Illicit Economies
The trillion-dollar shadow economy, and how it shapes politics, security, and migration patterns well beyond the countries it operates in.
(UN Office on Drugs and Crime (UNODC) range for 2009; contested, but later estimates fall in the same band)
(UNODC's 2009 estimate was 3.6%; the IMF's range is 2-5%; much higher in some countries)
(Global Financial Integrity's 2017 range, $426-652B; production is a small fraction of retail)
A note on framing. Illicit economies sit awkwardly between conventional economic and political coverage. They are too criminal for ordinary economic statistics, too economic for ordinary crime statistics, and too transnational for any single country's reporting to capture cleanly. The result is that one of the largest features of the world economy is consistently under-reported in mainstream coverage. This page tries to walk through the picture: what the categories are, where the money flows, and how it shapes the surrounding political environment.
What "illicit economy" covers
The broad category of illicit economies bundles several distinct activities, each with its own structure. Treating them as one phenomenon flattens distinctions that matter for understanding what is happening and what reform options exist.
Drug trafficking. The largest single category by retail value, with an estimated 331 million users in 2024 (UN World Drug Report 2026). Cocaine moves from Colombia, Peru, and Bolivia mostly to North America and Europe, with Brazilian and Venezuelan transit; output more than quadrupled in a decade, passing 4,000 metric tons in 2024. Heroin once moved from Afghanistan to Europe and from Mexico to North America; the Mexican fentanyl supply chain has now displaced most of the heroin market and reshaped the political economy of the US-Mexico border. Methamphetamine production is concentrated in Myanmar and Mexico, with newer suppliers in Afghanistan and parts of Africa. Cannabis remains large but has been partly legalized in jurisdictions accounting for substantial demand. Synthetic drugs (fentanyl, methamphetamine, and newer opioids such as nitazenes) are increasingly the growth segment because they require less land, less labor, and produce higher margins than plant-based drugs.
Human trafficking and smuggling. Distinct activities that are often conflated. Smuggling is moving people across borders for a fee, with their consent. Trafficking involves coercion or deception and continued exploitation. The combined economic value is harder to estimate than drugs, because there is no clear retail price and victims are often invisible to data. Even so, the International Labour Organization estimated in 2024 that forced labor and sexual exploitation generate roughly $236 billion a year in illegal profits - up 37% since 2014. Major flows: Central America to the United States (down more than 90% at the US border since fiscal 2024), North and West Africa to Europe, South Asia to the Gulf, intra-Africa, intra-Asia.
Illegal arms. Smaller in dollar value than drugs but more directly destabilizing. The post-Soviet weapons surplus, US firearms moving south to Mexico and Central America, illicit transfers from active conflict zones (Yemen, Libya, Syria, Ukraine), and the growing 3D-printed-firearm segment. Global Financial Integrity's 2017 estimate for small-arms trafficking is $1.7-3.5 billion a year; the murkier category of legally made arms diverted to illicit users is harder to value.
Wildlife and timber. Illegal logging, fishing, and wildlife products. Timber alone is estimated at $50-150 billion a year, with major flows from Central Africa, Southeast Asia, and the Amazon. Illegal fishing is smaller, at an estimated $15-36 billion a year, and badly under-tracked. Wildlife (rhino horn, elephant ivory, pangolin scales, exotic pets, traditional-medicine ingredients) is smaller in dollar value but ecologically catastrophic.
Illegal mining. Especially gold from West Africa and South America, cobalt from informal Congolese mines, lithium from undocumented Latin American operations, and a long list of others. Often interfaces with paramilitary groups, environmental destruction, and forced labor. The product sometimes enters legitimate global supply chains via complex laundering.
Counterfeiting and other intellectual-property (IP) violations. Pharmaceuticals, electronics, fashion, automotive parts. The OECD and the EU intellectual-property office put this category at $467 billion, or up to 2.3% of all world trade, in their most recent joint estimate. Much of it is not directly violent, but it carries serious safety consequences in pharmaceuticals and aviation parts.
