Skip to content
RegTech Codex
Directory

Fatf AML KYC Transaction Monitoring

Structuring & Smurfing Detection Models: CTR/TTR Threshold-Evasion Analytics in AML

Modern structuring extends far beyond $9,800 cash deposits—encompassing coordinated multi-account smurfing rings, instant-payment fan-in hubs, and EUR 950 crypto splits engineered to evade statutory reporting.

By Victoria Ashworth, LL.M, CAMSReview not recorded
Structuring & Smurfing Detection Models: CTR/TTR Threshold-Evasion Analytics in AML — RegTech Codex 2026 quantitative & supervisory data card
RegTech Codex (FATF Recommendations, CDD/EDD Beneficial Ownership & SAR Typologies): 31 USC 5324 sub-threshold clustering, Kolmogorov-Smirnov bunching tests, and N-to-1 mule graphs.Source: RegTech Codex Institutional Research DeskLicense: CC-BY-4.0

Consent orders from US FinCEN, the OCC, and FINTRAC Canada repeatedly expose how legacy AML scenarios anchored to narrow $9,000–$9,999 single-day cash bands fail against adaptive multi-day structuring and networked smurfing. Across retail banks, futures commission merchants governed by CFTC & NFA Rule 2-9 Supervision (such as AMP Futures), and crypto platforms governed by FATF Recommendation 16 Travel Rule Standards (such as Coinbase), syndicates systematically fragment transfers just beneath statutory reporting and verification cliffs. Review our FATF AML/KYC Compliance Playbook 2026 inside the FATF AML/KYC hub.

Four Statutory Reporting & Verification Cliffs Targeted by Structuring#

Effective surveillance maps every external regulatory cliff alongside internal firm EDD step-up gates (which launderers probe via micro-deposits and declined withdrawals):

Quantitative Distribution · Jurisdiction & Rule
Jurisdiction & RuleThreshold Value
01US Funds Travel / Monetary …31 CFR 1010.410…02EU Unhosted Crypto ProofReg (EU) 2023/1…03EU Cash Limit & Broker SoF …EU AMLR Art. 80…
Statutory / Internal CliffJurisdiction & RuleThreshold ValueTypical Evasion Ticket BandHigh-Precision Detection Feature
Cash CTR / LCTR / TTRUS FinCEN / FINTRAC / AUSTRAC>$10,000 (USD / CAD / AUD)$4,200 – $9,850 across branches/ATMsMulti-day rolling window + ATM geo-time travel impossibility
US Funds Travel / Monetary Inst.31 CFR 1010.410(e)-(f) & 1010.415$3,000 wire / $3,000–$10k cashier check$2,450 – $2,980 sequential wiresSub-cliff bunching density ([0.75T, 0.999T])
EU Unhosted Crypto ProofReg (EU) 2023/1113 Art. 14(5)> EUR 1,000 self-hosted walletEUR 880 – EUR 995 rapid UTXO/ERC20 burstsLinked destination-cluster graph aggregation
EU Cash Limit & Broker SoF GateEU AMLR Art. 80 / Internal Broker GateEUR 10,000 cash / $25,000 broker SoF4 × $6,100 ACH/card funding -> instant wire outPass-through velocity (Outflow / Inflow >= 0.90 in <72h)
Multi-layer structuring and N-to-1 smurfing detection pipeline combining bunching statistics and bipartite mule graphs
Multi-layer structuring and N-to-1 smurfing detection pipeline combining bunching statistics and bipartite mule graphsVerified Exhibit

Statistical Bunching Estimators vs Naive Hard-Band Rules#

Natural retail and SME commercial transactions follow smooth Benford and log-normal size distributions across $10,000 or EUR 1,000 boundaries. When an account exhibits 6 transactions inside [0.70T, 0.999T] and zero transactions inside [1.00T, 1.30T], exact binomial bunching probability drops below $p < 0.0016$—proving intentional threshold awareness even when individual amounts vary between $7,120 and $9,410.

