Política monetária em Cabo Verde e mudanças macroeconômicas: evidências empíricas

Abstract

This paper goal was to investigate important aspects about Cape Verde monetary policy and its transmission to the economy, during the 1991/2011 period with quarterly data. In which are included the Gross Domestic Product and the prices index as target variables, the liabilities operations rates over 91 days as a monetary policy instrument and the BCV´s nominal effective exchange rate index as the monetary policy intermediate goal. The implemented methodology was the VAR and MS-VAR models. As a result, we have that the modification of a VAR model to a MS-VAR brought information gains by allowing some non-linearity to the model. It was estimated two models for each methodology, in which the difference between then is due to the fact that one of them doesn´t include the exchange rate (simpler model). It was identified two regimes, in which the regime 2 appeared to be more persistent and is was also verified its exclusivity during the 1993:1 to 2006:2 period, which coincides with the period when it was separated the functions of the central and commercial banks and also was created two independent institutions, with a change in the exchange rate regime as well, that occurred in 1998. It was observed also that the regimes classification is sensitive to changes in the interest rate until some level and that the prevailing interest rate in Cape Verde is relatively high. This fact is due to the structure of the national financial market and the weak inter savings. In the second regime, the results from the impulse and response functions showed themselves coherents to the conventional theory, in the sense that a positive shock in the interest rates causes a reduction in the production and prices level. On the other hand, the obtained results from regime 1 are different, principally for the IPC response to a shock in the interest rate, once that this shock causes an increase in this variable. The introduction of the exchange rate in the model, caused a reduction on the impact of the Txjur on the IPC on the regime 1, when comparing to the previous model. Another registered alteration on the results was that the duration of the negative effect caused by the positive shock on the GDP appeared to be lower in both regimes.