Monetary policy and asymmetric effects of oil price shocks in South Africa.

Abstract

This study employs the Structural Vector Autoregressive (SVAR) model to examine the asymmetric effects of oil price shocks on monetary policy in South Africa using monthly data from 1994:1 to 2013:12. A seven variable SVAR with short-run restrictions on the contemporaneous relationships among the variables is constructed for the analysis. Consistent with the existing theory and empirical literature, the study finds that the response of output and inflation to an oil price shock is asymmetric. The nature of the asymmetry, however, is different to what has been reported in most empirical studies. This study finds that a negative oil price shock tends to stimulate output while a positive oil price shock has an insignificant effect on output. Similarly, it is observed that a negative oil price shock reduces inflation by a larger margin than a positive oil price shock increases the general price level. The study results also show that the response of interest rates, money supply, exchange rates, real output and inflation to shocks in the system may be larger or smaller depending on whether oil prices are increasing or decreasing. This underscores that the variables in the monetary policy transmission process respond to shocks asymmetrically. Further investigation reveals that the monetary policy response to increasing and decreasing oil prices is a highly likely source of the asymmetric relationships. It is observed that when crude oil prices are increasing, monetary authorities respond with greater speed and adjust interest rates by larger margins than when crude oil prices are decreasing, which probably explains why increasing crude oil prices tend to have a smaller impact on inflation than decreasing oil prices. It is further observed in the study that exchange rate variations account for a very large proportion of the fluctuations in inflation rates. It is also found that money supply accounts for a larger proportion of the fluctuations in real Gross Domestic Product (GDP) than exchange rates, indicating that money supply may be a more important intermediate target of monetary policy than exchange rates in the transmission of monetary policy where the policy goal is real GDP. In addition, it is found that the proportion of fluctuations in money supply explained by interest rates variations is larger than the proportion explained by exchange rates; and the impact of the interest rates on money supply is also observed to be larger when oil prices are decreasing than when they are increasing. We, therefore, conclude that that the money effect, interest rate and exchange rate are important channels of monetary transmission in South Africa. The study results further reveal that in South Africa, the contribution of the monetary policy goals (inflation and real GDP) to variations in money supply (an intermediate target of monetary policy) may be higher or lower depending on whether crude oil prices are increasing or decreasing. Fluctuations in the rate of inflation account for a higher proportion of the variations in money supply when oil prices are decreasing than when they are increasing; while real GDP fluctuations account for a higher proportion of the variations in money supply when crude oil prices are increasing than when they are decreasing. Similarly, it is observed that exchange rates respond asymmetrically to monetary policy in the event of an increase or a decrease in crude oil prices. The study also reveals that the largest proportion of the fluctuations in interest rates is explained by exchange rates, indicating that monetary authorities respond to exchange rate fluctuations by adjusting interest rates. The impact of this response is observed in money supply variations, as argued in the foregoing discussion. This shows that the monetary policy asymmetry observed in the operating tools and the goals of monetary policy can also be traced through the intermediate targets.