Central Asia Journal No. 66

Analyzing Pak-Iran Trade Dynamics in ECO (Economic Cooperation Organization) Region Using Gravity Model

Dr. Jahangir Achakzai*
                                                            .
Abstract

Using an international database on bilateral trade for 137 countries in 2005, a gravity model was estimated to address the question of whether Iran is a potential country for Pakistan’s trade and that whether country got lower share than its potential in bilateral trade. The results from the gravity model suggest that Iran is a potential country for Pakistan’s exports. It supports the hypothesis that bilateral trade is low and both countries trade less with each other than what is predicted by the model. As Iran has a common geographical border with Pakistan, and both are also member of ECO (Economic Cooperation Organization) the privilege of geography and the existence of trade preferences between the two countries could be expanded to cover potential trade between the two countries.

Introduction

Pak-Iran Relations

Iran situated in the west of Pakistan is characterized by common historical, cultural and economic affinities. It is not only geographical proximity that binds the two countries, but a deeper basis for the relationship is provided by their shared faith and belief in the eternal values of Islam. The affinities of sentiment, policy and stand between the two countries are reflected in their working hand in hand in the organizations of the Islamic Conference, United Nations and the Non Aligned Movement. Close ties between the Muslims of the sub continent and the people of Iran exist since centuries. After independence of Pakistan these ties were strengthened more. Iran lent its full, unconditional moral and material support to Pakistan, and also stood by the side of Pakistan in the hours of need. This deep-rooted bond of friendship between Pakistani and Iranian people remains stable and in fact has continued to grow. During the recent few-years both the countries have come closer to each other and are cooperating in the industrial, cultural and economic sectors.

Pak-Iran Trade Relations

Pakistan’s trade relations with Iran are very old. Both the neighboring countries had been trading with each other on the basis of barter trade. As part of their endeavors to expand trade relations the two countries have been examining the possibilities of special trade agreements. In order to achieve the objective, in 1964 Regional Cooperation for Development (RCD) was created. Under this treaty several projects have been completed. Later on, the regional integration was further strengthened by converting RCD into ECO (Economic Cooperation Organization). This step proved as a source for institutionalizing the centuries old relations. Under ECO several bilateral trade agreements have been signed which will increase the size of trade between these two brotherly nations. Although, the economic ties between Pakistan and Iran date back to centuries, the progress achieved in this respect remained marginal because of different political and economic reasons. Both the countries have the potential to cater for the demands of each other and trade in the products which are being imported from rest of the world. Keeping in view the above facts the present research paper is designed to chalk out the future prospects of trade between the two countries.

Empirical Analysis

Theoretical Basis of Gravity Model

The gravity model has been used frequently to analyze bilateral trade flows between countries . The study of the literature shows that Anderson (1979) attempted to provide a theoretical basis for gravity model. Anderson and Wincoop (2003) draw it from a model of monopolistic competition in differentiated products and Helpman et al. (2004) obtained it from a theoretical model of international trade in differentiated goods with firm heterogeneity.  
The important contribution of Anderson and Wincoop's paper has been to highlight that controlling for relative trade costs is crucial for a well-specified gravity model. Their theoretical results show that bilateral trade is determined by relative trade costs. In terms of the empirical gravity model this implies that after controlling for country size and bilateral distance, trade will be higher between country pairs that are far from the rest of the world than between country pairs that are close to the rest of the world. 
Another recent study that applies the gravity model to analyze regional integration agreements is by Frankel (1996). He estimates a gravity model using a sample of 63 countries for various years between 1965 and 1992. In its basic form, Frankel's model includes dummies for adjacency, common language and the traditional bloc dummies. 
His general conclusion is that the new wave of regionalism has resulted in a significant concentration of trade within different blocs all over the world. Even after holding constant for such natural determinants of bilateral trade such as size and distance, intra-regional concentrations of trade are appearing in various parts of the world .
While the model has done quite well in describing bilateral trade patterns, it has also been subject to criticism for its lack of theoretical foundation. The model, accused in the past of lacking theoretical foundations, has regained respectability and is now accepted as a well grounded tool to analyze bilateral trade flows . However, as Helliwell (1998) observes, the model in recent years has undergone a transformation from being "a theoretical orphan" to a model that can be derived from standard trade theories. Deardorff (1995) shows the consistency of the ­gravity model with the Heckscher-Ohlin theory of trade (both with frictionless and impeded trade) and Helpman (1984) and Bergstrand (1985) derive the model from theories of trade based on differentiated products.

