Abstract
We present a methodology for improving credit scoring models by distinguishing two forms of rational behaviour of loan defaulters. It is common knowledge among practitioners that there are two types of defaulters, those who do not pay because of cash flow problems (‘Can’t Pay’), and those that do not pay because of lack of willingness to pay (‘Won’t Pay’). This work proposes to differentiate them using a game theory model that describes their behaviour. This separation of behaviours is represented by a set of constraints that form part of a semi-supervised constrained clustering algorithm, constructing a new target variable summarizing relevant future information. Within this approach the results of several supervised models are benchmarked, in which the models deliver the probability of belonging to one of these three new classes (good payers, ‘Can’t Pays’, and ‘Won’t Pays’). The process improves classification accuracy significantly, and delivers strong insights regarding the behaviour of defaulters.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Alary D and Gollier C (2004). Debt contract, strategic default, and optimal penalties with judgement errors. Annals of Economics and Finance 5 (2): 357–372.
Anderson R (2007). The Credit Scoring Toolkit. Oxford University Press: New York.
Baesens B, Van Gestel T, Viaene S, Stepanova M, Suykens J and Vanthienen J (2003). Benchmarking state-of-the-art classification algorithms for credit scoring. Journal of the Operational Research Society 54 (6): 627–636.
Baesens B, Van Gestel T, Stepanova M and Vanthienen J (2005). Neural network survival analysis for personal loan data. Jounal of the Operational Research Society 56 (9): 1089–1098.
Bard JF and Jarrah A (2009). Large-scale constrained clustering for rationalizing pickup and delivery operations. Transportation Research Part B: Methodological 43 (5): 542–561.
Basu S, Bilenko M and Mooney RJ (2004). A probabilistic framework for semi-supervised clustering. In: Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining, KDD ’04, ACM: New York, pp 59–68.
Basu S, Davidson I and Wagstaff K (2008). Constrained Clustering: Advances in Algorithms, Theory, and Applications. Chapman & Hall/CRC: Florida, USA.
Benton M, Meier S and Sprenger C (2007). Overborrowing and undersaving: Lessons and policy implications from research in behavioral economics. Public and Community Affairs Discussion Papers 2007–4, Federal Reserve Bank of Boston.
Benzion U, Rapoport A and Yagil J (1989). Discount rates inferred from decisions: An experimental study. Management Science 35 (3): 270–284.
Block-Lieb S and Janger EJ (2006). The myth of the rational borrower: Rationality, behaviorism, and the misguided ‘reform’ of bankruptcy law. Texas Law Review 84 (6): 1481–1565.
Bravo C and Weber R (2011). Semi-supervised constrained clustering with cluster outlier filtering. In: CS Martin and S-W Kim (eds.) Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications, Lecture Notes in Computer Science Vol. 7042, Springer-Verlag: Berlin/Heidelberg, pp 347–354.
Bravo C, Maldonado S and Weber R (2013). Granting and managing loans for micro-entrepreneurs: New developments and practical experiences. European Journal of Operational Research 227 (2): 358–366.
Burks SV, Carpenter JP, Goette L and Rustichini A (2008). Cognitive skills explain economic preferences, strategic behavior, and job attachment. IZA Discussion Papers 3609, Institute for the Study of Labor (IZA).
Chabris CF, Laibson D, Morris CL, Schuldt JP and Taubinsky D (2008). Individual laboratory-measured discount rates predict field behavior. NBER Working Paper No. 14270, 2008. http://www.nber.org/papers/w14270, accessed 10 February 2012.
Davidson I and Ravi SS (2005). Clustering with constraints: Feasibility issues and the k-means algorithm. In: Proceedings of the SIAM International Conference on Data Mining, SDM 2005; April 21–23, New Port Beach, CA, pp 138–149.
De Waal DA, Du Toit JV and De La Rey T (2005). An investigation into the use of generalized additive neural networks in credit scoring. In: Proceedings of IX Credit Scoring & Credit Control Conference, Pollock Halls, University of Edinburgh, Scotland, 26–28 August, 2009, pp 1–10.
Dogan H and Guzelis C (2006). Gradient networks for clustering. In: Gknar IC and Sevgi L (eds). Complex Computing-Networks, Volume 104 of Springer Proceedings Physics; Springer Berlin Heidelberg, pp 275–278.