Cybercrime. Growing fast and overlapping with state activity. Ransomware, business email compromise, financial fraud, identity theft. Estimates run from hundreds of billions to several trillion in annual cost; the wide range reflects different definitions of cost (direct losses versus including downtime, response, and lost business). Covered separately on this site.
Industrial online fraud. The fastest-moving category is new enough that it does not fit the older list, because its product is fraud and its input is trafficked people. Scam compounds - the best-known variety runs the long-con investment fraud usually called pig butchering - grew out of casino towns in Myanmar, Cambodia and Laos and now operate at a scale that is hard to place alongside anything else here. The UN drugs and crime office put losses to online scams across East and Southeast Asia, Australia and New Zealand at between $88 billion and $114 billion in 2025. Its estimate for 2023, covering East and Southeast Asia alone, was $18 billion to $37 billion. Set that against the roughly half a trillion dollars of global drug retail sales in the key facts above: an industry that barely existed a decade ago is now a fifth of that size in one region alone.
Two features make it a different shape rather than a bigger version of what came before. The workforce is trafficked - credible estimates put at least 300,000 people inside these operations, recruited from at least 80 countries by fake job advertisements and held under debt bondage, which means most of the people committing the fraud are also victims of it. And the model travels: syndicates originally concentrated on the Mekong have set up in West Africa, southern Africa, Central America, Brazil, the Gulf and several Pacific island states. Drug production is tied to where a plant grows or a precursor can be bought. This needs only cheap buildings, internet, and a jurisdiction that will not look closely, which is a much shorter list of requirements and a much longer list of possible locations.
The major organizations
"Organized crime" is a broad term covering very different organizational shapes. Distinguishing them helps with thinking about why some are so durable.
Mexican cartels. The Sinaloa cartel, the Jalisco New Generation Cartel (CJNG), and several smaller groups dominate cross-border drug flows into the US. A war between Sinaloa cartel factions that began in September 2024 left nearly 2,000 dead in its first year, by local press counts, and in February 2026 Mexican forces killed CJNG's founder, "El Mencho." Mexican groups have evolved from purely transit organizations into integrated production-and-distribution networks with substantial fentanyl precursor sourcing from China. Estimated combined revenue is in the tens of billions annually; the political and territorial influence in parts of Mexico is substantial.
Italian organized crime. Cosa Nostra (Sicilian), Camorra (Neapolitan), 'Ndrangheta (Calabrian), Sacra Corona Unita (Apulian). The 'Ndrangheta has become the most powerful, with extensive global cocaine trafficking presence and a network across European countries. Italian groups have moved heavily into legitimate businesses, public-procurement fraud, and waste management.
Russian-speaking organized crime. The post-Soviet criminal class is no longer cleanly separable from elements of Russian state power, especially after 2000. Major activities include cybercrime, protection rackets, contract violence, and complex financial laundering. The relationship to Russian intelligence services is documented in some cases (the Litvinenko murder, several others) and contested in others.
Chinese triads and successor networks. Hong Kong-based triads, Taiwan-based groups, and mainland-based criminal networks. Activities include human smuggling, gambling, IP infringement, and, according to US Treasury findings, laundering money for Mexican cartels. The relationship to Chinese state policy is disputed: the state suppresses some criminal activity, and critics argue it tolerates or directs others when geopolitically useful.
Japanese yakuza. Long-established but in decline. Police counted a record-low 17,600 members and associates at the end of 2025, down from more than 80,000 in the late 2000s. Increased police pressure, public exclusion ordinances, and demographic decline have reduced their visible presence. The remaining capacity has shifted toward white-collar crime and sub-rosa relationships with parts of the construction and entertainment industries.
West African networks. Nigerian-led networks dominate several categories: drug courier operations, advance-fee fraud, human trafficking. The networks are decentralized, highly mobile, and integrated into the West African diaspora across Europe and the Americas.
