摘要：This paper investigates the implications of cross-country heterogeneity within
the euro area for the design of optimal monetary policy. We build an
optimization-based multicountry model (MCM) describing the euro area in which
differences between structural parameters across countries are allowed. Using
Bayesian techniques, we estimate the MCM and its areawide counterpart (AWM).We
then question which model is the most appropriate for monetary policy purposes.
Several results emerge. First, using an AWM induces relatively large and
significant welfare losses. Second, this is not the use of a rule based on
aggregated variables that is costly in terms of welfare, but rather the use of a
suboptimal forecasting model. Third, allowing for habit on consumption has
important implications for the dynamics of models, but taking into account
differences in price indexation has more drastic effects on welfare losses.
