Financial Crime, Sanctions and Trade-Based Money Laundering

Introduction and strategic context

International trade finance connects banks, exporters, importers, insurers, carriers, ports, customs authorities and payment systems. This network creates economic value but also creates opportunities to conceal the proceeds of crime, evade sanctions, finance proliferation and misrepresent the movement or value of goods. Trade-based money laundering (TBML) is particularly difficult because the financial institution may see only documents and payment instructions, while the underlying goods move through several jurisdictions and intermediaries.

A modern banking control framework must therefore combine customer due diligence, transaction monitoring, sanctions screening, export-control awareness, document analysis, supply-chain intelligence and professional judgment. No single alert, database or automated model is sufficient. Effective control depends on understanding the commercial purpose of a transaction and testing whether the parties, goods, routes, prices and payment terms are coherent.

The strategic challenge is to manage two risks simultaneously: under-control, which exposes the institution to legal, financial and reputational damage; and over-control, which causes unnecessary delays, de-risking and exclusion of legitimate customers. Proportionate, risk-based controls are therefore central.

Learning objectives

By the end of this chapter, the reader should be able to:

1. Explain the relationship between money laundering, terrorist financing, proliferation financing, sanctions evasion and TBML.
2. Design a risk-based KYC and customer-due-diligence process for trade-finance customers.
3. Distinguish sanctions screening from export-control screening and anti-boycott controls.
4. Identify red flags involving customers, counterparties, goods, vessels, documents, routes and payment patterns.
5. Analyse dual-use goods and vessel-related risks without treating every high-risk country or commodity as automatically prohibited.
6. Structure an investigation, escalation and suspicious-transaction reporting decision.
7. Evaluate the role and limitations of technology, data analytics and artificial intelligence.

1. The financial-crime landscape in international banking

1.1 Money laundering and related threats

Money laundering is commonly analysed as a process involving placement, layering and integration. Placement introduces criminal proceeds into the financial system; layering separates proceeds from their origin through transactions and structures; integration makes the wealth appear legitimate. Trade can support all three stages. For example, illicit funds may be used to pay for goods, transferred through apparently legitimate invoices and ultimately realised through resale proceeds.

Terrorist financing differs in an important respect: the funds may originate from lawful or unlawful sources, and the amounts may be small. The central issue is the intended use of funds rather than their criminal origin. Proliferation financing concerns the provision of funds or financial services connected with the development or acquisition of weapons of mass destruction and related delivery systems.

Sanctions are government or multilateral restrictions imposed for foreign-policy, security or legal purposes. They may target countries, governments, entities, vessels, aircraft, individuals, sectors, goods or types of activity. A sanctions breach can occur even where there is no money-laundering purpose.

Export controls regulate the transfer of specified goods, software, technology and services. They can apply to physical exports, re-exports, transfers of technology and, in some systems, assistance or brokering. A product need not be labelled as military equipment to be controlled. Civilian items with potential military, nuclear, chemical, biological or surveillance applications are often described as dual-use goods.

1.2 Why trade is vulnerable

Trade transactions contain multiple parties, documents and jurisdictions. This creates five structural vulnerabilities:

• Information fragmentation: the bank may know the applicant and beneficiary but not the ultimate buyer or end-user.
• Document dependence: banks commonly examine documents rather than physically inspecting goods.
• Valuation opacity: prices vary with quality, financing, freight, insurance, market conditions and related-party relationships.
• Jurisdictional complexity: a transaction may involve manufacture in one state, trans-shipment in another, payment through a third and final delivery elsewhere.
• Time pressure: letters of credit, vessel cut-offs and customs deadlines encourage rapid processing and can weaken challenge processes.

TBML is not limited to trade finance products. It can involve open-account payments, documentary collections, remittances, foreign exchange, guarantees, supply-chain finance and correspondent banking.

2. A risk-based control architecture

A defensible programme should be built as a chain of connected controls rather than as isolated checks.

