EVOPEDIA

← all systems · see it in The System

Cybercrime and its underground economy

Criminal activities carried out using digital devices and/or networks: crimes against computers (unauthorised access, damage to data or programs, sabotage, interception, espionage) and older crimes done with them (fraud, theft, extortion, illegal content, harassment) (src:wp:cybercrime); and the economy behind them, 'a loose federation of specialists selling capabilities, services, and resources explicitly tailored to the abuse ecosystem', where profit centers draw money from victims and support centers sell to other criminals (src:thomas-2015-underground). Walked as criminologists and security economists describe it: offences, support services, markets and forums, networks and roles, cash-out, and the purposes and costs, with law enforcement and defence as the partners that push back.

Properties published scienceour proposal

Sourced fields, and our own readings.

born at1970s (computer fraud; the Union Dime embezzlement began 1970); the economy from about 2004
born precisiondecade
boundaryInside: the offences done with or against computers and networks, the support services sold for them (infrastructure, malware, access, credentials, labour), the markets and forums where they are traded, the networks of people and roles who run them, and the cash-out that turns proceeds into usable money. Outside: the victims (people, firms, states), the legitimate infrastructure and payment systems it rides on, and the partners that push back (computer law, police, defenders, threat intelligence).
nested levelRuns on the Internet's services (L6) and on people (offenders' and victims' decisions, Human L7); a part of crime as a whole (about half of property crime is now online, src:anderson-2019-cost); the market its threat actors, malware, phishing, botnets and booters are sold and cashed out in.
scalesocial

Measures (45) published science

Share of property crime now online (by volume and value)about half fraction
Cost per citizen, traditional crimes now computer crimes (tax, welfare fraud)low hundreds EUR or USD per year
Cost per citizen, transitional crimes (payment fraud)tens EUR or USD per year
Cost per citizen, new computer crimestens of cents EUR or USD per year
Viagra-spam botnet revenue against the cost it causedabout $3m a year earned; about a hundred times that cost USD per year; ratio
Euro-area card fraud, 20161.8 (0.041% of transactions; card-not-present 1.32) billion EUR
Silk Road sellers' revenue (2012)slightly over 1.2 million USD per month
Silk Road operators' commission (2012)about 92,000 USD per month
Silk Road items examinedover 24,400 items
Item listing life (most items)less than three weeks
Seller life (majority)roughly three months; a core of 112 present throughout
Top 100 sellers' share of feedbackapproximately 60 %
Anonymous marketplace daily volumeup to 650,000; generally 300,000-500,000 USD per day (30-day average)
Cannabis, ecstasy and cocaine products' share of marketplace salesabout 70 %
Silk Road lifetime gross (US government calculation)214 million USD
Fake antivirus revenue, 2008-2010upwards of 130 million USD
Fake antivirus conversion: victims shown warnings who bought2.2 %
Liberty Reserve: alleged laundered funds6 billion USD
Compromised host price, US and UK100-180 USD per thousand hosts
Compromised host price, rest of worldless than 10 USD per thousand hosts
Share of pharmaceutical-affiliate payments by Visa and MasterCard95 %
Banks accepting payments for 95% of spam sales3 banks
ATM cash-out with duplicate cardsover 45 million in a matter of hours USD
Fake accounts back on sale after a platform purgewithin two weeks
MegaD botnet back after takedownwithin one month months
Advertised price, stolen card (2008)0.40-20.00 USD per card
Advertised price, bank account credentials (2008)10-100 USD per account
Advertised price, full identity (2008)1-15 USD per set
Phishing site life, known to the takedown firm vs unknown17 vs 112 hours (mean)
Spam fall after McColo was cut off (November 2008)almost 70 (temporary) %
Networks with three layers (core, enablers, mules)30 of 40 networks
Networks grown mainly through online forums10 of 40 (29 through offline ties) networks
Ransomware payments, 2013 to mid-2017 (lower bound, 35 families)12,768,536 (22,967.54 BTC) USD
Booter attacks after the HackForums market closurereduced for 13 weeks globally
Booter attacks after FBI Xmas2018down a third for at least 10 weeks
Countries nominated by at least one expert (World Cybercrime Index)97 countries
WCI overall score, Russia (rank 1)58.39 index (0-100)
Fraud and computer misuse share of the crime estimate, England and Wales, year to March 202153 %
Change on pre-pandemic: fraud / computer misuse / other crime (England and Wales)+24 / +85 / -19 %
IC3 complaints and losses, 2018351,937 complaints; 2.7 billion USD complaints; USD
Vulnerabilities across online devices, 2008-2014+17.75 %
Countries hit by ransomware, 2024153 countries
Budapest Convention ratifications (August 2025)81 states
UN Convention against Cybercrime: signatories; ratifications95; 3 (Qatar, Azerbaijan, Vietnam) states
Dutch panel surveyed on offender risk factors1,240 (aged 16-25) respondents

Open contentions (6) open contention

Where sources or readings disagree. Both stay on the map.

