The Impact of Emigration on Bilateral Trade Flows of Bangladesh

2nd International Conference on Humanities, Geography and Economics (ICHGE'2012) Singapore April 28-29, 2012 The Impact of Emigration on Bilateral Tr...
Author: Alisha Murphy
3 downloads 0 Views 933KB Size
2nd International Conference on Humanities, Geography and Economics (ICHGE'2012) Singapore April 28-29, 2012

The Impact of Emigration on Bilateral Trade Flows of Bangladesh Samai Haider 

that immigrants as well emigrants can potentially influence the bilateral trade between two partner countries. In a country such as Bangladesh, where the phenomenon of emigration is rampant, the expatriates abroad act as the major source of networking links for the country. The network theory hypothesizes that migrants promote bilateral trade flow between the country of destination and the country of origin. Reference [5] captures this occurrence with the term ‘network effect’, which functions through two main channels – the information effect and the preference effect. The information effect works through controlling the expenses of collecting information and staving off the uncertainty that is characteristic of international markets. When transportation and transaction costs appear to be major impediments to trade, migrants can act as trade intermediaries, by exploiting these lower costs. Emigrants utilize their contacts in the country of origin, and are better able to understand and provide knowledge about conditions in the source country. These range from information about the commercial situation to insights regarding the business and bureaucratic environments. Consequently, it is more expedient to enter into trading arrangements between the destination country and the country of origin, and emigrants are likely to trade more than their nonmigrant counterparts. The preference effect exists due to emigrants having a certain demand for home country goods – products that they are familiar with and have a preference for. Consequently, the source country is expected to experience an outflow of exports to the country of destination of the emigrants. This network theory is of paramount importance, and is supported by previous empirical studies. Amongst the supporting literature, [4], while studying international trade in differentiated products, asserts that limited or no information regarding prices prevents traders from carrying out their transactions optimally in international markets. The network theory view of trade is helpful in explaining certain micro-institutional features of trade. Focusing on this aspect of trade paves the way for study in the area of personal contacts and relationship building in determining the geographic distribution of economic activity. This paper uses the gravity equation to describe bilateral trade patterns, as it relates trade between two trading partners positively to both their incomes and negatively to the distance between them. Using this conceptual underpinning, the focus will be on the emigration of Bangladeshi workers, and its implications on bilateral trade flows. The following questions will be addressed in this paper: Do emigrant links to the country of origin, in this case Bangladesh, enhance bilateral

Abstract— The links between migration and bilateral trade flows have been explored in the past, however, empirical literature has little to offer in the field of emigration and bilateral trade flows. This paper uses a gravity model to study the impact of emigration on the bilateral trade flows of Bangladesh. It further examines whether religious ties bolster bilateral trade flows between countries. This study finds that while there is a clear and significant link between emigration and enhanced bilateral trade flows between Bangladesh and its trading partners, religion plays a limited role in determining trade flows. This paper also finds that the link between emigration and bilateral trade flows is particularly strong between Bangladesh and its Middle-Eastern and Asian trading partners, and hence explores emigration oriented policies which target the development of exportable manpower that are in particular demand in these nations. Keywords—Bilateral trade, Gravity model, International migration, Panel Data I. INTRODUCTION

A

migration becomes an increasingly common phenomenon, the concept of networking also comes into the limelight. The theory behind migration and networking is that social and business networks are formed by migrants who can help to enrich the information and knowledge stock, thereby eliminating informational barriers to trade. Migrant links can include knowledge of home country markets, language, preferences and business contracts. Migrant links suggest a “beneficial human-capital-type externality that enhances trading opportunities,” [1]. The impact of migration on bilateral trade flows has been explored at length in empirical economics. Reference [1] has studied the effect of immigrants on trade flows for the United States; [2] for Canada; [3] for the United Kingdom; and [4] for the Chinese community. All studies bear proof that immigration and the network theory is capable of inducing pro-trade effects, boosting bilateral trade flows. However, most papers have focused on the incident of immigration. The only exception to this is the study carried out by [5] which uses both immigration and emigration data from Italy. It is assumed S

Samai Haider undertook this study in the University of Essex, U.K. She was with HM Treasury, U.K. and is now with the Department of Education and Early Childhood Development, Melbourne 3002, Australia. (e-mail: [email protected]).