Figure: A connected, risk-based control architecture for trade finance.

2.1 Customer due diligence

Customer due diligence (CDD) establishes who the customer is, what the customer does and why the requested service is commercially reasonable. Enhanced due diligence (EDD) is appropriate where risk is higher, such as complex ownership, politically exposed persons, high-risk jurisdictions, opaque intermediaries, cash-intensive activity, unusual trade corridors or sensitive goods.

Core CDD elements include:

• legal name, registration, address and tax information;
• nature, ownership and control of the business;
• identification and verification of beneficial owners;
• expected account activity, products, currencies and trade corridors;
• source of funds and, where appropriate, source of wealth;
• management, directors, authorised signatories and major counterparties;
• licences, customs registrations and evidence of operational capacity.

A beneficial owner is the natural person who ultimately owns or controls a legal entity or for whom a transaction is conducted. Ownership percentages are useful but not always decisive; control can arise through voting rights, agreements, board influence or other means.

CDD is not a one-time event. It must be refreshed when ownership, business activity, risk indicators or relevant law changes. A bank should distinguish documentary completeness from substantive credibility: a customer can provide every requested document and still present a materially implausible business model.

2.2 Customer risk assessment

Risk assessment should consider inherent risk, control quality and residual risk. A conceptual representation is:

Here, inherent risk represents exposure before controls, control effectiveness is expressed as a proportion between 0 and 1, and residual risk is the remaining exposure. This is a management model, not a universal regulatory formula. It assumes that the factors are measured consistently and that multiplication is a reasonable approximation; institutions should not present the result as a precise probability of crime.

Relevant dimensions include customer, product, geography, channel, transaction, goods, counterparties and delivery route. Risk scoring should support judgment, not replace it. A low score must not override a decisive sanctions match or credible evidence of fraud.

3. Sanctions, export controls and anti-boycott controls

3.1 Sanctions screening

Sanctions screening should cover, as applicable, the customer, beneficial owners, directors, authorised parties, applicant, beneficiary, advising and confirming banks, issuing bank, carriers, vessel owners, charterers, consignees, end-users, ports and other parties in trade documents.

Screening requires more than exact-name matching. Effective systems address transliteration, aliases, spelling variation, name order, non-Latin scripts, dates of birth, addresses, registration numbers and vessel identifiers such as IMO numbers. A potential match must be investigated against reliable identifiers. False positives should be resolved and documented; true matches should be escalated under the institution’s legal and operational framework.

Sanctions regimes differ. A bank must determine which laws apply through its place of incorporation, branches, payment currency, correspondent relationships, employees, transaction location and contractual commitments. A prohibition may be asset-based, activity-based, sectoral, geographic or comprehensive. Licensing and exceptions must be interpreted by authorised legal or compliance specialists, not assumed from commercial necessity.

3.2 Export-control screening

Export-control analysis asks questions that are related to, but distinct from, sanctions screening:

1. What exactly is the item, software, technology or service?
2. What is its technical specification and classification?
3. Where did it originate, and from which jurisdiction is it exported or re-exported?
4. Who is the end-user and what is the stated end-use?
5. Is a licence required, and are there restrictions on the destination or end-user?

A harmonised system (HS) code may assist customs classification but does not necessarily determine strategic export-control classification. Technical parameters, functionality and intended use may be decisive. Banks are not normally engineering laboratories, but they should identify when specialist review, an export licence or customer evidence is necessary.

3.3 Anti-boycott controls

Anti-boycott rules prohibit or restrict participation in certain foreign economic boycotts that are not supported by the bank’s home government. A problematic instruction may request that a party certify it is not on a particular nationality list, that goods were not manufactured in a specified country, or that a prohibited discriminatory condition be inserted into a trade document.

Controls should examine letters of credit, purchase contracts, shipping instructions, certificates of origin and customer communications. Staff should not alter or conceal boycott language merely to process a transaction. The matter should be escalated to legal and compliance personnel, with reporting assessed where required.