Partners and parts published science

Built from / builds (7)

Sits inside / holds (7)

Trades with (valence) (16)

expressed by (17)

floor filled by (19)

Kinds (18)

Notes published science

Layers

The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.

cybercrime as criminologists and security economists describe it: the computers, data, money and people online it preys on, up to the profits and purposes it serves (reader's view)

  1. L1: The world it preys on: computers, networks, data, money and people online
  2. L2: The offences: unauthorised access, damage, sabotage, interception, espionage; fraud, theft, extortion, illegal content, harassment
  3. L3: Support services: infrastructure, malware and kits, installs, credentials, human labour (CAPTCHA solvers, mules)
  4. L4: Markets and forums: the underground economy where profit and support centers trade, with escrow, feedback and reviewers
  5. L5: Networks and roles: core members, enablers, money mules
  6. L6: Cash-out and laundering: proceeds turned into usable money
  7. L7: Profit and purpose; cybercrime's place in all crime and its cost to society

the boardroom (shared floors)

  1. L4: The underground economy's exchange floor: profit centers selling to victims and support centers selling to other criminals, through forums and anonymous marketplaces (Silk Road sellers about $1.2 million a month in 2012; marketplaces later $300,000-500,000 a day)
  2. L5: Networks and roles: core members who start and steer attacks, enablers who sell to many networks, money mules who are easily replaced; few strict hierarchies, mostly not organised crime by the formal definitions
  3. L6: Value realisation: cashiers, mules, exchangers and laundering services turn proceeds into money usable outside (Liberty Reserve allegedly laundered $6 billion)
  4. L7: Profit and purpose, and cybercrime's place in crime: petty crime on a global and industrial scale; about half of property crime now online; offenders whose risk factors partly differ from traditional offenders'

way in: Defenders: filters, blocklists, takedowns

  1. L1: Defenders' actions: spam filtering, takedown contractors, account purges
  2. L2: Takedown speed: sites the contractor knew of gone in 17 hours on average; unknown ones lasted 112 hours
  3. L3: The answer: purged accounts back on sale within two weeks; fast-flux hosting moves a site every few minutes

way in: Exposed systems and their vulnerabilities

  1. L1: Computer systems and networks with exploitable weaknesses; vulnerabilities across online devices rose 17.75% from 2008 to 2014
  2. L2: Exploitation: unauthorised access, damage, sabotage, interception, espionage (the UN Congress's five categories)
  3. L3: A compromised host with extractable value: bandwidth, IP address, CPU, storage, stored credentials and card numbers

way in: Forums, reputation and trust

  1. L1: Public IRC channels and black market forums where services, prices and deals are posted; markets self-police
  2. L2: Listings and mandatory buyer feedback, read as a record of past sales
  3. L3: Vetting: trusted reviewers check new products; newcomers interviewed and asked for references
  4. L4: Position in the forum's conversation network: high betweenness comes before vendor success

way in: Hosting, domains and cloud

  1. L1: Hosting and domain providers, including negligent ones that ignore abuse reports (the Russian Business Network until 2007)
  2. L2: Concentrated hosting for control servers: cutting McColo off (November 2008) dropped world spam almost 70 percent, for a while
  3. L3: Cheap cloud mail and computing replace dedicated servers and skills

way in: Law enforcement: investigations, takedowns, arrests

  1. L1: Investigations that start from an IP trace and the logs ISPs must keep
  2. L2: Takedowns and arrests: Silk Road shut October 2013; Operation Onymous November 2014; HackForums' booter market closed; FBI Xmas2018 operation
  3. L3: Read as a signal: volumes dip, then rebound to other markets (a botnet back within one month; booter attacks down a third for at least 10 weeks)

way in: Laws, treaties and safe havens

  1. L1: National laws, weak or absent in some countries; offenders abroad avoid extradition
  2. L2: Location hidden: offenders mask where they are behind nicknames and technical protections
  3. L3: Treaty cooperation, two regimes: the Budapest Convention (2001; 81 ratifications by August 2025) and the UN Convention against Cybercrime (adopted December 2024)
  4. L4: Geography: a relatively small number of countries house the greatest threats (Russia, Ukraine, China, the US, Nigeria, Romania top the index)

