留学生Essay代写|贸易开放度及其对经济增长的影响

留学生Essay代写|贸易开放度及其对经济增长的影响
TRADE OPENNESS AND ITS EFFECTS ON ECONOMIC GROWTH

选择南亚国家的贸易开放度及其对经济增长的影响：一个面板数据研究
研究了贸易开放度与时间1972-1985和检查方案1986-2007年前南亚区域合作联盟实施后四南亚国家的经济增长之间的因果关系。面板协整和FMOLS方法进行短期和长期预测。在1972-85短期的单向因果关系，GDP对开放性的发现而在1986-2007年GDP和开放性之间存在双向因果关系。长期来看，GDP弹性大小和开放性之间的1972-85包含负号表明存在长期的负相关关系。而在1986-2007年期间积极迹象表明GDP和开放性之间的正向的因果关系。因此可以得出结论：各国实施南盟整体情况好转。长期的误差项系数表明，短期均衡调整是由调整回到长期均衡驱动。

TRADE OPENNESS AND ITS EFFECTS ON ECONOMIC GROWTH IN SELECTED SOUTH ASIAN COUNTRIES: A PANEL DATA STUDY
The study investigates the causal link between trade openness and economic growth for four South Asian countries for period 1972-1985 and 1986-2007 to examine the scenario before and after the implementation of SAARC. Panel cointegration and FMOLS techniques are employed for short run and long run estimates. In 1972-85 short run unidirectional causality from GDP to openness is found whereas, in 1986-2007 there exists bi-directional causality between GDP and openness. The long run elasticity magnitude between GDP and openness contains negative sign in 1972-85 which shows that there exists long run negative relationship. While in time period 1986-2007 the elasticity magnitude has positive sign that indicates positive causation between GDP and openness. So it can be concluded that after the implementation of SAARC overall situation of selected countries got better. Also long run coefficient of error term suggests that short term equilibrium adjustments are driven by adjustment back to long run equilibrium.

INTRODUCTION 简介

NTERNATIOANL trade plays an important role in the development of any economy and assumed to be an engine of growth [1]. Trade is taking place not only in terms of commodities but also in terms of technology, flows of ideas and knowledge spillover.

International trade affects economy through different channels. It creates employment, generate capital formation that leads to better living standards in terms of higher level of GDP and GDP per capita. Over the past few years, the world trading system is becoming progressively open and competitive. Tariffs are reducing in both developed and developing countries and restrictions are eliminating. Economies are trying to adopt outward-looking economic policies, also looking for the ways to promote growth and employment through expanding export production and attracting inward investment.

The concept of trade openness and free trade is highly debated topic in economics. It is always assumed to be a very important source of economic growth. Trade openness can promote growth through several ways. It creates massive benefits, increase investments as a result of enlarged markets and economies of scale, flow of information, technology and knowledge spillovers. As, it creates efficient utilization of resources, improved technological efficiency and trade facilitation that returns in higher foreign exchange which is used to expand the less developed sectors of the economy. It is also supported by many economists in different studies. Some studies concluded that openness played effective role mostly in developed countries [2] whereas many studies concluded that openness can play significant role in less developed countries as well [3][4][5].

South Asia is economically one of the less developed regions of the world which accommodates more than 20 per cent of the world's population that is 1,542.95 million with the average GDP per capita of US $1,565[6]. The South Asian economies mostly followed protectionist trade policies during their initial phases of development. The prime principles behind the restrictive trade regimes were protection of the domestic industries from foreign competition and conservation of foreign exchange for balance of payments support [7]. Also, South Asia is assumed to be less integrated region of the world in terms of the trade of commodities, capital and ideas [8] whereas, Intraregional trade is very low for South Asia i.e. intraregional trade is less than 2 percent of GDP, compared to more than 20 percent for East Asia [8].

LITERATUTRE REVIEW 文献综述

The relationship between openness and economic growth has been extensively examined in the theoretical and empirical literature.

Dollar [9] used real exchange rate distortions to test that the law of one price holds in the long run. The study found a significant negative correlation between real exchange rate distortions and growth, which shows a positive trade-growth link. Harrison [10] investigated the association between openness and economic growth. The study concluded that the correlation between these two variables was strong. Frankel and Romer [11] examined the relationship between trade and growth and also considered geographic characteristics as an important ingredient in trade. The study concluded that trade has a large but moderately positive and significant impact on income of the country. Rodriguez and Rodrik [12] applied the Dollar [9] procedure to an updated version of the same data and found that the same regressions now yielded the opposite signed result. Ekanayake, Vogal and Veeramacheneni [3] checked the causal relationship between output level, inward FDI and exports for a cross-section of both developed and developing countries for period 1960-2001. The study concluded that there was bi-directional causality between export growth and economic growth. Dollar and Kraay [13] investigated the effects of trade on growth and poverty for 137 countries. The study concluded that at individual level and at cross country level the open regimes lead to faster growth and poverty reduction in poor countries. Din [14] examined the export-led growth hypothesis for the five largest economies of the South Asian region and found that long-run causality only existed in Pakistan and Bangladesh while all other countries had short run causality. Hassan and Kamrul [4] investigated the casual relationship between trade openness and economic growth and the structure of international trade for Bangladesh. The study explored that there was long-run uni-directional equilibrium relationship between trade openness and economic growth.

Sarkar [15] investigated the relationship between openness and growth. Study found no positive long-term relationship between openness and growth in majority of LDC's. Klasra [5] examined the long-run relationship between Foreign Direct Investment (FDI), trade openness and economic growth for Pakistan ad Turkey and found that there was bi-directional causality between openness and growth in Pakistan whereas for Turkey there existed bi-directional relationship between FDI and exports

Data and Variables 数据和变量

The analysis is based on annual data for four South Asian countries that are Bangladesh (BNG), India (IND), Pakistan (PAK) and Srilanka (SLK) for the sample period 1972 to 2007. The data is divided into two time spans that are from 1972 to 1985 and 1986 to 2007 to analyze the situation before and after the implementation of SAARC.

The variables used in the study are Gross domestic product (Current US $) as dependent variable. Whereas, the independent variables are the labor force, Gross fixed capital formation (Current US $), and openness. The variable openness is proxied by the ratio of imports plus exports to GDP. The data is taken from the World Development Indicators [6].

Model Specification

The following neoclassical production function is used to find out the effect of trade openness on economic growth

lnY= f{ln (OP, K, L)} (1)

The double Ln model is used to represent the growth model, to explain all the variables in growth terms.

The panel version of equation (1) can be written as follows:

Where i=1|..4 denote the countries, t=1972,|1985 and 1986|..2007 denotes time period. ï￥it is the error term with the usual statistical properties while  and ï￠ are coefficients.

The use of panel data has advantage that it can exploit both the time series and cross sectional dimensions of data and provide more efficient estimations of parameters by considering wider sources of variation.

METHODOLOGY 方法论

To estimate equation (2), panel Cointegration technique is used. The cointegration of panel data consists of four steps

Panel Unit Root Tests

The study uses unit root test to check the stationarity of the time series by using three different statistics proposed by Im, Pesaran, and Shin [16] , Maddala and Wu [17], and Levin, Lin, and Chu [18] panel unit root and stationary tests. Stationary series are integrated of order zero.

Cointegration Tests

After checking the stationarity of data and confirming that each series is integrated of the same order, the next step is to check whether these series can be combined together into a single series, which itself must be non-stationary, that is known as cointegration. Cointegrated series move in the same direction in long run and are in equilibrium relationship. So, the cointegration between openness and economic growth will explain that how these variables are related in the long run. For this heterogeneous panel cointegration test developed by Pedroni [19] and Kao [20] are employed.

Panel Fully Modified OLS estimates

When long run relationship among the variables is found then for the estimation of long run effects of openness on economic growth panel FMOLS is used, proposed by Pedroni [21].

