TESTING MONETARIST-STRUCTURALIST CONTROVERSY IN DETERMINING INFLATION IN BANGLADESH

Some studies hold monetary factors primarily responsible for inflation (Hossain, 1989; Mujeri, 2009) while others consider output constraint (Hoque, 1990) as the main reason. Side by side, a large body of literature finds the joint effect of monetarist and structuralist factors as responsible (Taslim, 1982; Momen, 1992; Amin and Rashid, 1994; Arif and Ali, 2012). Earlier studies mostly use annual and quarterly data for shorter span of time with mixed result. Unlike the other existing literature, we use long monthly data spanning from July 1977 to January 2016 from the sources of Bangladesh Bureau of Statistics (BBS) and Bangladesh Bank. Following the methodology suggested by Goswami and Sarker (2011), fixed weights of import are calculated using 1995/96 import figures of the 17 trading partners of Bangladesh. We multiply the weights with CPI of the respective partner countries and then sum them up to derive the weighted foreign CPI index and calculate foreign inflation. We then estimate a purely monetarist, structuralist and a mixed single equation dynamic model of inflation consisting of long run adjustment variable in the form of error correction. Both for monetarist and the mixed specification we alternatively use narrow money (M1) and broad money (M2) whereas we use two alternative specifications: output gap model and mark-up model for structuralist version. The estimation results clearly show that both the monetarist and the structuralist variables play a significant role in determining inflation in Bangladesh. However, a blended or mixed model performs better in explaining the inflation of the country compared to the other two types of models. The results hold even after several robustness checks in the form of split samples and controlling for outlier dummies. In addition to regular monetary and structural factors, inflationary expectations and foreign inflation are found to be strongly affecting inflation. That is why central bank should formulate policies in such a way that growth of broad money follows standard pattern over time and inflationary expectations can be minimized through policy interventions. The finding of the study is consistent with the Monetary Policy Statement (MPS) outlined by Bangladesh Bank, the central bank of Bangladesh twice a year. The targets set by Bangladesh Bank such as affecting broad money, reserve money, and interest rates should be broadened in order to achieve the dual goals of moderate inflation and sustainable r