way in: Offline social ties

  1. L1: Places people already know each other: the same neighbourhood, church, soccer club, prison
  2. L2: Recruitment through existing contacts; offline ties crucial to how networks start and grow
  3. L3: Online meeting places used even by offline-tied networks to recruit specialists
  4. L4: Forums: ten of forty networks grew mainly through them

way in: Payment rails

  1. L1: Banks, card networks, digital currencies and crypto currencies, all regulated to some degree
  2. L2: The bottleneck: few acquiring banks accept black-market payments (three banks for 95% of spam sales)
  3. L3: Bitcoin and pre-paid vouchers: ransoms of $100-400 (CryptoLocker); ransomware payments at least USD 12.8 million, 2013-mid 2017
  4. L4: Escrow held by the market: buyers pay the market, which pays the seller and takes its commission
  5. L5: Mules disguise the trail to core members

way in: People online as targets

  1. L1: People going about their online routines; routine activity theory predicts who is victimised (anyone who uses the Internet can be a victim)
  2. L2: Contact: scams, lures and harassment reach them; the computer is the tool, enlarging the pool of victims
  3. L3: A response that exploits human weakness: money paid, data handed over, a click (IC3: 351,937 complaints, $2.7 billion lost, 2018)

way in: Platform shifts

  1. L1: Platform change: mobile replacing the PC, Android replacing Windows, services moving to the cloud, social networks
  2. L2: New openings: cloud misconfiguration may cause as many breaches as phishing
  3. L3: New crimes: business email compromise, cryptocurrency crime

way in: Recruits and mules

  1. L1: Job ads in spam and on job sites offering work from home as a 'transaction processor'
  2. L2: Mules typically duped into accepting and forwarding stolen money
  3. L3: Easily replaceable; networks could use hundreds
  4. L4: The underground's labour market: affiliates paid by commission, free to move between programs

way in: Stolen data: cards, credentials, identities

  1. L1: Card and account data that businesses store for accounting and customer convenience; credentials people enter
  2. L2: Taken by breach, phishing or malware ('computer as target' as well as 'computer as tool')
  3. L3: Advertised in bulk to brokers: $0.40-$20 per card, $10-$100 per bank account, $1-$15 per identity (2008)

way in: The offender's own thought and emotion

  1. L1: The cyber places a person spends attention in
  2. L2: Movement: attention shifting from one cyber place to another (a cybercrime journey)
  3. L3: Online situational influences and risk factors (school, work, leisure, personality)
  4. L6: Different for cybercrime: higher self-control, protective against traditional offending, may open cybercrime opportunities

way in: Tools and labour supplied by other criminals

  1. L1: Code and capacity offered for sale: exploit kits, malware, botnets, pay-per-install, manual CAPTCHA solvers
  2. L2: Bought à la carte; installs priced by region ($100-180 per thousand US and UK hosts, under $10 elsewhere)
  3. L3: Combined with other bought components into a new scheme

way in: World events

  1. L1: Events that change daily routines (COVID-19 stay-at-home orders)
  2. L2: Routines online: not radically altered in early pandemic US data
  3. L3: Crime counts: in England and Wales fraud up 24% and computer misuse up 85% on pre-pandemic, other crime down 19%

way out: Cash-out into the legitimate economy

  1. L1: Proceeds spent in the legitimate economy
  2. L3: Mules forward stolen money; duplicate cards withdraw $45 million in hours
  3. L6: Laundering chosen: mules, exchangers, digital currencies

way out: Displacement: move market, method or country

  1. L1: New markets and hosts appear (Silk Road 2.0 by early November 2013)
  2. L3: Vendors and buyers move to other marketplaces; vendors keep shops on several; attackers shift across borders
  3. L4: Market participants read a takedown or block

way out: Extortion: ransomware and DDoS

  1. L1: Victims locked out; 153 countries hit by ransomware in 2024
  2. L2: Files encrypted or services flooded until paid; first documented ransomware 1989 (the AIDS trojan)
  3. L5: Decision to extort

way out: Illegal content and harassment

  1. L1: People harmed; laws differ by country (41 US states criminalise extreme online harassment)
  2. L3: Harassment, threats, offensive or illegal content
  3. L5: Content published or aimed at a person

way out: Recruitment

  1. L1: People drawn in, often duped
  2. L3: Job ads in spam and on job sites; forum recruitment
  3. L5: Networks need mules and specialists

way out: Sales to customers

  1. L1: Goods and services reach buyers: cannabis, ecstasy and cocaine products about 70% of marketplace sales
  2. L3: Spam-advertised products, fake antivirus, drugs shipped by post
  3. L4: Listings and storefronts

way out: Theft and fraud against victims

  1. L1: Money and data taken from people, firms and governments
  2. L3: Fraud executed: phishing, malware, bank fraud, carding, identity theft
  3. L5: A network commits to an attack on a financial institution or its customers (core members initiate)