Smaller but consequential. Albanian/Kosovar groups (heroin distribution in Europe, prostitution rings), Vietnamese gangs (cannabis cultivation in the UK, Australia), Brazil's Comando Vermelho and Primeiro Comando da Capital (cocaine distribution and prison-based control), Honduran and Salvadoran MS-13 and Barrio 18 (gang violence with diaspora connections), and many others.
What these organizations share is a willingness to use violence, an ability to recruit from populations facing limited legitimate options, networks that span jurisdictions, and a relationship to the formal state that ranges from outright opposition to functional partnership. The variation in those last two dimensions is what produces the variation in what they look like.
Narco-states and partial captures
In some countries, criminal networks and the state have become entangled enough that analysts apply the labels "narco-state" or "partial capture" - labels that are themselves contested.
Mexico. Contested. Some Mexican states are largely controlled by cartels at the local level. Federal institutions remain functional. In April 2026 US prosecutors charged Sinaloa's governor and nine other officials with aiding the cartel; he denies it, and Mexico's government calls the evidence insufficient. The López Obrador government's "abrazos no balazos" (hugs not bullets) approach was sharply criticized; the Sheinbaum government has shifted somewhat, and official homicide counts have roughly halved since 2024, though the registry of the disappeared has passed 130,000 names. The underlying problem is that fentanyl revenue, the price of keeping cartels at bay through confrontation, and the difficulty of prosecution have produced an equilibrium that falling homicide counts alone do not change.
Venezuela. US prosecutors charge that Venezuela's leaders used the state to move cocaine for more than 25 years; their January 2026 indictment calls the "Cartel of the Suns" a patronage system run by those at the top. US forces captured Nicolás Maduro on January 3, 2026. He has pleaded not guilty and faces trial in New York in June 2027.
Honduras. Former president Juan Orlando Hernández, extradited to the US in 2022, was convicted in 2024 of drug-trafficking and weapons charges and sentenced to 45 years. Prosecutors called him the center of "one of the largest and most violent drug-trafficking conspiracies in the world." President Trump pardoned him in December 2025, saying he had been "treated very harshly and unfairly."
Paraguay, Bolivia, parts of Peru and Colombia. Various degrees of partial capture, with documented cases of senior officials, judges, and police involvement in trafficking activity. Reform efforts have varied success.
Myanmar. Military-held areas, the contested "Special Regions," and the major ethnic armed organizations are all involved in different combinations of methamphetamine and opium production, casino-linked human trafficking ("pig-butchering" scam compounds), illegal mining, and drug taxation. UNODC now counts Myanmar as the world's main known source of illicit opium. The state-criminal-insurgent overlap is unusually dense.
Afghanistan. The Taliban, in power since 2021, banned opium in 2022; cultivation has since fallen by more than 95% (UNODC), constraining the global heroin supply. Methamphetamine production keeps growing. The Taliban's relationship with these economies is complicated by their own ideological commitments and the practical need for revenue.
Various African cases. Guinea-Bissau, Mali, parts of West Africa more broadly, where state capacity is low and trafficking flows are high. The line between state, military, militia, and trafficker becomes difficult to draw.
What these cases share is that the illicit economy is large enough relative to the licit one that controlling it would substantially reduce the state's revenue. The institutions that would normally check capture (independent judiciary, free press, opposition politics) are weakened by the same forces.
Why prohibition produces these structures
A feature of illicit economies worth stating plainly is that the shape of the criminal organizations is largely a function of the prohibition policy itself. This is not an argument that prohibition is wrong; it is an observation about the predictable consequences of choosing prohibition over alternatives.
When a substance or activity is prohibited and demand persists, supply organizes around prohibition. Costs of evading detection are passed through as price. Prices that include risk premiums attract entrants willing to take the risks - usually people with limited legitimate options or with specific advantages (state-protection, geographic position, existing organizational capacity). Violence becomes the dispute-resolution mechanism because legal contracts are unavailable. Economies of scale emerge in violence and in protection. Profits are routed through laundering structures because direct deposit is not an option. Each of these features is a predictable response to the legal environment.
