Illicit Network Dynamics: The Formation and Evolution of a Drug Trafficking Network

David Bright, Johan H. Koskinen, Aili E. Malm

Research output: Contribution to journalArticlepeer-review

59 Citations (Scopus)


Objectives: The project aims to: (1) investigate structural and functional changes in an Australian drug trafficking network across time to determine ways in which such networks form and evolve. To meet this aim, the project will answer the following research questions: (1) What social structural changes occur in drug trafficking networks across time? (2) How are these structural changes related to roles/tasks performed by network members? (3) What social processes can account for change over time in drug trafficking networks? 

Method: The relational data on the network was divided into four two years periods. Actors were allocated to specific roles. We applied a stochastic actor-oriented model to explain the dynamics of the network across time. Using RSiena, we estimated a number of models with the key objectives of investigating: (1) the effect of roles only; (2) the endogenous effect of degree-based popularity (Matthew effect); (3) the endogenous effect of balancing connectivity with exposure (preference for indirect rather than direct connections); (4) how degree-based popularity is moderated by tendencies towards reach and exposure. 

Results: Preferential attachment is completely moderated by a preference for having indirect ties, meaning that centralization is a result of actors preferring indirect connections to many others and not because of a preference for connecting to popular actors. Locally, actors seek cohesive relationships through triadic closure.

Conclusions: Actors do not seek to create an efficient network that is highly centralized at the expense of security. Rather, actors strive to optimize security through triadic closure, building trust, and protecting themselves and actors in close proximity through the use of brokers that offer access to the rest of the network.

Original languageEnglish
Pages (from-to)237-258
Number of pages22
JournalJournal of Quantitative Criminology
Issue number2
Early online date29 Mar 2018
Publication statusPublished - 1 Jun 2019


  • Drug trafficking
  • Social network analysis
  • Longitudinal social network analysis
  • Stochastic actor-oriented models
  • Dynamic


Dive into the research topics of 'Illicit Network Dynamics: The Formation and Evolution of a Drug Trafficking Network'. Together they form a unique fingerprint.

Cite this