Biases in Loan Recovery-Rate Estimation

Jaime Leyva and Tiago Pinheiro

Working Papers from Banco de Portugal, Economics and Research Department

Abstract: Sample selection and correlation between recoveries and the time to recovery are inherent to loan recovery datasets, leading to biased and inconsistent estimators of recovery rates. We characterize and quantify the biases of two common estimators, and we propose a class of estimators that corrects them. Simulations show that the bias increases with shorter observation windows and longer time to recovery. Real-world data on firm loans broadly confirms simulation results and shows that, even with sixty months of data, the biases of the two common estimators is at least ten percentage points and can be as high as twenty percentage points. These findings highlight the importance of addressing sample selection in recovery rate modeling for improved credit risk assessment and regulatory compliance.

Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.bportugal.pt/sites/default/files/documents/2026-04/WP202602.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:ptu:wpaper:w202602

Access Statistics for this paper

More papers in Working Papers from Banco de Portugal, Economics and Research Department Contact information at EDIRC.
Bibliographic data for series maintained by DEE-NTD ().