Table of Contents
Trade-based money laundering is one of the most persistent and least visible threats to the global financial system. Criminals manipulate international trade transactions to move value across borders, using the complexity and volume of global commerce as cover. By some estimates, TBML accounts for an estimated 87% of global illicit financial flows, and over 51% of compliance leaders view TBML as a top threat to their institutions.
This article examines how inconsistencies across invoices, contracts, shipping records, counterparties and payment flows can become meaningful trade-based money laundering indicators-and how compliance teams can move from checking individual documents to building a connected, evidence-based detection methodology.
What is trade-based money laundering (TBML) and why invoices matter first
Trade-based money laundering is the process by which criminal organisations disguise the proceeds of crime through legitimate trade transactions. Unlike other money laundering techniques that rely primarily on the financial system or cash couriers for physical movement of funds, TBML exploits the international trade system itself.
We have covered the problems related to TBML in Asia and the Middle East. This article goes in-depth regarding shadow exports and invoicing problems.
From single document checks to cross-document TBML detection
Many financial institutions still review trade documentation one document at a time-checking a scanned invoice for completeness, verifying a letter of credit against its terms. This approach catches obvious errors but routinely misses the relationship-level anomalies that characterise TBML schemes.
A data-driven approach works differently. Instead of examining each PDF in isolation, structured data is extracted from invoices, packing lists, bills of lading and payment messages, normalised across fields (product codes, quantities, currencies, dates, parties involved) and correlated across multiple documents and systems. This enables detection of patterns that no single document review would reveal.
The difference between document-level issues (a typo in an invoice date, a unit-of-measure mismatch) and relationship-level anomalies (recurring mismatches between invoices, bills of lading, SWIFT payment messages and customs declarations over months) is critical.
Financial institutions often lack visibility into full trade transactions because each department sees only its own slice-trade operations processes the documents, payments teams clear the wire, compliance screens the names. TBML detection depends on seeing the same transaction from different angles simultaneously.
Consider a case pattern resembling the publicly discussed Orion Global Trading scenario: an electronics exporter consistently invoices shipments at 35–40% above market value; packing list weights diverge from bill of lading declarations; the buyer is a newly incorporated entity with no visible business operations; payments arrive from unrelated third parties. Each of these, reviewed alone, might be dismissed. Together, they represent a textbook TBML profile-one that only cross-document analysis and transaction monitoring would surface.
Key TBML document sets: what should be consistent?
Global trade transactions involve multiple parties, complicating oversight-but also creating multiple data points that should logically align. The core document families in international trade include commercial contracts, pro forma and commercial invoices, packing lists, bills of lading or airway bills, certificates of origin, quality and insurance, customs declarations, and financial messages such as MT7xx and MT1xx series.
What should match across these documents:
Invoice vs. contract: price per unit, incoterm, currency, payment method, parties
Invoice vs. packing list vs. bill of lading: quantities, net and gross weights, packaging descriptions
Invoice vs. customs declaration: declared value, HS codes, country of origin, quantity
Payment records vs. invoice: payer identity (should match buyer or consignee), amount, timing relative to agreed terms
Common weak points exploited in trade-based money laundering TBML include differences between seller and shipper names (often obscured by intermediaries or shell companies), vague product descriptions that don’t align across documents (e.g., “spare parts” on the invoice, a specific HS code on customs forms), and unexplained amendments or “corrected” invoices that replace original documents with different values.
Falsely describing goods misrepresents their type or quality on documents and is a frequently observed tactic. Much of that loss traces back to inconsistent or falsified trade documents. Forged documents are one possibility, but inconsistent, incomplete or low-quality data is equally common in legitimate international trade-which is precisely why context, not just document verification, is essential.
Invoice, price and quantity inconsistencies as TBML signals
Invoice fraud detection in the AML context is less about asking “is this invoice fake?” and more about asking “does this invoice make commercial sense compared to other available data?” Discrepancies between invoice values and market prices are red flags, but they must be evaluated against product type, route, seasonality and incoterm to separate criminal activity from normal commercial variation.
