How are alternative approaches to credit risk assessment reshaping lending?

Long before the more known lending system originated about 3000 years ago in Ancient Greece and Rome, loans and lending existed in a few ancient civilizations such as India and Mesopotamia. In Ancient Rome, the Temples were at the heart of banking and lending – borrowers would then repay the money in the fear of the God. Subsequently, loans were either given out on personal references and trust or against a property (mainly land). A money lender (or later even a bank) would extend credit based on what he knows of the borrower, and often charged high-interest rates for the risk he undertook. 

With the evolution of modern branch banking, knowing a borrower personally became impossible. This new evolution demanded a new approach to credit risk assessment. The system of Credit Reports and Credit Scores, which look at all loans taken by an individual in the past and his repayment behavior on these loans, became the guiding post for credit underwriting. Over the years, the credit score has become the standard indicator of creditworthiness - not just for individuals, but also for businesses. A good credit score would mean you have a better chance of getting a loan in the future and at better rates. 

However, in today's complex world, there is a challenge. 

The credit bureaus such as CRIF High Mark today provide credit information on almost all who may have a loan or a credit card from any of the formal lenders across the country. However, there are individuals and small businesses in India itself that never would have a credit score as they have never availed any credit from a bank or an NBFC. Such individuals and businesses may be dependent on friends, families or local mostly unorganized money lenders. As “traditional” credit score improved access to credit, can there be an “alternate” credit score that can enable financial inclusion for the next set of such unserved and underserved people? 

What is “Alternate” credit scoring? 

Only data can help address such a problem and that too at a scale. Alternative data, which includes any type of data streams or approaches that credit bureaus traditionally do not tap, can provide insights into the behavior of those who do not have any past loans (commonly known as NTC: new-to-credit borrowers). 

In the quest of serving to NTC customers, FinTechs have been driving innovations in finding Alternative Data streams and approaches. The alternative data could be transactional data coming from mobile phones on expenses and incomes or could be one’s interactions on the social media platforms such as LinkedIn, Facebook, Twitter or Instagram. It could include information like what apps you use, how often use them, who we follow on Twitter on Instagram, what places we visit, which may seem totally unrelated when it comes to the matters of finance, but in fact throws up interesting behavioural and lifestyle indicators. 

While there are issues around privacy, the ubiquity of the internet and mobile devices have made it possible to capture and piece together what may seem to be unrelated pieces of information from your online activities to predict financial outcomes. But amidst this plethora of information, the focus of the alternative data for assessing creditworthiness still remains on the “human factor” – the intent, the behaviour and the character of the borrower – one of the five “C” s of credit risk. 

The alternative approaches are not limited to the exploiting new-age data sources, but also include leveraging a well-known practice such as psychometrics for a newer use case of credit risk assessment. Psychometric testing is being deployed to understand the behaviour and personality traits of an individual with an objective of assessing her intention to repay. For NTC borrowers who do not have smartphones or presence on social media, psychometric testing could be the leveller. 

Can alternate credit scoring reliably predict risk? 

Though there have been successes in linking alternative data and approaches to managing credit risk, there have been misses too. Like everywhere, the availability, quality, and reliability of data is the key to prediction. Though FinTechs are working on making sure the data captured through alternative mechanisms are consistent and reliable, there are still work-in-progress questions. 

Given these approaches are relatively new, is enough history available to make reliable predictions. The approaches have also not yet seen a few economic cycles to give a wide-spread confidence. It is also important to make sure that a borrower can work upon to influence these data points to indicate improved behaviour, these data points cannot be directly manipulated by the borrower or an intermediary. 

There will always be risks in the lending business, but the banks and the lenders are continuing to expand their business and reach while keeping the risk under control.  

The speed at which the data get generated today, the quantum of digital transactions that we all engage in and Alternative approaches are challenging the existing frontiers and creating newer avenues for both lenders and the underserved borrowers. Insights brought by alternative approaches on the borrower behaviour may not always be sufficient, but vital nonetheless, as those can fortify the existing approaches of credit risk assessment.

Original Source: Times Now

(Reprinted)