This paper characterizes the derivation and the assessment of design limits in the case of a regime-switching economy. The object of the analysis on design limits is to derive the restrictions on how feedback rules, the Taylor-type rules typically used in monetary economics, affect the frequency fluctuations underlying the state variable of interest. We extend the analysis in a structured context of model uncertainty where the uncertainty is described by the presence of different potential models whose probability of occurrence and switching is given by a known and ergodic Markov Chain transition matrix. The presence of switching modifies the characteristics of design limits in two main aspects. First, when the optimal variance minimizing rule is chosen, frequency specific restrictions appear more or less stringent with the respect to the linear case depending on the probability of switching: the higher it is, the worst is the performance in terms of frequency-specific fluctuations. Second, contrary to the linear case, design limits are also affected by the policy rule so that their role switches from a constraint to an externality that the policymaker may want to take into account.
C52 - Model Evaluation and Testing ; E6 - Macroeconomic Policy Formation, Macroeconomic Aspects of Public Finance, Macroeconomic Policy, and General Outlook
