Traditional credit scoring models have been the go-to method for lenders to assess risk and make lending decisions. However, these models have some major limitations, especially when it comes to serving thin-file consumers with little to no credit history.
In the Philippines, only around 20-30% of the adult population have sufficient credit information to generate a credit score. This results in a massive unserved segment that traditional models fail to evaluate properly.
To expand access to credit and make better risk decisions, lenders need to look beyond traditional scoring and adopt alternative data sources. In this blog, we explore the top 5 alternative credit scoring models that leading lenders are using today.
Limitations of Traditional Credit Scoring
Traditional credit scoring relies solely on historical repayment behavior and credit bureau data to generate a credit score. This works well for consumers with a lengthy credit history. But those without existing loans or credit cards fall into a "thin file" segment that cannot be assessed properly.
Some key limitations of traditional credit scoring models include:
Need for historical credit usage - not feasible for first time borrowers
Narrow focus on credit bureau data
No insight into true income or payment capacity
High rejection rates for thin file consumers
Lag in score updates - only reflects historical behavior
These drawbacks mean traditional scores fail to provide a complete risk profile, especially for thin file consumers like the unbanked, gig economy workers, and those without extensive bureaucracy records.
The Rise of Alternative Credit Scoring
To address these gaps, lenders are adopting alternative data sources that can evaluate a consumer's creditworthiness even without traditional credit history.
Some fast emerging alternative credit scoring models include:
Payroll Data Underwriting
Payroll details directly reflect a person's income streams and payment capacity. Earned Wage Access providers like Savii allow lenders to verify employment and income data via API for thin file applicants.
Using payroll data enables more accurate assessment of debt-to-income ratios. It also provides stability indicators like length of employment and frequency of pay.
Telco, Utility Payment Tracking
Timely utility and bill payments demonstrate financial responsibility even without formal credit history.
Finscore, as an example, collects such alternative data and provide insights into telco usage and payment patterns not found in credit reports. It enables lending to consumers with limited bureau footprints.
Bank Account Activity Analysis
Analyzing income streams and spending patterns from bank account data provides a detailed view of an applicant's cash flows.
This allows more accurate predictions of repayment ability, especially for gig economy and self-employed segments with variable incomes. An example of a company doing this is Brankas.
Employment History Verification
Longer tenures and more stable employment indicate lower income volatility risks. Verifying employment history through APIs like Smile provides income stability estimates for thin file consumers.
Social Media and Online Footprint Scoring
Alternative scores from startups like Lenddo analyze social connections and online behavior to derive creditworthiness predictors. These can evaluate first time borrowers with minimal formal credit history.
The growth of alternative data sources allows lenders to look beyond just historical repayment behavior for credit decisions. This is critical for expanding access to credit and driving financial inclusion.
Employment History Indicates Creditworthiness
Among alternative scoring models, employment history and income verification are emerging as the strongest indicators of creditworthiness. They provide direct signals on a borrower's ability to repay loans.
For example, longer employment tenures demonstrate income stability. Multiple jobs may indicate higher risk. Higher salaries and more frequent pay cycles suggest stronger payment capacity.
Employment details like these allow lenders to assess affordability and default likelihood even without traditional credit information.
Income Verification Complements Traditional Data
Using employment data and income verification as part of credit scoring provides more holistic risk assessment. Such alternative data complements traditional credit information.
For instance, open API solutions like Smile allow instant access to verified payroll and employment details. This can supplement bureau data for well-rounded evaluation of both thin file and existing borrowers.
Alternative scoring models are not meant to fully replace traditional credit information. Rather, the two can be combined to derive better insights and expand lenders' addressable market.
Adopting Alternative Credit Scoring Models
Here are a few tips for lenders looking to implement alternative credit scoring:
Start by understanding gaps not addressed by current systems
Identify additional data sources that can fill these gaps
Evaluate different scoring models before selecting one
Start with a hybrid approach combining alternative and traditional data
Test effectiveness before fully rolling out new models
Continuously improve model performance with machine learning
Alternative data is enabling smarter credit risk assessments, especially for previously unserved segments. The fintech ecosystem offers multiple options for lenders to get started.
With the right strategy, lending firms can harness these new models to make more accurate and inclusive decisions. The future definitely lies in looking beyond traditional scoring methods. Partnering with alternative data providers will be key to success.