December 2016 - Getting Serious About Diversity

TheMReport — News and strategies for the evolving mortgage marketplace.

Issue link:

Contents of this Issue


Page 27 of 67

26 | TH E M R EP O RT FEATURE T he vast majority of U.S. consumers can access affordable credit thanks to the widespread adoption of credit scoring by financial institutions during the past 25 years. However, there are more than 50 million consumers that don't receive a score today for two primary reasons. The first is that there are 25 million consumers who have not used any form of traditional credit and as a result are considered to be "credit invisible." The other consumer population has either not been able to maintain credit following a negative event or has elected to stop using credit. Many of these consumers rep - resent good credit risks for banks to bet on and, even though they do not fall squarely into tradition- al requirements or data sets, we should not overlook them. But can credit scoring models help lenders safely and responsibly extend credit to these consumers? The answer is yes for many in this population, but only when credit bureau data is augmented by the right alternative information. This approach is crucial to creating a diverse population of homeowners and allowing those who would traditionally be shut out of the market to become loan eligible. Why the Current Approach Isn't Cutting It for "Unscorables" S coring models that rely solely on sparse or old credit data do a poor job forecasting future performance. FICO developed a research score for approximately 7 mil - lion (25 percent) of the currently unscorable consumers, or those who have one or more collec- tions or adverse public records but no other credit account data. We calculated several standard predictive measures to evalu- ate score performance. For these sparse file consumers, the Gini index of the score–a measure of a score's predictiveness–was 0.147, significantly lower than the 0.600 to 0.800 Gini indices for scorable consumers. This means scores for this group are less predictive of future behavior, which leaves financial institutions less able to separate consumers representing good credit risks from bad. We also looked at scoring con - sumers with older bureau data, or those who had no credit account updated in the last six months. Using a research model with a recent national credit bureau sample, we compared the odds- to-score alignment of this group against a baseline of scorable consumers with at least one credit account updated in the last six months. The results showed that the risk level associated with a particular score begins to vary across successively more stale seg - ments of the population. What does this mean for our industry? To start, lenders work- ing to establish an underwriting strategy for borrowers at a given score cutoff could inadvertently accept consumers with mark- edly different repayment risk, depending on how long it's been since the credit file was updated. For example, our research model demonstrated that a credit score of 640 for consumers whose bu- reau data had not been updated in 21 months or more exhibits repayment risk roughly in line with a 590 score for the tradition- ally scorable population—an odds misalignment of about 50 points. The bottom line is that risk dis- crimination is weak when scoring on sparse or old bureau data. For lenders, use of a weak score could mean declining applicants they should have accepted, and vice versa–producing higher levels of delinquency and lower lend- ing volume than necessary. This could be particularly detrimental for consumers, who could either receive lower credit lines or loans than requested or higher than they can reasonably handle. This leaves many unscorable consumers stuck in a catch-22: To obtain credit, they have to be using credit, but without a reliable way to assess creditworthiness, lenders may not take a chance on them. Assessing Alternative Data A key reason there's an oppor- tunity to score more consum- ers today is the growing number of alternative data providers that have entered the market in recent Financial Inclusion: Alternative Data is the Key Credit scoring models that access alternative data can help to diversify homeownership by serving those who would be left out of mainstream credit. By Joanne Gaskin

Articles in this issue

Archives of this issue

view archives of TheMReport - December 2016 - Getting Serious About Diversity