Granger Causality Test

Finally, if the variables are cointegrated and long run relationship exists, next step is to apply the Granger causality test. For this purpose a panel-based error correction model (ECM) is used to explain the long-run relationship by using the Engle and Granger [22] procedures. The two-step procedure of Engle-Granger [22] is performed as: firstly, the estimation of the long-run model for Equation (2) in order to obtain the estimated residuals ï￥it. Secondly, to estimate the Granger causality model with a dynamic error correction:

The sources of causation between Y and OP are recognized by testing for the significance of the coefficients of the dependent variables in Eqs. (3) and (4). For short-run causality, study test H0: 12i,k = 0 for all i and k in Eq. (3) or H0: 21i,k = 0 for all i and k in Eq. (4). While, the long-run causality is tested by looking at the significance of the  , which is the coefficient of the error correction term, ï￥i,t-1. The significance of  indicates the long-run relationship of the cointegrated process, and so movements along this path can be considered permanent. For long-run causality, test H0: 1i =0 for all i in Eq. (3) or H0: 2i =0 for all i in Eq. (4) is used. Similarly, sources of causation between Y and other two variables (capital and labour) are identified.

Empirical results
Panel Unit Root Results

Panel unit root test results are reported in table 1-a and 1-b for 1972-85 and 1986-07 respectively. All tests results do not reject the null hypothesis of non-stationary at level with both individual effect and individual linear trend effect for both time periods.

Notes: LLC, IPS, MW and indicated the Levin et al. (2002), Im et al. (2003) and Maddala and Wu (1999) panel unit root and stationary tests. All tests examine the null hypothesis of non-stationary (unit root). The four variables were grouped into one panel with sample N= 4, T=14. The parenthesized values are the probability of rejection. Probabilities for the MW (ADF Fisher Chi-square) and PP (Fisher chi-square) tests are computed using an asymptotic distribution, while the other tests follow the asymptotic normal distribution.

However, all tests reject the null hypothesis of non-stationarity at first difference. This shows that all the variables Y, OP, K and L are integrated of order one, an I (1) process. So, as pooled data is stationary in first difference hence, the series can be cointegrated.

Notes: LLC, IPS, MW and indicated the Levin et al. (2002), Im et al. (2003) and Maddala and Wu (1999) panel unit root and stationary tests. All tests examine the null hypothesis of non-stationary (unit root). The four variables were grouped into one panel with sample N= 4, T=22. The parenthesized values are the probability of rejection. Probabilities for the MW (ADF Fisher Chi-square) and PP (Fisher chi-square) tests are computed using an asymptotic  distribution, while the other tests follow the asymptotic normal distribution.

Cointegration 协整

Table 2-a and 2-b present the results of Pedroni Cointegration for 1972-85 and 1986-2007 respectively. Pedroni provides seven statistics for tests of the null hypothesis of no cointegration in heterogeneous panels. Under this technique two models are developed model (1) with no deterministic trend and model (2) with deterministic intercept and trend. Results show that null hypothesis of no-cointegration is rejected for seven statistics for both models at 5 and 10 percent level showing evidence of cointegration for the group as a whole and individual countries of the panel for both time spans.

Note: This table reports Pedroni (2004) residual cointegration tests. The number of lag truncations used in the calculation of statistics is fixed at 1. The null hypothesis is no cointegration. Probability values are in parenthesis.

The results of Kao [21] residual cointegration test are reported in table 3 before and after the implementation of SAARC. The results show that null hypothesis of no cointegration is strongly rejected at one percent level of significance. So there exists a long-run relationship among Y, OP, K, and L for the panel of South Asian countries.

Notes: This table reports Kao (1999) residual cointegration test. The number of lag truncations used in the calculation of statistics is fixed at 1. The null hypothesis is no cointegration. Probability values are in parenthesis and computed using asymptotic Chi-square distribution.

FMOLS Estimates FMOLS评估

Tables 4-a and 4-b exhibit the results of the long-run elasticities for each country and a panel of these countries based on Pedroni's group mean FMOLS estimator for 1972-85 and 1986-2007 respectively. The results of regression equation in which Y was taken as the dependent variable show that the variables OP, K, and L, are statistically significant at 1 percent, 5 percent and 10 percent level of significance.

At country level, trade liberalization played negative role that is coefficient of OP is negative for three out of four countries in the time period of 1972-85. Openness played a positive role only for Pakistan before the implementation of SAARC and is statistically significant. One major reason for positive impact of OP on GDP for Pakistan is the green revolution. That took place in the late 1960's and led to the growth of agriculture products to double approximately. Whereas, due to the separation of East Pakistan (Bangladesh) from West Pakistan (Pakistan) badly affected the Bangladesh's economy as it was left with very few industries and was mainly an importer country.