4. TBML typologies and detection logic

TBML involves disguising the movement, source, ownership or value of illicit funds through trade transactions. Common typologies include:

• Over-invoicing: goods are priced above reasonable market value, allowing excess value to move to the seller or related party.
• Under-invoicing: goods are priced below value, transferring value to the buyer or allowing later resale at a profit.
• Multiple invoicing: the same shipment supports several payment claims.
• Over-shipment or under-shipment: the quantity differs materially from the documents.
• Phantom shipments: payment is made for goods that do not exist or were never dispatched.
• Misdescription: goods are described inaccurately to obscure controlled, high-value or restricted items.
• Carousel or round-tripping trade: goods pass through related entities or jurisdictions with little commercial purpose.
• Third-party payment: an unrelated party pays or receives funds without a credible explanation.
• Trade diversion: goods are routed through a trans-shipment location to conceal the true destination or end-user.
• Compensation or netting arrangements: obligations are offset outside transparent banking channels.

4.1 Red-flag framework

Red flags are indicators for inquiry, not proof of misconduct. They should be evaluated in context and combined with customer knowledge.

4.2 Vessel and maritime risk

Vessel risk analysis should use stable identifiers, especially the IMO number, rather than relying only on vessel name. Relevant information includes ownership and management, flag, classification society, port calls, voyage history, cargo, charterer, insurance and beneficial control.

Indicators requiring enhanced review may include unexplained automatic identification system (AIS) gaps, ship-to-ship transfers in sensitive areas, repeated changes of name or flag, unusual routing, inconsistent draft or cargo information, calls at sanctioned ports, and ownership links to designated persons. AIS data are informative but imperfect: equipment failure, technical limitations and legitimate security practices can create gaps. Therefore, an AIS anomaly should be corroborated with bills of lading, port records, satellite information, insurance data and customer explanations.

5. Investigation methodology

A consistent investigation prevents both superficial clearance and unjustified rejection. A practical sequence is:

1. Preserve the alert and evidence. Record the transaction, screening result, documents, system version and timestamps.
2. Establish the factual network. Map the customer, beneficial owners, counterparties, banks, vessel, ports, goods and payment flows.
3. Verify identities. Resolve name matches using dates, addresses, registration numbers, IMO numbers and ownership data.
4. Understand the commercial rationale. Compare the transaction with the customer’s stated business, historical activity and industry practice.
5. Test documents against one another. Compare invoice, packing list, bill of lading, certificate of origin, insurance document, inspection certificate, customs declaration and payment instruction.
6. Assess goods and end-use. Obtain technical descriptions, licences, end-user certificates and evidence of delivery where risk warrants.
7. Analyse value and payment flows. Examine price, quantity, currency, intermediaries, account activity and onward movement of funds.
8. Search reliable external information. Use official sanctions lists, corporate registries, customs data, shipping intelligence and reputable adverse-media sources.
9. Document the decision. State facts, analysis, uncertainty, controls applied, escalation and rationale for release, rejection, freezing, exit or reporting.

Suspicious-transaction reporting decisions must follow applicable law and internal policy. Staff must avoid tipping off a customer where prohibited. Confidentiality, privilege, record retention and secure case access are essential.

6. Worked application: identifying inconsistency without assuming guilt

Assume a bank receives a documentary collection for industrial pumps. The importer is a recently incorporated trading company. The exporter is in State A, the vessel is registered in State B, the bill of lading shows trans-shipment in State C, and the final consignee is in State D. The invoice describes “industrial equipment,” while a technical brochure attached to an earlier email indicates pumps capable of operating in a sensitive industrial facility.

The first alert is not automatically a violation. The bank should ask:

• Is the importer licensed and operational?
• Why does the importer require equipment inconsistent with its stated business?
• What is the precise technical specification and export-control classification?
• Is the final end-user different from the contractual buyer?
• Does the route reflect normal logistics or concealment?
• Are the exporter, consignee, vessel, owners and banks screened against relevant lists?
• Is the price consistent with comparable equipment?

