consistently positive. Pakistan has not yet reciprocated most favored nation (MFN) status for India and maintains a fairly narrow positive list (786 items) on goods that India may export to Pakistan (Khan, 2009). At the same time, India s tariff rates remain high, especially for goods of particular interest to Pakistan, such as textiles, leather, and onyx, and nontariff barriers are substantial (Khan, 2009). Unlike the cross Straits case where impediments to trade were reduced over time as the regional economy became more integrated around them, the lack of South Asian economic integration and interdependence means there is much less incentive to reduce barriers to trade or improve the poor infrastructure, lift bans on investment and ease people movement across the border. Since South Asian economic integration is lagging, dealings between India and Pakistan are primarily bilateral and do not have the positive sum economic element that is present in East Asia. That inevitably leads to the political security issues dominating bilateral dealings (and lack of positive economic news and events between the two in Figures 2 and 3). Political tensions have dominated the economic relations and economic relations have not been able to improve and have a positive effect on the political relationship. 17

Conclusions There are three sets of conclusions that might be highlighted from this discussion. First, East Asia is more integrated than South Asia, both in terms of intraregional trade and with the global economy. East Asia is also realising more of its intraregional trade potential than North America and Europe. In fact, East Asian economies in general achieve more of their trade potential among East Asian partners compared to their trade globally. South Asia, on the other hand, realises more of its potential trade with trade partners outside the region than within its own region. Key bilateral and country trade results add to the understanding of those regional level trade patterns. India is realising more of its trade potential with East Asia than with its own region and the rest of the world. Pakistani trade with India is realising much less of its potential than Pakistani trade overall. Trade between Japan and China over the period 1990 2006 and between China and Taiwan later in that period, are performing closer to potential than their overall trade and much more than the global average. Second, the effect on trade of political distance between countries is more pronounced for South Asia than it is in East Asia. This result is most likely due to the fact that East Asia is much more economically integrated than South Asia. The story in most of East Asia has been that of economic relations dominating the political relations and the positives of increased economic interdependence dampening the effects of the negative aspects of political relations. In South Asia the ups and downs of political relations have a larger negative effect on trade. Finally, the experience of the Japan China relationship shows the importance of countries integrating into the global economy and being part of the global trading system for bilateral relationships to prosper. The experience of the cross Straits relationship highlights the importance of intraregional economic integration for improving bilateral economic relations. India and Pakistan are neighbours in a region that has much lower achievement of potential trade than the global average and is the least economically integrated region globally. All bilateral economic dealings are swamped by bilateral political dealings which are dominated by negative sum or zero sum security issues. The experience of East Asia which shows that bilateral economic relations nested in a highly integrated region that is outward looking and globally oriented can help to dampen and even reverse the effects of political 18

conflict on trade. The non integrating East Asian economies of North Korea and Myanmar are important examples of closed economies associated with political conflict's dominating bilateral relations with their neighbours and beyond. A strategic inference from these conclusions and the argument here is that political problems that limit economic integration in South Asia are likely to become more tractable if the whole region is tied more closely into positive trade and economic relations trans regionally and globally and to full observance of the global rules of trade. References ADB (Asia Development Bank), 2008. Emerging Asian Regionalism: A Partnership for Shared Prosperity, Manila. Anderson, J.E. and van Wincoop, E., 2003. 'Gravity with gravitas: a solution to the border puzzle', American Economic Review, 93(1):170 92. Armstrong (2007). Measuring trade and trade performance: a survey, Asia Pacific Economic Papers, No. 368. Armstrong (2009) 2009. The Japan China economic relationship: distance, institutions and politics, Canberra, Australian National University, Ph D thesis. Armstrong, S. 2010. 'Taiwan's Asia Pacific Economic Strategies Post Economic Framework Agreement', East Asia Bureau of Economic Research Working Paper No. 63. Available at SSRN: http://ssrn.com/abstract=1767688 Armstrong, S. and P. Drysdale, 2011. The Influence of Economics and Politics on the Structure of World Trade and Investment Flows, in Armstrong, S. (ed) The Politics and the Economics of Integration in Asia and the Pacific, Routledge, forthcoming. Also available as working paper from http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1767687 Armstrong, Shiro Patrick, Drysdale, Peter and Kalirajan, Kaliappa P., Asian Trade Structures and Trade Potential: An Initial Analysis of South and East Asian Trade (2008). Available at SSRN: http://ssrn.com/abstract=1767686 Baldwin, R.E. and Taglioni, D., 2006. 'Gravity for dummies and dummies for gravity equations', NBER Working Paper No. W12516, National Bureau of Economic Research. Barbieri, K. and Schneider, G., 1999. Globalization and peace: assessing new directions in the study of trade and conflict, Journal of Peace Research, 36:387 404. Bandara, J.S. and W.Yu. 2003. How Desirable is the South Asian Free Trade Area? A Quantitative Assessment. The World Economy. 26 (9). pp. 1293 1323.] Batra, Amita (2004). India's Global Trade Potential: The Gravity Model Approach, Working Paper No. 151. Indian Council for Research on International Disdier, A. and K. Head, 2008. The Puzzling Persistence of the Distance Effect on Bilateral Trade, The Review of Economics and Statistics, Vol. 90, No. 1, Pages 37 48. 19

