Towards credibility from inflation targeting: the Brazilian experience

Abstract

This article analyses the use of the basic interest rate after the adoption of inflation targeting in Brazil and the credibility of this monetary regime through two indices that consider the Cukierman and Meltzer (1986) definition for credibility. It also shows the main theoretical and practical motives for changes in the conduction of the monetary policy in the 1970s; the way that inflation targeting strategy is inserted in rules vs. discretion analysis; and the main points that characterize the literature concerning inflation targeting. The findings denote that the strategy implemented in Brazil is not a good mechanism to develop credibility.

Notes

¹ See Romer and Romer (2002) for an analysis of stabilization policies in the USA after the 1950s.

² The strategy adopted by several countries (e.g. Germany, Japan and USA) was the use of a rule à la Friedman (1968) proposal.

³ During this time several papers by Lucas (1972 and 1973) and Sargent (1973) influenced in a large way the academic analysis.

⁴ Example of countries that adopted this monetary regime are: Australia, Brazil, Canada, Chile, Colombia, Czech Republic, Finland, Iceland, Israel, Korea, Mexico, New Zealand, Norway, Peru, Philippines, Poland, South Africa, Spain, Sweden, Thailand, United Kingdom.

⁵ For simplifying the analysis it is understood that the parameters a and b are constants, consequently ΔM_t has a constant value.

⁶ Goal independence refers to the freedom that the central bank has in defining the target to be reached. (Fischer, 1995).

⁷ It is important to highlight that although the literature reveals a narrow relation between central bank independence and inflation targeting an independent central bank is not necessary to implement the monetary regime.

⁸ Most of the time, the core of band is 2% allowing a fluctuation of 2% (Meyer, 2001).

⁹ The best example of transparency of central bank actions is the New Zealand case. In this country the government can dismiss the central bank governor if the inflation deviates 25% from the announced target. Other examples of benefits from transparency are Canada (1996) and England (1997). For a detailed analysis of these countries see Mishkin and Posen (1997). For an analysis of the countries that adopted inflation targeting in the 1990s see Bernanke et al . (1999).

¹⁰ The Real Plan was a stabilization program divided into three steps: (i) budgetary equilibrium; (ii) introduction of a new stable unity of account to align the most important relative prices in the economy; and (iii) the conversion of this unity (URV–real unity of value) to the new currency of this country (Real) with the parity semi-fixed with the dollar.

¹¹ Since August of 1999 the weight of administered prices on IPCA is 28.7%.

¹² The justification for the sensitivity of administered prices to the exchange rate is due to the fact that some prices are indexed to the General Price Index (IGP). This price index is more sensitive to the variation of exchange rate than to IPCA.

¹³ *(**) denotes rejection of H₀ at the 1% (5%) significance level. White (1980) t-statistics in parentheses.

¹⁴ ‘(…) the absolute value of the difference between the policymaker's plans and the public's beliefs about those plans’ (Cukierman and Meltzer, 1986, p. 1108).

¹⁵ The targets were at 6% for 2000, 4% for 2001, 3.5% for 2002, 4% for 2003 and 5.5% for 2004–accumulated annual variations by the year-end. Tolerance intervals of ±2% until 2002 and ±2.5% after were also defined.

¹⁶ The use of escape clauses in extreme situations implies that there is no loss in credibility when the target is not achieved, because changes in planned policies are not a result of time inconsistence problem, but a result of variables that cannot be foreseen.

¹⁷ The target in 2003 was changed from 3.25% to 4% and in 2004 the target is 3.75%. Besides this, the tolerance interval was changed from ±2% to ±2.5%. After this, a new change in targets occurred–the target for 2003 was 8.5% and for 2004 is now 5.5%.