The Model

The gravity model of trade has been widely used to examine trade flows between the nations . According to the model, there exists a positive relationship between the volume of trade and size of the economies (GDP) and inverse relationship of volume of trade with distance between the countries . 
The gravity model of trade states that trade between two countries is proportional to the product of their GDP and inversely related to the transportation cost. The model is further extended by adding dummy variables like common borders, language and membership of the common trading bloc.
Formally, the gravity equation can be written as:

Tij  =  β0 + β1* GDPi + β2* GDPj + β3* PCIi + β4* PCIj+  β5* DISTij

Where Tij is the trade volume between countries i and j, GDP and PCI are the respective gross domestic product and per capita income and DIST is the distance between them.
We expect trade to be positively affected by economic size (GDP) and negatively related to distance (DIST). The coefficients on per capita income (PCI)could be positive or negative . Since trade is expected to increase with the size of domestic economy (GDP), the expected sign of β1 is positive. Distance, in turn, may be seen as a general proxy for the costs of trade. To the extent that neighbouring countries can be expected to share many cultural traits, and that information from across the border is typically more readily available, a dummy for common borders, or adjacency, is normally also included in the gravity equation.
Finally, and for the same reasons, the sharing of a common language should also be included.
The Dummy for common language is used to capture the information costs.  Such costs are probably lower in trade between countries who share common language, whose business practices, competitiveness and delivery reliability are well known to one another.  Firms in countries with a common language or other relevant cultural features are likely to know more about each other and understand each other’s business practices better than firms operating in less similar environments. For this reason, firms are more likely to search for suppliers or customers in countries having common language where the business environment is familiar to them. For instance businessmen and traders in Pakistan feel more comfortable in dealing with Afghan businessmen speaking the same language Pashto than the Uzbak traders speaking different language.
Once all the above factors are considered, it is possible to assess whether a formal trade agreement is being effective or not in concentrating trade among its members. To this purpose, dummy variables of bloc membership are added to the basic equation. If bilateral trade exceeds (or lies below) the 'normal' levels of trade (normality being defined as the sample's average bilateral trade flows) the abnormality will be picked up by the bloc variables.

Estimation of the reference model

The Gravity equation used in the analysis is as follows:
 
ln(Tij)  =  β0+ β1 ln(GDPi)+ β2 ln(GDPj)+ β3 ln(PCIi)+ β4 ln(PCIj)+  β5 ln(DISTij)+ β6 (ADJij)+ β7 (LANGij)+ β8 (ECO)

Where ln is the Natural Log. Tij is the trade between country i and country j. GDPi  and GDPj is Gross Domestic Product of country i and j respectively. While Per Capita Income of the two countries are represented by PCIi and PCIj. Country i’s distance from that of j is denoted by DISTij. The dummy variable for common borders is ADJ, which is 1 if two countries have common border and 0 if they do not share common border. LANGij is the dummy variable for common langue. LANGij is 1 in the case where the two countries speak common langue and zero otherwise. ECO is the dummy for regional bloc between the countries. It takes the value of 1 when the two countries i and j are member of regional bloc and zero if not.
The model was estimated using the weighted least squares (WLS) technique. The technique was used to take into account the presence of hetroskedasticity. Hetroskedasticity arises when the variance of the error terms is not constant over all observations. In its presence the OLS estimators are unbiased and consistent, but not efficient estimators of the true variance of the estimated parameters. If this is the case, the statistical tests given by the OLS estimation are incorrect. Since one knows a priori  that hetroskedasticity, if present, will be related with the size of the countries, the appropriate correction for it is the use of the weighted least squares (WLS) technique, using as weights a measure of the size of countries. This is the procedure that was followed here, using as weights the logarithm of the exporter’s GDP.

The Dependent Variable

The size of a trade flow can be measured in two ways: at the point of export or at the point of import. Apart from the well-known differences in valuation - and apart from minor differences due to the time-lags between the recording of exports by the exporting country and the recording of the same flow as an import by the importing country, these two measurements normally gives the same results. This analysis uses mostly export data, most of them obtained from the UN COMTRADE database.

The Treatment of Zeros

The data on bilateral trade flows is bound to show some zeros. These may reflect either the absence of trade, or simply the presence of very small amounts of trade that for statistical reasons are reported as zeros. In order to deal with this problem the widely used method is simply to exclude the zero entries from the sample and estimate it with OLS . This is the simplest of all methods and has been widely used in previous estimations of the gravity model. This is the method followed here.

