Trade-led Growth and Growth-led Trade Hypotheses in Bangladesh: An ARDL Bounds Test Approach

This study investigates the relationship between foreign trade and economic growth using annual data of real GDP (proxied for economic growth), real exports and imports for the period of 1979 to 2014 in Bangladesh. The study operates the bounds testing cointegration procedure and ARDL-error correction model to examine the short-run and long-run relationships among the variables. The ARDL bounds test approach reveals that the long run cointegrations exist among the variables when GDP is dependent on export and import, and export is dependent on GDP and import. The long run coefficients of ARDL results show that export is positively and import is negatively related with GDP at 1% level of significance. Furthermore, GDP and import have a positive and highly significant long-run effect on export in Bangladesh. The short run dynamics along with the error correction term (ECT) results indicate that the coefficients of error correction terms of the two models are negative and highly significant which suggesting that the long run causalities are also directing from exports and imports to GDP, and GDP and imports to exports. The error correction terms imply that GDP requires about ten years to converge to equilibrium after being shocked, while export requires only about nine months. All of the short run coefficients of the models are significant and consistent with the long run results signifying that export-led growth, growth-led export and import-led export models are valid in Bangladesh. Although the role of imports on economic growth is negative, imports perform a noteworthy role to promote exports in Bangladesh both in the short run and long run. In conclusion, we can comment on this empirical study that trade liberalization which has been started in 1979 significantly benefits the economic growth in Bangladesh.