Financial institutions in Africa have historically relied on credit bureau scores, or customer financial footprints, to make lending decisions. In the absence of these, social demographics – such as gender, employment status, income level, etc. – are taken into account before granting loans.
One of the main drawbacks of these conventional lending methods is that they disadvantage millions of unbanked or informally employed people, which is why Africans have one of the lowest levels of access to credit in the world. world, especially those living in remote areas.
Over the past decade, new models based on artificial intelligence and machine learning tools have emerged as an alternative way to assess credit risk.
“Assessing creditworthiness without a formal history is a major issue in financial services in Africa when social demographics are not enough to make good credit decisions,” notes Eunice Gatama, Director of African Affairs at Yabx, l one of the fintechs behind alternative credit scoring. emerging market trends.
The startup, which is incubated by Comviva and part of India’s Mahindra Group, uses machine learning to analyze mobile money wallet records and combine them with other sources such as credit bureaus and bills. public services.
This data is used to build risk profiles for potential borrowers with no credit history and detects borrower activity, such as the amount of money invested in a small business or used for personal needs. This solution is available to banks and microfinance institutions in markets where credit bureau coverage may be limited.
“We enable financial service providers to create profitable unsecured portfolios that are accessible through easy loan application on mobile devices,” Gatama told TechCabal in an interview. “A lot of things can be deciphered from the collected data.”
In addition to micro and small consumer loans, Yabx offers small business loans, unsecured working capital loans for mobile money agents, smartphone purchase financing, and a savings product. Its solution enables instant lending decisions.
Yabx claims to have run the credit score of over 100 million borrowers, 50% of them African, across 15 emerging markets in Africa, Asia and Latin America. According to the company, it processes more than 100 billion data records in a month on partner networks.
In Africa, it operates in Tanzania, Uganda, Malawi, Somalia, Mauritania and Côte d’Ivoire, where it has partnered with leading players in telecommunications, e-commerce and services payment providers, banks and other financial institutions to deploy multiple digital lending products. . It earns money based on the performance of the loans it makes to banks, through revenue sharing agreements with other lending partners and white label services for banks that want to launch and market their own digital lending products.
Given the increasing penetration and use of mobile money, Gatama says that in Africa, mobile wallets are currently the best source of alternative credit scoring. In its latest annual State of the Industry report, the GSMA reveals that the value of mobile money transactions surpassed the $1 trillion mark in 2021, with Africa accounting for almost 70% of the total amount recorded transactions.
The ease of access to a mobile phone in Africa allows many people to access financial services that they otherwise would not have been able to access.
“More than half of Africans are still unbanked, but most own mobile phones which are also used to access financial services,” says Gatama. “This creates an opportunity to bring credit products to the mass market, especially to people that banks cannot currently integrate.”
Expansion into 11 more African countries is currently underway, with priority given to markets such as Kenya, Togo, Benin Republic and Zambia, where mobile money is widely used. “We find that countries where mobile money is established are easier to set up,” says Gatama.
She doesn’t see Yabx’s machine learning model as foolproof, despite its effectiveness so far in keeping lending partners’ non-performing loan (NPL) rates in the single digits, but expects further improvements to the system. ‘coming.
“The way machine learning works is that the more you train your model, the better it becomes at predicting,” says Gatama. “Over time, our models will strengthen as more information becomes available.”
Other than that, regulation, data privacy issues, and reliability of aggregated data are major concerns for Yabx as it seeks to achieve its grand vision of simplifying access to finance for the over $2 billion. unbanked people around the world using digital footprints through mobile devices.
If you liked reading this article, share it in your WhatsApp groups and Telegram channels.