More Complete Picture of Thin File Borrowers Now Available
Rejecting a loan applicant because there is not enough information on their credit file to determine their risk can mean giving away a creditworthy customer.
In today's disrupted marketplace, where customer-centricity and growth are crucial, there is a significant benefit to separating the likely good borrowers from the likely bad borrowers among those with limited credit history.
![](https://www.equifax.com.au/knowledge-hub/sites/default/files/Untitled%20design%20%284%29.png)
Among the approx. 4.5%1 of credit enquiries that are 'new to bureau' (new seekers of credit), there are borrowers who have the potential to honour their loan commitments with on-time repayments. But these new borrowers and other applicants with limited credit records (thin files) may be rejected because there is not enough information to separate them statistically from the pool of unsuitable borrowers.
It's now possible to achieve a better risk separation among thin file applicants by increasing the granularity of assessment using Equifax's latest generation credit score. When compared to our previous credit scores, Equifax One Score leverages advanced analytics, better matching capabilities and broader coverage to identify predictive behaviour patterns. Recent model analysis shows an increase in the score's predictive performance of up to 100%** for the new-to-bureau segment.
For lenders, this substantial uplift in risk prediction can translate to an increase in customer approvals without an associated hike in the rate of fraud or non-performing loans. Making credit available and at a fair price to these borrowers boosts financial inclusion and allows lenders to raise their volume of profitable loans.
Making the most of the data
The volume of data available to inform credit decisions continues to skyrocket. And with these highly diverse, giant consumer data sets comes the opportunity to improve scoring accuracy and better predict repayment behaviour. Equifax One score leverages a full 24 months of comprehensive credit reporting (CCR) data and five years of negative enquiries and defaults.
As the market leader in consumer credit information, Equifax holds the biggest collection of data, including CCR and Buy Now Pay Later (BNPL) data, than any other Australian credit bureau. This unrivalled coverage across the major banks, second and third-tier lenders is of immense value where limited credit history exists about an applicant. An individual who doesn't have an account with a large financial institution, for example, may well have accounts with smaller or non-bank lenders.
Current enquiry data and non-traditional sources like geo-demographic and BNPL are used as supplements to Equifax One Score when limited credit file information exists. Enquiry data can provide a red flag to either fraudulent activity or inability to repay debt. A borrower who applies for numerous mobile phone plan applications with multiple telcos within a short period, for example, may statistically indicate a higher risk.
BNPL data is beneficial for predicting credit risk among young, new to bureau applicants, many of whom are avid users of interest-free payment services. Generation Z, for example, accounts for a quarter of all BNPL enquiries, even though comprising only 5% of the working adult population2.
With the adoption of sophisticated machine learning analytical techniques, it's now possible to reach wider and deeper into these data sets, pulling out more insights than was previously possible. Equifax One Score uses an explainable AI (xAI) machine learning scoring methodology known as NeuroDecision™ Technology (NDT) to handle the complexity of data residing in different sources. NDT generates logical, actionable reason codes tailored to individual consumers. It's a groundbreaking approach that helps both consumers and regulators better understand risk models by providing insight into the 'why' behind the credit decision.
Better matching and linking
No discussion of thin files would be complete without commenting on data matching and linking. An applicant could present as a thin file simply because their identity doesn't seem to match any other current identity on file. But what if an improvement in technology allowed for more accurate identification and linking of customer information regardless of address changes, name changes or name variations. It might enable the discovery of new matching files associated with the applicant that can help round out their credit file.
This step-up in technology is happening right now. Substantial investment in this area has enabled Equifax to improve the accuracy of its automated data matching and record linkage and its scalability to large databases. From disconnected and multi-sourced data, the capability exists to quickly and accurately connect similar and dissimilar records. For example, you can suddenly recognise that the 'new' customer applying for your service isn't really a new customer. Instead, they're a past customer with an outstanding write-off balance that you may now be able to collect.
Uncovering bank transaction data potential
The enriching of traditional credit bureau information with bank transaction data is an exciting development made possible by Open Banking. Equifax is currently building a predictive model prototype based on transaction data, which searches for previously unrecognised positive and negative correlations in income and expense data.
Integration of new consumer-permissioned data assets will enable more granular assessments of affordability and expenditure and offer more predictive and inclusive credit scoring by adding another layer of up-to-date information. Every transaction tells a story and can reveal patterns and signals indicating credit repayment ability. These signals are all indicators that can help divulge more about where a thin file applicant sits on the scale of risk, assisting lenders in making more confident credit decisions.
For consumers, this combined data approach will improve an individual's ability to demonstrate their creditworthiness by enabling information that isn't currently used to be taken into consideration. This approach is designed to improve access and reduce prices for those with thin credit files, increasing financial inclusion.
Discover more potentially good borrowers among your thin file applicants. Speak to your account manager or talk to us to find out more about Equifax One Score.
1Applicable to the banking and finance segment. Approximated %
2 Source: Consumer Credit Demand by Equifax, Sept Qrt 2020
** The New to Bureau segment Gini increased from low 20s to mid 40s and made up circa 4.5% of the through the door population
Related Posts
The ongoing housing crisis remains a hot topic and one of the major concerns for many Australians. Figures from Equifax1 confirm a troubling trend: home ownership in Australia becomes a progressively distant aim for younger people.
![Data cleansing, fixing spelling errors in customer names, removing deceased individuals from datasets, identifying duplicate records or reconciling inconsistencies are key for high quality data.](https://www.equifax.com.au/knowledge-hub/sites/default/files/styles/intro_image/public/field/image/Deathcheck%20BLOG%20article%20image_0.jpg?itok=jRvaE6Rp)
Businesses collect and process vast amounts of data, and often overlook the importance of ensuring the data is accurate, reliable, relevant, complete, consistent and timely. For a data-critical sector like insurance, the process of cleaning data to fix inaccuracies, incomplete, duplicate or wrongly formatted records holds valuable benefits.