I came across an interesting blog post recently about how the untrained human brain is disadvantaged when it comes to recognising or thinking about big numbers.

These limitations, and reliance on gut feel and framing analysis via proportionality not only play into the hands of car dealers looking to upsize customers to the next model up with extra bells and whistles. It's also a factor when making financial planning decisions, from both the customer perspective, and potentially the bank's.

A study conducted by researchers from the Federal Reserve Bank of Atlanta, the University of Lausanne and Columbia University interviewed a set of subprime mortgage borrowers, asking them a series of questions designed to test their ability in mathematics. Then, they divided those subprime borrowers into four groups in order of their mathematical skills. The bottom two groups showed foreclosure rates of over 20 per cent, the top group had a foreclosure rate of 7 per cent.


So if banks ever take their current work on advanced pricing capabilities to their logical conclusion, they might be issuing maths tests along with mortgage application forms to enable them to segment more effectively, and offer more attractive rates to the customer who is less likely to default.

Of course, there are plenty of reasons why this isn't practical. But it's an interesting idea nonetheless.

Accenture quotes research showing that pricing is the number one driver of profitability for retail banks, followed closely by loan losses. Every 1 per cent improvement in these figures drives millions of dollars in bottom line improvement. 

Getting detailed personal level information, even down to the level of maths capability, for an individual could provide a bank with the opportunity to get really granular with its segmentation. That is, if its entrenched processes and IT systems allowed for more complex pricing structures of course.

In all the marketing spin about customer-centric propositions, this is the end game. If a bank knows everything about their customer, they can predict not only credit risk (based on past history, and in this instance a maths test, but there could be a whole range of other indicators) but also things like potential lifetime value of the client, and willingness to pay.

But from the consumer perspective, how comfortable are you with sharing this level of data with a bank? Perhaps if you were already confident that you qualified for one of their valuable segments you might want to share more personal data if you think you'll get a better price. But if you can barely remember long division let alone algebra, you would be understandably reluctant to open up.