The cleanest natural experiment is cannabis legalization in jurisdictions accounting for substantial demand. Where cannabis has been legalized with regulated commercial sales (parts of the US, Canada, Uruguay), the illicit cannabis economy has shrunk substantially, prices have fallen, organized criminal involvement has decreased, and tax revenue has increased. Where legalization has been partial or poorly executed (some US states, several European countries' decriminalization experiments), the illicit market has persisted alongside the legal one. The lesson is that prohibition's effects on the structure of supply are real, and so are the effects of removing prohibition - but execution details matter substantially.
For other categories - especially fentanyl-class synthetic opioids, methamphetamine, and the fully exploitative end of human trafficking - the policy alternatives are harder. The harm-from-the-substance is high enough that legalizing-and-regulating is not the obvious answer. The harm-from-prohibition is also real. The fairest summary in this area is that there is no clean policy that eliminates both kinds of harm; reasonable people of good faith disagree about how to weigh them.
The paths from here
Continued large illicit economies with shifting composition
The aggregate scale of illicit activity stays roughly stable as a share of global GDP. Its composition shifts - synthetic drugs displace plant-based, cybercrime grows, IP-counterfeiting rises with manufacturing decentralization. No major reform changes the overall picture.
Drug-policy reform spreads
Cannabis legalization continues across more jurisdictions. Limited harm-reduction approaches to opioids (regulated access, supervised consumption) expand in some countries. The illicit drug economy contracts somewhat as legal alternatives grow. Other categories of illicit activity remain mostly unchanged. So far the movement runs both ways: Oregon (2024) and British Columbia (2026) ended decriminalization experiments.
Fentanyl crisis triggers harder enforcement
The fentanyl mortality wave produces sustained political pressure for tougher enforcement. Some compounds become harder to acquire; the underlying demand persists; new compounds replace them. Much of the enforcement has happened: US terrorist designations of major cartels in 2025 and, since September 2025, strikes on about 70 alleged drug boats. Washington says it is in an "armed conflict" with the cartels; a UN special rapporteur, counting at least 223 dead through August 2026, says the strikes "plausibly constitute crimes against humanity." US overdose deaths have fallen about 40% since their 2023 peak.
Specific narco-states stabilize or destabilize
Mexico's trajectory is the highest-stakes question: lasting improvement or further deterioration could shift millions of lives. Other partial-capture cases (Honduras, Venezuela, Myanmar, Paraguay) follow their own trajectories with regional implications. Venezuela's trajectory changed abruptly with Maduro's capture in January 2026.
Anti-laundering enforcement closes onshore enabler infrastructure
The combination of beneficial-ownership transparency, sanctions enforcement, and AI-driven anomaly detection makes Western financial systems harder to use for laundering. Illicit proceeds shift toward jurisdictions outside Western reach (UAE, parts of Asia) or toward crypto. Total volume does not change much; geography does. The record is mixed: in 2025 the US dropped beneficial-ownership reporting for domestic companies, but also sought forfeiture of about $15 billion in bitcoin from an alleged scam-compound boss.
Where serious analysts disagree
Prohibition is the structural problem; legalize and regulate
The criminal-organization infrastructure exists primarily because demand persists under prohibition. Carefully designed legalization-and-regulation regimes (cannabis cleanest, opioids harder, others variable) would shrink the illicit economy substantially. Continued prohibition is choosing the violence and capture as a cost.
Held by: a substantial part of public-health and harm-reduction research, the Drug Policy Alliance, parts of the Cato Institute, and a strand of mainstream policy thinking. Empirical support is strongest for cannabis and weakest for fentanyl.
