Systemic risk in micro level: the case of “ cheques-as-collateral ” network

Download Presentation

Systemic risk in micro level: the case of “ cheques-as-collateral ” network

Loading in 2 Seconds...

- 104 Views
- Uploaded on
- Presentation posted in: General

Systemic risk in micro level: the case of “ cheques-as-collateral ” network

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Aristotle University, Mathematics Department

Master in Web Science

supported by Municipality of Veria

Systemic risk in micro level: the case of “cheques-as-collateral” network

MichalisVafopoulos,

vafopoulos.org

joint work with D. Soumpekas and V. Angelis

21/10/2011

- Financial crisis: a network explanation
- Why networks?
- Systemic risk and financial contagion
- The “cheques-as-collateral” network
- Data and model
- Results
- Further extensions

- 2007: Started from US sub-prime and disseminated rapidly to the global real economy
A reality: Regulation based on binary relations

- Government & bank
- Bank & customer
and a dogma: “too big to fail”

- Research on correlation and market risk
(VaR-like metrics)

Current risk systems cannot:

- Predict failure cascades.
- Account for linkages.
- Determine counterparty losses.

But the financial system (+info) is:

A global networked system

So,

+ “too interconnected to fail”

How to model it?

Networks!

- Easy to model and visualize relations
- Easy to calculate major statistics
- The study of the Web network help us to conclude that most of real networks are:
- Self-similar (Scale-free)
- Small worlds

Networktheory and related fields

Financial Network Analysis

Web Science

Social Network Analysis

NETWORK THEORY

Computer Science

Graph & Matrix Theory

Biological Network Analysis

- Define:
- Node (e.g. person, business)
- Link [directed or not] (e.g. friendship, commerce)
And if necessary:

3. Evaluation of node (e.g. score, potential)

4. Evaluation of link (weight) (e.g. trust)

0.54

4

5

Financial networks

Focused on banks, financial institutions etc.

Federal funds

Italian money market

Bech, M.L. and Atalay, E. (2008), “The Topology of the Federal Funds Market”. ECB Working Paper No. 986.

Iori G, G de Masi, O Precup, G Gabbi and G Caldarelli (2008): “A network analysis of the Italian overnight money market”, Journal of Economic Dynamics and Control, vol. 32(1), pages 259-278

What about trying model systemic risk directly from bank customers?

Financial systemic risk (definitions)

- The risk of disruption to a financial entity with spillovers to the real economy.
- The risk that critical nodes of a financial network fail disrupting linkages.
- Financial contracts with externalities.

- Nodes: cheque issuers & recipients
- Link ij : customer i issues cheque to customer j
- Weight of link: the fraction of the value of cheques that customer i have issued to customer j, to the total value of cheques in euros received by the bank
Cheque recipients use their

incoming chequesas collateral

to working capital credit.

(based on Martínez-Jaramillo et al., 2010).

Step 0

- Assume a set of criteria for the failure of every customer (c).
Here it is assumed that c=50% of the total amount of the unpaid cheques that drives every customer to failure.

2.For a given “cheques-as-collateral” network, calculate the weighted adjacency matrix (W).

Step 0

3. Calculate the failure threshold for every customer j:

It is assumed that this threshold remains constant in every stage k.

4. Assume a set of customers that initially fail to pay their cheques (Dk=0).

This set can be chosen by some relevant criterion. In our case, five customers with the highest weighted out-degree have been selected to collapse at stage k=0.

Step 1

- Calculate the sum of the defaulted exposures of failed customer i to j:

Step 1

2. Compare the calculated defaulted exposure failure threshold of customer j.

3. Update Dk with the failed customers.

Step 2

- Repeat Step 1 until Dk=Dk+1.

Stage 0

Number of failed nodes: 5

Decrease in total value: 17%

Stage 1

Number of failed nodes: 4

Decrease in total value: 27%

Stage 2

Number of failed nodes: 3

Decrease in total value: 38%

Stage 3

Number of failed nodes: 2

Decrease in total value: 41%

After the shock

Number of failed nodes: 14

Decrease in total value: 41%

- Assume that only a customer fails
- Ceteris paribus
- Calculate financial contagion
- Compare to others
- Weight factors like stage, sector etc
- So, variety of hypothesis for the stage-by-stage loss function

- decreasing stage-by-stage loss
- composite loss (e.g. weight)
- systemic risk assessment (e.g. cheque issuer)

- More data and metrics
- Model the initial shock
- Reverse logic: business development “multiplier” for banks
- and other sectors…
Thank you.

More at www.vafopoulos.org

Questions?

Our model is based on the idea of the Systemic Risk Network Model that accounts for bank failures in the financial system (Martínez-Jaramillo et al., 2010).

the total adjusted loss is calculated by weighting stage 0 loss with 0.5, stage 1 loss with 0.25 and stage 2 loss with 0.125.

taking into account her weight in the network.