Alternative Data to Develop a Credit Score

Today, credit for low-income clients is mainly provided through labor- and relationship intensive credit methodologies. Resulting high operational cost for providers and perceived riskiness of uncollateralized lending translate into high effective interest rates for clients. But high costs are not the only factors leading to high interest rates. Many low-income populations in developing countries are often not formally employed, lack a credit history, cannot provide sufficient proof of identity or other application requirements, do not possess an acceptable collateral, and often the lending institution has little experience with the clients’ economic activity leading to untailored loan products. The lack of information about the borrower’s payment capacity leaves many potential borrowers without access to credit, or access to credit with high interest rates, a way for banks to manage risk.

The increasing use of electronic payment methods by lending institutions (e.g., mobile banking, ATMs, banking agents, etc.) increases the ability to track and exploit clients’ payment and transaction histories to predict payment capacity and creditworthiness. For clients without an existing credit history and without formal employment in developing countries, such data can be a valuable resource for lending institutions to better assess risk and help currently unbanked clients develop a formal credit history.  Such data can include:
•       Bill payments (e.g., electricity, gas, water, etc.)
•       Phone bills (e.g., mobile and fixed, post- and pre-paid)
•       Rental payments
•       Transaction data (e.g., remittances, withdrawals, deposits, transfers, etc.)
 
Today, private credit bureaus of only 39 economies, out of 178 economies (22 percent) covered by the Doing Business Project of the World Bank, collect information on clients’ utility payments.  No public registry distributes such data.
 
CGAP’s Technology Program is preparing a study to learn more about the potential of alternative or non-traditional data (i.e., payment obligations such as phone, gas, electric, etc.) in developing countries to integrate these clients in the credit registry systems, but also how such data can be used by lending institutions to assess client repayment capacity and creditworthiness.