12

2nd International Conference on Humanities, Geography and Economics (ICHGE'2012) Singapore April 28-29, 2012

IOM shows that today there are up to 192 million migrants, which is 2.9% of the world population [6]. Its effects are expected to have great positive impacts on poverty levels, and also improve efficiency. The gains from migration are plentiful. Looking at it from an emigrant country’s perspective, the phenomenon not only reduces unemployment, but also boosts economic growth as it opens the doors to strategic inputs such as remittances and returning skills. The migrant workers have the added advantage of earning higher wages, and increasing their productivity in the capital intensive countries that they are immigrating to. This leads to the possibility of wages falling in the receiving countries, while wages in the sending countries experience an upward trend. As a result, labour costs balance out, which conforms to the standard theory of factor price equalization.

trade flows between home country and destination countries? Do religious ties play a role in boosting trade between the countries in question? Should Bangladesh actively seek out policies to facilitate emigration or should it remain passive? These issues are important in assessing the present and future consequences of emigration, a common phenomenon in a developing country like Bangladesh. This paper first looks at the trend in migration in Bangladesh and its trade history to ascertain the gains from emigration. It then outlines the theory, model and estimation methods before analyzing the results and implications. The paper then considers the policy options for Bangladesh. II. TRADE AND MIGRATION IN BANGLADESH A. Trends in Migration: The Bangladesh Context The overall trend in Bangladesh has seen a steady emigration of people driven out by political and economic problems. A large number of Bangladeshis migrated to other parts of the world, the Middle East in particular, in search of employment. The International Organisation for Migration [6] identifies the push pull factors that dictate migration. South Asian countries are riddled with push factors such as rapid population growth, inadequate economic opportunities and high unemployment rates that force people to emigrate. A labour intensive country, Bangladesh has considerable comparative advantage in the export of workers. Bangladesh falls amongst the nine largest Asian immigrant exporting countries. 3.98 million Bangladeshis went overseas with temporary work permits to provide services in the Middle-east, Asia, Africa and Europe during the period 1976 to 2004. In 1976 the official number of migrants going overseas was only 6087. The total number of migrant workers rocketed to 272958 in 2004. Currently, a considerable number of workers are relocating abroad, the figure standing at 331063 per year.1 The major recipients of this mass exodus of migrant workers are the Middle Eastern countries, and this has been the case since Bangladesh’s independence. The data also shows that Bangladesh exported 122252 unskilled workers in 2004, followed by 110177 skilled workers in the same year. The trend for Bangladesh so far shows that the volume of unskilled workers who relocate far surpass the number of skilled or semi skilled migrants. It has been established that owing to an abundance of labour in Bangladesh, the government encourages the export of manpower to foreign countries. Consequently it ensures that workers are provided with sufficient education and training so as to make them eligible for work abroad.

C. Trade in Bangladesh: A History Foreign trade is of utmost importance to the economy of Bangladesh, a reality that manifests itself in the financial year 1991 – 1992. Foreign trade’s contribution to government revenue exceeded thirty seven percent while export industries' share of employment in the manufacturing sector was sixty percent. In the first half of the 1990s, Bangladesh embarked on a hasty liberalization process. The country recorded one of the most rapid reductions in tariffs, as measured by the ratio of post-reform average tariffs to pre-reform rates, amongst a set of world economies that had undertaken trade liberalization measures (World Bank 1999). However, since the late 1990s, the pace of tariff liberalization in Bangladesh has slowed down moderately. Reference [7] pinpointed that trade liberalization policies pursued by Bangladesh over the last two decades have resulted in significant changes in the country’s trade policy regime, including the lowering of customs duty rate from 350 percent in 1992 to 37.5 percent in 2000, along with a drop in unweighted average tariff rates from 114 percent to 22 percent within a decade. There has also been a significant reduction of the number of commodities under Quantitative Restrictions (QR). In 1987 the number of commodities under the four-digit code subject to QRs was 550, which declined to 124 under the Import Policy of 1997-2002. In addition, there have also been moves towards a more market-determined exchange rate regime. Further, Bangladesh’s effort towards unilateral trade liberalization included the adoption of different export promotion measures, aiming to diversify exports, improve quality, encourage higher value added, and develop industries through backward linkages. Bangladesh has recognised the significance of trade liberalization and is actively seeking participation in regional Free Trade Associations which could be used as stepping stones for further multilateral liberalization. The benefits of increased trade to the country are multifaceted and Bangladesh realises the importance of taking steps to ensure increased trade, especially in promoting their exports.

