CoChat

Cascading Failures in Bi-partite Graphs: Model for Systemic Risk Propagation

Open access

The DOI matches its Crossref record; 25 of the 25 references it cites were found in a registry or an index.

Abstract

As economic entities become increasingly interconnected, a shock in a financial network can provoke significant cascading failures throughout the system. To study the systemic risk of financial systems, we create a bi-partite banking network model composed of banks and bank assets and propose a cascading failure model to describe the risk propagation process during crises. We empirically test the model with 2007 US commercial banks balance sheet data and compare the model prediction of the failed banks with the real failed banks after 2007. We find that our model efficiently identifies a significant portion of the actual failed banks reported by Federal Deposit Insurance Corporation. The results suggest that this model could be useful for systemic risk stress testing for financial systems. The model also identifies that commercial rather than residential real estate assets are major culprits for the failure of over 350 US commercial banks during 2008-2011.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/srep01219

Cites

25 of 25 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Bank-asset bipartite network model with banks as one node type and assets as the other node type. Link between a bank and an asset exists if the bank has the asset on its balance sheet. Upper panel: illustration of bank-node and asset-node. Bi,m is the amount of asset m that bank i owns. Thus, a bank i with total asset value Bi has wi,m fraction of its total asset value in asset m. si,m is the fraction of asset m that the bank holds out. Lower panel: illustration of the cascading failure process. The rectangles represent the assets and the circles represent the banks. From left to right, initially, an asset suffers loss in value which causes all the related banks' total assets to shrink. When a bank's remaining asset value is below certain threshold (e.g. the bank's total liability), the bank fails. Failure of the bank elicits disposal of bank assets which further affects the market value of the assets. This adversely affects other banks that hold this asset and the total value of their assets may drop below the threshold which may result in further bank failures. This cascading failure process propagates back and forth between banks and assets until no more banks fail.
Figure 1 Bank-asset bipartite network model with banks as one node type and assets as the other node type. Link between a bank and an asset exists if the bank has the asset on its balance sheet. Upper panel: illustration of bank-node and asset-node. Bi,m is the amount of asset m that bank i owns. Thus, a bank i with total asset value Bi has wi,m fraction of its total asset value in asset m. si,m is the fraction of asset m that the bank holds out. Lower panel: illustration of the cascading failure process. The rectangles represent the assets and the circles represent the banks. From left to right, initially, an asset suffers loss in value which causes all the related banks' total assets to shrink. When a bank's remaining asset value is below certain threshold (e.g. the bank's total liability), the bank fails. Failure of the bank elicits disposal of bank assets which further affects the market value of the assets. This adversely affects other banks that hold this asset and the total value of their assets may drop below the threshold which may result in further bank failures. This cascading failure process propagates back and forth between banks and assets until no more banks fail.CC BY-NC-ND · PubMed Central
Figure 2 Comparison of probability density functions (PDF) of weight of typical assets and equity ratios between all banks and FDIC listed failed banks for 2007. (a) PDF of the weight of loans for construction and land development in banks' total asset. (b) PDF of the weight of loans secured by 1-4 family residential properties in banks' total assets. (c) PDF of the weight of loans secured by nonfarm nonresidential properties in banks' total assets. (d) PDF of the weight of agriculture loans in banks' total assets. (e) PDF of banks' equity to asset ratios. Blue circles curves represents PDFs of all banks, red triangles represents PDFs of those banks that are on the FDIC failed bank list.
Figure 2 Comparison of probability density functions (PDF) of weight of typical assets and equity ratios between all banks and FDIC listed failed banks for 2007. (a) PDF of the weight of loans for construction and land development in banks' total asset. (b) PDF of the weight of loans secured by 1-4 family residential properties in banks' total assets. (c) PDF of the weight of loans secured by nonfarm nonresidential properties in banks' total assets. (d) PDF of the weight of agriculture loans in banks' total assets. (e) PDF of banks' equity to asset ratios. Blue circles curves represents PDFs of all banks, red triangles represents PDFs of those banks that are on the FDIC failed bank list.CC BY-NC-ND · PubMed Central
Figure 3 Fraction of survived banks after cascading failures as function of the initial loss of value of certain asset, with η = 0. Blue dashed lines represent the fraction of survived banks out of all banks, and the red solid lines represent the fraction of survived banks out of the 278 failed banks from FDIC failed bank list. The parameter α is changed from 0 to 0.1 by 0.01 to produce 10 lines for each case. (a) Initial shock is imposed to loans for construction and land development. The red solid lines are significantly lower than the blue dashed lines separating clearly the failed banks from the set of all banks. (b) Initial shock is imposed to loans secured by 1-4 family resid. properties. The red solid lines and blue dashed lines are entangled. (c) Initial shock on loans secured by nonfarm nonresid. properties. The red solid lines are distinguishably lower than the blue dashed lines, similarly as in the case under (a). (d) Initial shock on agricultural loans. The red solid lines are slightly higher than the blue dashed line, not showing clear distinction between failed and non failed banks.
