قوانین سیاست های پولی در یک اقتصاد باز کوچک: یک برنامه کاربردی به مکزیک

We estimate a small-scale macro model for the Mexican economy under the New Keynesian (NK) framework and alternative interest rate rules for Mexico. With these results we evaluate the performance of the Bank of Mexico against a set of optimality principles derived in the NK literature. Our system estimation results show that the Bank of Mexico holds a preference for stabilizing not only inflation around target, but also acts to achieve an output gap close to zero. Furthermore, we find that the central bank responds non-linearly to real exchange rate depreciations. We also find that the central bank has actively attempted to neutralize demand and supply shocks through monetary policy that is consistent with the Taylor principle.

Price stability is an explicit objective of central bank policy in Mexico. In the past, monetary policy centered on the use of the exchange rate as the economy’s nominal anchor, but when this resulted incompatible with macroeconomic conditions, the consequence often entailed a balance of payments crisis. Recently, Mexico’s monetary policy has evolved towards the establishment of a precise inflation target, favoringthe short-term nominal interest rate as policy instrument. Explicitly, the monetary authority aims to achieve the convergence of inflation to its target level in the medium-run; however, it is plausible that implicit additional objectives also guide policy. Hence, the comprehension of the factors to which the monetary authority responds, as well as the effect of policy decisions on the economy constitute a problem of interest. The effective identification of the transmission mechanism of monetary policy is a fundamental pillar of policy conduction. The purpose of the present paper is to model the behavior of the Mexican economy and the response of the central bank to economic conditions during the past decade. We model the economic restrictions faced by policy-makers through a small-scale macro model in the New Keynesian (NK) tradition. This class of models incorporates nominal rigidities and imperfect competition into the dynamic stochastic general equilibrium (DSGE) framework developed in the Real Business Cycle (RBC) literature. The assumption of nominal rigidities generates a structure where monetary policy is effectively non-neutral in the short-run, while maintaining its long-run neutrality. We use this modeling strategy as a reference to determine which factors account in greater part for inflation dynamics in Mexico, as well as to determine important aspects of the transmission mechanism. Furthermore, we model the conduction of monetary policy as endogenous –with the short-run nominal interest rate as instrument– through the use of the monetary policy rule methodology. This approach is attractive and has received plenty of attention in the literature. Its main strength lies in its capacity to systematically incorporate economic information in order to formulate a policy recommendation. Policy rules vary considerably in their essence and complexity, and can be broadly separated into simple and optimal rules. Simple interest rate rules tend to be characterized in the literature as linear functions of the inflation rate and the output gap. These rules are essentially ad hoc, although some baseline versions may be derived from standard theory as special cases.1 However, their simplicity makes them an attractive initial step in the evaluation of monetary policy, as they are able to capture some important aspects of policy conduction. Conversely, optimal rules are the solution to the explicit optimization of an objective function, the latter of which may be utility, subject to the constraints imposed by the structure of the economy as a whole. By using a larger information set and by modeling interactions between variables in a more sophisticated manner, optimal rules contain a more rigorous analysis. However, their complexity may reach considerable levels, making the communication of policy objectives to thegeneral public substantially more difficult. Simple policy rules are the focus of this paper because of their simplicity and capacity to describe actual policy. We use a short-term nominal interest rate as policy instrument, reflecting the practice of major central banks. Although the Bank of Mexico only started to use this policy instrument in 2008, the previous one, namely the corto, may be interpreted as a signaling mechanism through which the central bank indicated its preference for the market interest rate structure. Following Torres (2002) and Roldán (2005), our analysis focuses exclusively on monetary policy rules for the determination of the short-run nominal interest rate. We contrast two alternative policy rules and show that monetary policy in Mexico implicitly targets variables besides inflation, therefore operating under a “flexible” inflation targeting framework (IT). Under this regime, the central bank is additionally concerned with the evolution of macroeconomic variables besides inflation, and may pursue various goals simultaneously. In practice, it is reasonable to assume that central banks work towards achieving macroeconomic stability in a broad sense, assigning an important relative weight to price stability but not responding exclusively to it. The output gap and a measure of the exchange rate are among the additional variables targeted. Importantly, we show that the central bank’s response to exchange rate fluctuations is non linear, as previously modeled in the literature for Mexico. In fact, the analysis of a non-linear response to exchange rate fluctuations is imperative in order to square the interest rate rule with the facts. We then use our estimated results to evaluate policy against a set of optimality principles derived in the interest rate rule and NK literature. Our system estimation results suggest that the Bank of Mexico has established monetary policy as an effective nominal anchor, consistent with the Taylor principle, and that the Central Bank has in effect acted to stabilize inflation around its stated target, while at the same time responding to deviations in output and the real exchange rate from trend.2 The remainder of this paper is organized as follows. Section II introduces the macroeconomic model to be estimated. Section III presents a brief description of the data used in the exercise. Section IV carries out the estimation and presents some key results. Section V uses the econometric estimates of the key parametersof the model to perform an impulse-response analysis in the context of dynamic stochastic general equilibrium and compares the dynamics under alternative policy rules. Section VI concludes.

Monetary policy constitutes a powerful tool which may be used to influence the behavior of aggregate demand. In practice, its conduction is complex and may respond to broad considerations, beyond explicitly stated objectives announced by the central bank. Our results suggest some important points. First, even though monetary policy in Mexico is aimed at controlling inflation, it does not seem to exclusively respond to the behavior of that variable. The central bank seems to assign a positive weight to the behavior of real economic activity when making policy decisions, seeking to maintain output fluctuating around its potential level. Additionally we establish that, the central bank takes into account the behavior of the real exchange rate in its policy reaction function. Second, the estimation of the simple and augmented rules suggests that the central bank, between 1998 and mid-2008, did establish the conduction of monetary policy as the economy’s nominal anchor consistently with the Taylor principle. The finding that key parameters are higher than unity suggests that the central bank’s behavior has been sufficiently aggressive in pursuing price stability in the economy. Third, it is seen that for a small open economy like Mexico’s, the exchange rate does in fact provide an additional transmission mechanism for monetary policy. The exchange rate responds strongly to changes in monetary policy, inducing both first-round and second-round effects on output and inflation. Fourth, when contrastingthe model dynamics under alternative policy rules, we have found that the central bank response to demand and supply shocks seems more effective under the augmented interest rate policy than under the simple, taking less time and possibly being less costly for the economy to return to its steady-state equilibrium. Finally, using the criteria described in the literature under which optimal monetary policy should operate, results indicate that in Mexico, monetary policy does in fact responded aggressively to aggregate demand as well as aggregate supply shocks. Thus, the interest rate has been effectively used during the past decade to neutralize states of excess demand and also, to eliminate an environment of unemploymentinflation that follows inflationary exogenous shocks. Overall, monetary policy has constituted an effective demand management tool, satisfying this optimality criterion in the case of demand shocks. However, the results suggest that in the case of supply shocks, the nominal interest rate is not kept unchanged. While this response would not be considered optimal in the case of transitory supply shocks, it may be so when inflation is highly persistent and the risk of an inflation-recession environment is latent as is the case in Mexico as well as other economies. This last point merits additional attention in future research on the subject, particularly in the current context of adverse shocks to the economy. A relevant extension of this paper would consist of verifying the robustness of the estimated parameters to alternative estimation methods. In particular, the recent development of methods which allow for the simultaneous estimation of the system, through the use of Bayesian methods or conditional maximum likelihood, for example, could be useful in the verification of results here obtained. Finally, future research should be aimed at identifying the model’s structural parameters, allowing for the evaluation, in terms of welfare, of different policy regimes.