Help with RANK Condition not being satisfied

[Nice]

Hello,

I am getting the following error:

There are 4 eigenvalue(s) larger than 1 in modulus
for 5 forward-looking variable(s)

The rank condition ISN'T verified!

Error using print_info (line 45)
Blanchard Kahn conditions are not satisfied: indeterminacy
Error in stoch_simul (line 92)
print_info(info, options_.noprint, options_);
Error in liq_b (line 456)
info = stoch_simul(var_list_);
Error in dynare (line 223)
evalin('base',fname) ;

I have tried using Model_diagnostics but it cannot identify any problems with the .mod file

I have a feeling the problem could be with equation 8 and 9 as they can be collapsed into one equation but then I am not sure which endogenous variable to eliminate. Could someone kindly provide some guidance.

Many Thanks!

[jpfeifer]

That is generally hard to tell. Most probably there is still a timing problem or a problem with the parameterization. I would recommend simplifying the model by adding lump sum taxation to balance the budget and see whether it runs. Only then add the financing rules. This assures that it is not a matter of the parameterization of the fiscal rules.

[Nice]

Thank you Jpfeifer!

[hassen]

dear all
i have the same problem.
could anyone help me
please
this is the model

[jpfeifer]

Dear hassen, please check the timing of all of your equations. It is most probably a timing error.

[dynare16]

Hello,

I also have the same message error when running my code.

I checked the timing of my variables many times and cannot see where the problem comes from.

As I have one more forward-looking variables in comparison with the number of explosive eigenvalues, I changed the timing of the risk-free rate that I have in my model such that it is no longer predetermined, and then the model is running.

But in my view this does not make sense economically, as the risk-free rate is a predetermined variable, so it should be written in Dynare in t when received in t+1, right ? But then, the Blanchard-Kahn conditions are not satisfied.

Could it be that in my model, default on risk-free bonds can occur ex-post, even though I didn't explicitly allow for default ?

Thanks a lot

[jpfeifer]

@dynare16 Without the equations, it is hard to tell what you are talking about. But in most models the risk-free rate between time t and t+1 is not predetermined. The reason is that at time t the shocks first realize and only then the interest rate is chosen. This implies that the risk-free rate is chosen at time t, making it a regular jump variable.

[hassen]

thank u so much jpfeifer
i'll check

[dynare16]

Thanks for your answer !

Here is my mod file.

[jpfeifer]

The file runs and looks sensible, so what it the problem?

[dynare16]

The code is running because the risk-free rate is not predetermined here. When it is predetermined, Blanchard-Kahn conditions are not satisfied. So I just want to make sure that a non-predetermined risk-free rate is consistent in my model.

Thanks

[jpfeifer]

Now I see what you mean with predetermined risk-free rate. Indeed your timing for the risk-free rate in the model looks wrong.
The underlying reason why the wrong timing makes the model run might be the same as in http://www.dynare.org/phpBB3/viewtopic.php?f=1&t=3564

[dynare16]

Thanks for your help !

Then what should I conclude ? The risk-free rate which is paid in period t should be written r(-1) in Dynare, right ? And then the problem with the Blanchard Kahn conditions comes from another variable ? But I can't see which one.

[jpfeifer]

Yes, the risk-free rate which is paid in period t should be written r(-1).
Please follow the above link I gave you and check it out, because I think your problem is related to one stock (debt) that is predetermined being split up within the period for two uses. In this case, the latter are not predetermined anymore.

[dynare16]

It seems to me that the stock of debt is not predetermined : in period t, households demand debt, firms demand debt and market clears, and both sectors use that debt to fund projects in period t already. And then the risk-free rate is paid in period t on debt borrowed in period t-1 but it is already determined in period t-1 at the time market clears. Am I wrong ?

In the model, I used one variable for households' demand for debt, one variable for firms' demand but only one variable for debt supply from patient households. Is this wrong ? I considered that, as both debts are riskless, they are bought on a unique market with a unique borrowing rate. Maybe I should use two distinct variables for debt supply from patient households : one for firms and the other one for impatient households, and then two market-clearing conditions ?

Thanks !