Regarding Brazil there are some studies on

Regarding Brazil, there are some studies on Central Bank reaction function. Minella et al. (2002, 2003)’s reaction function captured the effect or the lagged effect of the interest rate over aggregate demand. This effect can be seen as a weighted average of the deviations of present and future inflation expectations. The major objective of this reaction function was to see how long the effect of actual interest rate policy lasts. The weighted average of the deviation of expected inflation from the target for this year may be losing relevance when looking the lagged ones. However, forward looking measures of this variable may be gaining importance. This innovative way of viewing inflation expectation is represented by the following reaction functionwhere Dj is the deviation between the expected inflation from the inflation target, and the nominal interest rate is a function of the lagged output gap and the lagged interest rate. The reaction function (10) is estimated for the Aminoallyl-dCTP - Cy3 July 1999 to June 2002. The authors main results are (i) the COPOM adjusts the interest rate gradually, since the smoothing coefficient is around 0.8; (ii) the coefficient of the output gap is not statistically significant when using inflation expectations of the market and has an inverted signal when using inflation expectations; (iii) the coefficient of the deviations of inflation expectations in relation to the inflation target are far superior to the unit; (iv) when exchange rate was included, its coefficient was not significant. Therefore, the authors point out that during the period analyzed, the BCB – Brazilian Central Bank policy showed a forward-looking attitude, i.e., responding quickly to deviations of inflation expectations from the target previously established. In sum, the Central Bank credibility was not so high because the inflation expectation of the target was superior to the unit. This indicates that agents were expecting inflation to rise in the future. Using a model close to the ones represented by Eqs. (1)–(4) and with Eq. (5), Freitas and Muinhos (2002) estimate a model based on three equations. An IS curve, a Phillips curve and interest rate rule a la Taylor, which can be divided into two, one traditional Taylor rule and a rule called optimal rule. The authors obtained the following results: (i) the lagged interest rate impacts negatively the output gap; (ii) the lagged output gap affects the actual inflation rate negatively; (iii) the two lagged period of quantity of money affects inflation positively not the actual one; (iv) the Phillips curve has a direct effect on inflation rate, but it is not influenced by the exchange rate policy; and lastly, (v) the reaction function with optimal rule did not do well compared to the traditional basic Taylor rule; the last one presents more favorable results than those obtained via a optimal rule in explaining interest rate. The optimal rule study did return in the paper written by Almeida et al. (2003). By using dynamic programming techniques, they derived a rule for optimal monetary policy conduct using an IS curve, a Phillips curve and a reaction function for a closed economy and an open economy. Estimates for the reaction function suggest that the BCB has to calibrate the rate of interest intensively to contain the rise of inflation compared to developed countries. When the reaction function considers the exchange rate, the authors suggest that the cost to curb inflation rate is lower compared to a closed economy. Thus, exchange rate is an important mechanism to help price stability in Brazil. The importance of the exchange rate is also studied by Holland (2005). Empirically the author analyzed whether emerging countries, specifically Brazil, respond to exchange rate shocks via its reaction function. Inspired by the work of Clarida et al. (1998), the author assume that the interest rate is a function of the expected inflation, the output gap and the exchange rate, as can be seen in the following equation: