We propose a structural approach to model the reasoning underlying credit decisions. By moving beyond black-box approaches, our model provides transparency and interpretability, enabling a deeper understanding of decision-making processes in credit allocation. In addition, we contribute to fairness assessment methodologies by proposing a new metric for evaluation. Rather than conventional parity measures or confusion matrix-based metrics, we leverage cumulative distribution comparisons to assess fairness across racial groups. Specifically, we utilize Lorenz curves and the Gini index to provide a transparent analysis of disparities in credit decisions. Together, these contributions offer a rigorous framework for evaluating fairness of credit lending decisions.
Giudici et al. (Thu,) studied this question.