Sources (23) published science

Caballero, Grier, Kreibich, Paxson 2011, Measuring Pay-per-Install: The Commoditization of Malware Distribution (USENIX Security)
Cybercrime (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01)
Soska K, Christin N (2015) Measuring the Longitudinal Evolution of the Online Anonymous Marketplace Ecosystem. USENIX Security 2015 (behind its links)
Thomas K, Huang DY, Wang D, Bursztein E, Grier C, Holt TJ, Kruegel C, McCoy D, Savage S, Vigna G (2015) Framing Dependencies Introduced by Underground Commoditization. WEIS 2015 (behind its links)
Christin N (2013) Traveling the Silk Road: A Measurement Analysis of a Large Anonymous Online Marketplace. WWW 2013 (arXiv:1207.7139) (behind its links)
Moore T, Clayton R, Anderson R (2009) The Economics of Online Crime. Journal of Economic Perspectives 23(3):3-20, doi:10.1257/jep.23.3.3 (behind its links)
Leukfeldt, Lavorgna & Kleemans 2017, Organised Cybercrime or Cybercrime that is Organised? An Assessment of the Conceptualisation of Financial Cybercrime as Organised Crime, European Journal on Criminal Policy and Research (open access), doi:10.1007/s10610-016-9332-z (behind its links)
Hadjichrysanthou C, Broom M, Rychtář J (2018) Models of kleptoparasitism on networks: the effect of population structure on food stealing behaviour. J Math Biol 76:1465-1488, doi:10.1007/s00285-017-1177-7 (behind its links)
Anderson, Barton, Boehme, Clayton, Ganan, Grasso, Levi, Moore and Vasek, Measuring the Changing Cost of Cybercrime, WEIS 2019 (behind its links)
Paquet-Clouston, Haslhofer & Dupont 2019, Ransomware Payments in the Bitcoin Ecosystem, Journal of Cybersecurity (arXiv:1804.04080) (behind its links)
Collier, Thomas, Clayton, Hutchings 2019, Booting the Booters: Evaluating the Effects of Police Interventions in the Market for Denial-of-Service Attacks, IMC 2019 (doi:10.1145/3355369.3355592) (behind its links)
Bruce M, Lusthaus J, Kashyap R, Phair N, Varese F (2024) Mapping the global geography of cybercrime with the World Cybercrime Index. PLoS ONE 19(4):e0297312, doi:10.1371/journal.pone.0297312 (behind its links)
Correia SG (2022) Making the most of cybercrime and fraud crime report data: a case study of UK Action Fraud. Int J Popul Data Sci 7(1), doi:10.23889/ijpds.v7i1.1721 (behind its links)
Budapest Convention on Cybercrime. Wikipedia (WME snapshot). (behind its links)
United Nations Convention against Cybercrime. Wikipedia (WME snapshot). (behind its links)
Bekkers LMJ, Moneva A, Leukfeldt ER (2026) Distinct group, distinct traits? A comparison of risk factors across cybercrime offenders, traditional offenders and non-offenders. Psychiatry, Psychology and Law 33, doi:10.1080/13218719.2025.2546311 (behind its links)
Patsakis, Arroyo & Casino 2024, The Malware as a Service ecosystem (book chapter, arXiv:2405.04109) (behind its links)
Phishing (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Khan S et al. (2022). A systematic literature review on cybercrime legislation. F1000Research, doi:10.12688/f1000research.123098.1 (PMC11384205). (behind its links)
Hawdon J, Parti K, Dearden TE (2020) Cybercrime in America amid COVID-19: the Initial Results from a Natural Experiment. Am J Crim Justice 45:546, doi:10.1007/s12103-020-09534-4 (behind its links)
Boekhout HD, Blokland AAJ, Takes FW (2024) Early warning signals for predicting cryptomarket vendor success using dark net forum networks. Sci Rep 14:16336, doi:10.1038/s41598-024-67115-5 (behind its links)
Moneva A, Ruiter S, Bernasco W (2026) The topology of cyberspace and cybercrime journeys: a framework for analyzing online offender mobility. Crime Science, doi:10.1186/s40163-026-00294-w (behind its links)
Ransomware (Wikipedia, via Wikimedia Enterprise snapshot) (behind its links)