An Econometric Analysis of Investment, Trade Openness and Economic Growth in Bangladesh

Abstract

Investment (domestic plus foreign investment) is the emotive force for economic growth of a country. Domestic investment is positively associated with domestic capital formation. FDI complements the process of attaining the saving investment gap by supplying foreign capital while trade openness maintains an important channel for country’s investment and the economic growth. This study, however, attempts to find out the impact of domestic investment, FDI and trade openness on economic growth and their causalities in Bangladesh. It further examines the influences of different components of the core variables (domestic investment, FDI, trade openness) on them and analyzes the short and long run causal relationships among them at the disaggregated level. Doing so, it undertakes the sample of 42 annual observations covering the period of 1972 to 2013. In this context, the objectives of this study are: to assess the current states of domestic investment, FDI, trade openness and economic growth in Bangladesh; to assess the influences of different components of domestic investment, FDI, trade openness on these core variables at the disaggregated level; and to examine the short and long run causal relationships associated with domestic investment, FDI, trade openness and economic growth in Bangladesh. In order to meet these objectives, a disaggregated econometric analysis has been carried out in this study. The tabular and graphical techniques have been used to assess the current states of the variables. They indicate that the variables of the core function (e.g. economic growth is the function of domestic investment, FDI and trade openness) have the upward trends over the time but individually they all are stable in Bangladesh. The variables at the disaggregated level are also suffering with the instability problem. In every case, the instability index is higher during pre-liberalization than that of the post-liberalization period (1990). The theoretical frames with model specifications of the issue and the variables of the respective functions with evidence in the literature have been discussed in this study. In econometric analysis, the time series properties of the data of domestic investment, FDI, trade openness and GDP growth functions have been justified successively. In this context, the unit root tests such as, the correlogram test, the ADF test, the D-F (GLS) test, and the Phillips-Perron test have been applied in the study. For each of the functions the tests provide the same results that is, the null hypotheses of unit root problem have been accepted at the level form but they have all been rejected at the first

difference. Hence, the data of the variables of the domestic investment, FDI, trade openness and growth functions for Bangladesh have been found non-stationary at the level form but they have been stationary at the first difference. Thus, the variables of the functions have been integrated of order one I(1). Results of the cointegration test (the trace and max-eigen value tests) confirm that there are 2 (two) long run cointegrated stable relationships between domestic investment and its various components, FDI and its different factors, trade openness and its major components while same results have also been found for economic growth of Bangladesh with stock of labour, domestic investment, FDI as well as trade openness of the country. In order to assess the influence of the components of different functions, OLS method has been applied for estimations. Results indicate that domestic investment of Bangladesh is however influenced by its different factors but financial intermediation and human capital have significantly negative effects while GDP growth rate, FDI, real export and domestic credit have positive impact on domestic investment in Bangladesh. The OLS estimated coefficients of the FDI function indicate that FDI in Bangladesh is no doubt influenced by its various factors but gross capital formation significantly negatively affects it while GDP growth rate and wage rate in Bangladesh positively affect FDI. Again, FDI in Bangladesh is negatively influenced by the GDP, stock of labour and trade openness but they are insignificant. Results further show that trade openness in Bangladesh is positively influenced by its different components (GDP, real export, real import, and real exchange rate).The terms of trade and the real inflation have significantly negative effect on trade openness in Bangladesh. Finally, the estimated coefficients of the growth function indicate that GDP of Bangladesh is definitely influenced by its different components. The stock of labor and domestic investment positively affect economic growth in Bangladesh of which, the impact of domestic investment on economic growth is significant. That is, domestic investment positively affects GDP by 70 percent while the impact of labour is insignificant. GDP of Bangladesh is negatively influenced by FDI and trade openness but they are insignificant. This may be due to insignificant contribution of them on the domestic economy of Bangladesh. The Wald test confirms that the coefficients of GDP growth, FDI and trade openness functions are jointly significant while the coefficients of domestic investment are not jointly significant but some of them may be individually significant. For domestic investment, the VECM result shows that there is short run dynamics to the long run equilibrium among GDP growth rates, real exports, human capital and domestic investment while there is long run causality but with a divergence among FDI, financial intermediation, domestic credit availability and domestic investment in Bangladesh. Result further shows that there is short run dynamics to the long run equilibrium among GDP, gross capital formation, stock of labour and the wage rate to FDI while there is long run causality but with a divergence relation among GDP, trade openness and FDI in Bangladesh. Again, there is short run dynamic adjustment to the long run equilibrium among GDP, real import, terms of trade and trade openness in Bangladesh while there is long run causality but with a divergence among real export, real exchange rate, real inflation and trade openness. Results further show that there is short run dynamics to the long run equilibrium between domestic investment and GDP growth while there is long run causality but with a divergence relation among stock of labour, FDI, trade openness and GDP growth in Bangladesh. The VAR estimation shows the elasticities of the functions that is, the coefficients of real exports and domestic credit availability are significant for domestic investment in Bangladesh while others are insignificant. The short run positive elasticities of GDP growth rate, FDI and financial intermediations are statistically significant while others are inelastic for domestic investment in Bangladesh. The long run significant elasticities exist between gross capital formation and FDI while the short run significant elasticities exist among GDP, gross capital formation, stock of labour and wage rate in Bangladesh to FDI, they may either be positive or negative. Again, the long run significant elasticities exist among real export, terms of trade and trade openness while the short run significant elasticities exist among real imports, real exchange rate and trade openness in Bangladesh. The VAR result finally shows that the long run significant elasticities exist among stock of labour (negative), domestic investment (positive) and GDP growth while the short run significant but negative elasticities exist among stock of labour, FDI and GDP growth in Bangladesh. The Granger causality test indicates that there are short run bidirectional causalities between pair-wise real export and domestic credit availability to domestic investment; otherwise unidirectional causalities exist between the pair-wise variables of the domestic investment function in Bangladesh. In case of FDI function, there are bidirectional causalities between FDI and GDP growth rate; otherwise unidirectional causality exists between the pair-wise variables of the FDI function in Bangladesh. On the other hand, there are bidirectional causalities between trade openness and real exports; otherwise,

