Skip to content

FRM Part II · FRM Exam Part II · Credit Scoring and Retail Credit Risk Management

A retail lender uses a logistic regression scorecard to approve credit card applications. Compliance finds that a variable based on the applicant's postal code strongly predicts default, but it is also highly correlated with the applicant's ethnic group, which is a protected class. Which action is most consistent with fair lending practice?

The lender should test for disparate impact and look for a less discriminatory alternative with similar predictive power. Fair lending considers outcomes, not intent, so predictive strength alone does not justify a proxy for a protected class, and treating applicants differently by class would be disparate treatment.

  1. AKeep the variable because it improves predictive power and the lender has no intent to discriminate
  2. BTest for disparate impact and, if present, assess whether a less discriminatory alternative with similar predictive power existsCorrect
  3. CRemove the variable only for protected-class applicants and keep it for all others
  4. DKeep the variable but lower the cutoff score for the affected postal codes without further analysis

Explanation

Fair lending frameworks recognise disparate impact: a neutral variable that disproportionately harms a protected group must be justified by business necessity, and a less discriminatory alternative with comparable accuracy should be used if one exists. Intent is irrelevant, so keeping the variable on predictive power alone is wrong. Treating applicants differently by class is itself disparate treatment.

Did you get it right without looking?

One question tells you little. A timed set on Credit Scoring and Retail Credit Risk Management shows your real accuracy, how long you take and where you lose marks.

Practise Credit Scoring and Retail Credit Risk ManagementFree diagnostic mock

More Credit Scoring and Retail Credit Risk Management questions