The Fairness of Credit Scoring Models

References

• Adel T, Valera I, Ghahramani Z, Weller A (2019) One-network adversarial fairness. Proc. Conf. AAAI Artificial Intelligence, vol. 33 (AAAI Press, Palo Alto, CA), 2412–2420.Crossref, Google Scholar
• Alesina AF, Lotti F, Mistrulli PE (2013) Do women pay more for credit? Evidence from Italy. J. Eur. Econom. Assoc. 11(S1):45–66.Crossref, Google Scholar
• Arrow KJ (1973) The theory of discrimination. Aschenfelter O, Rees A, eds. Discrimination in Labor Markets (Princeton University Press, Princeton, NJ), 3–33.Google Scholar
• Athey S, Imbens G (2016) Recursive partitioning for heterogeneous causal effects. Proc. Natl. Acad. Sci. USA 113(27):7353–7360.Crossref, Google Scholar
• Barocas S, Hardt M, Narayanan A (2023) Fairness and machine learning: Limitations and opportunities. Accessed November 10, 2023, http://www.fairmlbook.org.Google Scholar
• Bartlett R, Morse A, Stanton R, Wallace N (2022) Consumer-lending discrimination in the fintech era. J. Financial Econom. 143:30–56.Crossref, Google Scholar
• Bartlett RP, Morse A, Wallace N, Stanton R (2021) Algorithmic accountability: A legal and economic framework. Berkeley Tech. Law J. 36(3):675–736.Google Scholar
• Bayer P, Ferreira F, Ross SL (2018) What drives racial and ethnic differences in high-cost mortgages? The role of high-risk lenders. Rev. Financial Stud. 31(1):175–205.Crossref, Google Scholar
• Becker GS (1957) The Economics of Discrimination (University of Chicago Press, Chicago).Google Scholar
• Berg T, Fuster A, Puri M (2023) FinTech lending. Annu. Rev. Financial Econom. 14(1):187–207.Crossref, Google Scholar
• Berg T, Puri M, Rocholl J (2020a) Loan officer incentives, internal rating models, and default rates. Rev. Finance 24(3):529–578.Crossref, Google Scholar
• Berg T, Burg V, Gombović A, Puri M (2020b) On the rise of FinTechs: Credit scoring using digital footprints. Rev. Financial Stud. 33(7):2845–2897.Crossref, Google Scholar
• Bergstra J, Bengio Y (2012) Random search for hyper-parameter optimization. J. Machine Learn. Res. 13:281–305.Google Scholar
• Bhutta N, Hizmo A (2021) Do minorities pay more for mortgages? Rev. Financial Stud. 34(2):763–789.Crossref, Google Scholar
• Bono T, Croxson K, Giles A (2021) Algorithmic fairness in credit scoring. Oxford Rev. Econom. Policy 37(3):585–617.Crossref, Google Scholar
• Butler AW, Mayer EJ, Weston JP (2023) Racial disparities in the auto loan market. Rev. Financial Stud. 36(1):1–41.Crossref, Google Scholar
• Calders T, Verwer S (2010) Three naive Bayes approaches for discrimination-free classification. Data Mining Knowledge Discovery 21:277–292.Crossref, Google Scholar
• Calders T, Kamiran F, Pechenizkiy M (2009) Building classifiers with independency constraints. Proc. IEEE Internat. Conf. Data Mining Workshops (IEEE Computer Society, Washington, DC), 13–18.Google Scholar
• Calem PS, Longhofer SD (2002) Anatomy of a fair lending exam: The uses and limitations of statistics. J. Real Estate Finance Econom. 24:207–237.Crossref, Google Scholar
• Calmon FP, Wei D, Vinzamuri B, Ramamurthy KN, Varshney KR (2017) Optimized pre-processing for discrimination prevention. Von Luxburg U, Bengio S, Wallach H, Fergus R, Vishwanathan S, Garnett R, eds. Proc. 31st Internat. Conf. Neural Inform. Processing Systems, Advances in Neural Information Processing Systems, vol. 30 (Curran Associates Inc., Red Hook, NY), 3995–4004.Google Scholar
