Research paper
Transnational cocaine and heroin flow networks in western Europe: A comparison

https://doi.org/10.1016/j.drugpo.2015.04.016Get rights and content

Highlights

  • Country-to-country cocaine and heroin flow networks in western Europe are compared.

  • The cocaine network is more compact with fewer structural redundancies than the heroin network.

  • This may be because the source of cocaine is focused while heroin has multiple sources.

  • Interdiction efforts will need to be tailored to these structural differences.

  • The methods of this paper can be applied to other locations and drugs.

Abstract

Background

A comparison of the properties of drug flow networks for cocaine and heroin in a group of 17 western European countries is provided with the aim of understanding the implications of their similarities and differences for drug policy.

Methods

Drug flow data for the cocaine and heroin networks were analyzed using the UCINET software package. Country-level characteristics including hub and authority scores, core and periphery membership, and centrality, and network-level characteristics including network density, the results of a triad census, and the final fitness of the core-periphery structure of the network, were computed and compared between the two networks.

Results

The cocaine network contains fewer path redundancies and a smaller, more tightly knit core than the heroin network. Authorities, hubs and countries central to the cocaine network tend to have higher hub, authority, and centrality scores than those in the heroin network. The core-periphery and hub–authority structures of the cocaine and heroin networks reflect the west-to-east and east-to-west patterns of flow of cocaine and heroin respectively across Europe. The key nodes in the cocaine and heroin networks are generally distinct from one another.

Conclusion

The analysis of drug flow networks can reveal important structural features of trafficking networks that can be useful for the allocation of scarce drug control resources. The identification of authorities, hubs, network cores, and network-central nodes can suggest foci for the allocation of these resources. In the case of Europe, while some countries are important to both cocaine and heroin networks, different sets of countries occupy positions of prominence in the two networks. The distinct nature of the cocaine and heroin networks also suggests that a one-size-fits-all supply- and interdiction-focused policy may not work as well as an approach that takes into account the particular characteristics of each network.

Introduction

Cocaine and heroin users account for roughly 39 million to 55.5 million people worldwide (United Nations Office on Drugs and Crime (UNODC), 2012). Despite increased law enforcement activities, consumption of these two drugs continued unabated (Wiessing, Olszewski, Klempová, Vicente, & Griffiths, 2009). At a time when drug policy is increasingly taking an integrated turn and policy and enforcement organizations are increasingly adopting strategies that cut across drug types, the comparative study of drug-related phenomena such as trafficking is becoming increasingly important (European Union, 2012). The need for such analysis is compounded by the phenomenon of drug trafficking organizations diversifying their portfolios from a single drug into multiple drug categories, as in the case of Mexico's Sinaloa cartel, which now deals in cannabis, heroin, and methamphetamine (Keefe, 2012).

The magnitude of the illegal drug trade and its resulting problems have led governments and the drug policy research community to invest in producing large and comprehensive datasets on a variety of phenomena relating to this traffic (European Monitoring Centre for Drugs and Drug Addiction, 2008, UNODC, 2013, United States Department of Justice, 2014). These datasets contain information on prices, seizures, and flows of drugs, which lend themselves to a variety of interesting and potentially valuable analyses. Yet, with a few recent exceptions (Boivin, 2013, Boivin, 2014a, Boivin, 2014b), given the quantity and content of the available data, surprisingly little systematic research has been conducted at the country level that applies the methods of network analysis to the available data on these drugs. The main objective of this paper is, therefore, to use network analysis to compare the properties of drug flow networks for cocaine and heroin in a group of 17 western European countries, inferred using data on wholesale prices for these drugs, with a view to understanding their similarities and differences and to examining the implications of these findings for drug policy. Data permitting, the analytic methods presented in this paper can be applied to the analysis of other illegal drugs, including cannabis and amphetamine-type stimulants (ATS), in particular, and to other illegally trafficked goods and markets, including but not restricted to those for weapons, people, and wildlife.

The network approach to the analysis of drug flows is potentially rewarding on a number of levels. At the level of a single network, it can provide insights into how and why the network displays a particular set of structural properties, with implications for drug control policies including resource allocation and interdiction. Notions including but not restricted to centralization, core–periphery structure, path-redundancy, and membership can be brought to bear on these issues. At the level of an individual country, network analysis can identify the position and importance of each country in a network and the specific paths through which a drug enters, transits through, or exits a specific country. In doing so, such analysis can inform strategies for interventions by providing insights into locations in the network where the allocation of resources is most likely to be effective, both at the country level and at the level of the entire network. Comparisons of networks can likewise shed light on the degree to which an integrated approach to drug control is likely to work, and the degree to which customization for specific drugs and trafficking routes is warranted. Comparisons of the positions of specific countries in different trafficking networks can yield insights into whether the trade in one drug operates in a similar manner to the trade in another drug, which in turn can offer clues about the degree to which trafficking of one drug uses the same infrastructure as another drug.