unidirectional causality exists between the pair-wise variables in the trade openness function. Results of Granger Causality test further show that there are bidirectional causalities between stock of labour and the GDP growth in Bangladesh. On the other hand, domestic investment causes GDP to grow but GDP in Bangladesh does not do so. FDI in Bangladesh does not Granger cause GDP to grow but GDP causes FDI to inflow. Again, both trade openness and GDP Granger cause each other to grow at the same direction. Thus, bidirectional short run causalities exist between pair-wise labour and trade openness to GDP while unidirectional causality exists between pair-wise domestic investment and FDI to GDP growth in Bangladesh.Results of Impulse Response Analysis of domestic investment function indicate thatthe response of all variables is either positive or negative in the short run but in the long runthey all are responded towards domestic investment in Bangladesh. It further confirms thatthe response of all variables of the FDI function is either positive or negative in the shortrun but in the long-run they all are responded towards FDI in Bangladesh. Whereas, thediversification of responses of GDP, real exchange rate as well as real inflation to opennessare very high in the short run but they all have been responded towards the same path in thelong run. The results finally confirm that the diversification of responses of stock of labour, domestic investment, FDI, as well as trade openness is very high in the short run but they all have been responded towards the same path of GDP in the long run. The variance decomposition outputs indicate that the changes in domestic investment are mainly caused by its own variation. The volatility of domestic investment is mainly caused by its own variation. The variance of foreign direct investment is always caused by 100 percent by itself and the share of FDI subsequently decreases over the year while the volatility of FDI is mainly caused by above 80% of its own variation. The variance of trade openness is always caused by 100 percent by itself in the first year and decreases gradually in the subsequent years. Again, the GDP is decomposed into its own variance by stock of labour, domestic investment, FDI, and trade openness while the share of GDP in explaining the variance decomposition decreases gradually. Finally, the model diagnostics (the L-M, the B-G, the WGH, the CUSUM and the CUSUMSQ tests) confirm the model stability and they have made the findings consistent, robust and valid. From the findings of the study, it has been imperative for government of Bangladesh to formulate human development policy to increase managerial skills, technological know how and efficiency of labour. It should also adopt policy to create more avenues towards the capital formation through instigating national savings for domestic investment. It should adopt policy to attract FDI inflows by abolishing the constraints regarding inward FDI. Finally, government should formulate improved exportled growth policy for favourable external balance as well as to increase exports of Bangladesh by reducing trade barriers. Interestingly, with a great perception about Bangladesh of its great potential in absorbing FDI into the country, it shows that FDI has not really aided the economic growth in Bangladesh. This perception might be ascribed to investment constraints like, corruption, bad governance and the decay within the economy of Bangladesh. Thus, the government needs to work out all of its institutional frameworks to enhance and monitor the inflow of the FDI. So that it could significantly contribute to the economy of Bangladesh. But, the first emphasis should be given on the enhancement of domestic investment through instigating the process of domestic capital formation. Government should also take proper initiatives in reducing different constraints for stimulating private domestic investment in Bangladesh for sustainable economic growth.