• Caton S, Haas C (2024) Fairness in machine learning: A survey. ACM Comput. Survey 56(7):1–38.Crossref, Google Scholar
• CFPB (2022) CFPB acts to protect the public from black-box credit models using complex algorithms. CFPB (May 26), https://www.consumerfinance.gov/about-us/newsroom/cfpb-acts-to-protect-the-public-from-black-box-credit-models-using-complex-algorithms/.Google Scholar
• Chernenko S, Scharfstein D (2024) Racial disparities in the paycheck protection program. J. Financial Econom. 160:103911.Crossref, Google Scholar
• Chiappa S (2019) Path-specific counterfactual fairness. Proc. Conf. AAAI Artificial Intelligence 33(1):7801–7808.Crossref, Google Scholar
• Chouldechova A (2017) Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. Big Data 5(2):153–163.Crossref, Google Scholar
• Coston A, Rambachan A, Chouldechova A (2021) Characterizing fairness over the set of good models under selective labels. Proc. 38th Internat. Conf. Machine Learn., Proceedings of Machine Learning Research, vol. 139 (PMLR, New York), 2144–2155.Google Scholar
• Cowgill B, Tucker CE (2019) Algorithmic fairness and economics. Preprint, submitted April 4, http://dx.doi.org/10.2139/ssrn.3361280.Google Scholar
• D’Acunto F, Ghosh P, Rossi AG (2021) How costly are cultural biases? Evidence from FinTech. Preprint, submitted January 8, http://dx.doi.org/10.2139/ssrn.3736117.Google Scholar
• Dobbie W, Liberman A, Paravisini D, Pathania V (2021) Measuring bias in consumer lending. Rev. Econom. Stud. 88(6):2799–2832.Crossref, Google Scholar
• Dua D, Graff C (2019) UCI Machine Learning Repository. University of California, Irvine, School of Information and Computer Sciences, Irvine.Google Scholar
• EBA (2020) Report on Big Data and Advanced Analytics (European Banking Authority, Paris).Google Scholar
• EBA (2021) Discussion Paper on Machine Learning for IRB Models (European Banking Authority, Paris).Google Scholar
• EC (2020) White Paper on Artificial Intelligence: A European Approach to Excellence and Trust (European Commission, Brussels, Belgium).Google Scholar
• EEOC (1978) Uniform Guidelines on Employee Selection Procedure (Equal Employment Opportunity Commission, Washington, DC).Google Scholar
• Evans CA (2017) Keeping Fintech Fair: Thinking About Fair Lending and UDAP Risks (Consumer Compliance Outlook, Federal Reserve System, Washington, DC).Google Scholar
• Evans CA, Miller W (2019) From Catalogs to Clicks: The Fair Lending Implications of Targeted, Internet Marketing (Consumer Compliance Outlook, Federal Reserve System, Washington, DC).Google Scholar
• Fay PM, Hunsberger SA (2021) Practical valid inferences for the two-sample binomial problem. Ann. Statist. 15:72–110.Google Scholar
• Fuster A, Goldsmith-Pinkham P, Ramadorai T, Walther A (2022) Predictably unequal? The effects of machine learning on credit markets. J. Finance 77:5–47.Crossref, Google Scholar
• Fuster A, Plosser M, Schnabl P, Vickery J (2019) The role of technology in mortgage lending. Rev. Financial Stud. 32(5):1854–1899.Crossref, Google Scholar
• Galhotra S, Brun Y, Meliou A (2017) Fairness testing: Testing software for discrimination. Proc. 11th Joint Meeting Foundations Software Engrg. (ACM, New York), 498–510.Google Scholar
• Geiger G (2021) Court rules Deliveroo used discriminatory algorithm. VICE (January 5), https://www.vice.com/en/article/court-rules-deliveroo-used-discriminatory-algorithm/.Google Scholar
• Gillis TB, Spiess J (2019) Big data and discrimination. Univ. Chicago Law Rev. 86(2).Google Scholar
