In modern commercial lending and anti-money laundering (AML) compliance, detecting artificial revenue inflation and fund diversion is paramount. Borrowers or commercial syndicates occasionally attempt to inflate banking turnover or disguise NPA stress by cycling funds between interconnected entities—a practice known as circular transactions or round-tripping.
With modern algorithmic underwriting platforms like CreditCore Ten processing raw banking telemetry, underwriters can instantly pinpoint transaction looping, layering, and fictitious counterparty flows. This guide explains how banking forensics systems identify, trace, and categorize circular fund movements.
1. What Constitutes a Circular Transaction?
A circular transaction occurs when funds leave Account A, travel through intermediate Accounts B and C, and return to Account A (or a sister company) within an abnormally compressed timeframe with little to no underlying economic value creation or genuine trade exchange.
While on paper the bank statements show cumulative turnover of nearly ₹40 Lakhs across the accounts, the net economic liquidity injected into the business was precisely zero.
2. Core Forensic Red Flags Audited by Underwriters
Same-Day In-and-Out Velocity
Large funds credited and debited within minutes or hours without resting in the account to support payroll, vendor payments, or inventory purchases.
High-Value Round-Figure Transfers
Frequent transfers of exact round sums (e.g., ₹5,00,000, ₹15,00,000, ₹25,00,000) that contrast sharply with legitimate operational invoices containing GST and fractions.
Month-End Balance Window Dressing
Spikes in account deposits on the 29th, 30th, or 31st of the month, followed by immediate withdrawals on the 1st or 2nd of the next month, designed solely to inflate Average Monthly Balance (AMB).
Common Directorship & Related Parties
Transactions conducted between companies having identical registered office addresses, common DINs, shared mobile numbers, or cross-holding shareholder structures.
3. Graph Neural Networks and Cycle Detection in CreditCore Ten
Modern credit engines employ directed graph analysis algorithms (such as Tarjan's strongly connected components and cycle detection) to analyze multi-bank statement batches:
- Node Representation: Every unique bank account / IFSC / VPA is assigned a graph node.
- Edge Weights: Transaction volume, timestamp differentials, and frequency form the directed weighted edges.
- Closed Loop Identification: Closed loops that return to origin within a threshold window (e.g., ≤ 48 hours) are automatically highlighted with an anomaly risk score.
4. Consequences of Circular Transaction Flags
When a borrower's banking exhibits circular transaction behavior, underwriting engines enforce automatic defensive actions:
- Haircut on Banking Turnover: All detected circular volumes are mathematically subtracted from total business turnover, resizing loan eligibility strictly to genuine operational cashflows.
- Mandatory CA Certificate: The file is referred for specialized forensic auditing with mandatory submission of audited 3CD tax reports and vendor invoices.
- Fraud Registry Warning: Severe, repeated circular routing can lead to categorization under Section 138 / PMLA internal bank caution lists.
Summary
Transparent accounting and legitimate operational banking are essential to maintain institutional borrowing capacity. Automated forensic technology ensures that only genuine cash flows qualify for commercial credit sanctions.