Inflation, inflation uncertainty and relative price variability in Bangladesh

Developing countries are often busy with the rate of inflation and its effects on the economy. Although the monetary policy of developing countries is concerned with business fluctuations and its effects on stability, recent studies are giving importance on the relationship between inflation and relative price variability (RPV). In recent macroeconomic theory, RPV generates the fundamental distortions of inflation, which disrupts the informational content of nominal price. It has long been popularly believed that the relationship between RPV and inflation is positive and stable. Using disaggregated monthly CPI data for Bangladesh from 2002:7 to 2013:6, this study tries to tackle the following problems: (1) whether the relationship is linear? (2) whether the relationship is sensitive to the models of inflation forecasting? (3) whether the model is stable? This study finds that the relationship is not linear, which contrasts with the earlier works on RPV-inflation relationship. Our semiparametric estimations show that the relationship is U-shaped. The estimation of the parametric quadratic function shows that the model of inflation forecasting is sensitive to this relationship which makes that it is the unexpected inflation which matters for RPV. Although the equation is specified, but it is not stable over time. The rolling regression analysis and breakpoint test show that there is a breakpoint during the sample period. The instability also comes from the food inflationary shock and poor macroeconomic policy management. This stability of the relationship is important for determining the threshold level of inflation, which is crucial to minimize RPV.