• Givens AR, Schellmann H, Stoyanovich J (2021) Opinion: We need laws to take on racism and sexism in hiring technology. New York Times (March 17), https://www.nytimes.com/2021/03/17/opinion/ai-employment-bias-nyc.html.Google Scholar
• Haas C (2019) The price of fairness: A framework to explore tradeoffs in algorithmic fairness. Proc. 40th Internat. Conf. Inform. Systems (AIS eLibrary).Google Scholar
• Hardt M, Price E, Srebro N (2016) Equality of opportunity in supervised learning. Proc. 30th Internat. Conf. Neural Inform. Processing Systems, Advances in Neural Information Processing Systems, vol. 29 (Curran Associates Inc., Red Hook, NY), 3315–3323.Google Scholar
• Hertzberg A, Liberti JM, Paravisini D (2015) Information and inventives inside the firm: Evidence from loan officer rotation. J. Finance 65:795–828.Crossref, Google Scholar
• Hofmann H (1994) Statlog (German Credit Data) (UCI Machine Learning Repository, Irvine, CA).Google Scholar
• Howell ST, Kuchler T, Snitkof D, Stroebel J, Wong J (2022) Lender automation and racial disparities in credit access. J. Finance 79(2):1457–1512.Google Scholar
• Iosifidis V, Fetahu B, Ntoutsi E (2019) FAE: A fairness-aware ensemble framework. Proc. IEEE Internat. Conf. Big Data (IEEE Computer Society, Washington, DC), 1375–1380.Google Scholar
• Jansen M, Nguyen H, Shams A (2024) Rise of the machines: The impact of automated underwriting. Management Sci., ePub ahead of print April 30, https://doi.org/10.1287/mnsc.2024.4986.Google Scholar
• Jiang H, Nachum O (2020) Identifying and correcting label bias in machine learning. Proc. Internat. Conf. Artificial Intelligence Statist. (PMLR, New York), 702–712.Google Scholar
• Kairouz P, Liao J, Huang C, Vyas M, Welfert M, Sankar L (2022) Generating fair universal representations using adversarial models. IEEE Trans. Inform. Forensics Security 17:1970–1985.Crossref, Google Scholar
• Kamiran F, Calders T (2009) Classifying without discriminating. Proc. 2nd Internat. Conf. Comput. Control Comm. (IEEE Computer Society, Washington, DC), 1–6.Google Scholar
• Kamiran F, Calders T (2012) Data preprocessing techniques for classification without discrimination. Knowledge Inform. Systems 33(1):1–33.Crossref, Google Scholar
• Kamiran F, Calders T, Pechenizkiy M (2010) Discrimination aware decision tree learning. Proc. IEEE 13th Internat. Conf. Data Mining (IEEE, Piscataway, NJ), 869–874.Google Scholar
• Kamiran F, Karim A, Zhang X (2012) Decision theory for discrimination-aware classification. Proc. IEEE 12th Internat. Conf. Data Mining (IEEE, Piscataway, NJ), 924–929.Google Scholar
• Kilbertus N, Rojas-Carulla M, Parascandolo G, Hardt M, Janzing D, Schölkopf B (2017) Avoiding discrimination through causal reasoning. Proc. 31st Internat. Conf. Neural Inform. Processing Systems (Curran Associates Inc., Red Hook, NY), 656–666.Google Scholar
• Kim S, Lessmann S, Andreeva G, Rovatsos M (2023) Fair models in credit: Intersectional discrimination and the amplification of inequity. Preprint, submitted August 1, https://arxiv.org/abs/2308.02680.Google Scholar
• Kleinberg J, Mullainathan S, Raghavan M (2017a) Inherent tradeoffs in the fair determination of risk scores. Papadimitriou CH, ed. 8th Innovations Theoret. Comput. Sci. Conf. (ITCS 2017), Leibniz International Proceedings in Informatics (LIPIcs), vol. 67 (Schloss Dagstuhl - Leibniz-Zentrum für Informatik, Wadern, Germany), 43:1–43:23.Google Scholar
