Impacts of monetary policy shocks on the Ethiopian macro economy autoregressive approach

Abstract

This thesis analyses the impacts of monetary policy shocks on Ethiopian macro economy, and the channels through which these shocks will be transmitted to the overall economy using annual data running from 1981/82-2011/12.In order to achieve this objective; unit roots (ADF and PP), cointegration (Johansen procedure), Vector Autoregressive (VAR), Error correction model and Granger Causality (Wald Coefficient restriction test, X2and F-statistics) are carried out. The results of the unit root tests show that the series of interest are stationary in first differences. Cointegretion and causality between the variables are tested and compared using tri-variate vector Autoregressive model (VAR). The results of co-integration tests on the tri-variate models show that two Granger-causality alternative models (VAR in first difference and Error Correction Models) fit for this study. In both types of models, Granger-causality (unidirectional or bidirectional causation) between monetary policy instrument on one side and price and economic growth on the other side are tested using Chi — square and F — statistics from Wald coefficient restriction test. The results of cointegration test shows that there is a long run equilibrium relation between the variables. The error correction model and the Wald Granger causality indicates the existence of short run bidirectional causality between money supply and price level, and unidirectional causation running from output to price level that shows that price level is a monetary phenomenon in the short run.Based on the short run causation results, all the foregoing discussion leads to put exchange rate as an important channel of monetary policy instrument to be considered as a supplementary, through which shocks in monetary policy transmits to the real economy.