While, the results are mixed for L and K for all four countries. The sign of the coefficients of L is positive for three out of four countries except India before the implementation of SAARC but for Sri lanka L played positive but insignificant role. Whereas, after the implementation of SAARC in the period of 1986-2007 labor played positive and statistically significant role for three out of the four countries. L responded negative only for Bangladesh.

After the implementation of SAARC openness played positive and statistically significant role for all the four countries.

From the panel results of estimated regression the coefficients can be interpreted as long-run elasticities for group of countries. The results suggest that 1 percent increase in openness leads to approximately 0.13 percent decrease in GDP for time period 1972-85, whereas after the implementation of SAARC in the period of 1986-2007 the overall situation got better as a 1 percent increase in openness leads to 0.03 percent increase in GDP. Whereas, the role of capital also got better after SAARC that is a 1 percent change in capital leads to 0.91 percent instead of 0.50 percent. But the role of labor decreased from 0.40 to 0.12 that is because of the reason that technological advancements took place of labor.

Notes: Wald Chi-square tests reported with respect to short-run changes while error term coefficient as long-run changes. Parentheses values are the probability of rejection of Granger non-causality. Asterisks * and ** indicate statistically significant at 1 % and 5% level respectively.

The results find that there exists significant unilateral causal relationship between Y and OP in the short-run before SAARC i.e. in 1972-85 running from Y to OP. This shows that Y caused OP through error correction term. Hence, H2 is verified. Also, there exists bi-directional causality between OP and K, and uni-directional causality between OP and L and between K and L running from OP to L and from K to L. For GDP equation, the estimated coefficient on the error correction term is negative and statistically significant. It shows that short-term adjustments to equilibrium are driven by adjustment back to long-run equilibrium through error correction term.

Notes: Wald Chi-square tests reported with respect to short-run changes while error term coefficient as long-run changes. Parentheses values are the probability of rejection of Granger non-causality. Asterisks * and ** indicate statistically significant at 1 % and 5% level respectively.

The results for time period after the implementation of SARRC find that there exists significant bilateral causal relationship between Y and OP, between Y and K and also between Y and L in the short run. This shows that both of the variables in each set caused each other through error correction term. Hence, H2 is verified. While, there exists uni-directional causality between OP and K, OP and L, L and K running from K to OP and from L to OP and from L to K.

For GDP equation, the estimated coefficient on the error correction term is negative and statistically significant. It shows that short-term adjustments to equilibrium are driven by adjustment back to long-run equilibrium through error correction term. For OP equation, the estimated coefficient on error correction term is negative and statistically significant indicating that OP is responsive to adjustments back to equilibrium. It specifies long-run feedback between Y and OP.

Summary and policy implications 摘要和政策的影响

The goal of this study is to determine the direction of causal relationship between openness and economic growth in four South Asian countries for two time spans that is from 1972-85 and from 1986-2007 to examine the scenario of economic growth before and after the implementation of SARRC.

The panel cointegration technique and panel based error correction models (ECM) are used to find out the causation results. Also, fully modified ordinary least squares (FMOLS) technique is used to find the long-run relationship.

The results of the study have important policy implications. There exists short run unidirectional causality running from Y to OP but not vice versa in the time period of 1972-85. While, negative relation exists between the two in the long-run whereas, in 1986-2007 there exists short-run bi-directional causation between Y and OP. The FMOLS results explored positive sign which show that there exists a long-run positive causation between these two variables. The magnitude of long run elasticity is not very high after the implementation of SAARC but the good point is that it shows positive responsiveness in GDP due to Openness. The results show that a one percent increase in OP will lead to 0.03 percent increase in GDP.

To increase this long run responsiveness magnitude the South Asian countries should introduce export oriented policiesto enhance more and more exports that will help in the earnings of foreign exchange and will lead to the economic growth rapidly. Also these countries should try to switch from the exports of raw material and semi manufactured goods to final product. It is essentially needed to change the export and import patterns in the region. Furthermore there is need of technological advancement, production of capital intensive commodities instead of more labor intensive commodities and also there must be proper vocational institutes to train and increase the number of skilled labor force which can effectively contribute towards the trade sector as well as GDP of the region.