Institutionalizing explainability in credit scoring

Kevin Bauer, Lucia Franke, Andrej Gill and Katja Langenbucher

No 114, SAFE Policy Letters from Leibniz Institute for Financial Research SAFE

Abstract: Machine learning credit scoring expands the informational frontier of retail lending, particularly for thin file borrowers, yet it also erodes the practical meaning of disclosure duties that anchor consumer protection and prudential oversight. The central financial implication is that explainability is no longer a peripheral communication task. It is a market structuring variable that can reshape access, pricing efficiency, competition, and the distribution of compliance burdens across incumbents and challengers. The central regulatory gap is that current regimes articulate rights and obligations but remain under specified on what constitutes a sufficient explanation, how fidelity can be verified, and how opportunistic framing can be prevented when explanation techniques permit multiple plausible narratives. The most effective policy response is institutional rather than purely technical, achieved by creating a governed intermediary layer that can translate proprietary model behavior into standardized consumer facing and supervisor facing disclosures while preserving legitimate confidentiality.

Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.econstor.eu/bitstream/10419/341103/1/1971023256.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:zbw:safepl:341103

Access Statistics for this paper

More papers in SAFE Policy Letters from Leibniz Institute for Financial Research SAFE Contact information at EDIRC.
Bibliographic data for series maintained by ZBW - Leibniz Information Centre for Economics ().