Code with Financial Accelerator

[bigbigben]

Hi, Guys:
Do you know someone using Dynare to code Financial Accelerator in a DSGE model like BGG model? I want to learn the stuff but I can’t find a reference code. I want to use the code to do some exercise.

Best

[EnzoCerletti]

I wish I knew any example I could lay my hands on!
I’m working myself in a model based on BGG, but the clock keeps ticking, so finding a close reference would make my life much easier

[fabio_portugal]

Here is the code for BGG model.

I hope it could be useful to you.

Best Regard

Fabio
BGG.mod (1.92 KB)

[bigbigben]

[quote=“fabio_portugal”]Here is the code for BGG model.

I hope it could be useful to you.

Best Regard

Fabio[/quote]

That helps a lot. Thank you very much.

[dilsat]

Hello Fabio, I have been working on your financial accelerator code for my thesis. My model is a little bit more complicated than this but yours helps me very much to understand the steps I should follow. So, first of all, thank you for that. I have a minor question about the code: I see that you wrote all the equations that is written on BGG(1999) but the evolution of net worth equation which is eqn(4.21) in the paper is different from the code! As far as I see this is not an algebraic manipulation. Why did you do such change? Secondly, variables r, k, n and i are written with one period lag although they are not in the paper. Do you have any purpose for doing that?

I am a beginner at using Dynare and these questions may be too easy for you but if you help me I would be glad. In my model, there is more than one interest rate and I do not know how to deal with them.

Thank you in advance.
Dilşat

---

[quote=“fabio_portugal”]Here is the code for BGG model.

I hope it could be useful to you.

Best Regard

Fabio[/quote]

[poelser]

Hi Fabio, hi community,

does anyone know of a not log-linearized version of the BGG99 financial accelerator?

Edit: And maybe a question regarding the intuition for the steady state: If I would switch off any financial frictions, the external finance premium equals zero and the risk free interest rate is equal to the expected return on capital. If I would now switch the frictions on and the premium is positive, shouldn’t the return on capital equal the risk free rate plus the external finance premium? Just to make sure I understood the steady state mechanics.

Thanks for your reply!
Best

[jpfeifer]

Please see

Steady state in BGG 1999

Has anybody ever replicated the steady state in the financial accelerator paper from BGG (1999)? According to BGG they want a steady state threshold value that implies an annual bankruptcy rate of 3%. Given that the threshold is log normally distributed with the moment conditions they describe this should give a threshold value of 0.4936 (adjusted so that we can work with quarterly rates, as in the paper). BGG also declare that the annual spread between the return on capital and the risk free r…

And yes, your intuition seems correct (if we are talking about net rates up to first order)

[poelser]

Dear Prof. Pfeifer, dear forum,

thank you for your reply. Either you misunderstood my question or I did not understand the thread you linked. I was not able to find a code with the BGG99 financial accelerator where the model equations are not linearized.

So I build one myself (finacc.mod (3.6 KB)). However my previous question regarding the returns and interest rate remains: Why is the return on capital in steady state greater than the sum of interest rate (risk-free) and external finance premium?

I also checked this result in the model from Christiano, Motto, Rostagno (2010) obtained from Fabio Verona’s website ( CMR_FA.mod (14.1 KB)). And it holds. I edited the model to make the external finance premium an additional variable and added Int1 (sum of interest rate and EFP) and Int2 (non-default interest rate from cutoff-value equation). Not only that they are not identical (from my understanding they should) they are both smaller than the return on capital.

Would that not provide an opportunity for arbitrage? I could borrow funds to acquire one more unit of capital stock with expected returns higher than the interest rate… It seems I am missing a fundamental ingredient of the story. I appreciate any hints!

Thank you for your time

[jpfeifer]

1. The thread that I linked above suggests that people have failed calibrating the nonlinear version of the original BGG model.
2. You need to be more precise about which objects we are talking. Is the return to capital the one after you have subtracted depreciation of capital?

[poelser]

Dear Prof Pfeifer, dear forum,

1. I understand, thank you.

2. I was referring to the return on capital after depreciation. In finacc.mod it is Rk = (MPK + (1-delta) * Q) / Q(-1) and in CMR_FA.mod it is 1+RkXU (equation A.10).

From my understanding the interest rate for entrepreneurs should equal risk-free rate plus external finance premium (finacc.mod: R + mu * G * Rk * Q(-1) * K / (Q(-1) * K - N) where I calculate the external finance premium identical to BGG99 as default costs over amount borrowed.

First observation: The entrepreneurial interest rate is not equal to the non-default loan rate Z (as specified in eq. 3.3 in BGG99): Z(+1) = omegabar * Rk(+1) * Q * K(+1) / (Q * K(+1) - N(+1). I do not know why.

Second observation: Both, the entrepreneurial interest rate and the non-default loan rate are below the return on capital
R+EFP < Rk (finacc.mod) and ReXU + EFP = Int1 < RkXU (CMR_FA).
AND
Z(+1) < Rk (finacc.mod) and Z(+1) = Int2 < RkXU (CMR_FA)
As far as I understand this would leave room for arbitrage.

Your help is much appreciated as I know this question might not directly relate to dynare but might be caused by my ignorance of some mechanism of the model I do not see.

Thank you!