• Kleinberg J, Ludwig J, Mullainathan S, Rambachan A (2018) Algorithmic fairness. AEA Paper Proc. 108:22–27.Crossref, Google Scholar
• Kleinberg J, Lakkaraju H, Leskovec J, Ludwig J, Mullainathan S (2017b) Human decisions and machine predictions. Quart. J. Econom. 133(1):237–293.Crossref, Google Scholar
• Kozodoi N, Jacob J, Lessmann S (2022) Fairness in credit scoring: Assessment, implementation and profit implications. Eur. J. Oper. Res. 297(3):1083–1094.Crossref, Google Scholar
• Lael B (2021) Supporting Responsible Use of AI and Equitable Outcomes in Financial Services (Board of Governors of the Federal Reserve System, Washington, DC).Google Scholar
• Lakkaraju H, Kleinberg J, Leskovec J, Ludwig J, Mullainathan S (2017) The selective labels problem: Evaluating algorithmic predictions in the presence of unobservables. Proc. 23rd ACM SIGKDD Internat. Conf. Knowledge Discovery Data Mining (ACM, New York), 275–284.Google Scholar
• Lee M, Floridi L (2021) Algorithmic fairness in mortgage lending: From absolute conditions to relational tradeoffs. Minds Machine 31:165–191.Crossref, Google Scholar
• Lessmann S, Baesens B, Seow H-V, Thomas LC (2015) Benchmarking state-of-the-art classification algorithms for credit scoring: An update of research. Eur. J. Oper. Res. 247(1):124–136.Crossref, Google Scholar
• Liu LT, Simchowitz M, Hardt M (2019) The implicit fairness criterion of unconstrained learning. Proc. 36th Internat. Conf. Machine Learn., vol. 97 (Proceedings of Machine Learning Research, New York), 4051–4060.Google Scholar
• Madras D, Creager E, Pitassi T, Zemel R (2018) Learning adversarially fair and transferable representations. Proc. 35th Internat. Conf. Machine Learn., vol. 80 (Proceedings of Machine Learning Research, New York), 3384–3393.Google Scholar
• Mayson SG (2019) Bias in, bias out. Yale Law J. 128(8):2218–2300.Google Scholar
• Menon AK, Williamson RC (2018) The cost of fairness in binary classification. Friedler SA, Wilson C, eds. Proc. 1st Conf. Fairness Accountability Transparency, vol. 81 (Proceedings of Machine Learning Research, New York), 107–118.Google Scholar
• Miller T (2019) Explanation in artificial intelligence: Insights from the social sciences Artificial Intelligence 267:1–38.Crossref, Google Scholar
• Mitchell S, Potash E, Barocas S, D’Amour A, Lum K (2021) Algorithmic fairness: Choices, assumptions, and definitions. Annu. Rev. Statist. Appl. 8(1):141–163.Crossref, Google Scholar
• Molnar C (2024) Interpretable machine learning. Accessed July 31, 2024, https://christophm.github.io/interpretable-ml-book/.Google Scholar
• Morse A, Pence K (2021) Technological innovation and discrimination in household finance. Rau R, Wardrop R, Zingales L, eds. The Palgrave Handbook of Technological Finance (Springer Books, Berlin, Heidelberg), 783–808.Crossref, Google Scholar
• Mullainathan S, Spiess J (2017) Machine learning: An applied econometric approach. J. Econom. Perspective 31:87–106.Crossref, Google Scholar
• Munnell AH, Tootell GMB, Browne LE, McEneaney J (1996) Mortgage lending in Boston: Interpreting HMDA data. Amer. Econom. Rev. 86(1):25–53.Google Scholar
• Nabi R, Shpitser I (2018) Fair inference on outcomes. Proc. 32nd AAAI Conf. Artificial Intelligence (AAAI Press, Palo Alto, CA).Google Scholar
• Nabi R, Malinsky D, Shpitser I (2019) Learning optimal fair policies. Proc. 36th Internat. Conf. Machine Learn. (PMLR, New York).Google Scholar
