Drug Smuggling Seizures: The Effects of Reporting Consistency and Quality on the Observed Transnational Structure

• Author: Gisela Bichler,Ivette Jimenez
• DOI: http://doi.org/10.1177/00220426221107550
• Published date: 01 January 2023
• Date: 01 January 2023
• Subject Matter: Articles

Article

Journal of Drug Issues

2023, Vol. 53(1) 159–179

© The Author(s) 2022

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/00220426221107550

journals.sagepub.com/home/jod

Drug Smuggling Seizures: The

Effects of Reporting Consistency

and Quality on the Observed

Transnational Structure

Gisela Bichler

1

and Ivette Jimenez

2

Abstract

If it is possible to overcome signif‌icant data challenges, social network analytics could be used to

expose structural vulnerabilities in transnational drug smuggling operations, offering clear targets

for crime control efforts that aim to disrupt transhipment. This study explores the extent to

which data inclusion decisions might distort the emergent structure of nation-to-nation smuggling

networks mapped with aggregate intelligence using United Nations Off‌ice on Drugs and Crime

(UNODC) incident level seizure data (2010–2016). Bivariate exponential random graph models

(ERGM) show that relaxing data inclusion standards exposes illicit backchannels (reciprocity) and

a more complete picture of major transhipment activity (activity and popularity spread) than

would be otherwise undetected. Relaxed data inclusion standards may help to adjust for the data

limitations associated with the detection of rare events and inconsistent reporting practices, if

usage rules are followed.

Keywords

data f‌ilters, drug seizures, drug smuggling, transnational crime, network analysis

Introduction

Tracing the movement of illicit drugs and precursor chemicals between nations offers an op-

portunity to understand the structure of transnational supply networks (e.g., Boivin, 2014a;2014b;

Giommoni, Berlusconi, & Aziani, 2021). Specif‌ically, investigating transit routes and smuggling

techniques can expose how asymmetric transportation and border control capabilities are ex-

ploited (e.g., Aziani, Berlusconi, & Giommoni, 2021;Caulkins, Burnett, & Leslie, 2009;

Fathurrohman & Bichler, 2021). Uncovering system vulnerabilities is pivotal to designing policy

1

School of Criminology and Criminal Justice, California State University San Bernardino, San Bernardino, CA, USA

2

Center for Criminal Justice Research, California State University San Bernardino, San Bernardino, CA, USA

Corresponding Author:

Gisela Bichler, School of Criminology and Criminal Justice, California State University San Bernardino, 5500 University

Parkway, San Bernardino, CA 92407, USA.

Email: gbichler@csusb.edu

and crime control strategies that disrupt or constrain drug smuggling operations, as well as other

illicit supply chains using the same mechanisms or channels.

Network approaches to detecting system vulnerabilities typically describe actor positions and

overall network characteristics, or test hypotheses about the f‌lexibility and resilience of drug

smuggling operations, using networks generated from police intelligence, investigative f‌iles, or court

records and transcripts (e.g., Bichler et al., 2017;Bright et al., 2014;Bright & Delaney, 2013;Duijn

et al., 2014;Morselli & Petit, 2007). While this body of work stands to advance efforts to identify

local patterns that characterize drug smuggling, scientists still do not fully understand how network

generation protocols intersect with source limitations to affect observations of supply chains. For

example, systematically missing information about interpersonal or inter-group connections can

obscure central f‌igures andthe complexity of their drug market involvement, adversely affecting the

impact of targeted crime control policy.

Responding to recent calls to better understand and contextualize f‌indings in consideration of

data constraints (Bright, Brewer, & Morselli, 2021;Divi´

ak, 2019;Faust & Tita, 2019), the current

study explores the extent to which data inclusion decisions might distort the emergent structure of

nation-to-nation smuggling networks mapped with aggregate intelligence using United Nations

Off‌ice on Drugs and Crime (UNODC) incident seizure data (2010–2016). Following a brief

review of criminal justice data constraints, this paper describes the data source and justif‌ies study

aims with an illustration. After describing the methods, the narrative reports two sets of results—

univariate effects and the emerging structural properties revealed by two sets of exponential

random graph models (ERGM). The discussion considers the implications of using data f‌ilters and

offers preliminary data usage rules and directions for future research.

Data Constraints

A social network framework could be used to expose structural vulnerabilities in international

drug smuggling operations, offering interdiction points for crime control efforts that aim to disrupt

transhipment (e.g., Giommoni et al., 2021;Calderoni, 2012). If, that is, it is possible to overcome

signif‌icant data challenges. Detracting from the applied potential of network analytics are the

related issues of data consistency and quality. Efforts to model social processes assume complete

network data is available (Wasserman & Faust, 1994;Knoke & Yang, 2008). That is, all relevant

actors are identif‌ied, and all relations among them observed. Incomplete network information

results in biased parameter estimates and may result in misdirected crime control policy. Missing

data is a signif‌icant challenge to network criminology because despite our best efforts to map

criminal networks, operations are covert. Actors function within a hostile setting, deploying tactics

to shield their activities from law enforcement and criminal competitors (Morselli, 2009). Thus,

some amount of missing data is inevitable.

As reported in Table 1, operational priorities, detection and investigatory capacity, and data

permanence and access, intersect with methodological issues to raise serious concerns about the

completeness and accuracy of observed drug smuggling networks mapped from intelligence data,

particularly when data are extracted from a single jurisdiction. These data constraints can pose

challenges to the advancement of networked criminology by limiting our ability to use network

models to theorize about social processes that shape criminal behavior (Bright et al., 2021;

Campana & Varese,2012;Divi ´

ak, 2019;Faust & Tita, 2019). Though not expressly discussed in a

recent systematic review of applied network analysis (Bright et al., 2021), transactional data are

also likely to raise similar concerns regarding sampling, actor and tie identif‌ication, directionality

of activity, reproducibility, and data accessibility (e.g., D´

ecary-H´

etu, Paquet-Clouston, & Al-

dridge, 2016;Duxbury & Haynie, 2018;Norbutas, 2018).

1

160 Journal of Drug Issues 53(1)