COMPATABILITY MODE

It looks like you are running Internet Explorer in compatability mode, please turn this off to benefit from all functionality on this website.

To do this, click the blue broken page icon in your browser address bar (example below).

compatability mode

JAVASCRIPT DISABLED

It looks like you aren't running java, please turn this on to get the best user experience

Cookies & Privacy

By continuing to use this site, you consent to our use of cookies to optimise your browsing experience.

Learn More

Unintended consequences

Posted in 'Credit Score' by Barry Stamp

20 August 2019

When it was lawful to include gender in a credit scorecard, it was universally included. This is because gender is predictive of the risk of default, indeed at the time it was ranked in the top 6 of all predictive characteristics.

As the machinations of scorecards are kept very secret, only very few knew exactly how gender affected a credit score. The generally held view, never challenged, was that the inclusion of gender in credit scorecards and in other automated decision processes was damaging to women. Even government regulators at the time believed that the existing decisioning systems prejudiced women.

So when equality law required the removal of gender from scorecards, the truth came out. Women are more disciplined at managing money and have fewer defaults than men. They are more creditworthy than men. The removal of gender from the decision process meant that more women were declined credit than had been the case beforehand. The previous practice of including gender in the decision mix had actually helped more women get the credit they deserved.

Forty years on, ageism in credit scoring is now discussed in much the same way as gender was discussed in the early 1980’s. Age is the most predictive characteristic in scorecards. Generally, and up to a point, the older you are, the less likely you are to default on a credit agreement.

If age is removed from scorecards, credit scorecards will be re-weighted to take account of the loss of predictiveness of that component. They will still do a much better job than humans at predicting who will and won’t pay. But less people will get credit, and those that do will probably pay more for it.

Barry Stamp

Barry is a Chartered Banker and a Fellow of the Chartered Institute of Credit Management. He has a degree in Statistics and Business Economics from the Open University. Barry writes mostly on news from the worlds of banking and mortgages.

Related ARTICLES

To Top