Discussion Paper

No. 2017-66 | September 25, 2017
Policy experiments in an agent-based model with credit networks

Abstract

In this paper the authors build upon Assenza et al. (Credit networks in the macroeconomics from the bottom-up model, 2015), which include firm-bank and bank-bank networks in the original macroeconomic model in Macroeconomics from the bottom-up (Delli Gatti et al., Macroeconomics from the Bottom-up, 2011). In particular, they extend that framework with the inclusion of a public sector and other modifications in order to carry out different policy experiments. More specifically, the authors test the implementation of a monetary policy by means of a standard Taylor rule, an unconventional monetary policy (i.e. cash in hands) and a set of macroprudential regulations. They explore the properties of the model for such different scenarios. Their results shed some light on the effectiveness of monetary and macroprudential policies in an economy with an interbank market during times of crises.

JEL Classification:

C63, E51, E52

Assessment

  • Downloads: 183

Links

Cite As

Tiziana Assenza, Alberto Cardaci, Domenico Delli Gatti, and Jakob Grazzini (2017). Policy experiments in an agent-based model with credit networks. Economics Discussion Papers, No 2017-66, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2017-66


Comments and Questions


Anonymous - Referee report 1
October 11, 2017 - 21:02

see attached file


Tiziana Assenza, Alberto Cardaci, Domenico Delli Gatti, and Jakob Grazzini - Reply to Referee Report 1
October 26, 2017 - 16:39

See attached file


Hazem Krichene - Comment
October 30, 2017 - 14:02

See attached file


Tiziana Assenza, Alberto Cardaci, Domenico Delli Gatti, and Jakob Grazzini - Tiziana Assenza, Alberto Cardaci, Domenico Delli Gatti, and Jakob Grazzini - Reply to Comment
November 09, 2017 - 16:05

See attached file


Sander van der Hoog, Bielefeld University, Germany - Comment
November 13, 2017 - 09:39

I would like to comment on the issue of bankruptcy by many banks in Fig. 3. These banks all seem to have negative equity when they go bankrupt. For normal, non-financial firms, a negative net worth would indeed be a point at which they should definitely stop their productive business. ...[more]

... For banks however this is a bit different. Banks should stop all new loan business when they are below their regulatory requirements. Having a capital buffer a la Basel II or III therefore would mean that banks already default when they reach their minimal regulatory capital. So then the credit crunch that results from these banks going into default should already kick in a bit earlier even.

This may sound like bad news, but there is also some good news: I have a suggestion on how to improve it! What this model needs is loan loss reserves that form a buffer against expected losses. This buffer should protect the banks' own residual equity from the bad debt write-offs by the firms. If you could include such a dynamic loan loss provisioning by the banks, this would work counter-cyclical; at least that's what the buffer is supposed to do. Also, the relation you have made with SIFI's is relevant here, because the regulator could increase the regulatory capital buffer just for the large banks, in order to prevent such systemic failures.

In another context, we have made good experience with this, and it seems to resolve the issue of a sudden credit crunch that affects a large majority of firms. Instead, the banks will build-up their loan loss reserves endogenously, and prevent such sudden, systemic, credit market failures. Another solution would be to cut off credit to the most risky firms. Such a Minsky-inspired regulation would quickly let all speculatively financed and Ponzi financed firms default, before they can build up a lot of risk exposure on the banks' balance sheets. But this is more a borrower-targeted policy that focuses on the risk of the firms, rather than a lender-targeted policy which focuses on the banks. Both seem to be important for financial stability.


Tiziana Assenza, Alberto Cardaci, Domenico Delli Gatti, and Jakob Grazzini - Reply to Comment
November 30, 2017 - 18:33

We are grateful for your very useful suggestion. We have already discussed the possibility of introducing a similar kind of mechanism, precisely because, in line with your comment, we think that the presence of a loan-loss reserve buffer would prevent the system from recording such a large number of bank ...[more]

... defaults. The second type of solution you suggest – that is the Minsky-inspired one – is instead something that we might be going to implement in future expansions of the model.
Hence, once again, we appreciate very much your suggestions, which we will surely take into account in the revision of the model.