Alternative Credit Scoring to Include the Poor in Financial Lending
In India, a considerable portion of society falls under the credit-invisible category. This is because many workers either don’t have a bank account, are paid wages in cash, or have no loan/credit history. Traditional credit score are not available for credit-invisible individuals since they have no typical record to prove their creditworthiness. In such a case, we are heavily missing out on lending loans to ‘deserving yet deprived’ candidates. The non-accessibility of conventional lending channels often forces these people to turn to alternative options with high-interest rates and a high risk of getting exploited. Lending institutions here have a moral responsibility to revive the unbanked section with the use of alternate data. This can prove to be a surefire way for lenders to not only conduct proper credit scoring but also tap opportunities at the bottom of the pyramid. Let’s understand the issue with the help of an example:
Mr. Raju hails from a small town but has migrated to Delhi for work. He stays in a makeshift shanty and works as a laborer with a contractor who pays him a fixed monthly salary in cash. Now he needs to buy a vehicle for daily transportation to work for which he has saved half the amount but requires an additional loan to fulfill his need. Now, conventional lenders would seldom lend Mr. Raju a loan, citing multiple reasons such as:
- Non-availability of bank statement as proof of income
- Non-availability of credit history
- No rent agreement
- No collateral, etc
But what they don’t know is that Mr. Raju is a sincere & hardworking person. He spends diligently, sends money to his family back at home, and has a steady employment record. With his current skills and learning abilities, he has very little chance of being unemployed for long. So how to bridge this information gap between lenders and borrowers? The answer lies in the use of alternative data and analytics to create what can be called – an alternative credit score.
Use of Alternative Credit Data to Calculate an Alternative Credit Score
Alternative credit scoring refers to the use of data from digital platforms and applications on consumer behavior for credit risk assessment. Traditionally, credit bureaus have been the only providers of consumer credit information, which lending institutions still use to reduce bad debt and market risk. On the other hand, alternative credit scoring takes into account data from multiple sources, like telephone, TV & radio, mobile wallets, geo location, bill payment history, and social media usage. These data points can greatly help lenders get a sense of an individual’s financial reliability.
Using Machine Learning to Analyze Alternate Data
Machine learning combines & reads data in a way that leads to stronger Predictive Analytics in Banking for credit scoring. For instance, going back to the case of Mr. Raju, we can fetch details of his current residence and the places he visits from the past several weeks using geo-data by integrating Google’s API with the ML application. Using ML, we can calculate his employability index to assess his job opportunities. We can also conduct tests and scrutinize the answers under the lens of an ML algorithm to determine Mr. Raju’s willingness to repay the loan amount. These are just a few possibilities that can be, the scope is rather immense. We can accommodate micro segmentation from thousands of segments and determine micro patterns on a frequently updated basis.
As digital touch points continue to expand at a rapid pace, the use of alternate data is deemed to rise. Financing companies that want to reach credit-invisible consumers and make their lending decisions more inclusive, should make faster, more reliable decisions with alternative data. CRIF – an RBI authorized Credit Bureau in India – is also a leading provider of expert and statistical approaches to support needs in risk management, marketing, and decisions with their Behavioral Scorecards, Predictive Analytics, Advanced Analytics, and Risk Consulting. For more information, contact CRIF, today!