The Conventional Wisdom
Your transaction monitoring system flags tens of thousands of alerts each month, most of which are false positives. Your team spends its days clearing low-value transactions that rarely escalate to Suspicious Activity Reports. When leadership looks to cut costs, the solution seems clear: raise your alert thresholds. Focus on big-dollar wires, structuring patterns, and high-risk customers moving large sums.
This approach is common in AML operations. Small-dollar transactions from retail customers are often seen as low risk. Teen checking accounts? Even less so. The reasoning appears sound: if you're going to miss something, better to miss a $200 anomaly than a $200,000 one.
Why We Disagree
This cost-benefit analysis may work for fraud and money laundering, but it fails for human trafficking.
The Atlanta FBI field office handles about 43,000 child trafficking cases annually. Your monitoring system likely misses them because trafficking doesn't follow your existing typologies. Traffickers don't structure deposits to avoid Currency Transaction Report thresholds or layer funds through shell companies. They move small amounts through low-risk accounts, exactly the profile your system is set to ignore.
When a teenager's checking account shows unusual patterns, your current rules treat it as low-risk. The account holder is a minor, the transactions are small, and there are no sanctions matches or geographic red flags. The alert clears automatically or gets a brief analyst review.
However, trafficking-related transactions are typically small and don't involve high-risk customers. A $50 Venmo payment, a $200 ATM withdrawal in an unexpected location, or a series of small cash deposits may represent a pattern you've never defined as suspicious.
The Evidence
At the FinScan FinCrime Forum in July, survivor advocates, FBI agents, and state investigators emphasized that financial institutions are built to catch fraud, not trafficking.
Trafficking doesn't look like the cinematic version of abduction and force. Traffickers build relationships, exploit vulnerabilities, and the financial footprint reflects that: small payments that resemble normal peer-to-peer activity, cash movements that stay below alert thresholds, and account behavior deemed low-risk because risk is defined by dollar amounts and customer segments.
Sextortion cases follow a similar pattern. A teenager receives what seems like romantic interest online, shares explicit images, and then faces demands for payment. The amounts often exceed what a teen can access, but the transaction pattern goes undetected by your system. Some cases end tragically, yet none trigger your high-dollar Transaction Monitoring Rules.
The gap isn't theoretical. Financial institutions process millions of transactions daily from accounts that could show trafficking indicators, but those indicators aren't in your rule library. You're not looking for them because you've optimized for efficiency, which has come to mean focusing on material financial risk.
What to Do Instead
Develop trafficking-specific typologies in your transaction monitoring rules. This isn't about lowering thresholds across the board. It's about creating new detection scenarios that recognize trafficking patterns as distinct from fraud or traditional money laundering.
Start with account behavior that doesn't match the customer profile. Look for a teen account with sudden geographic mobility, peer-to-peer payments to multiple recipients, or cash withdrawals in locations inconsistent with school or home. Consider an adult account holder with small, regular payments to the same recipient combined with other indicators of control or coercion.
Collaborate with organizations that understand the problem firsthand. Operation Light Shine connects survivor leaders with law enforcement. The FBI's field offices can describe what financial activity looks like in real investigations. Your team needs to understand that a $200 transaction can signal serious harm, even if it doesn't impact your bottom line.
Train your review teams differently for these alerts. Trafficking indicators require context that fraud analysts aren't trained to recognize. A payment described as "rent" or "phone bill" might be genuine, or it might be a trafficker maintaining control over a victim's finances. The dollar amount won't tell you which.
Implement specific review protocols for accounts held by minors or young adults. These aren't your lowest-risk customers. They're your highest-risk population for a crime type your current framework ignores. Alerts from these accounts need human review, not auto-clearance based on transaction size.
When the Conventional Wisdom IS Right
Raising thresholds to reduce false positives makes sense for the typologies you're built to detect. If you're seeing 50,000 structuring alerts a month and 49,800 clear without action, your rules need tuning. High-dollar wires from high-risk jurisdictions do warrant more scrutiny than domestic ACH payments between retail customers.
The conventional wisdom also holds when allocating resources for complex money laundering schemes. A $5 million trade-based laundering case requires different expertise and more time than a $500 anomaly. Prioritizing based on potential harm and institutional exposure is sound risk management.
But that logic breaks down when the crime you're trying to detect doesn't scale with dollar amounts. Trafficking, child exploitation, and sextortion cause massive harm through small financial transactions. Your monitoring system's inability to see them isn't a resource constraint problem. It's a typology gap.
You can't fix this by hiring more analysts or buying better technology. You fix it by recognizing that some of your lowest-dollar alerts represent your highest-obligation cases. The Bank Secrecy Act requires you to report suspicious activity, not just large suspicious activity. When your system is blind to an entire category of financial crime because it doesn't fit your fraud-and-laundering framework, you're not meeting that obligation.
The cost of building these typologies is real. The cost of not building them is 43,000 cases a year in one city alone, most of them invisible to the institutions processing the transactions.