Key inconsistency types to look for:
Unit price significantly above or below market value for the relevant HS code and date
Large unexplained discounts or surcharges not matching sector norms
Quantities on invoices vs. bills of lading vs. packing lists that don’t reconcile
Repeated use of round numbers or identical values across multiple shipments, suggesting fabricated transaction documents
The classic trade-based money laundering typologies include:
Typology | Mechanism | Objective |
|---|---|---|
Over invoicing | Invoicing above fair market value | Export excess funds to justify illicit payments |
Under invoicing | Billing goods below their actual market value | Transfer value offshore or enable capital flight |
Multiple invoicing | Issuing more than one invoice for a single transaction | Generate additional payment justifications for the same transaction |
The Black Market Peso Exchange model is a common TBML technique that combines trade misinvoicing with currency exchange to repatriate drug proceeds.
Price variance thresholds of 15–25% above or below trade benchmarks are useful screening tools, as referenced in FATF–Egmont indicators. However, legitimate commercial rationales exist: distressed sales, related-party transfer pricing, promotional discounts.
Product description and classification mismatches
Trade-based money laundering relies on ambiguity. Vague or shifting product descriptions can hide high-value goods, dual-use items or even phantom shipments-documentation for goods that are never shipped.
Key red flags around product descriptions include invoices using generic labels (“spare parts,” “electronics,” “foodstuff”) while customs forms list precise HS codes, or vice versa; changes in HS codes for similar counterparties and routes without a clear business reason; and declared quality or grade not aligning with price level. High-risk products like scrap metal are red flags-for example, “scrap copper” priced at the same level as refined copper cathodes should trigger immediate scrutiny.
FATF–Egmont indicators specifically identify inconsistent or inaccurate HS classification as a common TBML indicator, especially in commodities, precious metals and pharmaceuticals. Physical plausibility checks add another layer: does the declared weight match the volume? Would the listed quantities realistically fit into the declared containers? A 40-foot container allegedly holding thousands of heavy machinery units is implausible on its face.
Shipping, routing and logistics discrepancies
Unusual shipping routes or destinations may signal TBML, and mismatched trade routes can indicate TBML activity even when individual documents appear authentic.
Key inconsistencies between documents and trade context include:
Invoice origin or destination vs. bill of lading ports and transhipment hubs that do not align
Implausible or uneconomic routes-routing regional trade through distant free trade zones without commercial logic
Shipping dates that conflict with production capacity, contract dates or seasonal factors
Even authentic bills of lading can support sham trades if they reference empty containers, low-value cargo or recycled document numbers. Ghost or duplicate shipments-involving the re-use of container IDs, bills of lading numbers or warehouse receipts-are a well-documented tactic. Misalignment between incoterms in the contract and where risk or payment transfers actually occur in SWIFT or payment records is another signal.
TBML often involves false shipments and mismatched documentation, and some high-risk patterns highlighted by FATF and national regulators (e.g., repeated stopovers in certain free trade zones) should be treated as higher-priority alerts.
Counterparties, beneficial ownership and commercial profile anomalies
Misaligned counterparty information is a core TBML signal. Who is buying, who is selling, who ships and who pays should form a coherent commercial story. When they don’t, the trade counterparties warrant closer examination.
Key anomaly types include importers or exporters whose stated business profile doesn’t match the traded goods (a small consulting firm exporting bulk agricultural commodities), frequent changes in trading partners in high-risk sectors or corridors, and repeated use of newly incorporated or thinly capitalised entities with no visible trading history. Shell companies are frequently used as intermediaries to obscure beneficial ownership and enable artificial pricing between related parties involved in round-tripping and layering of illicit funds.
Cross-checking against sanctions lists, adverse media and previous internal risk assessments is essential-tools like AML screening of companies and counterparties and entity risk assessment platforms support this workflow. Machine learning builds networks of customer trading relationships, helping investigators identify connections that manual reviews would miss.
When commercial capacity doesn’t match transaction scale-a micro-business booking multi-million-dollar high-value transactions without infrastructure, warehouses or employees-enhanced due diligence and potentially on-site verification should follow.
Payment behaviour and financial-flow inconsistencies
Many TBML schemes are only visible when trade documents are reconciled with actual payment flows observed through AML transaction monitoring systems. The payment side of a trade transaction often reveals what documents alone cannot.
Specific payment red flags linked to TBML include:
Third-party payments not mentioned in contracts, especially from unrelated countries or sectors and from unrelated third parties
Split payments for a single shipment routed through multiple banks or accounts
Early or very late payments compared to standard terms without reasonable justification
Frequent partial prepayments followed by cancelled or heavily amended shipments
Detecting circular fund flows-repeated round-trip transfers among related entities-is particularly relevant here.
