The term “romance scam” does not adequately describe what the U.S. Institute of Peace documented in its 2024 report on cyber scam operations in Southeast Asia. “Romance scam” implies an individual — a person running a con, probably from their home, targeting one victim at a time. It implies opportunism. It implies a ceiling on scale that the actual operations exceed by orders of magnitude.

What USIP documented, what FTC Consumer Sentinel data reflects in $1.3 billion in reported U.S. losses in 2022 alone, and what the State Department’s Trafficking in Persons report addresses in its forced labor sections is an industry. Compounds in Myanmar, Cambodia, and Laos operating at the scale of mid-sized corporate call centers, staffed in significant part by trafficked workers who were recruited through deceptive job postings and then coerced into running fraud operations. The operational sophistication includes fraud scripts, performance management, training programs, and regional specialization. It is not a scam. It is an economy.

The terminology

“Sha zhu pan” translates from Mandarin as “pig butchering plate” or “pig slaughter plate.” The victim is the pig, fattened over weeks or months of relationship investment before slaughter — the extraction of money. The term is used internally by the fraud operations and has been adopted by USIP and other researchers as the most accurate English descriptor. The U.S. State Department and FBI use variants of the English transliteration. The FTC uses “romance investment scam,” which captures the compound nature of the fraud — the romance element that precedes the investment solicitation — but misses the industrialized scale.

The operational reality described by all these terms is the same: a compound fraud that combines a romance fraud vector (the relationship, developed over time to establish trust) with an investment fraud vector (the solicitation, presented as an opportunity rather than a request). The two-stage structure is not incidental. It is the design.

The structural anatomy

USIP’s 2024 report documents the three-phase operational structure of sha zhu pan fraud.

Phase one: trust-building. This phase extends over weeks to months. The initial contact is typically through a messaging platform — LinkedIn, WhatsApp, Instagram, dating applications — often appearing as a wrong number or an accidental connection. The persona is carefully constructed: attractive, professional, often representing a high-status occupation (finance, medicine, technology). Communication is attentive, consistent, and emotionally invested. This phase requires sustained labor: the fraudster maintains dozens of these relationships simultaneously, following scripts but also personalizing conversations to establish the sense of unique connection.

The trust-building phase operates entirely within legitimate communication channels. There is no platform-specific behavior that would flag it as fraud. The messages are appropriate. The interest seems genuine. The relationship feels real because the behavioral investment in the relationship is substantial — it simply belongs to someone operating under a false identity with an ulterior purpose.

Phase two: the ask. The transition to the investment solicitation is presented not as a sales pitch but as an opportunity being shared within a trusted relationship. The persona mentions their own investment returns — spectacular, consistent, obviously from a proprietary platform or insider connection. An offer to help the target access the same returns. The framing is as a favor between people who trust each other, not as a solicitation. By this point, weeks or months of relationship investment have created the cognitive conditions in which accepting the offer feels like participating in the relationship, not engaging with a financial product.

Phase three: extraction. The fraudulent investment platform displays fabricated returns, creating the impression of successful investment. Initial withdrawals may be permitted — this is documented in both USIP and FTC reports — specifically to reinforce the legitimacy of the platform and encourage larger deposits. The platform then blocks withdrawals, typically requiring additional “fees,” “taxes,” or “security deposits” that are themselves extracted. The extraction continues until the target stops depositing. The persona then disappears.

The forced labor dimension

The USIP (2024) report documents a dimension of sha zhu pan operations that distinguishes them from prior models of organized fraud: the workforce is substantially composed of trafficked labor. Workers are recruited through deceptive job postings advertising customer service, data entry, or translation roles in Southeast Asia, often posted in Chinese-language media targeting workers from mainland China, Taiwan, and Chinese-speaking diaspora communities. On arrival, workers find themselves in compounds — in Myanmar’s Shan State, in Cambodia’s Sihanoukville, in Laos’s border regions — where passports are confiscated, movement is restricted, and the work is running fraud operations under threat of physical violence.

The State Department’s 2023 Trafficking in Persons Report identifies these operations as a form of labor trafficking. The workforce is not choosing to commit fraud. They are being coerced into it, often by criminal organizations with documented ties to organized crime networks in China and Southeast Asia.

This dimension of the sha zhu pan economy has two implications for how the fraud should be understood. First, the scale: a compound with hundreds of coerced workers running dozens of simultaneous fraud relationships can generate tens of millions of dollars per year. The FTC’s $1.3 billion figure in reported U.S. losses represents a fragment of global losses, and reported losses represent a fraction of actual losses, since fraud victims frequently do not report. Second, the moral frame: the fraud has two categories of victims — the financial victims whose money is extracted, and the labor trafficking victims who are forced to conduct the extraction. Policy responses that address only the financial fraud miss half the victim population.

The scale

FTC Consumer Sentinel data, published annually, reported that romance fraud — a category that includes but is not limited to sha zhu pan operations — accounted for reported losses exceeding $1.3 billion in 2022, with a median loss per victim of $4,400. The FBI’s Internet Crime Complaint Center logged approximately 19,000 complaints in the romance fraud category in that year.

Both figures are floor estimates. FTC research on fraud reporting rates suggests that a substantial majority of fraud victims do not report their losses to federal agencies. FBI IC3 figures reflect only complaints submitted through the IC3 reporting portal. The actual scale of romance fraud losses in the U.S. is estimated to be significantly higher than reported figures indicate.

Why platform-level solutions are insufficient

The trust-building phase of sha zhu pan operations is designed to occur through normal communication channels. An operation that begins on a dating application will migrate to WhatsApp or Telegram within days of initial contact. The fraudulent investment platform does not appear until the relationship has been established and trust has been developed — by which point the victim’s communication is occurring across multiple standard platforms.

Platform-level fraud detection tools — profile verification, bot detection, behavioral flags for suspicious messaging patterns — can disrupt some initial contact attempts. They cannot address the trust-building phase, which is indistinguishable from legitimate relational communication until the investment solicitation appears. By the time the investment fraud vector activates, the user has typically moved the conversation off the original platform.

This is a structural feature of the fraud’s design, not a gap in platform moderation. The operational sophistication of sha zhu pan compounds includes explicit training on migrating conversations to unmonitored channels before the solicitation appears.

References


  1. Federal Trade Commission. Consumer Sentinel Network Data Book 2022. FTC; 2023.

  2. Federal Bureau of Investigation Internet Crime Complaint Center. 2022 Internet Crime Report. FBI IC3; 2023.

  3. United States Institute of Peace. Cyber Scam Operations in Southeast Asia. USIP; 2024.

  4. U.S. Department of State. Trafficking in Persons Report 2023. State Department; 2023.