Recurrence, Scale, and Risk-Based Monitoring in Money Laundering: Temporal Displacement under Cumulative Enforcement

How AML reshapes the timing of laundering

Modern anti-money laundering (AML) practice evaluates transactions in a behavioral context – not as isolated events but as parts of a customer’s transaction history. Transaction monitoring systems aggregate activity, update risk classifications, trigger enhanced due diligence (EDD), and escalate scrutiny when concerns persist. That institutional design makes recurrence itself costly: repeated interaction with monitored institutions increases review burdens, the likelihood of deeper investigation, and the risk of relationship termination. When monitoring intensity mechanically rises with accumulated activity, frequency of interaction with financial institutions becomes a strategic margin that illicit actors can manage alongside transaction size.

A model of frequency and scale under cumulative monitoring

Standard economic models of money laundering treat the frequency of interaction as fixed and focus on how criminals alter transaction size, method, or channel to avoid detection. The model summarized here endogenizes laundering frequency: a launderer chooses both transaction size and expected number of interactions (persistence ) inside a relevant aggregation window.

Monitoring frictions have two parts:

1. scale-dependent frictions tied to transaction size and
2. recurrence-dependent frictions that grow convexly with the number of interactions.

Detection probability falls with larger transactions and with greater persistence; penalties and per-transaction costs complete the payoff environment.

The key mechanism is simple and robust. As recurrence-based monitoring intensity increases, the marginal cost of an additional interaction rises. Launderers therefore reduce laundering frequency to limit cumulative scrutiny. If monitoring technologies or practices link persistence and scale – for example when transaction size becomes more informative under persistent activity or when monitoring costs combine both margins – then the reduction in frequency encourages consolidation: fewer, larger transactions and greater organizational investment to sustain repeat access. If scale and recurrence do not interact, frequency falls while transaction size may remain unchanged.

Bastian Schwind-Wagner _Follow

"Recurrence-based monitoring makes the timing of laundering a strategic choice – when repeated interaction raises scrutiny, actors reduce how often they interact with monitored institutions and may concentrate value into fewer, larger transactions. Observed declines in transaction counts therefore need careful interpretation because they can indicate temporal consolidation rather than genuine deterrence.

Effective AML evaluation and policy require tracking frequency, transaction size, persistence indicators, and organizational complexity together. Without coordinated measures across channels and clear sanctions, tightening recurrence-based monitoring can shift activity to less-linkable venues or incentivize greater organizational sophistication."

Temporal consolidation versus fragmentation – a different kind of displacement

This analysis identifies a form of temporal displacement that complements the familiar notions of fragmentation (smurfing ) and channel substitution. Under size-based monitoring (where scrutiny is driven primarily by transaction amount), the optimal response is typically fragmentation: divide funds into many small transfers to stay below detection thresholds. Under recurrence-based monitoring, frequent small transactions themselves trigger escalating scrutiny, so actors economize on the number of monitored interactions. The result can be temporal consolidation – fewer transactions that are larger on average – rather than fragmentation.

This distinction matters for interpretation. Observed declines in transaction counts should not automatically be read as deterrence. A fall in counts accompanied by larger transaction sizes and increased organizational complexity may instead indicate temporal consolidation under intensified recurrence-based monitoring. In other words, fewer observed transactions do not necessarily imply reduced laundering volume.

Organizational structure and sorting across actors

Fixed organizational costs – incorporation, intermediaries , nominee arrangements – act as entry costs for persistent laundering rather than as investments justified by transaction size alone. Professional laundering organizations with lower effective fixed costs and greater capacity to manage compliance friction are better positioned to adopt persistent, consolidated strategies. Opportunistic or small-scale offenders, who face higher organizational costs and weaker ability to navigate monitoring systems, are relatively more likely to prefer episodic laundering despite higher per-transaction exposure.

This sorting mechanism helps explain observed heterogeneity across crime types. Activities that generate steady streams of illicit revenue (large trafficking networks, systematic fraud rings) are more likely to organize persistence and invest to manage cumulative scrutiny. One-off or episodic offenses are more likely to be handled with single large transactions or by relying on non-financial channels.

Channel displacement and the limits of recurrence-based monitoring

Recurrence-based monitoring is most effective where continuity of identity and transaction linkage are strong, as in regulated banking systems. Where linkability is weak – informal remittances, cash-based businesses, or parts of the crypto ecosystem with fragmented identity linkage – actors have cheaper ways to reset or split identity, reducing the marginal cost of persistence. Thus, intensifying recurrence-based monitoring in banks can push activity toward channels where persistence is harder to detect. The model therefore implies potential spillovers: recurrence-sensitive supervision in one part of the financial system can increase laundering incentives in other, less-linkable channels.

Empirical implications – what to measure and where to look

The framework generates specific empirical predictions that differ from size-based enforcement models. Stronger recurrence-based monitoring should reduce transaction frequency per beneficial owner. If scale–recurrence interactions are present, average transaction size should increase conditional on total illicit volume. Importantly, declin­ing transaction counts coupled with increases in transaction size and organizational complexity are consistent with temporal consolidation rather than deterrence.

Data sources suited to test these predictions include financial intelligence unit (FIU) microdata and suspicious activity report (SAR) panels that enable entity-level longitudinal analysis. Supervisory datasets that record enhanced due diligence flags, account restriction events, and escalation histories are particularly valuable because they directly reflect cumulative monitoring. Enforcement case records that document organizational arrangements and repetition of access provide complementary evidence on the link between persistence and structural complexity.

Policy implications – interpret carefully, coordinate broadly

Two interpretive points follow directly.

1. Declines in transaction counts alone are insufficient evidence of effective deterrence under modern AML regimes that heighten recurrence-based monitoring. Policymakers and supervisors should jointly examine frequency, transaction size, persistence, and indicators of organizational complexity to distinguish deterrence from temporal displacement.
2. Recurrence-based monitoring in isolation can produce unintended channel spillovers. Effective AML strategy requires coordinated attention to scale, frequency, and channel linkability – ideally aggregated at the beneficial ownership level rather than at account or transaction level only.

Practical takeaways for practitioners and supervisors

Supervisors should make recurrence an explicit metric when assessing AML effectiveness. This means tracking not just volume and amounts but also counts per beneficial owner, the incidence and timing of enhanced due diligence, and account-level escalation trajectories. When regulators tighten recurrence-sensitive measures (shorter look-back windows , lower escalation thresholds, or mandated behavioral scoring ), they should monitor whether activity migrates to less-linkable channels or whether organizational complexity increases. Law enforcement’s credibility also matters: when penalties and credible sanctions are weak, recurrence-driven escalation may mainly encourage cooling-off tactics or channel switching rather than genuine deterrence.

Conclusion

Treating persistence as a strategic margin makes clear that AML enforcement reshapes not only how laundering is done but how often illicit actors are willing to interact with monitored institutions. Cumulative, actor-based monitoring raises the shadow price of recurrence. The dominant adaptation may be temporal consolidation and increased organizational investment rather than fragmentation. Distinguishing temporal displacement from deterrence demands richer, longitudinal data and coordinated policy responses that consider scale, frequency, and channel linkability together.

Dive deeper

• Research ¦ Endre J. Reite, Recurrence, scale, and risk-based monitoring in money laundering: Temporal displacement under cumulative enforcement, Journal of Economic Criminology, Volume 12, 2026, 100221, ISSN 2949-7914, https://doi.org/10.1016/j.jeconc.2026.100221. ¦ Link ¦ licensed under the following terms, with no changes made: CC BY 4.0