ENSURING PRICE STABILITY WITH AN INTEREST RATE RULE* Bernardino Adão** Isabel Correia** Pedro Teles** 1. INTRODUCTION The primary concern of monetary policy is to ensure price stability. Such is the mandate of the ESCB established by the Maastricht Treaty, but the objective, spelled out in various ways, is common to every other central bank. In a somewhat old fashioned language, the objective of price stability requires that monetary policy provide a nominal anchor, that it anchor expectations. For once, the objective seems easier to attain in practice than in theory. Central banks in developed countries have been very successful in the last twenty five years in targe- ting low inflation. The success has been attributed to a somewhat mechanical interest rate rule where the short term nominal interest rate is set in response to deviations from trend of inflation and econo- mic activity, a Taylor rule, named after John Taylor who first estimated it (Taylor, 1993). It turns out that no policy rule of this type is able to achieve in a monetary model what it appears to achieve in reality. The same models that give very reasonable answers to other questions, generate multiple equilibria when monetary policy is conducted with an interest rate feedback rule whether it may respond to future, current or past inflation. There is an extensive literature on this issue of central importance to monetary policy making, dating back to Sargent and Wallace (1975) who showed that a policy that targets the interest rate gives rise to multiple equilibria. Most of the later literature has focused on conditions for local determinacy, meaning that, while the multiplicity of equilibria remains, there might be only one equilibrium in a particular ne- ighborhood of interest. McCallum (1981) is the main responsible for triggering this literature, showing that there are indeed interest rate feedback rules that guarantee local uniqueness. Technically, this has been very useful because it has allowed economists to abstract from a problem that was not easy to solve, and concentrate on other issues focusing on the unique local equilibria. Unfortunately, as po- inted out by Benhabib, Schmitt-Grohe and Uribe (2001, 2002), the same policy rules that ensure lo- cally determinacy typically generate global indeterminacy, so that the alternative equilibria can converge to other steady states or cycle around the original one. In this note, and based on Adao, Correia and Teles (2006), we discuss how interest rate rules can be used to implement a unique equilibrium with stable prices. We first consider an economy with a finite horizon and show that nominal interest rates are not a sufficient policy instrument. In a finite horizon economy a finite number of equilibrium variables is restricted by a finite number of equilibrium conditi- ons. If policy specifies restrictions for the nominal interest rates only, then there are more unknowns than equations and there are multiple equilibria. This result does not depend on whether there is an exogenous target for the interest rate or whether it responds to endogenous variables. The number of equations is the same. Similarly, whether prices are flexible or sticky also does not matter. Economic Bulletin | Banco de Portugal Articles | Summer 2007 57 * The opinions are solely those of the authors and do not necessarily represent those of the Banco de Portugal. ** Economics and Research Department.