If the customer supplies a credible end-user certificate, an export licence where required, independently verifiable shipping evidence and a coherent explanation for the route, risk may be reduced. If the customer refuses to identify the end-user, changes the consignee after screening, provides inconsistent specifications and requests payment through an unrelated third party, escalation is warranted. The correct outcome depends on the applicable law and evidence; a bank should neither process blindly nor declare guilt merely because the goods are dual-use.

7. Illustrative case study: Northstar Components

Background

Northstar Components Ltd, incorporated six months earlier, requests a large import letter of credit for electronic components. Its stated business is agricultural machinery repair. The seller is a company in a major manufacturing hub, and the goods are to be shipped through a free-trade zone before reaching Northstar’s jurisdiction.

Problem

The application contains three concerns: the value is much larger than Northstar’s projected turnover; the goods’ technical description is vague; and the proposed vessel recently changed name and flag. A screening system also generates a possible match for a director of the seller.

Analysis

The bank first resolves the possible match using date of birth, address and corporate identifiers. It determines that the match is a different person, but records the reasoning. It then obtains Northstar’s ownership information and discovers that its beneficial owner also controls a logistics company receiving the goods. The logistics company was not disclosed during onboarding.

The bank compares the invoice with a packing list and finds inconsistent quantities. An independent commercial database shows that the unit price is materially above typical prices, although some premium could be explained by specialised testing and warranty terms. The vessel’s name change is lawful, but the bank identifies an unexplained AIS gap near the trans-shipment location. The customer cannot provide a satisfactory end-user statement or export-control classification.

Approach and solution

The bank pauses processing under its internal escalation procedure. It requests technical specifications, purchase contracts, end-user details, export licences or classification opinions, evidence of financing capacity, and an explanation for the related logistics company. Compliance consults legal counsel and the relevant sanctions and export-control specialists. The bank does not rely on the initial false-positive screening result, but it treats the cumulative inconsistencies as material.

The customer later provides an export licence, a genuine end-user certificate, corrected shipping documents and evidence that the logistics company is an ordinary group affiliate. Independent checks corroborate the explanation. The transaction is approved with enhanced monitoring and a requirement for post-shipment evidence. In a different factual outcome—particularly if documents were forged or the end-user were designated—the transaction would be rejected or blocked and reporting considered.

Learning points

1. A false-positive sanctions alert should be resolved, not ignored.
2. Risk often arises from the combination of moderate indicators.
3. Dual-use goods require technical and end-use analysis.
4. Vessel anomalies are investigative leads, not conclusive proof.
5. Good compliance records explain both the concern and the decision.

8. Governance, technology and professional judgment

A strong programme requires board oversight, a documented risk appetite, independent compliance testing, staff training, clear escalation authority and adequate resources. The three-lines model can be useful: business units own first-line risks, compliance provides second-line challenge and monitoring, and internal audit provides independent assurance.

Technology can improve screening and detection through fuzzy matching, network analysis, entity resolution, graph databases, geospatial data, vessel analytics, optical character recognition and anomaly detection. A model may identify a customer whose trade volume, counterparties and routes differ sharply from peers. Graph analysis can reveal common addresses, directors, vessels, intermediaries or payment accounts.

Artificial intelligence requires governance. Training data may be incomplete or biased; false positives can overwhelm investigators; explanations may be weak; sensitive customer information may be exposed; and generated conclusions may be inaccurate. Institutions should maintain human accountability, model validation, audit trails, access controls, data-quality testing, performance monitoring and procedures for correcting erroneous records. Automation should prioritise and assist investigation, not make legally consequential decisions without appropriate review.

