I am trying to solve Gali and Monacelli (2005)'s small open economy model using Dynare (see the attached .mod file). However, Dynare says that the Blanchard-Kahn conditions are not satisfied (6 eigenvalues larger than 1 in modulus for 7 forward-looking variables).
What worries me is that, the calibration and equations I use are identical to the ones given in Monacelli’s own Gauss code (irnew3.prg) which is given on his website. I have run his code and it works fine. He uses the Anderson-Moore algorithm to solve the system.
I’ve attached a working version of the Gali Monacelli model. It reproduces the impulse responses in Figure 1 of the NBER Working Paper.
Not sure why Dynare has trouble with the UIP condition. I have a hunch: from equation (18) and Appendix 1 of the paper, we know that the exchange rate is pinned down uniquely; but Dynare can’t read Appendix 1. So when you tell it “e(+1) - e = r - r_star”, it thinks that “e” isn’t determined. (But then why would Monacelli’s AIM be able to solve it? — and given that, why would Dynare’s AIM fail…?)
In any case, as the authors explain, UIP is redundant if you have a risk-sharing equation. So I followed that route.
A replication file for the Gali/Monacelli (2005): Monetary Policy and Exchange Rate Volatility in a Small Open Economy", Review of Economic Studies 72, pp. 707-734 is available at github.com/JohannesPfeifer/DSGE_mod
He uses the Anderson-Moore algorithm to solve the system.