Is it important to incorporate in a small open economy model an interest rate that is debt elastic à la SGU (2003) in a perfect foresight model ? I reckon that this is done to induce stationnarity in the model, but perfect foresight does not need stationary to perform simulation.
Note that my model is not a RBC model, but rather looks at the very long run (2100) with a dynamic general equilibrium model. I was also wondering if I could just have a trade balance in my model and no foreign bonds to close this trade balance.
It depends on what exactly you are trying to do. Even in stochastic models, you would not necessarily have to induce stationarity. The issue in your case may be that you need to be able to compute the terminal steady state, which is easier with stationary assets.
Thank you. I first did not have foreign bonds or bonds in my model, because this is not what I’m interested in in my model. I do have a trade balance because I’m interesting in flows of different types of goods.
I was thinking in taking out this foreign bonds and keeping that trade balance. Is it not too restrictive ?
I am not sure I understand. Having a non-zero trade balance always implies movements in the net foreign asset position. So there needs to be some asset.
Am I obliged to have a CPO for foreign bonds demand from households ?
I just want my net assets to close the trade deficit and fix the foreign interest rate without going through the SGU stationarity debt-elastic interest rate.
I already have the value at SS from my equations for TB.
So having B_foreign(+1) = (1+interest_rate)*B_foreign+TB, it’s pretty straight forward to find B_foreign by having a fix interest rate.
All I am saying is that there will always implicitly be an NFA. But in many models, you may not need to model it explicitly. The steady state trade balance will imply a particular NFA position.