Alternate Underwriting — The Why, The What , The How and The Where — Part 1
For an industry (financial services) that is certainly amongst the hottest and one of the most rapidly advancing in terms of technology. Is dependency on rudimentary Credit Bureau assessment finally going to end?
“Ignoring technological change in a financial system based upon technology is like a mouse starving to death because someone moved their cheese”
-Chris Skinner
The Why
Ever since the advent of Credit Rating Agencies like Experian, TransUnion, etc. our Banks, NBFCs (Non-Banking Financial Corporations) and even FinTech(s) have looked up to them as the “Messiah” for deciding the credit worthiness of an applicant. However, very little is known about the scoring models/algorithms that these agencies use, what parameters do they consider, how accurate they actually are?
To put things into perspective for non-financial services background peeps. A credit score (or CIBIL Score more popular in India) is a score generated on the basis of one’s past repayment history on loans availed, the degree of delinquencies (defaults), the type of loans availed, enquiries for loans and much more.
But wait, what if you have never taken a loan before? Well then you would be scored as -1 (or “New to Credit”) as the industry likes to call it. Lending institutions integrate Credit Bureau APIs or manually fetch these report for each applicant which is then underwritten by machine or a Credit manager/analyst in most cases.
Well, it is certainly true that we have come a long way from end to end manual loan processing to semi/fully digital products. But, in actuality Credit Bureaus are still the “Big Daddies” of the lending landscape. Lending institutions build their credit models based on the data provided by these bureau(s) and decide upon each application accordingly. But, the real issue is still “How accurate are these bureau scores and reporting?”
Each Bank and NBFC has to mandatorily report the performance of each loan in their portfolio to these bureaus, which is then updated at their end and the score is revised. But, there are many such instances where these loans go unreported, have inaccuracies or like more recently due to the COVID crisis when “Moratoriums” were awarded to customers, the whole idea of reporting went for a toss thus, raising eyebrows on the dependencies of these Credit Bureau Reports.
To add to this, the per application cost to access the Credit reports is a function of the volume that the institution guarantees. So, for a large bank this could be lower than Rs. 5 per hit and for a small scale NBFC this could go well above Rs. 100 per hit. This is similar for fintech as well since, they are mostly working with NBFCs and have to use their credentials for accessing these reports. This shoots up the sourcing costs by a considerable amount thus, making an “Alternate underwriting” approach more lucrative for such players.
The What
Data sources that fall outside of the traditional scope are generally referred to as “alternative,” despite their growing use for a variety of purposes across the loan life cycle. Examples of alternative data include a consumer’s payment history on items not included in a traditional credit report, such as rent and cell phone, utilities, or other bills; a consumer’s use and repayment of certain alternative loans; and bank account transaction data. In addition, lenders may consider data that is not as closely tied to financial behavior, such as educational background, occupation, and social media or other online activity.
In the recent past multiple alternate approaches have been used by new age “Fintech players”. Bank statement analysis is now a widely adopted “alternate” approach. Unlike, Credit Bureau reports a bank statement is something which sheds more light on the overall financial health of the customer while also giving us a fairer representation of existing obligations , bounces, spending patterns, additional income and much more.
In a Mobile application driven product a bunch of “Permissions” are taken from the customers and a whole new plethora of “alternate/surrogate inputs” are born. SMS data, App usage data, Location data, Social and demographic profiling are a few examples of the same.
The whole idea is to create an “Alternate profile” for the customers, which can then be used as a surrogate for Credit bureau reports for lending.
The How and The Where — A sneak peek (Much more in the next set of posts)
Recently, a lot of middlemen aka “Data aggregators” have come into play. These are essentially B2B or B2C startups who have amassed a considerable amount of data, using which a sustainable underwriting mechanism for their specific industry is possible. For example, Zomato has recently ventured into “Restaurant loans” in partnership with a well renowned NBFC to offer credit facilities to the businesses it already has onboard.
Now, this is alike any Business loan or a loan to professional but here the lender essentially has unfettered access to the applicant’s business health. From revenues, to inventories, to operational cost, etc. “you name it they have it”.
Such an approach is a byproduct of multiple small stacks/developments being made over the last decade. Be it, the advent of E-KYC, C-KYC , Video KYC, etc. or Digital agreements and Mandates (auto-debit) available in multiple forms. The “digital stack” is bound to improve and a huge Indian audience which has a “lack of access to affordable credit” will be able to benefit from this.
The future of fintech is surely exciting, it’s probably hotter than the e-commerce space back in 2010s.
In the next set of posts, I am going to try and give a deeper insight into “how” alternate underwriting functions and also discuss “where” it fits both in the current and future scheme of things with respect to lending in India.
Part 2 coming soon!!
If you have any suggestions, please write in the comments or DM me at:
Twitter: https://twitter.com/prakhar_1296
