How do banks decide whether to give you a loan?
Two things determine the outcome of any loan application: available data on the customer’s creditworthiness and the methods used to process the data.
“How much do you earn?”
“Do you have a stable source of income?”
“Have you taken a loan before?”
“Did you repay your last loan?”
“How do we make our money back if you cannot pay?” must be asked and answered by any reasonable credit loan officer. These are the type of questions that determine who gets credit and how much they get.
Getting this process right is also important for economic stability.
The 2009 Great Recession was triggered by poor risk assessment measures adopted by large banks when giving out loans to individuals. Banks skipped the questions above and granted loans to anyone that walked through the door. As they intended to aggregate and sell these loans to third parties, they did not care whether the borrowers were creditworthy.
They stopped caring about the data.
These days, financial institutions—and the regulators that oversee them—pay more attention to data on prospective borrowers. That just creates another problem, though: whether the data is available.
Today, Nigerian banks are facing this problem head-on as they try to accurately predict the risk of a borrower defaulting on a loan. If you are wondering how our banks have survived for so long without fixing this problem, it is quite simple.
For a long time, Nigerian banks focused on giving loans to the people and institutions that they were sure were creditworthy. Generally, that meant that loans were mainly given to government agencies, multinationals, and high net-worth individuals. For a typical bank, it was much safer to lend $10 million to 5 high net-worth individuals they had done business with in the past than to lend $10 million to 100,000 salaried earners that they knew nothing about.
Of course, banks could try and get data on these 100,000 people, but they generally opted not to for two reasons. First, existing loan origination and credit underwriting systems struggle to process large volumes of small loans. It is expensive to manage.
The human underwriters who assess the creditworthiness of borrowers must manually go through thousands of applications and decide on each of them. It is slow, expensive, and often unprofitable. So, lenders stuck to the market they knew.
The second reason is the data itself. Remember, two things are needed to decide on a loan application: data on the borrower and a method to process the data.
Banks have some data, but this mainly covers the people already within the financial system, i.e. the banked. Moreover, this data is usually lagged, meaning that it can tell you whether the borrower had a high credit risk in the past, but not whether they will repay a loan in the future.
Yet, the bigger challenge is that financial institutions in Nigeria know very little about most of the population. Data from EFInA in 2018 shows that more than 35% of Nigerians—about 40 million adults—are completely excluded from the formal financial system. These people face a Catch-22: banks refuse to lend to them because they have no credit histories, and they are unable to build credit histories because nobody will lend to them. When they do manage to get loans, the interest rates are exorbitantly high.
Once the data is available, lenders also need to be able to process it. To accurately predict creditworthiness, high-quality decision modelling from data experts is sorely needed.
Faced with all these problems, Nigeria’s biggest lenders eschewed the retail lending market. In the end, banks decided it wasn’t worth it; until the Fintechs arrived. The demand and supply mismatch, i.e. many people looking for loans and only a few lenders ready to provide them, attracted Fintechs to the space.
Fintechs are built with entirely different business models and have been able to make headway into the market. Now, banks want to close in on this gap, but they need the technology, innovation, and efficiency that Fintechs can boast of if they want to compete.
This is where players like Indicina come in.
Indicina provides a unified lending platform that digitises the entire credit value chain. Using their platform, African lenders (banks and non-banks) can apply a data-driven approach to credit underwriting at scale, extract new insights about their consumers, and de-risk their unsecured loans.
The platform touches on the six main stages of the credit value chain, including identity verification, application processing, underwriting, decision making, disbursement, and monitoring.
Now, you may be wondering, with so many Fintechs out there offering solutions in the credit value chain, why Indicina?
It comes down to people.
As is becoming clearer, the future of lending is a battle between Fintechs innovating in finance and financial services players figuring out technology.
Indicina’s founding team of Yvonne, Jacob, and Carlos masterfully combine both. Yvonne entered the technology space early, completing a first degree in Computer Science, before going on to get an MBA and work at Merrill Lynch.
The big change came when she moved back to Nigeria and spent eight years leading the strategy team at First Bank of Nigeria. In time, she met her co-founder, Yemi, while investing in Andela. Conscious of the need to deepen the technical expertise in the founding team, they onboarded an experienced software engineer who had served as the Lead Engineer on Konga Pay and as the CTO at Supermart.ng.
The pieces were not quite in place, though as the team had a data expertise gap. Step forward, Carlos del Carpio, an economist with over a decade of experience building and leading global analytic startups. Carlos was a founding member of EFL (now acquired), where he built the Data Science team from scratch. He has a deep understanding of risk management in emerging markets.
This eclectic group has achieved phenomenal success by combining statistical and machine learning techniques to create credit stories for their clients’ applicants. Indicina has analysed over 25,000 customer profiles and generated a 15% decline in default rates for their clients.
Over the years, Indicina has discovered that the most useful data points for assessing credit risk are the financial and socio-demographic data (e.g. name, age, income bracket, etc.). Moreover, Indicina found that applying their risk prediction models to these data points reduced the likelihood of a client adding a bad loan to its portfolio.
As the company expands, they are targeting banks and technology companies that require support to understand the growing lending market better. Between 2012 and 2017, the size of the retail banking revenue pool in Nigeria increased from $20 billion to $35 billion.
According to the World bank, “Today and tomorrow belong to those who are able to play in retail banking. The drivers of any sustainable retail lending business model include digitization and data-driven decisions.”
It is a privilege to invest in this unique team that is rebuilding credit infrastructure in Nigeria. We are excited to be building the future of such an important market.
We welcome Indicina to the Future Africa community as we look forward to what the future holds.