Drysdale, P.D., 1967. Australian Japanese trade, Canberra, Australian National University. Ph D. thesis. Drysdale, P.D. and Garnaut, R., 1982. Trade intensities and the analysis of bilateral trade flows in a many-country world: a survey, Hitotsubashi Journal of Economics, 22(2):62-84. Drysdale, P.D. and Xu, X., 2004. 'Taiwan's role in the economic architecture of East Asia and the Pacific', Pacific Economic Papers No.343. Drysdale, P.D., Huang, Y. and Kalirajan, K.P., 2000. 'China's trade efficiency: measurement and determinants', in P. Drysdale, Y. Zhang and L. Song (eds), APEC and liberalisation of the Chinese economy, Asia Pacific Press, Canberra. Frankel, J.A., Stein, E. and Wei, S., 1997. Regional Trading Blocs in the World Economic System, Institute for International Economics, Washington, D.C. Gary King; Will Lowe, 2003, "10 Million International Dyadic Events", http://hdl.handle.net/1902.1/fyxlawzria UNF:3:dSE0bsQK2o6xXlxeaDEhcg== IQSS Dataverse Network [Distributor] V3 [Version] Kalirajan, K.P. and Findlay C.C., 2005. Estimating potential trade using gravity models: a suggested methodology, Foundation for Advanced Studies on International Development, Tokyo. Kang H. and Fratianni, M., 2006. International trade efficiency, the gravity equation, and the stochastic frontier, Working Papers 2006-08, Indiana University, Kelley School of Business, Department of Business Economics and Public Policy. Kemal, A. R., M. K. Abbas, and U. Qadir. 2002. A Plan to Strengthen Regional Trade and Cooperation in South Asia. In Trade, Finance and Investment in South Asia, ed. T. N. Srinivasan. New Delhi: Social Science Press Khan, M. 2009. "India Pakistan Trade: A Roadmap for Enhancing Economic Relations," Policy Briefs PB09 15, Peterson Institute for International Economics. Kumar, R and Singh, M, (2009) India s Role in South Asia Trade and Investment Integration, ADB Working Paper Series on Regional Economic Integration, No. 32, July. Mansfield, E.D. and Pollins, B.M., 2003. Economic Interdependence and International Conflict: New Perspectives on an Enduring Debate (eds), University of Michigan Press, Michigan. Polachek, S.W., 1980. Conflict and trade, Journal of Conflict Resolution, 24:55 78. Raseen Sally (2011, PAFTAD): Reuveny, R. and Kang, H., 1996. International trade, political conflict/cooperation, and Granger Causality, American Journal of Political Science, 40:943 970. Reuveny, R. and Kang, H., 1998. Bilateral trade and political conflict/cooperation: do goods matter?, Journal of Peace Research, 35:581 602. Taneja, N. 2006. India Pakistan Trade. Working Paper 182. New Delhi: Indian Council for Research on International Economic Relations. 20