The harm of the substances themselves is the structural problem
Substances like fentanyl and methamphetamine, and the trafficking of people, cause harms that are not adequately addressed by legalization. Even imperfect prohibition produces less aggregate harm than legal availability would. The right approach is better, more humane, and better-resourced enforcement, not capitulation.
Held by: parts of the law-enforcement and treatment community, parts of the conservative criminology tradition, and serious harm-from-substance researchers. Empirical support varies by substance.
Demand-reduction is the underweighted variable
Most policy attention goes to supply-side enforcement (interdiction, prosecution, sanctions on traffickers). Demand-side investment - treatment, mental-health care, social services, harm reduction - has higher empirical evidence of impact per dollar. The supply-side bias reflects political optics rather than effectiveness.
Held by: Carl Hart (Columbia), the public-health and addiction-treatment community, Portugal's policy architects, and a growing body of mayor-level US policy thinking. The case is empirically strong and politically difficult.
Illicit-economy estimates are inflated
The widely cited UNODC figures involve substantial estimation uncertainty and tend toward upper-bound numbers because that serves the funding interests of anti-trafficking organizations. The actual scale, while real, may be meaningfully smaller than headline figures suggest. Better data would help; the institutional incentives to produce it are weak.
Held by: Peter Andreas (Brown, "Smuggler Nation"), and a strand of academic crime-statistics research. The case is partly correct - estimates are uncertain - but the directional picture is robust across multiple methods.
The state-criminal-elite overlap is the most consequential part of the picture
The interesting question is not what cartels do, but where the line between cartels and states actually runs, and how Western financial and political systems handle the resulting wealth. Mexican cartels, Russian-state-adjacent crime, Chinese-state-tolerated activity, Myanmar's military-criminal economy, and the Venezuelan case are all examples of this fusion. Coverage that treats them as ordinary crime misses the larger picture.
Held by: Moisés Naím ("Illicit"), Misha Glenny ("McMafia"), Tom Wainwright ("Narconomics"), parts of the Organized Crime and Corruption Reporting Project's investigative network. Their case is supported by case studies across multiple regions.
What the available evidence actually supports: illicit economies are real, large, and important, and their composition is shifting toward synthetic and cyber categories. Prohibition policy is one significant variable among several in shaping their structure. The state-criminal overlap in specific countries is more consequential than the public conversation usually conveys, and the Western financial-and-legal infrastructure that processes the proceeds is part of the system rather than separate from it.
What this means for you
If you read political coverage
Some of the political stories that confuse coverage make more sense once illicit-economy dynamics are visible. Mexican election violence, parts of the migration story, certain African and Southeast Asian conflicts, and specific US border-policy debates all involve illicit-economy structures that mainstream coverage often understates. Reading specialized sources alongside general news is useful.
If you have personal exposure to substance use
The fentanyl supply has made the underlying drug supply much more dangerous than it was a decade ago. Whatever one's policy view, the practical implication is that drugs sold as one thing increasingly contain others, and the safety margin for ordinary recreational use has narrowed. Naloxone availability, fentanyl test strips, and not using alone are evidence-based harm-reduction measures.
If you do business in higher-risk environments
Specific industries (extractives, construction, logistics, hospitality) interact with illicit-economy structures more than others. Local norms about informal payments, security arrangements, and regulatory navigation often rest on assumptions that are not visible from an outside perspective. Investing in proper local knowledge is part of operating responsibly.
If you make consumer choices about traceable goods
Specific products carry meaningful illicit-economy exposure: gold (especially from artisanal sources), cobalt (in batteries), tropical timber, certain seafoods, fashion and electronics counterfeits. The supply-chain certifications are imperfect but generally better than nothing. The "fair trade," "responsibly sourced," and certified-by-respected-bodies categories are not marketing alone.
If you vote on drug or migration policy
The empirical evidence base on these policies is more developed than political debate usually implies. Voting on these issues with awareness of the empirical record - rather than only on cultural intuition - is one of the more impactful forms of civic engagement on subjects directly tied to illicit-economy dynamics.
