Two-country DSGE model: common monetary policy

[loesverstegen]

Dear all,

For my research I have set up a two-country DSGE model with standard features such as habit formation, price and wage rigidities and a government sector. The two countries are (for the moment) supposed to be symmetric, and the model is close to Kolasa(2009). The model runs fine as long as I assume that each country sets its own interest rate, when I want to make it a monetary union by imposing a common interest rate set by a common central bank, then there are only 21 eigenvalues larger than 1 for the 22 forward-looking variables, so that the rank condition is not verified.

To me it seems that the only way to solve this is to make sure that the interest rates are not exactly the same, however I would really like to have one common interest rate in this model. I have attached the dynare code, the equations for monetary policy are to be found on line 374-383. Does anyone of you have a solution to this issue, or a way to work around it?

I am looking forward to your responses!

Kind regards,
Loes

[costa]

I think you could impose some correlation between the interest rates or monetary policy shocks. Just an idea, not sure it suits you or works.

[jpfeifer]

This indicates something more fundamental as the way you seem to close your model is not compatible with a unique stable solution. You need to find out how to correctly close the model when moving to a currency union. Did you keep both interest rates in the model or did you eliminate one by replacing both interest rates with the common one?

[loesverstegen]

Thank you for your comments!

I have been able to find out how to close the model correctly: the common interest rate had to be determined by an interest rate rule that takes into account both the lagged union-wide inflation and output gap, as well as the differential inflation and output gap (contemporaneous minus the lagged output gap). Then it works, if you also make sure that the coefficient of the tax rates (consumption and capital income tax rates) with respect to the government debt level are not too high. Just for your info

Thanks for the help!

[sanga]

Hello,
I have two country model which run perfectly with independent monetary policies, but when im trying to run the model with a single monetary policy in order to have a monetary union(the taylor rule contains the weighted average of output and inflation of both countries), the following message appears:

[i]There are 19 eigenvalue(s) larger than 1 in modulus
for 20 forward-looking variable(s)

The rank conditions ISN'T verified![/i]

Someone can help me please, because i got stuck now? Here the mod file;

[jojokre]

Hi:
I ran into the same problem and, as loesverstegen recommended, experimented a bit with the arguments of my Taylor rule. However, so far I didn't find a way to fix my model. I wonder if there shouldn't be some more general fix to it, as jpfeifer said, some more fundamentally different way of closing the model.
Sanga, did you find a way of fixing your model? Or can someone maybe point out some literature which might help in this context?
I would really appreciate any help!
Best, Johanna

[Hugo]

Dear contributors & moderators,

I would be interested in knowing whether you have found a solution to the instability issue arising from a two-country monetary union framework. I have the same issue as discussed in several posts
viewtopic.php?f=1&t=5948&p=16277&hilit=rank+condition+monetary+union#p16277
viewtopic.php?f=1&t=5781&p=15714&hilit=rank+condition+monetary+union#p15714

That is when I build a monetary union framework with common monetary policy, I have 9 eigenvalues larger than 1 and 10 forward-looking variables, i.e, one eigenvalue less than what is needed for stability. It seems that several of us came across this issue, but I do not see any solution on the forum.

Have you found a solution to get as many eigenvalues larger than one as forward looking variables when in a monetary union framework?

All the best,

Hugo

[jpfeifer]

One problem obviously are specification errors. See my reply to your original post at http://www.dynare.org/phpBB3/viewtopic.php?f=1&t=7058#p20692