Figure 3 Fraction of survived banks after cascading failures as function of the initial loss of value of certain asset, with η = 0. Blue dashed lines represent the fraction of survived banks out of all banks, and the red solid lines represent the fraction of survived banks out of the 278 failed banks from FDIC failed bank list. The parameter α is changed from 0 to 0.1 by 0.01 to produce 10 lines for each case. (a) Initial shock is imposed to loans for construction and land development. The red solid lines are significantly lower than the blue dashed lines separating clearly the failed banks from the set of all banks. (b) Initial shock is imposed to loans secured by 1-4 family resid. properties. The red solid lines and blue dashed lines are entangled. (c) Initial shock on loans secured by nonfarm nonresid. properties. The red solid lines are distinguishably lower than the blue dashed lines, similarly as in the case under (a). (d) Initial shock on agricultural loans. The red solid lines are slightly higher than the blue dashed line, not showing clear distinction between failed and non failed banks.CC BY-NC-ND · PubMed Central
Figure 4 ROC curves of the prediction of failed banks by our cascading failure model when the loans for construction and land development are initially shocked (top figures) and when the loans secured by nonfarm nonresidential properties are initially shocked (bottom figures), based on 2007 data. Each point of the ROC curves corresponds to one combination of parameters (α, η, p). (a)(e) ROC curve of predictions made by the entire cascading failure process, (b)(f) of predictions made by the first cascading failure step and (c)(g) of predictions made by the other than the first cascading steps. The color of a dot represents the number of failed banks correctly identified by the model with the corresponding parameters combination. (d)(h) For fixed false positive rates of 5%, 10%, 15%, and 20%, we find parameter combinations with maximum true positive rates in fig. (a), and show the number of failed banks identified by the first step (red) and the number of failed banks identified by the other steps in the cascading failure process (white). The black dots in (a)(b)(c) show the positions of four combinations respectively.
Figure 4 ROC curves of the prediction of failed banks by our cascading failure model when the loans for construction and land development are initially shocked (top figures) and when the loans secured by nonfarm nonresidential properties are initially shocked (bottom figures), based on 2007 data. Each point of the ROC curves corresponds to one combination of parameters (α, η, p). (a)(e) ROC curve of predictions made by the entire cascading failure process, (b)(f) of predictions made by the first cascading failure step and (c)(g) of predictions made by the other than the first cascading steps. The color of a dot represents the number of failed banks correctly identified by the model with the corresponding parameters combination. (d)(h) For fixed false positive rates of 5%, 10%, 15%, and 20%, we find parameter combinations with maximum true positive rates in fig. (a), and show the number of failed banks identified by the first step (red) and the number of failed banks identified by the other steps in the cascading failure process (white). The black dots in (a)(b)(c) show the positions of four combinations respectively.CC BY-NC-ND · PubMed Central
Figure 5 ROC curves of predictions of failed banks by our cascading failure model when (a) loans secured by 1-4 family residence properties and (b) agricultural loans are initially shocked respectively. The straighter the ROC curve is, the closer it is to random case, meaning the less predictive power in regard to the failure of the commercial banks during the 2007–2011 financial crisis.
Figure 5 ROC curves of predictions of failed banks by our cascading failure model when (a) loans secured by 1-4 family residence properties and (b) agricultural loans are initially shocked respectively. The straighter the ROC curve is, the closer it is to random case, meaning the less predictive power in regard to the failure of the commercial banks during the 2007–2011 financial crisis.CC BY-NC-ND · PubMed Central
Figure 6 Survival rate of banks when asset 0 (loans for construction and land development) is initially shocked as function of one parameter with the other two parameters fixed. Average over 300 independent realizations with 95% confidence interval. Left panel: parameter combination α = 0.14, η = 0.26 and p = 0.6; right panel: parameter combination α = 0.35, η = 0.2 and p = 0.6.
Figure 6 Survival rate of banks when asset 0 (loans for construction and land development) is initially shocked as function of one parameter with the other two parameters fixed. Average over 300 independent realizations with 95% confidence interval. Left panel: parameter combination α = 0.14, η = 0.26 and p = 0.6; right panel: parameter combination α = 0.35, η = 0.2 and p = 0.6.CC BY-NC-ND · PubMed Central
Figure 7 Phase diagram for parameter alpha and p, when η = 0.26. The network is stable in region I. Significant part of banks in the network would still survive after cascading failure. In region II, almost all the banks in the network fail after cascading failure.
Figure 7 Phase diagram for parameter alpha and p, when η = 0.26. The network is stable in region I. Significant part of banks in the network would still survive after cascading failure. In region II, almost all the banks in the network fail after cascading failure.CC BY-NC-ND · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 25 references

