Examining the Exchange Rate Overshooting Hypothesis in Bangladesh : A Cointegration and Causality Analysis

Following the abandonment of the Bretton Woods Agreement in 1973, the exchange rate has become an important macroeconomic variable that can be linked to the performances of multidimensional sectors within an economy. It has been empirically acknowledged that countries participating in rigorous bilateral and multilateral trade activities are often vulnerable to unanticipated exchange rate movements following which the economy as a whole could encounter adverse impacts. The focal point of this paper is to identify the causal associations between Nominal Exchange Rate (NER) of Bangladesh and its macro fundamentals using relevant annual time series data from 1980 to 2015. A multivariate linear log-log model is used in which NER is expressed as a function of Money Supply (MS), GDP, Domestic Interest Rate (INT) and Local Rate of Inflation (INF). As part of the methodology, Johansen Cointegration test is employed to identify the long run association between the variables considered in the model. In addition, the results from the Vector Error Correction Model (VECM) approach and Granger Causality test provide the directions of causalities between NER and its determinants. The results reveal that GDP and INT are effective in influencing NER in the long run while MS and INF are found to be ineffective. Thus, in the long run, macroeconomic variables are partially effective in stimulating exchange rate movements in Bangladesh. However, in the short run monetary policy is totally ineffective in influencing NER movements