Divergent currencies, inconsistent use of letters of credit versus open account, or sharp shifts from formal trade finance to unsecured transfers can be material TBML detection clues.
From discrepancies to investigative red flags: building a TBML risk picture
The concept of “signal stacking” is central to effective TBML investigation. Multiple, independently suspicious inconsistencies-each potentially explainable in isolation-convert a routine alert into a high-priority case when they converge.
Scenario 1: An exporter of used vehicles consistently invoices at 30% above market value. Payments arrive from a different jurisdiction via an unrelated third party. Shipments route through a free trade zone with no commercial reason. The exporter was incorporated six months ago with minimal capital.
Scenario 2: A textile importer’s invoices and bills of lading misalign on quantities across multiple shipments. Related shell companies appear as intermediaries. Payments are split below reporting thresholds across multiple lenders and accounts.
Investigators should verify data quality first-eliminating simple clerical errors, currency conversion issues and unit-of-measure misalignments. Next, check internal and external reference data: trade benchmarks, company registries, adverse media via tools like adverse media screening. Finally, assess whether a plausible commercial explanation exists that is consistent across documents and behaviour.
Discrepancies are indicators for further review-they do not themselves prove trade-based money laundering. TBML is difficult to detect due to high volumes and complexities of trade, and investigations should be documented, proportionate and risk-based. Semantic search analyzes unstructured trade documents for TBML detection, helping analysts surface relevant precedents.
Operationalising TBML detection: data, tooling and workflows
Practical steps to strengthen detection:
Digitise and structure trade documents using OCR and data normalisation
Link trade records with payment transactions and customer profiles using consistent identifiers (entity IDs, tax IDs, LEIs)
Implement trade data analytics to compare invoice values against market or customs datasets
Calibrate TBML-specific rules and scenarios in AML transaction monitoring systems
RegTech and financial-crime intelligence tools can help close these gaps. ZIGRAM’s ecosystem addresses several dimensions: PreScreening.io and Entity Hero support entity risk assessment and counterparty profile checks; Transact Comply enables AML transaction monitoring that incorporates trade-context scenarios; Due Diliger provides due diligence reports for risk assessment of opaque or high-risk entities.
Governance, regulatory expectations and information sharing
Global standards set clear expectations. FATF’s 40 Recommendations require monitoring for TBML as part of broader anti-money laundering obligations. The Bank Secrecy Act, the Sixth Anti-Money Laundering Directive, and the Wolfsberg Group’s Trade Finance Principles provide operational standards and mandate TBML SAR filings for banks providing significant trade finance services.
What is generally required versus recommended practice:
Mandatory customer due diligence and ongoing monitoring for trade finance transactions and relevant correspondent relationships
Risk-based escalation and SAR/STR reporting when trade-based money laundering is suspected
Use of TBML typology and red-flag guidance from FATF, Egmont Group and national regulators such as HMRC’s TBML Handbook or FinCEN
Governance elements include a clearly defined TBML risk appetite and policies, documented procedures for investigating invoice discrepancies and forged documents, and regular training for trade finance, operations and AML compliance teams on TBML typologies and data inconsistencies. The World Bank estimates $1.5 trillion is paid in bribes annually-a figure that underscores the scale of illicit financial flows that legitimate businesses and financial intelligence units must contend with.
Conclusion: connecting documents, data and behaviour to protect legitimate trade
Data inconsistencies are investigative signposts. Financial institutions must apply judgement, understand commercial context and distinguish genuine operational errors from deliberate manipulation of trade flows. Suspicious transactions are starting points, not endpoints. The quality of the investigation that follows-documented, contextual, proportionate-determines whether controls actually protect both the financial system and legitimate global trade.
Strengthening TBML controls reduces compliance risks and de-risking pressures, allowing more confident engagement with higher-risk markets and trade corridors. Operationalise cross-document detection through integrated AML solutions. ZIGRAM’s financial-crime intelligence ecosystem offers entity screening, transaction monitoring, due diligence and adverse media capabilities designed for exactly this challenge. Reach out to schedule a discovery call if you are evaluating how to connect these workflows in your institution.
Combating trade-based money laundering is not a one-time project. It requires continuous learning, calibration and collaboration-across teams, across systems and across borders.