Fudenberg D and Tirole J (1991). Game Theory. MIT Press: Massachusetts, USA.
Green L, Fry AF and Myerson J (1994). Discounting of delayed rewards: A life-span comparison. Psychological Science 5 (1): 33–36.
Guiso L, Sapienza P and Zingales L (2010). The determinants of attitudes towards strategic default on mortgages. Economics Working Papers ECO2010/31, European University Institute.
Hastie T, Tibshirani R and Friedman J (2009). The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd edn. Springer: New York, USA.
Hosmer D and Lemeshow H (2000). Applied Logistic Regression. John Wiley & Sons: Massachusetts, USA.
Jokivuolle E and Peura S (2000). A model for estimating recovery rates and collateral haircuts for bank loans. Research Discussion Papers 2/2000, Bank of Finland.
Levy M and Sandler M (2008). Structural segmentation of musical audio by constrained clustering. IEEE Transactions on Audio, Speech, and Language Processing 16 (2): 318–326.
L’Huillier G, Weber R and Figueroa N (2009). Online phishing classification using adversarial data mining and signaling games. SIGKDD Exploration Newsletter 11: 92–99.
Lin S, Ansell J and Andreeva G (2011). Predicting default of a small business using different definitions of financial distress. The Journal of the Operational Research Society 63 (4): 539–548.
Littwin A (2007). Beyond usury: A study of credit-card use and preference among low-income consumers. Texas Law Review 86 (3): 451–506.
Moffat PG (2005). Hurdle models of loan default. The Journal of the Operational Research Society 56 (9): 1063–1071.
Patil G, Modarres R, Myers W and Patankar P (2006). Spatially constrained clustering and upper level set scan hotspot detection in surveillance geoinformatics. Environmental and Ecological Statistics 13 (4): 365–377.
Siddiqi N (2006). Credit Risk Scorecards: Developing and Implementing Intelligent Credit Scoring. John Wiley and Sons: New Jersey, USA.
Stiglitz JE and Weiss A (1981). Credit rationing in markets with imperfect information. American Economic Review 71 (3): 393–410.
Thomas LC (2000). A survey of credit and behavioural scoring: forecasting financial risk of lending to consumers. International Journal of Forecasting 16 (2): 149–172.
Thomas LC, Crook J and Edelman D (2002). Credit Scoring and Its Applications. SIAM: Philadelphia, USA.
Vatsa V, Sural S and Majumdar A (2005). A game-theoretic approach to credit card fraud detection. In: Jajodia S and Mazumdar C (eds). Information Systems Security. Lecture Notes in Computer Science Vol. 3803. Springer-Verlag: Berlin/Heidelberg, pp 263–276.
Wang Y (2007). Combining data mining and game theory in manufacturing strategy analysis. Journal of Intelligent Manufacturing 18 (4): 505–511.
Wette HC (1983). Collateral in credit rationing in markets with imperfect information: Note. The American Economic Review 73 (3): 442–445.
Yamashita S and Yoshiba T (2010). Analytical solution for expected loss of a collateralized loan: A square-root intensity process negatively correlated with collateral value. Discussion Paper 2010-E-10, Bank of Japan.
Zhang GP (2007). Avoiding pitfalls in neural network research. IEEE Transactions on Systems, Man, and Cybernetics, Part C 37 (1): 3–16.
Zhu H, Beling PA and Overstreet GA (2001). A study in the combination of two consumer credit scores. The Journal of the Operational Research Society 52 (9): 974–980.
Acknowledgements
The first author acknowledges CONICYT for the grants that support this work (AT-24110006, NAC-DOC: 21090573) and the PhD in Engineering Systems, Universidad de Chile. All authors acknowledge the support of the institution that provided the data. The work reported in this paper has been partially funded by the Institute of Complex Engineering Systems (ICM: P-05-004-F, CONICYT: FBO16) and the Finance Center of the Department of Industrial Engineering, Universidad de Chile, with the support of Bank Bci.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Bravo, C., Thomas, L. & Weber, R. Improving credit scoring by differentiating defaulter behaviour. J Oper Res Soc 66, 771–781 (2015). https://doi.org/10.1057/jors.2014.50
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1057/jors.2014.50