Machine learning is transforming lending

In June 2015, the World Economic Forum published The Future of Financial Services - How disruptive innovations are reshaping the way financial services are structured, provisioned and consumed. The paper defined a taxonomy for disruption in financial services. It identified six broad areas and within them, 11 clusters of disruptive innovation.

The six areas are Payments, Market Provisioning, Investment Management, Insurance, Deposits & Lending and Capital Raising. Of the 11 clusters in the report, the combination of Alternative Lending and Shifting Customer Preference clusters are most relevant to today's article. Within this combination, the Banking-as-Platform movement is playing a remarkable intermediary role in matching lenders and borrowers via a seamless, transparent, user-friendly standard API based model which leverages advances in processing power, artificial intelligence, user interfaces and design-thinking.

Traditional lenders, i.e. banks, usually find it difficult to match the agility and granular understanding that P2P Lending players have. Banks' credit evaluation and origination platforms are often industrial-age processes with a large quantum of human intervention. The operational and statistical models are based on predictability of borrowers' behaviour across a large enough population set. For example, there are historical models that predict the likelihood of default of male white-collar workers who are married with one or more children and have been at their job for 18 months or more. However, there is very little statistically significant differentiation at more granular levels. For example, is there a way to predict if a person is credit-worthy based purely on their social media profile?

Also, while most practitioners in consumer and business lending in banks fully appreciate the rapidity with which change is happening, legacy investments in technology and the historical "bank-hall" mindset make transformation a challenge.

Similarly, senior credit analysts and risk officers usually pride themselves on their mathematical prowess and instinct. However, like experienced pilots in an era of self-flying aircraft, could these skill-sets require a change of mindset and orientation?

Jinand Shah, founder promoter of CapitaWorld (www.capitaworld.com), understands this really well. Shah is actively looking at the UAE as a beachhead global market for bringing in CapitaWorld's lending technology platform. The platform has proven itself in the tough and competitive India market. Shah is working closely with the Abu Dhabi Global Market's (ADGM) Reglab to comply with the requirements in the UAE.

First a little more about the UAE itself. At the end of 2016, loans outstanding net of provisions were Dh1574.0 billion (www.centralbank.ae/en/pdf/dataroom/UAEMonthlyBankingIndicators_Dec16_En.pdf). Emirates NBD continues to lead with close to 20 per cent market share (from the UAE Banking Market Pulse 2016 report by Alvarez and Marsal). Based on the quarterly reports of 10 leading banks in the UAE, the Market Pulse 2016 study indicates overall worsening operational efficiency and thereby a downward pressure on profitability.

Separately the highly readable and informative Q1 2017 Credit Sentiment Survey of the Central Bank of the UAE reflects generally more positivity in both the business and consumer segments i.e. there is more demand for loans now than a year before and this looks set to continue.

Shah sees this as an excellent opportunity for banks in the UAE to tap the growth in demand without the downward pressure of lower operating efficiency. CapitaWorld's platform achieves this in three distinct areas. These are the front-end, the operational work-flow and credit decisioning itself.

The front-end is a platform that hosts one common online form which in turn becomes the source of several linkages to biometric identification databases, electronic KYC and FATF checks, autofill data from other platforms, credit bureau checks, National ID checks, social media scoring as well as a proprietary SME scoring module. The front-end can either be an independent market-place or a white-label platform for a bank to do its own branding and customisation. In its market-place avatar, the CapitaWorld platform potentially becomes a vastly superior information source than run-of-the-mill financial comparison websites which today are more form than function.  

The front-end provides APIs for connectivity to the banks' own operational processes. This is where CapitaWorld's operational efficiency model also claims strengths. The fully digital form with inbuilt validated information creates efficiency through reduction in human-resource intensive processes. The queue time reduces from weeks to hours. For banks competing with each other as well as with startups, this is a critical differentiator.

In addition, the platform enables video-conferencing with prospective borrowers. The video-conferencing has the capability to use artificial intelligence to interpret borrowers' emotions and body-language. It can also measure biometric markers that could help in the decision process.

And finally the credit decision process itself. The model is based on machine learning. Prior decisions and rules as well as portfolio performance are captured by the platform. The vastly superior computing power today enables multiple hypotheses building and analysis. This in turn sets up new decision outcomes. What this also does is that pricing and risk decisions can be taken on much smaller sets of customers and even at an individual level. It is a step away from a standard Annualized Percentage Rate model. Imagine if your credit card interest rate was specific to you, based on your past behaviour. If you are a non-borrower, you may very well be open to borrow on a really low interest rate if there is an opportunity to start that little coffee roastery that you have always wanted to.

Shah is a 33-year-old US qualified CFA and an India-qualified chartered accountant. He comes from a family of finance professionals who are fully involved with and supportive of this venture.

The writer is a Partner at BridgeDFS, a bespoke financial advisory firm (www.bridgeto.us). He's a digital banking and digital banking financial services evangelist, practitioner, advisor and consultant. Views expressed are his own and do not reflect the newspaper's policy. He can be contacted at ves@vyashara.com.

Source: .khaleej Times