What Happens at the Convergence of Machine Intelligence & Online Lending

Credit scoring and approval rates changed substantially with the arrival of alternative lenders, mainly due to the adoption of new practices in collecting and analyzing potential borrower data. Alternative data has played its role in expanding horizons for financial institutions and for creating an opportunity to enter the financial sector fir technology startups and data-rich international companies.

While social media, for example, as a source of data for creditworthiness assessment is still at a nascent stage, certain startups are already claiming to have incorporated information from social networks into their frameworks. In the quest to reinvent the way to assess consumer-related risk (as well as extend credit to unscored and questionable), startups were found more imaginative than traditional institutions.

Alternative data requires alternative approach to data analytics, which wide adoption of machine learning and artificial intelligence brought.

The use of alternative data sources, big data and machine learning technology, and other new artificial intelligence models could reduce the cost of making credit decisions and/or credit monitoring and lower operating costs for lenders. FinTech lenders could pass on the benefits of lower lending costs to their borrowers. – Fintech Lending: Financial Inclusion, Risk Pricing, and Alternative Information

LenddoEFL, for example, the company that enables financial institutions to do predictive analytics to service new client segments, in addition to collecting applicant data from traditional sources (credit bureau data and financial transactions, if available), also collects non-traditional data from such sources as psychometric tests, telecoms, browser history, mobile data (i.e., geolocation), social networks, and e-commerce transactions. HBS shares that using advanced machine learning techniques, LenddoEFL’s predictive algorithms look at 12,000 variables and come up with a credit score, called LenddoScore, which ranges from 0 to 1000, with higher scores representing a lower likelihood of default. According to HBS, financial institutions using LenddoScore have, on average, increased their approval rates by 50% and decreased their default rates by 12%.

There is also an interesting example of Upstart, the P2P online lender using machine learning algorithms to make underwriting decisions. Machine learning allows the company to determine if applicants are telling the truth about their income by looking through their employment history and comparing their data with that of similar clients. It can also find hidden patterns that might favor an applicant. The company managed to automate 25% of its less risky loans, allowing lenders to save resources. The technology is planned to be available to banks, credit unions and even retailers that are interested in providing low-risk loans to their customers, CoinTelegraph shares. As of December 2017, 47% of all Upstart loans were fully automated.

Avant, founded in late 2012, has raised over $600 million of equity capital and has provided access to over $4 billion of loans on the platform. But that’s not the most interesting part. In July 2016, Avant turned its attention to middle-class borrowers with AvantCredit. Using big data and machine learning algorithms, AvantCredit analyzes thousands of trends and patterns in lieu of the traditional one-size-fits-all credit model to offer competitive rates on an individual basis. The company’s technology takes into account many more variables in order to better suit each borrower’s needs rather than looking at an individual as just a credit score.

ZestFinance is another interesting example. In February 2017, the company announced the Zest Automated Machine Learning (ZAML™) Platform for credit underwriting. The platform uses Google-like algorithms to analyze tens of thousands of data points to provide a richer, more accurate understanding of all potential borrowers. ZAML enables lenders to analyze vast amounts of non-traditional credit data to increase approval rates and reduce the risk of credit decisions, particularly for thin-file and no-file borrowers like millennials. The platform also provides the ability to explain data modeling results to measure business impact and comply with regulatory requirements.

Traditional underwriting works well when evaluating borrowers with long credit histories, but when there is limited data, it can’t differentiate between creditworthy and high-risk applicants. Machine learning fills those gaps by analyzing a vastly broader set of data. – Douglas Merrill, Founder & CEO of ZestFinance

What’s interesting, the newly launched platform was fine-tuned based on the experience Zest had working with the search engine Baidu in China, where only 20% of the population has any known credit history, MIT Review reports.

Studying 21 different factors such as how people search and the way they traverse between Web pages, Zest discovered patterns in Baidu’s data that could be used to decide whether to make small loans to those customers for purchases like clothing. Among the things Zest evaluated was how well a person’s self-reported income matched up against their modeled income, what Zest calculates that person actually earned based on other behavior. Just as important as how much discrepancy there is between reported and modeled income is when they report the inflated income (in other words, income that’s higher than what the model implies they’re actually making) and how much they inflated it. – Nanette Byrnes, An AI-Fueled Credit Formula Might Help You Get a Loan, MIT Review

In two months, the edition shares, Baidu, which has a small lending business, was approving 150% more borrowers with no increased losses on their loans, and the company has made hundreds of thousands of loans since.

Artificial Intelligence is the new electricity. It’s transforming industry after industry, and financial services are particularly ripe for innovation. – Andrew Ng, Chief Scientist at Baidu