Section snippets

Literature review

In this section, we briefly describe two literatures on which this study is based. The first is the literature on drug markets, prices, price data, and the rich variety of information that is contained in these data – the network data on which this study is based are constructed using drug price data. The second literature is the relatively small body of work that quantitatively or qualitatively analyzes the organizational structure of drug networks, using concepts from fields including network

Analytic approach

The data used in this paper represent two networks of transnational drug flows (i.e., cocaine and heroin) for 17 countries of western Europe. The most common visualization of a network is that of a collection of nodes (or vertices), representing countries, and lines or ‘edges’ that connect nodes, representing drug flows between pairs of countries. While network flows can in theory be bi-directional (i.e., country A sends to country B and vice versa), our analysis adopts the UNODC convention in

Results

The results of the triad census, shown in Table 1, reveal some interesting contrasts between the structure of the heroin and cocaine networks in Europe. The cocaine network contains many more empty triads and fewer triads with one or more connections than does the heroin network. In other words, there are many more paths for the flow of heroin through Europe than for cocaine. Interestingly, there is one type of non-empty triad, the directed line (see Fig. 1), for which the cocaine network

Discussion and conclusions

In light of the size of the cocaine and heroin problem in Europe and worldwide and the integrated nature of drug policy making and research, the comparative study of drug flow networks can provide important insights into issues of relevance to drug control efforts. Methods of network analysis originally developed in the fields of engineering and sociology can capture and characterize important features of networks that may not be visible upon casual observation. For example, in this study, it

Conflict of interest

The authors declare that there is no conflict of interest.

References (62)

  • S.P. Borgatti et al.

    UCINET for windows: Software for social network analysis

    (2002)
  • J. Caulkins et al.

    What price data tells us about drug markets

    Journal of Drug Issues

    (1998)
  • J.P. Caulkins et al.

    How drug enforcement affects drug prices

    Crime and Justice

    (2010)
  • J.P. Caulkins et al.

    Marijuana price gradients: Implications for exports and export-generated tax revenue for California after legalization

    Journal of Drug Issues

    (2012)
  • S. Chandra et al.

    What the price data tell us about heroin flows across Europe

    International Journal of Comparative and Applied Criminal Justice

    (2013)
  • S. Chandra et al.

    Inferring cocaine flows across Europe: Evidence from price data

    Journal of Drug Policy Analysis

    (2011)
  • S. Chandra et al.

    How powdered cocaine flows across the United States evidence from open-source price data

    Journal of Drug Issues

    (2014)
  • P.A. Chouvy

    Opium: Uncovering the politics of the poppy

    (2010)
  • P. Clawson et al.

    The Andean cocaine industry

    (1996)
  • K.W. Clements

    Pricing and packaging: The case of marijuana

    Journal of Business

    (2006)
  • J. Csete et al.

    Telling the story of drugs in West Africa: The newest front on a losing war

    (2013)
  • L. De Benedictis et al.

    The world trade network

    The World Economy

    (2011)
  • S.H. Decker et al.

    Drug smugglers on drug smuggling

    (2008)
  • F. Desroches

    Research on upper-level drug trafficking: A review

    Journal of Drug Issues

    (2007)
  • European Monitoring Centre for Drugs and Drug Addiction

    Monitoring the supply of heroin to Europe

  • European Union

    EU drugs strategy 2013–20

    Official Journal of the European Union

    (2012)
  • M.G. Everett et al.

    The centrality of groups and classes?

    The Journal of Mathematical Sociology

    (1999)
  • G. Farrell et al.

    Cocaine and heroin in Europe 1983–1993: A cross-national comparison of trafficking and prices

    The British Journal of Criminology

    (1996)
  • P. Gootenberg

    Andean cocaine: The making of a global drug

    (2008)
  • R.A. Hanneman et al.

    Introduction to social network methods

    (2005)
  • S. Howell et al.

    West Africa and the transnational trade in illegal drugs: Physical properties, policing, and power

    Africa Review

    (2015)
  • Cited by (22)

    • Review of the most common chemometric techniques in illicit drug profiling

      2019, Forensic Science International
      Citation Excerpt :

      Additionally, at an international level, network analysis could identify countries that are pivotal to drug markets. This could aid the suggestion of strategic targets for disrupting international drug trade [145]. It is possible to receive an indicator of how well implemented strategies are disrupting drug markets by analysing the extent of certain CCs at a given time.

    • The determinants of heroin flows in Europe: A latent space approach

      2017, Social Networks
      Citation Excerpt :

      The results show that some characteristics of the importer (e.g. risk of arrest), when viewed in conjunction with the position of the country within the global trafficking network, can predict wholesale price mark-ups (Boivin, 2014b). Chandra and colleagues provide the second key contribution to the network analysis of drug trafficking flows (Chandra et al., 2011; Chandra and Barkell, 2012; Chandra and Joba, 2015; Chandra et al., 2016). The authors use wholesale prices reported by the UNODC between 2000 and 2008 to infer the presence and direction of transnational flows of heroin (Chandra and Barkell, 2012) and cocaine (Chandra et al., 2011) for seventeen western European countries.

    View all citing articles on Scopus
    View full text