Data

This analysis uses mostly export data of 137 countries for the year 2005.  The data was taken from UN COMTRADE database. GDP and PC GDP figures were taken from World Bank’s (2005) World Development Indicators. The French government Centre for Exploratory studies and International Information data was used for distance between capital cities, countries sharing common borders and common languages.

Pak-Iran Trade

Before going through the gravity model results for the bloc coefficient it is important to have a look at the Pak-Iran trade figures. It is on the basis of these figures that the public opinion, and also decision makers, appraise the relative success of trade negotiations: If trade is growing among the countries, then it must be because of the negotiations. If it lags, the negotiations went wrong. Intuitively such qualifications make sense. After all, trade negotiations are undertaken so as to foster trade among a group of nations. If this trade grows, what other indicator of success does one need? The problem with intuition, however, is that it seldom tells the whole story and, by omitting important variables, it may actually tell the wrong one. That is why it is interesting to compare what intuition tells and what the gravity model does.
Figure – 1 shows Pak-Iran bilateral trade during the period 1996-2005.
                            
Fig. 1. Pak-Iran Bilateral Trade during 1996-2005 (Mln US $)
The figure in general shows an increasing trend during the ten years period of 1996-2005, except the years 1997, 1998 and 2000. After a fall of the country's trade with Iran in 1998, when Pakistan met serious economic problems as a result of nuclear explosion, political instability and a severe draught, the trade started to grow more rapidly from 2000 onward. There was a sharp increase in the imports from Iran to Pakistan during the year 1999 reporting the highest rise in the imports from Iran during the ten years period. In 2005 the bilateral trade was 638 million US dollars showing more than a hundred percent rises in Pakistan's trade with Iran compared to the year 1996 figure of 302 million dollars.

Estimation Results

According to the results of the model, the level of significance for the standard gravity variables like GDP, GDP Per Capita and Distance is 5%, which is highly significant. The same significance level 5% is reported by the model in case of common border and language variables. The model registers the expected signs for all the variables.

Table 1.    Estimation Results of the Gravity Equation

In Table 1 total exports are taken as dependant variable. Weighted Least Square (WLS) method is used to get the results where weight is the Log (GDPi).
The results of the model are quite favorable in comparison with previous studies. It states that with the rise in both domestic and foreign GDP as well as Per Capita Income trade also increases. While in case of distance, as expected, the result was otherwise.
The significant values of the coefficients for GDP validate that there exist a positive relationship between international trade and income of the trading partners. When there is a rise in income the international trade among the nations also increases.
The model reports estimated coefficient of 1.1 in the case of log of the exporting country GDP, showing that there was a 1.1% increase in trade as GDP increased by 1%. While 0.86% increase was registered by trade when GDP increased by 1% in the case of importing country where the coefficient is 0.86.
The statistically significant values of the GDP per capita coefficient shows that the richer the country the more would be its trade and vice versa.
By comparing the results of the present study with that of the previous ones in the case of GDP and per capita GDP more or less the same figures are reported. For instance, a study "Asian regionalism and its effects on trade" conducted by Clarete, Edmonds, and Seddon in the year 2002, reports the same coefficients 1.1 in case of the exporting country’s GDP and 0.8 regarding the GDP of importing country.
Another study by Frankel, upon analyzing the data for sixty three countries in 1996 reports the value of 0.93 for the coefficient of GNP and 0.13 for per capita GNP.
Regarding the value of the coefficient of log of distance, the results registered by the model were higher in comparison to other studies. It was -1.27 showing that trade between the two countries fell by 1.27% whenever their distance increases by 1 %. The value of the distance coefficient is large, reflecting that transportation and communication among most member countries are generally more costly and act as a significant barrier to trade. This heavy impact of distance is in line with findings by other studies. Frankel (1996), with a sample of 63 countries between 1965 and 1992 finds a coefficient for distance of -0.77 in the year 1992. Baldwin (1994), controlling for adjacency and using the great circle distance between capitals finds coefficients -0.88.Bikker (1987), without controlling for adjacency  reports a coefficient for distance of  -1.1. And Boisso and Farrantino (1995) finds coefficient -1.5 without controlling for adjacency.
The figure for coefficient of adjacency dummy was 1.06 showing that two countries sharing a common border were trading about 190% more compared to the nations with no common border. 
The result reported by the model in the case of common language dummy clearly depicts the impact of this important dummy upon trade. The coefficient value of the common language dummy was registered as 0.92.
As regards the estimated value of the bloc coefficient of ECO, the model registers a statistically significant value of 1.1, showing that holding the values of GDP, proximity, and other gravity variables constant, two members states of ECO regional bloc were trading 210 percent more among themselves, than two otherwise similar countries would [exp(1.1) = 3.1].
When the values of bloc dummy of the present study are compared with the previous study of Clarete, Edmonds, and Seddon, which records the figure of 1.7 for ECO dummy coefficient, it strongly validates the result of the present study. Their findings pinpointed that countries in ECO bloc tended to trade more intensely among themselves at the expense of trade with the rest of the world. Their results further prove that trade in ECO regional area was higher in 1995 and 2000 than would be expected if the countries were not members of ECO.

