Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Cash Flow Underwriting with Bank Transaction Data: Advancing MSME Financial Inclusion in Malaysia

Chun Chet Ng, Wei Zeng Low, Jia Yu Lim, Yin Yin Boon

TL;DR

This paper addresses the financing gap faced by Malaysian MSMEs by moving beyond traditional credit bureau data to bank statement based cash flow underwriting. It presents an end-to-end workflow that ingests bank statements, extracts structured transaction data, and generates features for credit risk scoring, evaluated on a novel dataset of 611 MSME loan applicants. The results show that bank statement derived features substantially enhance predictive power, with logistic regression achieving an AUROC of 0.782 on validation and gains when combining statement and application data. A production ready deployment framework with feature stores, CI/CD, model registry, and Champion-Challenger monitoring is outlined, aiming to enable scalable and auditable credit decisions that improve financial inclusion for new-to-credit MSMEs in Malaysia.

Abstract

Despite accounting for 96.1% of all businesses in Malaysia, access to financing remains one of the most persistent challenges faced by Micro, Small, and Medium Enterprises (MSMEs). Newly established businesses are often excluded from formal credit markets as traditional underwriting approaches rely heavily on credit bureau data. This study investigates the potential of bank statement data as an alternative data source for credit assessment to promote financial inclusion in emerging markets. First, we propose a cash flow-based underwriting pipeline where we utilise bank statement data for end-to-end data extraction and machine learning credit scoring. Second, we introduce a novel dataset of 611 loan applicants from a Malaysian lending institution. Third, we develop and evaluate credit scoring models based on application information and bank transaction-derived features. Empirical results show that the use of such data boosts the performance of all models on our dataset, which can improve credit scoring for new-to-lending MSMEs. Finally, we will release the anonymised bank transaction dataset to facilitate further research on MSME financial inclusion within Malaysia's emerging economy.

Cash Flow Underwriting with Bank Transaction Data: Advancing MSME Financial Inclusion in Malaysia

TL;DR

This paper addresses the financing gap faced by Malaysian MSMEs by moving beyond traditional credit bureau data to bank statement based cash flow underwriting. It presents an end-to-end workflow that ingests bank statements, extracts structured transaction data, and generates features for credit risk scoring, evaluated on a novel dataset of 611 MSME loan applicants. The results show that bank statement derived features substantially enhance predictive power, with logistic regression achieving an AUROC of 0.782 on validation and gains when combining statement and application data. A production ready deployment framework with feature stores, CI/CD, model registry, and Champion-Challenger monitoring is outlined, aiming to enable scalable and auditable credit decisions that improve financial inclusion for new-to-credit MSMEs in Malaysia.

Abstract

Despite accounting for 96.1% of all businesses in Malaysia, access to financing remains one of the most persistent challenges faced by Micro, Small, and Medium Enterprises (MSMEs). Newly established businesses are often excluded from formal credit markets as traditional underwriting approaches rely heavily on credit bureau data. This study investigates the potential of bank statement data as an alternative data source for credit assessment to promote financial inclusion in emerging markets. First, we propose a cash flow-based underwriting pipeline where we utilise bank statement data for end-to-end data extraction and machine learning credit scoring. Second, we introduce a novel dataset of 611 loan applicants from a Malaysian lending institution. Third, we develop and evaluate credit scoring models based on application information and bank transaction-derived features. Empirical results show that the use of such data boosts the performance of all models on our dataset, which can improve credit scoring for new-to-lending MSMEs. Finally, we will release the anonymised bank transaction dataset to facilitate further research on MSME financial inclusion within Malaysia's emerging economy.
Paper Structure (35 sections, 8 equations, 4 figures, 1 table)

This paper contains 35 sections, 8 equations, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Limitations of Traditional Credit Scoring
  4. Credit Scoring with Alternative Data
  5. Machine Learning Models in Credit Scoring
  6. Agentic Workflow in Credit Scoring
  7. Bank Statement Cash Flow Underwriting
  8. Web Layer (Customer Onboarding)
  9. Application Layer (Bank Statement Analyser)
  10. Key Information Extraction Module
  11. Transaction Table Extraction Module
  12. Fraud Analysis Module
  13. Network Analysis Module
  14. Cash Flow Analysis Module
  15. Data & Scoring Layer (Cash Flow Underwriting)
  16. ...and 20 more sections

Figures (4)

  • Figure 1: Proposed end-to-end cash flow underwriting workflow leveraging bank statement transaction data for credit scoring.
  • Figure 2: Overview of AI modules with their objectives and example outputs within the Bank Statement Analyser.
  • Figure 3: ROC curves of all models evaluated on the validation split of the proposed dataset.
  • Figure 4: Results of ablation studies on IV and AUROC performance on the proposed dataset. Best viewed in colour.