B. Gains from Emigration Bangladesh is afflicted with the distinct developing country characteristics of rapid population growth, inadequate economic opportunities and high unemployment rates. These epitomise the pull factors which fuel migration. Recent data by 1

Data is in absolute values. Data source: Bureau of Manpower, Employment and Training (www.bmet.org.bd) 13

2nd International Conference on Humanities, Geography and Economics (ICHGE'2012) Singapore April 28-29, 2012

study. The limitations with regards to the distance variable, is that it changes across countries but not across time, so there could be some multicollinearity. In this case, it could cancel out the between country heterogeneity in trade and migration. The small sample size of a twenty five countries may also lead to unrealistic observations.

III. METHODOLOGY A. Theory and Model This section examines the model that will display the mechanisms of the impact of emigration on trade. The gravity model is used in empirical economics to predict how countries compare with regards to international trade. It predicts that the volume of trade between two countries will be directly proportional to the product of their economic masses (as measured by GDP) and inversely proportional to the distance between them. Borrowing the notation used in [8], where, Vij is the bilateral volume of trade between countries i and j, si and sj are the shares of the countries in world spending. Distance may be used to capture the cost of trading while the GDP serves to explain the differences in demand and supply. Thus, founded on previous empirical works, it is possible to develop the following equation for this particular study, augmented by other variables that are related to bilateral volume of trade: Vit = GDP it β1DISTANCE it β2 EMIGRANT it β3 β2 RELIGION it β3 + εk

B. Estimation Method The primary focus of this empirical analysis is to examine emigrant-link effects on Bangladeshi bilateral trade flows both over time and across countries, thus requiring a panel data analysis. The dataset has been divided in two cases where, Case 1 is a pooled Ordinary Least Squares (OLS) regression model. In this case it has been assumed that all the countries in the study have the same slope and the same intercept. Case 2 is country fixed-effects model. The empirical testing begins by examining the main hypothesis - emigrant networks have pro-trade effects. This has been done using the pooled OLS estimation, as carried out in the study by [5]. Bangladesh has been a predominantly active trading partner of the twenty five countries included in this study, a phenomenon that could render country or temporal effect statistically insignificant. Consequently, the need to use a Tobit estimation, as carried out in [2] are eliminated. Thus, equation (1) will be estimated by the Ordinary Least Squares (OLS) method. The expected signs of the coefficients in the gravity model laid out in equation (1) are as follows: GDP - the size of the economy is expected to have a positive effect on bilateral trade; EMIGRANT: In the event that the network effect comes into play and there is a positive relationship between migration and bilateral trade flows, the coefficient should have a positive sign. Emigrant preferences can have a positive impact on the value of bilateral trade flows, via the preference effect or the information effect3. DISTANCE - Distance is expected to have a negative effect on trade, as it poses as a barrier to trade and involves transportation costs. As a result, the overall transaction costs of bilateral trade increase with distance. Case 2, or a country fixed effects model, has constant slopes but intercepts that differ according to cross-sectional, i.e. country. While there are no significant temporal effects, there are significant differences among countries in this type of model. Fixed effects regression is the model to use when it is necessary to control for omitted variables that differ between cases but are constant over time. It allows for changes in the variables over time to estimate the effects of the independent variables on the dependent variable, and usually is the main technique used for analysis of panel data. Fixed-effects is an appropriate specification if the focus is on a specific set of countries e.g. Middle Eastern countries. In order to test for fixed effects, the pooled regression model is used as the baseline for comparison. The country effects are first tested with an F test. To test for country effects the null hypothesis is that all dummy parameters except one are zero. This is done by the F-Test, which is based on loss of

