Dear all,
I have a question about running perfect foresight simulations in Dynare when studying a fiscal consolidation that is implemented via an exogenous shock path rather than through a fiscal feedback rule.
In my stochastic model, I include a fiscal rule (e.g., taxes responding to deviations of debt from steady state) to ensure debt stationarity and satisfy the BK conditions.
However, for a deterministic policy experiment, I want to:
- Start from a higher debt level than steady state,
- Impose a consolidation path using a sequence of shocks (e.g., higher taxes or lower spending),
- Simulate the transition under perfect foresight,
- Without relying on a fiscal feedback rule (i.e., setting the feedback parameter to zero).
My first question is:
In a perfect foresight setup, is it acceptable to shut down the fiscal rule and model consolidation purely through a pre-specified shock path, provided that the imposed policy path ensures debt eventually stabilizes? Or does Dynare still require an active fiscal feedback rule in the model to guarantee transversality and solvability?
Related to this, I also have a question about estimation:
When the fiscal rule is shut down, the stochastic version of the model does not satisfy the BK conditions and estimation fails. Does this mean that:
- The model should be estimated with the fiscal rule in place (to ensure determinacy),
- And then deterministic consolidation experiments can be run separately (possibly shutting down the rule)?
Or is the standard practice to keep the fiscal rule active both for estimation and for perfect foresight simulations?
More generally, what is best practice when one wants to estimate a model with a stabilizing fiscal rule, but run deterministic policy experiments that are implemented via exogenous shock paths rather than via that rule?
Thank you very much for any guidance.