Stability problems in a two country model

[pens]

Dear All,

I am simulating a two country, two goods model, with unilateral transfers from one country to the other, and a contemporaneous exogenous TFP increase in the latter country. I start from an initial steady state with zero transfer and a given TFP differential between the 2 countries. The model ends in a final steady state with transfers and a new (lower) tfp differential. I modelled the TFP increase by using a logistic map. I paid attention to chose numerical values such that the logistic map has a well behaved dynamics.

The model works fine and is able to compute the transitional dynamics, in the case of no capital mobility.

However, as soon as I try to introduce capital mobility, Dynare is not able to pass from one steady state to the other. Apparently, the system does not comply with Blanchard-Kahn conditions. I have tried several modifications and tricks, and verified that even for different specifications of the shock the model does not satisfy the rank condition, even though in some of those attempts B-K are verified.

I typically works with Dynare 3.065, but I have tried the model also on Dynare 4.0.2 with no success. I enclose the code for the capital-mobility case.

Where could the problem be?