Predicting Credit Card Delinquency: A Fundamental Model of Cardholder Financial Behavior
Year: 2019
Type: dissertation
Study design: Observational Study
Abstract: This thesis proposes a model of credit card customer delinquency based on\ntheoretical advancements in financial decision-making. As follows, this thesis\nhas two main research purposes.\nFirst, credit card delinquency is modeled explicitly, incorporating mechanisms\nfrom mental accounting and financial decision-making. This allows\nfor more realistic modeling of cardholder behavior, while simultaneously inspecting\nthe validity of these theoretical concepts.\nSecond, the modeling specification advances previous research in the\nbehavior scoring literature. Accounting for individual-level heterogeneity,\ndynamic effects are assigned as individual lag weights using a segmented\napproach. Hence, potentially different behavioral patterns between non-delinquent\nand eventual delinquent cardholders are modeled directly.\nUsing a comprehensive dataset combining credit and debit transactions\nof cardholders between June 2008 and June 2011 from a Norwegian bank,\nsupport is found for the following three hypotheses related to mental accounting\nand present bias. First, increased payment decoupling leads to a higher\nlikelihood of delinquency, when continued borrowing is promoted by reduced\nsalience of past expenses. Second, the results show that behavior consistent\nwith persistence of decision-making ineptitude also increases the likelihood\nof delinquency. Some cardholders habitually spend excessively, refusing to\naccommodate consumption to a financially reasonable level. Third, a lower\nconcern for future consequences also increases the likelihood of delinquency.\nPresent-biased individuals tend to discount future credit card repayments at\na higher rate and consistently spend at perilously high rates.\nFurther, the results reveal how the structure of dynamic effects improves\nprediction of delinquency. Capturing the heterogeneous effects of previous\nfinancial status leads to a more precise understanding of cardholder behavior.\nThe proposed model has greater predictive performance than machine\nlearning algorithms that are frequently applied to credit scoring data.
Author: Huse, Håvard
Language: English
Subfield: General Decision Sciences
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W3006023901