Are Higher Suspicious Transaction Report Volumes Associated with More Future Money-Laundering Convictions?

TEXT START: Almost all countries require banks to file suspicious transaction reports (STRs) to help authorities detect and prosecute money laundering.

The Dissection

This paper is an autopsy of a bureaucratic proxy. It tests whether more STRs predict more future convictions, then finds that the apparent relationship weakens to insignificance once country-specific and common time effects are included.

Its real function is to sever reporting volume from AML effectiveness. STR counts measure institutional activity, not necessarily criminal detection, successful prosecution, or control of laundering. The paper’s strongest result is negative: a widely cited metric does not survive basic contextualization.

The Core Fallacy

The residual fallacy is treating convictions as the final truth of AML. Convictions are another institutional output. They reflect what authorities detect, process, prove, and prioritize—not the total volume of laundering or the degree of control over financial crime.

In Discontinuity Thesis terms, the paper is still reading administrative dashboard signals as evidence of system control. It correctly exposes one bad metric but leaves the deeper assumption intact: that reporting and conviction pipelines remain reliable proxies for power as cognitive and financial activity becomes increasingly automated.

Hidden Assumptions

• STR volumes are measured consistently across countries and years.
• Higher reporting reflects genuine suspicious activity rather than compliance incentives, defensive filing, or changing thresholds.
• Population, shadow-economy estimates, and police-force size capture the major confounders.
• One- to five-year lags represent the relevant enforcement pipeline.
• Conviction volumes are sufficiently comparable across legal systems.
• More convictions indicate greater AML effectiveness rather than different enforcement priorities or institutional throughput.
• Fixed effects adequately remove the unobserved forces driving both reporting and convictions.

These assumptions turn two layers of bureaucracy—reports and convictions—into a supposed measurement system for an underlying phenomenon it never directly observes.

Social Function

Classification: Partial truth, with a transition-management function.

The paper punctures KPI worship without challenging the legitimacy of the broader enforcement architecture. It gives institutions a technically respectable warning that one metric is weak, while leaving them free to replace it with more elaborate metrics, dashboards, and evaluation regimes. That is useful diagnosis, but not structural escape.

The Verdict

This is a useful demolition of STR volume as a standalone AML score, not proof that AML works and not proof that convictions track laundering. It shows that paperwork is not control. Under the DT lens, the paper diagnoses dashboard failure while leaving the underlying ownership, automation, and institutional-capacity problem untouched.