Grameen Bank was started in 1983 by micronfinance pioneer Muhammad Yunus:
“The micro lender had pioneered a group lending system where it lends to an individual only when he or she belongs to a 5 member group. The group was not the guarantor for its member’s loans but further credit was not extended to a group when any of the members defaulted. This created societal pressure on the delinquent to repay the loan and reduced fraud.” Lenddo – The Google of Lending Algorithms
Default rates for microfinance lenders were so successful that default rates were capped at about 2%. Compared to online lenders in developed economies – like Prosper and Lending Club – who incorporate FICO Credit Scores and deep financial data, have default rates in excess of 5%. To bring billions into the financial services ecosystems requires us to develop next-gen credit scoring criteria using social incentives and non-traditional data sources.
With the advent of digital, we are ready to see Grameen 2.0. There is an emerging group of pioneers that use new data sources – social, mobile, etc. – to credit score individuals in developing economies who have no/limited access to financial services.
Lenddo is one such model.
“Lenddo has emerged as the pioneer in the science to evaluate all these unique factors and has leveraged social media and email to bring about a massive upheaval in the market.” Lenddo – The Google of Lending Algorithms
For the launch of the Lenddo Score in 2013, the company created a Facebook-like UI (user interface) that allowed users to connect digital accounts – Facebook, LinkedIn and Email – to come up with a Lenddo Score.
The Lenddo Score – from 0 to 1,000 – would determine how much money an individual, in one of the respective developing markets, could borrow. One of the keys to the company’s success is that they would loan to people who didn’t meet the minimum Lenddo Score threshold with their own microfinance model. Users with low scores would have to invite their friends on the platform, and if the individual with a poor score failed to payback the loan, the Lenddo Score of all their connected friends would lose points off of their Lenddo Score and hurt their ability to borrow money. Similar to the Grameen model, this created a set of social incentives to repay loans collectively. With version 1 of their product, they created the algorithms and tested them on their own loan portfolio to prove that they worked.
“In January 2015, after having proven that the company can underwrite loans successfully without using traditional data, the company started offering its algorithm to third parties for lending and verification purposes . The company has started working with brick and mortar banks, p2p lenders, telecom companies and other financial institutions that are looking to make better and faster decisions for under banked clients” Lenddo – The Google of Lending Algorithms
They demonstrated that their approach had a greater predictive power than traditional measures in developing countries:
“Gini Coefficients are used in the banking world to evaluate the predictive power of credit scoring tools. A Gini Coefficient is merely a scale of predictive power from 0 to 1. A higher Gini means more predictive power, a lower Gini means less predictive power. The country had a sample Gini Coefficient of .25 to .32, whereas the Lenddo model had a Gini of .32 to .39 using non traditional data.” Lenddo – The Google of Lending Algorithms
The company’s model has evolved over the years. Where it once began as a blend between social and mobile data to compute the Lenddo Score, they now rely ONLY on a combination of Android mobile data and email behavioral data. Social media (ie. Facebook) is only for verification. As proof that this model could be the future of credit scoring, the company has recently signed an agreement with FICO to prototype a hybrid score in the Indian market:
“The collaboration aims to combine Lenddo’s dedicated knowledge in the use of non-traditional data for scoring and online verification with FICO’s strength in credit scores.” FICO and Lenddo Join Hands
Could this same science be applied to millennials?
The need is massive. In a global economy where quantitative easing has made fixed assets unaffordable for the average millennial and failed macro policies have made good, high-paying jobs difficult to secure (#sharingeconomy), Millennials are failing massively in the Credit Scoring category (US data).
Clearly new mechanisms need to be designed to determine whether millennials are bankable …
While many in the developed world would consider a service like Lenddo invasive, people in the developed world demonstrated that they would trade some privacy for the ability to access financial services. Given that so many millennials in the developed world have credit scores that technically limit them from accessing financial services at prime lending rates, we believe that an approach like Lenddos could work to help bring millennials up the ladder towards ‘creditworthiness’ and asset ownership.
This will require new trust signals beyond what Lenddo has done, as collecting Android data and email habits will likely rub many millennials in developed economies the wrong way from both a privacy and believability perspective. But there is clearly big space for signals that use non-traditional data to determine creditworthiness such as rent payments, sharing economy platform data, and behavioral questions. We believe that the whole model needs to move towards as Trust as an Asset ™.
Combining new data sources with new social incentive models will give people the ability to buy assets in the future by banking trust in the present.
Next-gen credit scoring is one of the key steps in this journey.