17,000 new credit cards issued by ICICI linked to wrong users

More data means more confidence, precise lending decisions, and more support to every individual in the economy. When a credit risk analysis is included, alternative sources like real-time income and employment data offer a truer, more comprehensive view of a borrower’s finances.

According to the World Bank’s estimates, nearly 1.7 billion adults remain unbanked globally, particularly in developing economies such as Bangladesh, India, Mexico, China, Indonesia, Pakistan, and Nigeria. This means that 1.7 billion adults do not have any accounts with financial institutions or mobile money providers.

Extending financial access to these invisible groups of people is a major challenge that continues to hold the interest of fintech and other banking institutions globally. As we inch closer to changing the very framework of our formal financial system, the process of lending and borrowing has also been undergoing significant changes.

The downside of traditional credit systems

Traditionally, an individual’s creditworthiness was defined by their credit score and the data contained in their core credit files. Lenders would make any lending-related decision based on the individual’s credit history, credit utilisation, terms and performance on previous loans, and types of credit. However, the problem with this system is that the largely unbanked set of our population has a short or non-existent credit history, thereby disqualifying them from borrowing on the basis of credit.

In India, especially, a lack of credit scores is the major obstacle loans face. The vast majority of working people in India are denied access to financial services and have been rejected because of a low or non-existent credit score. These individuals find it difficult to build their wealth with their resources and purchase any productive assets, leading to an imbalance of financial power in our society, as it lacks opportunities for people to prove their credit-worthiness.

Traditional data is also insufficient in terms of its inability to verify a customer’s current and past employment, current and past income, and reflect an accurate credit score or the working person’s actual ability to repay a loan. Copies of bank statements and self-reported income are also susceptible to fraud.

Financial institutions are increasingly open to adopting alternative credit data as the primary source of determining creditworthiness to tackle the limitations in existing underwriting processes and provide credit access more fairly and improve financial inclusion. Turning to alternative data points can fill the substantial gaps in credit and employment records and offer a fairer assessment of workers’ overall financial health.

What is alternative data?

Alternative data is any data indirectly related to a customer’s credit conduct. These include several non-traditional data sources that are a part of a person’s regular financial commitments; some examples could include rental payments, mobile bill payments, cable TV payments, small loans, travel history, e-commerce, government transactions, property records, demand deposit account information such as recurring payroll deposits and payments, social media usage, information such as a person’s education or employment background, and so on.

Smartphones can help reduce the cost of lending by generating massive amounts of data through texts, emails, social media posts, GPS coordinates, retail receipts, etc. This indirect access can help financial institutions keep track of consumer behaviour patterns in repayment or default. Before lending money to the individual, comprehensive credit risk management models are used to leverage alternative data and develop credit scores. This credit risk analysis gathers data points that include the loanee’s habits, behaviour, character, preferences.

An assessment of one’s history with money can help establish whether the customer is trustworthy and reliable enough to extend credit or other financial services. Further, the customer data can enable providers to develop innovative digital financial products to fit the needs of the invisible unbanked persons and help them build their credit history and scores.

A meticulous alternative credit data source should have broad and consistent coverage, should be specific in terms of the information gathered, should be updated accurately and frequently, should be predictable in terms of consumer behaviour, must abide by existing regulations for consumer credit, and should complement traditional bureau data so that it would improve the accuracy of traditional credit score systems.

Alternative data is available in two forms— the less expensive but comparatively less efficient aggregated data sets and the more expensive and comparatively more efficient raw data fed through an application programming interface (APIs).

Why is alternative data better for financial institutions?

Alternative data allows a more accurate analysis of a loan applicant’s risk status by enriching a fintech provider’s data source. This can help lenders make more profitable loans that align with a given risk appetite. They can also lower transaction costs, exclude high-risk applicants, expand data indicators of risk borrowers, identify low-risk applicants for automatic loan pre-approval, increase competitive interest rates, issue loans and credit faster and more efficiently without the tedious process of applying for credit. In general, when more credit or loans are provided to individuals, economies will perform better.

Fintech’s are beginning to realise the power of alternative data sources. The right types of alternative data can significantly improve predictive models, including risk modelling, fraud detection, and lead scoring. Alternative data provides a more comprehensive, well-rounded picture of who prospects and potential borrowers are—which in turn, helps fintechs make better decisions about who to target, extend credit to, and strive to retain over the long run.

Alternative data processing can be streamlined with the help of artificial intelligence, deep learning, data analytics, and anonymous metadata. AI-enabled fraud detection solutions can also help lenders identify fraudulent persons early in the lending process at the pre-qualification stage.

Fintechs and banking institutions can limit the negative impact by targeting lower-risk leads who are more likely to pay back loans on time. Enhancing existing datasets with real-time, relevant information enables the creation of new lead scoring models that will accurately qualify leads and identify those with low-risk profiles.

Financial institutions are also more mindful of protecting sensitive data and user privacy. By making payment history from alternative data sources user-permissioned, customers are empowered to control whom they wish to share their data with and how it is used.

The future of alternative data

More data means more confidence, precise lending decisions, and more support to every individual in the economy. When a credit risk analysis is included, alternative sources like real-time income and employment data offer a truer, more comprehensive view of a borrower’s finances. By taking advantage of these solutions, lenders can unlock access to affordable credit for millions of overlooked workers and tap into new markets and new business opportunities.

New and out of the box, alternative data sources promise unique advantages. The roaring success of their option is witnessed in the fact that alternative data is actively being used to forecast major financial events of the future, such as an economic slump.

(The article is written by Meet Semlani, COO & Co-Founder, Tartan, views are personal)