(1)

Where i = 1,…, 25 countries, t = 1980, …2004 years. The dependent variable, Vit represents bilateral trade flows between Bangladesh and its trading partners. Similar to Marina and Pistori (2006), this is the total volume of Bangladeshi exports or imports with its trading partners. The data on bilateral trade flows has been obtained from the CEPII databases available online. The exogenous variables in the model are explained as follows: GDP -the trading partner’s Gross Domestic Product. This is a proxy for the size of the economy. This study uses GDP (constant 1995 U.S. $) as a proxy, extracted from the World Bank’s World Development Indicators 2004; DISTANCE - the distance (in kilometres) between Dhaka, the capital of Bangladesh and the capital city of its trading partner, accumulated from the CEPII databases; EMIGRANT- the stock of emigrants from Bangladesh to its trading partner. The data has been collected from the Bangladesh Bureau of Manpower, Employment & Training (BMET). One of the key contributions of this paper is to study the effect of RELIGION, a dummy variable that takes the value of 1 when it is an Islamic nation and 0 otherwise. This variable has not been previously explored in empirical literature, and given the fact that Bangladesh has been engaged in significant trade and emigration with other Islamic states, it is a relevant factor to take into account in studying whether religion plays a significant role in determining emigration and bilateral trade flows. Countries are deemed as an Islamic state if the majority of its population is Muslim2. A list of the 15 Islamic nations used in this study is available in Table 1 of the Appendix. Other variables commonly used in gravity models, such as language, colonial ties etc. have been excluded from this study as it does not apply to the Bangladesh context. Due to limited data availability, the dataset ranges from 1980 to 2004, thereby limiting a recent and comprehensive

3 2

The data has been collated from the CIA World Factbook (2008). 14

This study will not examine the two effects separately.

2nd International Conference on Humanities, Geography and Economics (ICHGE'2012) Singapore April 28-29, 2012

goodness-of-fit. If the null hypothesis is rejected, you may conclude that the fixed group effect model is better than the pooled OLS model. Unlike reference [5] this study includes a lagged dependent variable amongst its regressors. Borrowing from [2] and [1], and given that this study uses data over twenty four years, it is viable to take into account a lag in emigrants’ potential decision and production lags. Thus, a lag of emigrants has also been used. In order to examine how emigration and bilateral trade flows between Bangladesh and other predominantly Muslim states eventuates. This study runs a separate regression on the 15 Islamic trading partners. The 25 countries in this study have been further segregated by regions, Middle East, Asia, Africa and Europe, (a complete list of the countries include in this study is laid out in table II of the appendix), in order to examine which regions Bangladeshi migrants have the most impact on. As there happens to be extraordinary levels of migrants moving to Middle Eastern countries, a set of regressions are run based on whether it is a Middle Eastern country or not. Each region has been assigned as a dummy variable, where if the country is a part of the continent it is assigned a value of one, and zero otherwise. Separate pooled OLS and country Fixed Effects regressions, with and without lagged emigrant variables are run in order to gauge the impact of emigrants on bilateral trade flow. As there is no data showing emigrants to the American and Australasian continents, those regions have been excluded.

TABLE I REGRESSION RESULTS FOR ALL 25 COUNTRIES Panel A: Regression Results without Religion (Dependent Variable= Vit, Bilateral Trade Flows) Number of Observations: 625 Panel Cases Case 1

Constant

GDP

Distance

Emigrant

R2

Specification Test

2.38 (6.50)

2.13 (4.45)

-0.14 (-2.79)

0.36 ( 2.07)

0.54

-

Case 2

3.49 (11.62)

1.19 (3.28)

dropped

0.52 (2.17)

0.23

F= 0.69 P value

Suggest Documents