Inside the Scam-Centre Economy: Southeast Asia's Industrial Fraud Complex and the Global Response

What began as isolated compounds along the Mekong has evolved into an industrialised fraud economy. Scam centres across Cambodia, Myanmar, Laos, and the Philippines now operate at a scale and sophistication that rivals legitimate BPO operations — and their output is measured in victims rather than tickets closed. The compounds combine call-centre infrastructure, cryptocurrency on-ramps, and professional social engineering at a scale that was unimaginable just a few years ago, and they are increasingly operated like franchised businesses with shared tradecraft across multiple jurisdictions.

The UN UNODC, in its most-cited assessments, has described the phenomenon as one of the most serious transnational organised crime threats in Asia. For banks and payment providers, the scam-centre economy sits at the intersection of fraud, AML, sanctions, and human-rights risk. In practice, a single investigation into a pig-butchering case often threads through retail fraud, correspondent banking, sanctions, and virtual assets — which makes siloed compliance programmes structurally unable to respond effectively.

Ignoring the interconnectedness of these risks is no longer a tenable strategy. Scam centres generate pig-butchering losses in the West, funnel proceeds through regional corridors, and rely on coerced labour — a rare convergence of financial crime and modern slavery that regulators are beginning to treat as a single problem. The institutions that will fare best in the next supervisory cycle are those that can demonstrate a unified view of how scam-centre proceeds move across their products, rails, and geographies.

Regulatory, Enforcement, and Market Context

Singapore's MAS and the Hong Kong Monetary Authority have both sharpened expectations around scam detection, victim protection, and cross-border information sharing. The Australian Transaction Reports and Analysis Centre (AUSTRAC) has highlighted scam-centre typologies in recent strategic analysis briefings. Supervisors have also begun asking banks to evidence specific detection capabilities for scam-centre typologies rather than generic fraud red flags, and recent examination reports have called out institutions that could not show such capabilities.

Enforcement actions are escalating. Operations coordinated by Interpol, the FBI, and regional police forces have disrupted several compounds, with asset seizures running into hundreds of millions of dollars. However, the Egmont Group and FATF have both warned that enforcement alone cannot match the pace at which new compounds emerge. The ongoing regional cooperation between ASEAN financial intelligence units, Interpol, and Western agencies has accelerated since 2024, and we are now seeing regular public updates on joint operations and seizures.

Regulators are also beginning to link scam-centre investigations with virtual asset service providers, given the heavy reliance on stablecoins in the cash-out chain. This is creating clearer expectations for private-sector participation and, in some markets, the beginnings of formal information-sharing frameworks that banks can actually use.

What the Data Is Showing

Chainalysis has traced billions of dollars in illicit flows linked to 'pig-butchering' and investment-fraud typologies, with a significant share tied to Southeast Asian compounds. Sumsub's global fraud index shows sustained double-digit growth in scam-related identity abuse across the region. The scale of crypto outflows is particularly telling: a significant share of scam-centre losses ultimately exits via stablecoin corridors, underscoring how closely scam detection, VASP oversight, and sanctions screening are now interlinked.

Reuters reporting suggests that as many as 100,000 people may be working in such centres under varying degrees of coercion — a figure that reframes the problem from fraud-as-crime to fraud-as-organised-human-exploitation. The human dimension — particularly the growing evidence of coerced labour inside compounds — has pushed the issue onto the agenda of labour, migration, and human-rights regulators alongside traditional financial crime authorities.

Implications for Financial Institutions

Banks need integrated detection that fuses scam signals, mule account behaviours, and sanctions-adjacent exposure. Treating scam losses as a retail-fraud KPI misses the underlying financial-crime structure and underestimates regulatory risk. Investment in detection should be paired with investment in customer protection — because even the best detection capability cannot fully prevent losses once a victim is emotionally committed to the scammer's narrative.

Institutions should review correspondent and VASP relationships in higher-risk corridors to ensure due diligence captures exposure to scam-centre networks. Silence from partner banks is not assurance. Internally, banks should consider whether their scam-related KPIs (loss rates, intervention effectiveness, recovery timelines) receive the same board-level visibility as more traditional fraud and AML metrics.

Finally, customer-protection programmes must evolve. Warning labels and friction at the point of payment can meaningfully reduce victimisation, and regulators are now explicitly looking for evidence of such controls. Finally, scam-centre exposure should feature explicitly in enterprise-wide risk assessments — not as a subsidiary concern of the retail fraud team, but as a cross-cutting financial crime theme.

Conclusion

Southeast Asia's scam-centre economy is a case study in how financial crime now operates as a vertically integrated industry. Dismantling it requires the same level of cross-border coordination, data sharing, and analytic sophistication that banks deploy for sanctions or terror finance. Banks that treat this as a strategic risk — with dedicated analytics, executive ownership, and cross-border partnerships — will be far better positioned than peers still treating scam losses as a business-as-usual fraud KPI.

Suggested Next Steps

• Build an internal typology map linking pig-butchering, romance fraud, mule networks, and stablecoin cash-outs.

• Engage with industry bodies and law enforcement on coordinated intelligence-sharing in high-risk corridors.

• Strengthen customer warnings and real-time friction on high-risk investment payments.

• Reassess VASP counterparties with credible exposure to Southeast Asian scam operations.

Sources: UN UNODC assessments, Reuters, Chainalysis, Sumsub, MAS and HKMA notices, AUSTRAC strategic analysis, Interpol and FBI operational updates, Egmont Group and FATF statements.

TrustSphere helps financial institutions design and deploy intelligent fraud and financial crime detection solutions. Visit www.trustsphere.ai