Few numbers are more important to the financial lives of consumers than the credit score. It gives lenders a working idea of whether a loan will be paid back and is used to attach prices to mortgages, credit cards, and car loans. A good credit score makes the American Dream more achievable.

Their use by financial services firms and creditors seems inevitable and unquestioned. But are the scores accurate? In other words, are they the best predictors of whether a borrower will repay?

The Powerlytics database of financial information contains a comprehensive and anonymized financial view of over 200 million adults, down to the zip+4 level. We have proven time and time again the ability to help lenders improve their ability to identify borrowers likely to default at the time of origination far more accurately than the traditional measures. In one recent example that reflects projects we’ve done for numerous other clients, we used a loan-level data file obtained from the customer to identify more than 30 Powerlytics variables that are more predictive of consumer default than the FICO score.

One variable, the total three-year percent change in other gains and losses, was 573 percent more powerful than the FICO score when it came to predicting whether a borrower would default on a loan. The top five variables were more than nine times more predictive.

For perspective, the average difference in credit score between a defaulter and non-defaulter was about 5 percent across all income bands. In other words, Powerlytics most powerful variable was 115 times more predictive of default than the FICO score, one of the most iconic credit models in consumer finance.

In almost all the cases, the variables that accurately predicted default examined how different types of a borrower’s income or expenses changed in relation to those in his or her surrounding community, or how they changed over time.

Those changes aren’t reflected in the credit score. Instead, credit scores usually rely on a record of whether a borrower has made payments on time. This drawback can mask significant changes in creditworthiness. People see their income decline or expenses increase, and use small loans or other sources of money to maintain their standard of living, masking risk. On the other hand, young graduates often have trouble getting credit cards, even though they may have recently started working a lucrative new job. With over 1,000 consumer financial indicators Powerlytics show the momentum of the borrower, not just how they performed in the past.

Powerlytics’ proprietary data set of anonymized tax information has been used by major companies to solve a wide range of business problems, from customer retention to marketing and risk analysis, and yes, other loan default models. The beauty of the data is it doesn’t require any information from the consumer, so it’s non-invasive. The only information that is needed is a consumer’s nine digit zip code.

We may be able to add one to the list: Improving how lenders assess credit worthiness.

Leave a comment