Predicted Trade of Pakistan with Iran

In this part of the paper, an attempt is made to predict Pakistan's exports to Iran. For this purpose we use the gravity equation estimates generated by the model. The trade volumes predicted by the model are then compared with the actual trade volumes of the member countries.
The following equation is used to predict Pakistan's export to Iran. 

ln(Xij)  =  -27.93 + 1.1ln(GDPi) + 0.86ln(GDPj) + 0.08 ln(PCIi) + 0.08 ln(PCIj) -1.27ln (DISTij) + 1.06(ADJij) + 0.92(LANGij) + 1.13(ECO)

In order to predict Pakistan’s export to Iran in particular and other member countries of ECO bloc in general we insert the values of GDP, PCI etc in the gravity equation to get “normal” trade flows. This will provide us an indication of the predicted trade volumes which prevail between these countries.

Table 2.    Predicted exports of Pakistan with Iran and eight member states of ECO bloc.
(Million US$)

As can be seen from Table 2, Pakistan's actual exports to ECO member countries were below the levels predicted by the model in all but except one of the cases examined. The exception is found for Pakistan's exportsto Turkey, where the actual level is 12 percent higher than the predicted value for 2005. While on the other extreme, in case of Tajikistan the exports are only 3 percent of the predicted value and still 97 percent potential exist in case of Pakistan's exports to that country. Afghanistan, being the second biggest market for Pakistan's exports after Turkey and having common border with Pakistan broadly matches the predicted value. The country received 97 percent of the exports which the model predicts for it. Among the Central Asian countries, Azerbaijan is the major market for Pakistan's exports during the year 2005 which meets 20 percent of the predicted exports.
In the case of Iran, the country being a close neighbor of Pakistan, hardly matches ten percent of the potential exports predicted by the gravity model of trade. The results clearly indicate that there is considerable scope for an increase in Pakistan's exports to Iran.

Conclusions

The gravity model of trade was used to compute Pakistan’s predicted trade with Iran and other member countries of ECO bloc. The estimated result shows that Iran and all the six Central Asian States have come out as the potential countries for bilateral trade of Pakistan. They are receiving less than fifteen percent of Pakistan's bilateral exports, except in the case of Azerbaijan which receives 20 percent of the exports. While the remaining two members like Afghanistan and Turkey fully matches the potential of bilateral trade.
The elasticity estimates produced by the model clearly show that Pakistan’s exports to Iran at present are only one tenth of the bilateral trade potential. Hardly ten percent trade is taking place between the two neighboring countries and a large volume of bilateral trade is waiting to be exhausted yet. The size of trade predicted by the model in comparison to actual trade is very large showing that Iran can prove a potential country for Pakistan’s exports in future. It also supports the question posed at the beginning of the paper that Pak-Iran bilateral trade is low and both countries are at present trading less with each other than their inherent potential. As both the countries share common geographical border with each other, and are also member of the same regional bloc of ECO, the privilege of geography and the existence of trade preferences between them could be expanded to cover potential trade between the two countries.

