Changes in GDP's measurement error volatility and response of the monetary policy rate: Two approaches

Document downloaded from http://zl.elsevier.es, day 07/01/2015. This copy is for personal use. Any transmission of this document by any media or format is strictly prohibited. Ensayos sobre Política Económica 32 (2014) 41–47 Ensayos sobre POLÍTICA ECONÓMICA www.elsevier.es/espe Changes in GDP’s measurement error volatility and response of the monetary policy rate: Two approaches夽 Julian A. Parra-Polania ∗ , Carmiña O. Vargas Banco de la Republica, Bogota, Colombia a r t i c l e i n f o Article history: Received 19 March 2014 Accepted 25 August 2014 Available online 11 November 2014 JEL classification: D81 E52 E58 Keywords: Prudence Robustness Measurement error Optimal monetary policy a b s t r a c t Using a stylized model in which output is measured with error, we derive the optimal policy response to the demand shock signal and to changes in the measurement error volatility from two different perspectives: the minimization of the expected loss (from which we derive the ‘standard’ policy) and the minimization of the maximum possible loss across all potential scenarios (from which we derive the ‘prudent’ or ‘robust’ policy). We find that (1) the prudent policymaker reacts more aggressively to the shock signal than the standard one and (2) while the standard policymaker always mitigates her reaction if the measurement error volatility rises, the prudent one may even increase her response if her risk aversion is very high. When we incorporate forward-looking expectations, the second result is preserved but, in this case, the prudent policymaker is less aggressive than the standard one in responding to the shock signal. © 2014 Banco de la República de Colombia. Published by Elsevier España, S.L.U. All rights reserved. Cambios en la volatilidad del error de medición del PIB y respuesta de la política monetaria: dos enfoques r e s u m e n Códigos JEL: D81 E52 E58 Palabras clave: Prudencia Robustez Error de medición Política monetaria óptima Usando un modelo estilizado en el que el producto se mide con error, determinamos la respuesta de política óptima a la señal del choque de demanda y a los cambios en la volatilidad del error de medición desde dos perspectivas diferentes: la minimización de la pérdida esperada (de la que derivamos la política “estándar”) y la minimización de la pérdida máxima en todos los escenarios posibles (de la que derivamos la política “prudente” o “robusta”). Observamos que: (1) el tomador de decisiones de política prudente reacciona de manera más agresiva a la señal de choque que el decisor estándar y (2) mientras que el decisor estándar siempre atenúa su reacción si aumenta la volatilidad del error de medición, el decisor prudente puede aumentar incluso su respuesta si su aversión al riesgo es muy alta. Cuando incorporamos las expectativas futuras, el segundo resultado se mantiene, pero, en este caso, el decisor prudente es menos agresivo que el estándar en su respuesta a la señal de choque. © 2014 Banco de la República de Colombia. Publicado por Elsevier España, S.L.U. Todos los derechos reservados. 夽 The authors thank Hernando Vargas, participants at the economics seminar of Universidad Javeriana and the members of the Macroeconomic Models and Programming and Inflation Departments at the Banco de la Republica for their comments. The views expressed in the paper are those of the authors and do not represent those of the Banco de la Republica or its Board of Directors. ∗ Corresponding author at: Cra 7 #14-78, Bogota, Colombia. E-mail addresses: (J.A. Parra-Polania), (C.O. Vargas). http://dx.doi.org/10.1016/j.espe.2014.08.002 0120-4483/© 2014 Banco de la República de Colombia. Published by Elsevier España, S.L.U. All rights reserved. Document downloaded from http://zl.elsevier.es, day 07/01/2015. This copy is for personal use. Any transmission of this document by any media or format is strictly prohibited. 42 J.A. Parra-Polania, C.O. Vargas / Ensayos sobre Política Económica 32 (2014) 41–47 1. Introduction In general, increments in the volatility of an estimated variable could be attributed to an increase in the volatility of the actual variable or to an increase in the volatility of its measurement error. In practice, increments in the error volatility of estimated macroeconomic variables hinder the process of taking the appropriate economic policy measures. In the decision-making process, the first step is to try to determine what proportion of the observed increase in volatility is the result of measurement error and, once this is established, the second step is to determine what is the optimal policy reaction given these circumstances. This paper focuses on the analysis of the second step assuming that the increase in volatility corresponds exclusively to measurement error. Specifically, the paper analyzes from two different perspectives how monetary-policy actions should change when there is higher volatility in the measurement error of the aggregate economic activity. The first perspective corresponds to the standard problem of loss minimization by the monetary policymaker. The second perspective has had good reception recently in economics and corresponds to the concept of “robustness” in which the policymaker seeks to minimize the maximum possible loss across all potential conditional1 scenarios as a form of prudence to avoid huge losses. Previous literature has explored the effect of uncertainty on optimal decision making in the framework of optimizing models. Aoki (2003), in a model with nominal price stickiness and asymmetric information, concludes that a central bank that seeks to minimize its expected loss but that faces uncertainty about the actual state of the economy, would exhibit some degree of cautiousness, that is, the central bank would not respond too strongly to noisy indicators of the economy. In the context of a model with symmetric information, Svenson and Woodford (2003) conclude that the optimal response to the imperfect observation of output depends on the noise contained in its indicator.2 Orphanides (2003) shows that inefficient policy rules are followed if the noise in signals of economic variables is not taken into account, resulting in excessively activist policy. In that sense, policy reactions should be cautious and less sensitive to unfiltered data. A general result in this literature is that policymakers should recognize the existence of measurement errors in the information at their disposal, and therefore they should act cautiously in the sense of avoiding overreaction. This result is found assuming that the policymaker knows the distribution function of all possible events and minimizes the expected loss of those events, that is, all the abovementioned papers study the problem of noisy indicators from what we call the standard perspective. With regard to the second perspective, attention has being brought to the fact that in most instances policymakers do not know the probabilities of all relev (...truncated)