10 Jan In the Zoom – The Use of Mobile Phone Data for Alternative Credit Scoring
In the advent of online and digital lending, any individual could get information about various loan facilities and offerings right at their fingertips. Through an online web search, one can choose the best-suited loan program based on their needs, budget, and convenience. The availability of customer feedback and public reviews provide great insights and credibility to these loan facilities, creating better competition and striving to make satisfy their customers and provide the best experience.
For traditional loan providers, extending loans means putting in mind the 5 Cs of Credit – character, capacity, condition, collateral, and capital. Having these quantifiable indicators means creating traditional credit scores which verifies customers identify and assess their ability and willingness to pay. Thus, extending loans are limited to traditional banked and carded consumers.
But what about the majority lower-income segment –people who do not have financial records to prove creditworthiness as a result of having NO bank accounts and/or have not experienced borrowing from banks or regular financial institutions? Given the reality of their financial needs in medical emergencies, making ends meet, trying to invest in education or start a small business – their lack of option leads them to costlier informal loan providers. Per BSP’s 2018 3Q Financial Inclusion survey, only 3% of the adult population with outstanding loans borrowed from the bank, while a huge 39% were from informal and unregulated channels.
Yes, the government and several institutions have been pushing for Financial Inclusion to the unbanked and underbanked for years now. With its share of struggles in terms of reach and growth, hope springs with Fintech (Financial Technology) companies coming up with solutions such as alternative data credit scoring using simple but potent data sources. Nowadays, mobile phone data is widely available. It is accessible and dynamic, enough to provide indicators of identity, location, and financial activity. Given also the 1:1 ratio of a Filipino to having mobile phone, the volume of mobile phone data will continue to accumulate and increase over time.
- Use of data and reliability – Raw mobile data includes actual calls, the length, the location where it happened, numbers contacted, top up loads, etc. Telcos and fintech companies use big data analytics to create bespoke models with predictive patterns of creditworthiness of their subscribers. The signature algorithm is constantly evolving, with new telco-based data being available.
- Credit Cycle Disruption – Traditional credit philosophy would often rely on scores from socio-demographic information and/or face to face KYC and/or field visit. To include scores from mobile data will require mobile phone numbers as mandatory info to be provided by loan applicant. This is a great innovation from methods used by old school financial institutions and lending companies.
- Regulation and control – It takes some time for some lenders to keep at pace and adjust to numerous compliance circulars and memos from regulators. Using scores from mobile phone data will require consent from each loan applicant that the respective mobile data will be used for credit scoring. The specific consent is stipulated in the loan application form, which means additional clause in the lender’s T&Cs. The use of mobile technology makes it easier and faster for subscribers to get information about the data they are sharing, and eventually, give their consent for its use.
Quality of Data – Unlike some static socio-demographic data, mobile phone data is moving and dynamic. Accumulated data will always have fresh information of calls, locations and multitude of numbers contacted. More volume means more opportunity for risk score development.
Innovative solution – For lenders, it is not only the benefits of having new scores generated from consumer mobile use but having new approaches on early fraud detection. Mobile phone data contains unique KYC elements including identity of borrower based on the use of SIM number, declared addresses during application which should appear in the geo-location portion of the mobile data, and contact number references/co-maker/guarantor as frequently contacted.
Growth and Saves –Penetrating the unbanked and underbanked will require readiness on the part of lenders to confront absence of financial literacy and overcome fraud and default while managing the cost of including the segment and maintaining them in the financial ecosystem. Traditional data will not be enough as this segment has more activities in their mobile phones from which data can be used for credit risk profiling. With the right profile and volume of customers, repeat loan customers will increase overtime with a lower cost of credit to maintain on the part of the lenders. No risk, no reward.
The use of mobile phone data for alternative credit scoring is a disruptive innovation. The disruption leads to progression and will bring good effects in the long run. The key part is in the fusion of traditional scores and scores from alternative data sources like mobile phone data – creating a more powerful and predictive model for creditworthiness. In the future, it is ideal that a credit bureau with massive traditional data and mobile phone data generate an overall synthesis consumer score. Now that’s worth the innovation.
About the Author:
Elmer is currently the Chief Collections Officer of FinScore. Prior to that, he headed the collections department of CC Mobile Financial Services, pioneering the best collection practices for digital lenders. His 20 years in the banking industry established his expertise in traditional credit lending landscape before diving into Fintech.
Are you interested to know more about alternative data credit scoring? We’d like to speak to you!