FICO has had a huge impact on the everyday lives of millions of Americans. Would you be able to get the mortgage to afford the new house? Will you be approved for that credit card? All of this is directly influenced by the consumer’s FICO score. It is important to understand what does a FICO score entail and how it is scored. The below representation is for an average customer and this may change for customers with different lengths of credit history.
FICO basically takes into account the credit report supplied by credit bureaus. Experian, Equifax and Transunion are the three largest bureaus in the United States and they utilize credit data like repayment of loans, credit card usage and other relevant statistics to create a summary of consumer’s financial actions. But it has been roundly criticized by many experts for not providing a complete picture and having an adverse impact on minority and low-income borrowers. It is also essential to analyze the phenomenon of “credit invisible”, those who do not have a credit history or have a limited credit history rendering them unscorable. A study by Consumer Financial Protection Bureau in 2015 revealed that 26 million Americans do not have a credit history and 18 million Americans are unscorable because of thin credit files. Almost 20% of American adults are therefore rendered ineligible for credit because of no fault of theirs. Also, minorities and low-income consumers were reported to be more likely to be credit invisible.
The solution for credit invisible is leveraging alternative data to model and predict the behavior of the user. FICO in collaboration with Equifax and LexisNexis Risk Solutions has launched FICO Score XD and TransUnion has launched CreditVision Link to enhance its coverage of the American adult population. “ In testing, FICO XD allowed more than half of credit card applicants who were previously unscorable to receive a score”, said Jim Wehmann, executive vice president for scores at FICO. TransUnion reported that CreditVision allowed it to score 95% of Americans and resulted in 24% more loan approvals in a pilot with an auto lender. So what is the new data which has resulted in such a sharp jump in “scorables”.
Going even further, PRBC.com offers a great alternative to FICO’s XD product. PRBC uses monthly-bills payment history, income data, cash flow data, account receivables, account payables, payroll, and outside investments to provide a complete picture of a financial profile. This allows PRBC to evaluate all Americans that pay regular bills. Unlike FICO, a consumer can self-register on PRBC.com and report their own scores.
Companies like Envestnet-Yodlee and ID Analytics have launched alternative data solutions as well. Envestnet® | Yodlee® Risk Insight claims to incorporate data elements that are not available through the credit bureaus. ID Analytics has Credit Optics®, an FCRA compliant credit score which uses a repository of consumer behavior data sourced from a wide range of industries. The cross analysis of the alternate data with the traditionally available information helps solution providers present unique insights to lenders and others users.
Mortgages, Credit Card utilization, and auto loans represent the traditional data. Telecom, cable and utility bills and payments represent the new paradigm for credit analysis and scoring. Checking and saving accounts can be a particularly rich source of information for credit bureaus. Companies will also start taking into account short term loans, alternate loans and even qualitative factors like how often the consumer changes his physical address. All this information will translate into a complete 360-degree view of a prospective borrower versus the one-sided review from the current FICO score. This approach will also be a blessing for credit Invisibles who are destined to have a good credit if given the chance.
Start-ups have already jumped on this bandwagon a decade ago when they started utilizing parameters like time of application, use of upper case in the online forms and amount sought as a loan. Lenddo, a fledging start-up in the alternative lending space has gone further by assigning “Lenddo Scores” to its applicants. The company uses no traditional data but instead depends on its invention- a graph variable methodology called “Archano Score”, which is a PageRank-like scoring algorithm, incorporating attributes of each member and their social group. It uses more than 12,000 data points and analyses social profiles, communications and emails of the applicant to base its decision on whether to lend to the applicant. The social underwriting experience has been a success with the company helping lenders increase approval rates by 50% and reducing risk by 12%.
Limitations of alternative data
The new credit rating system might be a boon for some, but not for all. Credit challenged individual dependent on short-term loans or families behind on their utility dues because of peak bills in winters will be heavily penalized. Currently, the utilities only reported the serious delinquents to the credit bureaus, but this move will certainly push down the credit ratings of many Americans. Also, there is fear among consumer advocates that lenders can selectively target this data to exclude traditionally credit challenged minorities. Another lament is on the accuracy of the data provided by third party providers like telecoms which do not have strong consumer protection histories and are often in dispute with their customers due to practices of the surcharge, cramming etc.
The new normal
Alternative data is going to be the new normal is a foregone conclusion. Easier accessibility to large troves of consumer information and our digital imprint are the next frontier in the never ending quest to enhance the world of credit. But it is important to recognize that alternate data is a double-edged sword. A white salaried American with no credit history will be welcomed with open arms but a Hispanic businessman 30 days overdue on rent can be pushed to higher interest rates or even no credit. It will be good for the society if the new data sources are used to include those not covered earlier rather than exclude those arbitrarily deemed to be too risky for credit.