References

Anderson, J. and E. van Wincoop (2003), ‘Gravity with Gravitas: A Solution to a Border Puzzle', American Economic Review, Vol. 93, No. 1, pp. 170-192.
Baldwin, R. (1993) 'A Domino Theory of Regionalism', NBER Working Paper4465, Cambridge, Mass.: NBER.
Bank Markazai Jomhouri Islami, Iran, Annual Review (various issues), Tehran,
Bank Markazi Jomhouri Islami, Iran, Economic trends (various issues), Tehran,
Bhagwati, J. (1992) 'Regionalism and Multilateralism: An Overview', in de Melo and Panagariya (eds.) New Dimensions in Regional Integration, CEPR, New York, Cambridge University Press.
Bhagwati, J. and A. Panag ariya (1996) 'Preferential Trade Areas and Multilateralism: Strangers, Friends or Foes?' in Bhagwati, J. and A. Panagariya (eds) The Economics of Preferential Trade Agreements, AEI Press, Washington
Brada, J.C and   J.A. Mendez (1985) 'Economic Integration Among Developed, Developing and Central Planned Economies: A Comparative Analysis',
Brown, D. K., Deardorff, A. V. and Stern, R. M. (2003) 'Multilateral, Regional and Bilateral Trade Policy options for US and Japan' The World Economy 26:6, 803-828.
Clarete, Edmonds, and Seddon (2002) 'Asian regionalism and its effects on trade in the 1980s and 1990s' Asian Development Bank.
Deardoff, A. (1995) 'Determinants of Bilateral Trade: Does Gravity Work in a Classical World?' NBER Working  paper 5377.
Dee P. and J. Gali (2003) 'The Trade and Investment Effects of Preferential Trading Agreements', NBER Working Paper 10160.
De Benedictis, L. and C. Vicarelli (2005), 'Trade potential in gravity panel data models', Topics in Economic Analysis and Policy (forthcoming).
Economic Cooperation Organization (ECO) Secretariat database, (2006). Theren Iran.
Egger, P. (2004), 'Estimating Regional Trading Bloc Effects with Panel Data', Weltwirtschaftkliches Archive/Review of World Economics, Vol. 140, No. 1.
Export Promotion Bureau 'Foreign Trade of Pakistan' (2004-05) Government of Pakistan
Federal Bureau of Statistics 'Pakistan Statistical Year Book', (Various Issues), Government of Pakistan. Islamabad, Pakistan.
Fernandez, R. (1997) 'Returns to Regionalism: An Evaluation of Non­ Traditional Gains from Regional Trade Agreements', The World Bank, Policy Research Working Paper 1816.
Finance Division, Economic Advisor’s Wing, 'Economic survey' (2004-05) Government of Pakistan, Islamabad. Pakistan.
Frankel, J.A. (1996) 'Regional Trading Blocs in the World Economic System', Institute for International Economics, Washington.
French Centre for Exploratory studies and International Information(Le CEPII, Centre d'Etudes Prospectives et d'Informations Internationales). Georges Pitard – 75740 Paris Cedex 15
Helpman E., M. Melitz and Y. Rubinstein (2004) 'Trading Partners and Trading Volumes',
Institute of Regional Studies (1995) 'Pakistan and Regional Economic Cooperation in SAARC – ECO'. Islamabad, Pakistan.
Joseph Frankel, (1964) 'International Relations', (New York: Oxford University Press, 1964).
Kamal A. R. (1995) 'Significance of SARRC and ECO for Pakistan' Institute of Regional Studies Islamabad. Pakistan.
Khan, Ashfaque H.(1995) ' Intraregional Trade in South Asia Problems and Prospects ' Institute of Regional Studies Islamabad. Pakistan.
Krugman, P. (1992)'Regionalism vs. Multilateralism: Analytical Notes', in de Melo and Panagariya (eds.), New Dimensions in Regional Integration, Center for Economic Policy Research, New York, Cambridge University Press.
Maqbool A. Bhatty, (1992) 'ECO Challenge and Opportunities in the 1990’s', Strategic Studies No. 2: 16.
Maqbool Ahmad Bhatty, 'Pakistan and ECO', Regional Studies (Winter 1995), p. 103.
Mukherjee Reed, summer (1997) 'Regionalism in South Asia: Theory & Praxis', Pacific Affairs, Vol.70, no 2, pp.235-251.
Naseem Zehra,(1992) 'ECO challenges and opportunities in the 1990’s', Strategic Studies No 2 (1992) : 31.
Roberta Piermartini and Robert Teh (2006) Economic Research and Statistics Division
World Trade Organization Centre William Rappard Geneva.
Saeed Khawaja Amjad. (2005)  'Economy of Pakistan'. Lahore, Pakistan.
Shahi, Agha. (1992) ' The ECO, prospects and problems’, Khalij Times.
Soloaga I, and A. Winter s, (1999) 'Regionalism in the Nineties: What Effect on Trade?' The World Bank, Policy Research Working Paper 2156.
Sayed Rifaat Hussain, 'Political Economy of Pakistan’s Relations with Central Asian Republics, ' Regional Studies (Autumn 1992), p. 4.
UN Commodity Trade Statistics Database (COMTRADE), United Nations Statistics Division.
World Bank, (2005) World Development Indicators, Washington, D. C.
WTO (1995) 'Regionalism and the World Trading System' WTO, Geneva.