Academic journal article Journal of Money, Credit & Banking

Has Monetary Policy Been So Bad That It Is Better to Get Rid of It? the Case of Mexico

Academic journal article Journal of Money, Credit & Banking

Has Monetary Policy Been So Bad That It Is Better to Get Rid of It? the Case of Mexico

Article excerpt

MANY LATIN AMERICAN COUNTRIES are considering adopting the U.S. dollar as legal currency, and some, like Ecuador, have taken concrete steps in that direction. Proponents of dollarization generally hold the view that domestic monetary policy has been the primary cause for the economic instability experienced by these countries in the past three decades. Yet, at least for Mexico, very few empirical studies have tried to identify the role of monetary policy.

The existing empirical literature on Mexican monetary policy consists mainly of single equation estimations (see Calvo and Mendoza 1996 and Kamin and Rogers 1996), or of reduced-form vector autoregressions (see Copelman and Werner 1995 and Hernandez 1999).(1) The first class of models is silent on the impact of monetary policy on the rest of the economy. The second class of models, by definition, cannot identify monetary policy. In addition, all previous literature has either ignored the issue of changes in regime, or has confined itself to the study of monetary policy within regimes. This despite the fact that some of Mexico's major crises occurred during the passage from one regime to another. A proper evaluation of the impact of monetary policy on the Mexican economy requires that these critical transition periods are considered.

This paper attempts to fill the existing gap by estimating an identified vector autoregression (VAR) model for the Mexican economy. The model is a "small open-economy" VAR, in the vein of Cushman and Zha (1997). This approach makes it possible to account for the impact of foreign shocks on the Mexican economy. We use monthly data from 1976 through 1999. We develop a methodology that extends the standard identified VAR framework in order to take into account: the changes in monetary policy regime that occurred during this period.

The paper addresses two specific questions. First, how large has the impact of exogenous shocks to monetary policy been on economic activity? Second, did the endogenous response of monetary policy mitigate or exacerbate the impact of disturbances originating elsewhere? If erratic movements in monetary policy have been the source of large fluctuations in real activity, proponents of dollarization can argue that adopting the U.S. dollar may bring more stability to the economy. Likewise, if we find that the lack of an adequate response on the part of the Banco de Mexico to domestic or external shocks has been detrimental for the economy, then adopting a different "policy rule," for instance, the one imposed by dollarization, may be welfare enhancing for Mexico.(2)

In regard to the first question, we find that exogenous shocks to monetary policy have had a small impact on real activity, as well as on prices. Friedman's view that erratic movements in monetary policy are the primary source of business cycle fluctuations finds little empirical support. This finding is consistent with the evidence obtained for the U.S. (see Leeper, Sims, and Zha 1996 and Sims 1998b) and for other industrialized countries (Kim 1999). Our results also indicate that the U.S. business cycle, and not only movements in commodity prices, have had a strong impact on the Mexican economy.

In regard to the second question, we study the implications of alternative policy regimes by means of a counterfactual example, following the approach in Bernanke, Gertler, and Watson (1997), Leeper and Zha (1999), and Sims (1998a). In particular, we ask whether the 1994 crisis could have been averted had the monetary policy regime not changed in December 1994. The experiment suggests that it would have been averted.

The paper continues as follows: section 1 illustrates the model and the identification structure; section 2 describes the data; section 3 analyzes the results; section 4 concludes.


The model is a "small open-economy" vector autoregression, similar to the one successfully adopted by Cushman and Zha (1997) in their study of Canadian monetary policy. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.