Multiple Equilibrium and BK condition

Perfect foresight simulations
1

Dear all，

I am trying to make a model with multiple equilibrium base on BGG.
One equilibrium BK conditions are satisfied while the another are not(no stable equilibrium).
I want to study two equilibrium transition path using perfect foresight simulations.
when I simulate the equilibrium(BK conditions not satisfied) as initial value and the another(BK conditions satisfied) as end value is fine.

But when I simulate the equilibrium(BK conditions satisfied) as initial value and the another(BK conditions not satisfied) as end value, there are a lot of problem.
some variable like bankruptcy rate become negative, and it can simulate only with linear_approximation option.

Can someone help me with this?

Thank you!

2

I am not sure I understand the exercise. If there is no stable equilibrium, the final steady state will not be an attractor. As soon as you leave the steady state, the system will diverge.