Sub-Threshold Bunching & Network Fan-In Structuring Score#

Surveillance data scientists score structuring and smurfing risk ($S_{\text{struct}}$) over rolling window $W$ relative to statutory threshold $T$:

Formula or model from the articleNot independently verified
B_bunch = ( N_[0.65T, 0.999T] - N_[1.00T, 1.35T] ) / ( N_[0.65T, 0.999T] + N_[1.00T, 1.35T] + 1 )
  • Vdrain = min(1.0, Outflow48h / max(1, Inflow_W))
  • Sstruct = 100 * min(1.0, 0.40 * max(0, Bbunch) + 0.30 * min(1, SumW / (2.0 * T)) + 0.30 * Vdrain * min(1, Kuniquesenders / 5))
  • Where:
  • N_[a, b] = Count of transactions falling within interval [a, b] over 14-day window W
  • Sum_W = Cumulative dollar/euro volume across account or shared-device/wallet cluster
  • Kuniquesenders = Number of distinct originators feeding the hub account (1 for solo structurer; >=5 for smurf hub)
  • Worked Case (8 Transfers Between $7,200 and $9,600 From 6 Unrelated Senders Totaling $67,400, 94% Wired Out in 19 Hours):
  • T = $10,000 -> Bbunch = (8 - 0) / 9 = 0.889, SumW / 2T = 1.0, Vdrain = 0.94, Kunique_senders/5 = 1.0
  • > S_struct = 100 * (0.3556 + 0.30 + 0.282) = 93.8 / 100 (High-Confidence Smurfing SAR Escalation)
Share of structuring and smurfing SAR filings across cash ATMs, P2P instant rails, and sub-threshold crypto splits
Share of structuring and smurfing SAR filings across cash ATMs, P2P instant rails, and sub-threshold crypto splitsVerified Exhibit

Geo-Temporal Impossible-Travel & Smart-ATM Recycler Controls#

Modern cash-smurfing syndicates distribute cloned deposit cards or QR deposit tokens across 15 to 40 smart-ATM recyclers in multiple cities simultaneously. Calculating physical travel speed between consecutive ATM cash-in timestamps (Distance_km / Delta_hours > 250 km/h) deterministically proves coordinated multi-mule card sharing on a single retail account.

Linking Structuring Alerts to SAR Narratives, Tipping-Off Rules & Model Tuning#

Because branch staff or relationship managers who casually tell a client "if you deposit under $10,000 we don't have to file paperwork" commit criminal counseling/tipping-off violations, pair automated sub-threshold detection with tipping-off prohibitions and account exit SLAs, FinCEN SAR and goAML XML narrative audits, and SR 11-7 transaction monitoring model validation.

Frequently asked questions

Is structuring illegal in the United States if the cash comes from a legitimate small business?

Yes. Under 31 U.S.C. 5324, breaking up cash deposits or withdrawals with the intent to evade a domestic financial institution’s $10,000 CTR reporting obligation is a federal crime regardless of whether the cash originated from lawful commerce or criminal proceeds.

What is the operational difference between "structuring" and "smurfing" in AML typologies?

Structuring refers broadly to splitting a large financial value into smaller sub-threshold tranches over time or branches by one actor or business. Smurfing scales that evasion horizontally across a coordinated network of third-party money mules ("smurfs") who deposit or wire sub-threshold amounts that rapidly converge into a central beneficiary hub.

How do banks aggregate multiple cash transactions for a FinCEN Currency Transaction Report?

Under 31 CFR 1010.313, a financial institution must treat multiple currency transactions as a single transaction if it has knowledge that they are by or on behalf of any person and result in either cash-in or cash-out totaling more than $10,000 during any one business day across all branches and ATMs.

How does sub-threshold structuring manifest on non-cash brokerage and crypto rails?

Although brokers prohibited from accepting physical cash do not file cash CTRs, traders and VASPs routinely encounter structuring below the $3,000 US Travel Rule floor, EUR 1,000 EU TFR unhosted-wallet proof gate, SGD 1,500 MAS Travel Rule floor, or internal $25,000 automated Source-of-Funds review triggers.

Publication record

Published
14 November 2026
Updated
14 November 2026
Author
Victoria Ashworth, LL.M, CAMS
Review
Review not recorded
Automated fact check
Passed in the stored pipeline record
Source URLs
3 recorded

AI assistance recorded: anthropic/claude-opus-4.6. This field does not establish human review or source verification.

Academic & Institutional Attribution

Cite This Research & Embed Live Data Card

Open Attribution License (CC BY 4.0)

Standard academic & institutional research paper citation format.

Victoria Ashworth, LL.M, CAMS. (2026, November 14). Structuring & Smurfing Detection Models: CTR/TTR Threshold-Evasion Analytics in AML. RegTech Codex. https://regtechcodex.org/briefings/structuring-smurfing-ctr-threshold-evasion-detection-models
Explore Fatf AML KYC Transaction Monitoring Hub (20) →

RegTech Codex Data, Benchmarks & Contributor Records

AML Structuring & Smurfing Detection Models Guide (2026) | RegTech Codex