ARGEM: A Dynamic Stochastic General Equilibrium Model for Argentina

The last few years have seen an explosion of Dynamic and Stochastic General Equilibrium (DSGE) models built for policy analysis and forecasting in industrialized countries. The set of papers presented to the recent joint U.S. Federal Reserve Board-European Central Bank-IMF conference: "DSGE Modeling at Policymaking Institutions: Progress & Prospects" is a significant sample. The need for better microfounded models that can contribute to policy analysis is also experienced by developing country Central Banks. On top of the many difficulties encountered in developed countries in building, calibrating and/or estimating these models, those who seek to construct models that can be relevant in the developing country context find various additional difficulties. One of these stems from the fact that the models built for industrialized countries typically assume a freely floating exchange rate and hence can avoid modeling exchange rate policy. Most developing countries do not have a pure exchange rate float and their Central Banks regularly intervene in the foreign exchange market with varying degrees of intensity and frequency. While the opposite "corner" of a pure interest rate float with a monetary policy based on determining a path for the nominal exchange rate is not difficult to model, one of the challenges faced by developing country modelers is to incorporate intervention in the foreign exchange market as an additional tool available for a Central Bank that also intervenes in the "money" market (typically by determining an operational target for the short run interest rate). This is one of the main objectives of this paper, which on this topic builds on previous analysis by the author (see Escudé, 2006). The paper benefits from various recent developments in monetary macroeconomic modeling, including Christiano, Eichenbaum and Evans (2001) (CEE), Smets and Wouters (2003), Woodford (2003), and Adolfson, Laséen, Lindé and Villani (2005) (ALLV), to mention but a few. The model has been calibrated for the Argentine economy, and solved using Klein's (2000) generalized Schur decomposition methodology.