Important performance measures include alert quality, investigation timeliness, unresolved-case age, sanctions-match resolution time, recurrence of documentation defects, post-approval exceptions, training completion and independent-testing findings. A reduction in alert volume is not necessarily an improvement if it results from excessive threshold increases.

9. Limitations and contemporary developments

No control system can eliminate financial crime. Beneficial ownership information may be unavailable or unreliable; trade prices are difficult to benchmark; sanctions lists change rapidly; data may be fragmented across jurisdictions; and criminals adapt when controls improve. Correspondent banks may also face conflicting legal duties, particularly where one jurisdiction prohibits a transaction and another permits it.

Contemporary developments include greater attention to proliferation financing, maritime shadow fleets, virtual assets used alongside trade payments, sanctions on sectors and services, beneficial-ownership transparency, supply-chain concentration and the misuse of free-trade zones. Regulators increasingly expect institutions to demonstrate effectiveness rather than merely possess policies.

The most durable principle is proportionality. Controls should be sufficiently rigorous to identify and manage genuine risk, but sufficiently evidence-based to avoid arbitrary exclusion. Decisions should be explainable, reviewable and consistent with applicable law.

Chapter Summary

International trade finance creates vulnerabilities because banks must make decisions across fragmented information, multiple jurisdictions, complex documents and rapidly changing sanctions and export-control rules. KYC and beneficial-ownership analysis establish the customer’s identity, ownership, business purpose and expected activity. Transaction due diligence then tests whether a particular trade is consistent with that profile.

Sanctions screening identifies prohibited parties, jurisdictions and activities; export-control screening addresses controlled goods, technology, end-users and destinations; anti-boycott controls address prohibited discriminatory or boycott-related instructions. These controls are related but legally distinct.

TBML detection requires analysis of price, quantity, documentation, routes, counterparties, vessels, end-users and payment flows. Red flags are prompts for inquiry rather than automatic proof. Effective investigations preserve evidence, map the transaction network, resolve identity issues, test commercial rationale, assess goods and end-use, and document the final decision.

Technology, data analytics and AI can improve entity resolution and prioritisation, but they cannot replace expert judgment, governance, data-quality controls or legal interpretation. The strategic objective is a risk-based, proportionate and auditable system that protects the institution and the integrity of international commerce.

Key Terms

Anti-boycott control: A control designed to prevent prohibited participation in specified foreign economic boycotts.

Beneficial owner: The natural person who ultimately owns or controls an entity or for whom a transaction is conducted.

CDD: Customer due diligence used to identify, verify and understand a customer and its activities.

Dual-use goods: Items, software or technology that can serve civilian and military or other controlled purposes.

EDD: Enhanced due diligence applied to higher-risk customers, products, jurisdictions or transactions.

Export control: Regulation of specified goods, technology, software, services or transfers across borders.

KYC: Know-your-customer processes for identifying and understanding customers.

Money laundering: Concealing or disguising criminal proceeds so that they appear legitimate.

Proliferation financing: Provision of funds or services connected with prohibited weapons or delivery systems.

Red flag: An indicator requiring further inquiry; it is not necessarily proof of wrongdoing.

Residual risk: Risk remaining after controls are applied.

Sanctions: Legal restrictions imposed on persons, entities, countries, sectors, goods or activities.

TBML: Trade-based money laundering through manipulation or misuse of trade transactions.

Third-party payment: Payment by or to a party not clearly connected to the underlying contract.

Trans-shipment: Transfer of goods from one vessel or transport mode to another during transit.

Ultimate end-user: The final person or entity that will use or receive the goods.

Vessel AIS: Automatic identification system data used to track maritime identity and movements.

Vessel IMO number: A stable identification number assigned to a vessel, generally remaining unchanged throughout its life.

Review Questions

Short Answer Questions

1. Why is TBML difficult for banks to detect?
2. Distinguish sanctions screening from export-control screening.
3. What is the role of beneficial-ownership identification in KYC?
4. Give four examples of TBML red flags.
5. Why should an AIS gap not be treated as conclusive proof of sanctions evasion?