Research works printed but not linked: 5

  1. Newman, M. E. J. et al. Large Scale The Structure and Dynamic of Networks (Princeton University Press, Princeton, NJ, 2006).
  2. Smart, A. G., Amaral, L. A. N. & Ottino, J. M. Cascading failure and robustness in metabolic networks. Proc. Natl. Acad. Sci. USA 99, 1511215117 (2008).
  3. Wells, S. UK interbank exposures: systemic risk implications. Bank of England Financial Stability Review December, 175–182 (2002).
  4. Merton, R. C. On the pricing of corporate debt: the risk structure of interest rates. The Journal of Finance 29, 449–484 (1974).
  5. Gray, D. F., Merton, R. C. & Bodie, Z. Contingent claims approach to measuring and managing sovereign credit risk. Special Issue on Credit Analysis. Journal of Investment Management 5, 5–28 (2007).

Cited by

Items 201–235 of 235

Metadata

Authors
Huang, X., Vodenska, I., Havlin, S., & Stanley, H. E.
Affiliations
Boston University (US)
Corresponding author
Vodenska, Irena
Venue
Scientific Reports
Published
2013-02-05
Volume
3
Issue
1
Pages
1219-1219
ISSN
2045-2322 (ISSN-L)
OpenAlex
W1987285546
Cited by
238
References
23
Citations by year
2026: 3 · 2025: 9 · 2024: 14 · 2023: 16 · 2022: 21 · 2021: 20 · 2020: 23 · 2019: 24 · 2018: 14 · 2017: 19 · 2016: 29 · 2015: 19 · 2014: 21 · 2013: 3
FWCI
41.46
Type
journal article
Language
en
Update notices
erratum (2013-04-08)
Open access
Licence
Locations
  • Scientific Reports (publishedVersion, cc-by-nc-nd)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • OpenBU (Boston University) (submittedVersion, cc-by-nc-nd)
  • arXiv (Cornell University)
Funders
  • European Commission (255987)
  • Office of Naval Research (N00014-12-1-0548, N00014-12-1, N00014-09-1-0380, N00014)
Topics
  • Social Sciences › Economics, Econometrics and Finance › Finance › Banking stability, regulation, efficiency
  • Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Complex Network Analysis Techniques
  • Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Complex Systems and Time Series Analysis
Keywords
cascading failures, systemic risk, financial networks, bipartite graphs, bank failure prediction, financial contagion, 2008 financial crisis, commercial real estate, residential real estate, bank balance sheets, stress testing

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.