Fighting financial fragility through inclusive credit scoring tools

The European Commission estimates that around 20% of the EU population (over 100 million people) faces some level of financial exclusion, which includes under-banking*. Not having access to mainstream financial services increases greatly the risks of financial exclusion and inequality. It also limits personal growth, innovation, and the ultimate goal of being financially independent.

There is an opportunity to address the needs of the underbanked populations with innovative credit score models including alternative data sources that go beyond just analyzing traditional credit histories.

The start-ups in the space can be split into these groups

• AI-based models start-ups focus on gathering alternative data sources, such as mobile phone usage, utility bill payments, rental payments, e-commerce transactions, social media activity, and even psychometric data. They then use these new data sources to build models to identify patterns, correlations, and predictive indicators of creditworthiness. Credit Kudos is a company that I am following in this space.
• Open banking and transaction data startups focus on accessing detailed transaction histories and account data. This information can provide insights into an individual's financial behavior and help assess creditworthiness. Tink, a Sweden-based start-up will be a good example of a company in this category.

What I like about it

• ?It is a huge market both in developed markets as well as emerging countries.
• ?Inclusive credit scoring is mainstream and it offers quick wins for certain segments, for example, contractors.
• The traditional scoring system base scorings historical banking data instead of focusing on the lifetime value of their clients
• Huge opportunities in some industries lack professional scoring systems and processes and they will welcome solutions in the field (real estate tenant scoring will be a great example of nonprofessional systems).

Some challenges of the model

• Financial institutions and other sophisticated clients will need to see a clear outcome in terms of increasing their scope but also in terms of less delinquency.
• ?As in other intensive data business models, data bias, interpretability, and privacy may end up being relevant concerns.

This is a category we are looking at actively with an angle of how AI can play a major role in developing tools that make more inclusive scoring systems that increase the reach of potential customers and make credit accessible to more people.

Source*: https://ec.europa.eu/eurostat/web/products-eurostat-news/-/ddn-20220915-1