Long Answer Questions

1. Explain how a bank should design a risk-based trade-finance control architecture.
2. Discuss the relationship between CDD, EDD, transaction monitoring and suspicious-transaction reporting.
3. Analyse the principal TBML typologies and explain how documentary inconsistencies may reveal them.
4. Compare sanctions, export controls and anti-boycott controls, including their operational implications for banks.
5. Evaluate the benefits and limitations of AI and data analytics in financial-crime compliance.

Analytical / Application Questions

1. An importer with no prior trade history requests a large letter of credit for dual-use equipment through a complex trans-shipment route. Identify the information and evidence the bank should obtain before deciding whether to process it.
2. A vessel is not designated, but it has changed names twice, shows unexplained AIS gaps and is linked to a counterparty making unrelated third-party payments. Construct an investigation plan and explain possible outcomes.
3. A customer’s invoice price is substantially above market price, but the customer provides evidence of specialist testing, extended warranty and unusual freight costs. Explain how the bank should distinguish a legitimate commercial explanation from over-invoicing risk.

Multiple Choice Questions with Answers

Question 1

Which activity is the primary purpose of customer due diligence?

A. To guarantee that no customer will ever commit a crime

B. To identify and understand the customer, ownership and expected activity

C. To determine the market price of every traded commodity

D. To replace all transaction monitoring

Correct Answer: B. To identify and understand the customer, ownership and expected activity

Question 2

Which statement best distinguishes export controls from sanctions screening?

A. Export controls concern only payment currency

B. Export controls assess controlled goods, technology, end-users and destinations

C. Sanctions screening applies only to vessels

D. There is no operational distinction between them

Correct Answer: B. Export controls assess controlled goods, technology, end-users and destinations

Question 3

Which is an example of multiple invoicing?

A. Using one invoice for a shipment

B. Paying a supplier in the invoice currency

C. Using the same shipment to support several payment claims

D. Obtaining cargo insurance

Correct Answer: C. Using the same shipment to support several payment claims

Question 4

Why is an IMO number useful in vessel screening?

A. It is generally a more stable identifier than a vessel name

B. It proves that the cargo is lawful

C. It identifies the vessel’s beneficial owner in every case

D. It replaces all sanctions screening

Correct Answer: A. It is generally a more stable identifier than a vessel name

Question 5

A red flag should generally be interpreted as:

A. Automatic proof of money laundering

B. A reason to close every customer relationship

C. An indicator requiring contextual inquiry

D. Evidence that a transaction must always be approved

Correct Answer: C. An indicator requiring contextual inquiry

Question 6

Which document would most directly assist analysis of the final intended recipient of sensitive goods?

A. An end-user certificate

B. A bank advertisement

C. A staff attendance record

D. A general exchange-rate table

Correct Answer: A. An end-user certificate

Question 7

A third-party payment is most concerning when:

A. The relationship and commercial rationale are documented

B. The payer is an unrelated party and the customer cannot explain the connection

C. The payment is made electronically

D. The payment is made in the customer’s home currency

Correct Answer: B. The payer is an unrelated party and the customer cannot explain the connection

Question 8

What is a major limitation of automated sanctions screening?

A. It cannot produce any alerts

B. It may generate false positives and requires identity resolution

C. It eliminates the need for legal advice

D. It can assess physical cargo without supporting data

Correct Answer: B. It may generate false positives and requires identity resolution

Question 9

Which control is most directly associated with prohibited discriminatory boycott instructions?

A. Anti-boycott control

B. Inventory control

C. Interest-rate control

D. Credit-limit control

Correct Answer: A. Anti-boycott control

Question 10

If inherent risk is 0.8 and control effectiveness is 0.75, using the chapter’s conceptual model, residual risk equals:

A. 0.05

B. 0.20

C. 0.60

D. 1.55

Correct Answer: B. 0.20