• Noriega-Campero A, Bakker MA, Garcia-Bulle B, Pentland A (2019) Active fairness in algorithmic decision making. Proc. AAAI/ACM Conf. AI Ethics Society (ACM, New York), 77–83.Google Scholar
• O’Neil C (2017) Weapons of Maths Destruction: How Big Data Increases Inequality and Threatens Democracy (Wiley, New York).Google Scholar
• Petrides G, Moldovan D, Coenen L, Guns T, Verbeke W (2022) Cost-sensitive learning for profit-driven credit scoring. J. Oper. Res. Soc. 73(2):338–350.Crossref, Google Scholar
• Petrović A, Nikolić M, Radovanović S, Delibašić B, Jovanović M (2022) FAIR: Fair adversarial instance re-weighting. Neurocomputing 476:14–37.Crossref, Google Scholar
• Phelps E (1972) The statistical theory of racism and sexism. Amer. Econom. Rev. 62:659–661.Google Scholar
• Philippon T (2019) On Fintech and financial inclusion. NBER Working Paper No. 26330, National Bureau of Economic Research, Cambridge, MA.Google Scholar
• Pleiss G, Raghavan M, Wu F, Kleinberg J, Weinberger KQ (2017) On fairness and calibration. Guyon I, Von Luxburg U, Bengio S, Wallach H, Fergus R, Vishwanathan S, Garnett R, eds. Proc. 31st Internat. Conf. Neural Inform. Processing Systems, Advances in Neural Information Processing Systems, vol. 30 (Curran Associates Inc., Red Hook, NY).Google Scholar
• Pope DG, Sydnor JR (2011) Implementing anti-discrimination policies in statistical profiling models. Amer. Econom. J. Econom. Policy 3(3):206–231.Crossref, Google Scholar
• Prince A, Schwarcz D (2020) Proxy discrimination in the age of artificial intelligence and big data. Iowa Law Rev. 105:1257–1301.Google Scholar
• Qian J, Strahan PE, Yang Z (2015) The impact of incentives and communication costs on information production and use: Evidence from bank lending. J. Finance 70:1457–1493.Crossref, Google Scholar
• Seber GAF (2013) Statistical Models for Proportions and Probabilities (Springer, Berlin, Heidelberg).Crossref, Google Scholar
• Shah RD, Peters J (2020) The hardness of conditional independence testing and the generalised covariance measure. Ann. Statist. 48:1514–1538.Crossref, Google Scholar
• Valera I, Singla A, Gomez-Rodriguez M (2018) Enhancing the accuracy and fairness of human decision making. Proc. 32nd Internat. Conf. Neural Inform. Processing Systems (Curran Associates Inc., Red Hook, NY), 1774–1783.Google Scholar
• Verma S, Rubin J (2018) Fairness definitions explained. Proc. IEEE/ACM Internat. Workshop Software Fairness (ACM, New York), 1–7.Google Scholar
• Vigdor N (2020) Apple card investigated after gender discrimination complaints. New York Times (November 10), https://www.nytimes.com/2019/11/10/business/Apple-credit-card-investigation.html.Google Scholar
• Xu D, Yuan S, Zhang L, Wu X (2018) Fairgan: Fairness-aware generative adversarial networks. Proc. IEEE Internat. Conf. Big Data (IEEE, Piscataway, NJ), 570–575.Google Scholar
• Xu D, Wu Y, Yuan S, Zhang L, Wu X (2019) Achieving causal fairness through generative adversarial networks. Proc. 28th Internat. Joint Conf. Artificial Intelligence (AAAI Press, Palo Alto, CA).Google Scholar
• Yang J, Soltan AAS, Eyre DW, Cifton DA (2023) Algorithmic fairness and bias mitigation for clinical machine learning with deep reinforcement learning. Natural Machine Intelligence 5(11):884–894.Crossref, Google Scholar
• Yeh IC, Lien C (2009) The comparisons of data mining techniques for the predictive accuracy of probability of default of credit card clients. Expert Syst. Appl. 36:2473–2480.Google Scholar