Organized Crime Research Highlights Number 2

Criminal cooperation between organized crime groups

Empirical research using Canadian data shows that criminal organizations are ethnically and culturally diverse.

Most research on offenders offending together (co- offending) in criminal organizations is concerned with individuals involved in illicit markets. This study is a first step towards understanding co-offending at the group level; it uses a research methodology known as social network analysis to examine whether criminal organizations have a tendency to co-offend only with ethnically and sub-culturally similar criminal organizations.

The data used in this study were drawn from the 2007 RCMP 'E' Divison Provincial Threat Assessment, which includes data on criminal organizations involved in organized crime between 2004 and 2006 in British Columbia and the Yukon Terrority. This research used the following widely-used categories of recognized criminal organizations: Asian-Canadian, Eastern, European, Italian, Hispanic, Middle-Eastern, Outlaw Motorcycle Groups (OMG), street gangs, and unclassified groups.

The authors found that OMG and members of unclassified criminal organizations had the lowest co-offending rates with other members of a similar category. Asian- and Indo- Canadian crime groups had the highest rates of co- offending within their own category of criminal organization.

Yet, compared to the other groups examined, Asian- and Indo-Canadian groups have significantly higher co-offending rates with people who had not been not assigned to a recognized criminal organization when they were admitted to prison. Interestingly, individuals without known group membership comprised 43% and 33% of the co-offending networks in Asian and Indo-Canadian criminal organizations, respectively.

The authors suppose that individuals who were not identified as members of certain criminal organizations in the threat assesment may have been excluded because they did not fit into the ethnic or sub-cultural categories which had been commonly used by law enforcement.

OMG share offenders with other chapters of their organization as well as with non-OMG organizations. The authors posit that OMG's tendency to work with a variety of criminal organizations makes them particularly resilient to having their criminal business disrupted by law enforcement efforts.

The authors suggest that this research supports the contention that an ethnically-based classification of criminal organizations may be misleading, a finding that Organized Crime Rresearch Highlights Number 2 is being recognized in the law enforcement community. This research is subject to certain limitations. In particular, law enforcement threat assessments were used to identify co-offending activity, and co-offending was measured by co-arrest. As such, some co-offending within criminal organizations was not recorded or used in the analysis.

Malm, Aili, Gisela Bichler and Rebecca Nash. (2011) "Co-offending between criminal enterprise groups." Global Crime, vol. 12 no. 2, 112-128.

Do all gangs pose equal security risks within correctional facilities?

Data from Canadian federal penitentiaries show that not all gangs pose equal risks to correctional systems.

Previous research has consistently shown that gang members threaten safety and security in correctional systems because of their involvement in misconduct and violence. The current study examines whether all gangs pose the same types of risks or threats to correctional systems.

Correctional Services Canada (CSC) data on 1,636 gang members and 1,636 randomly selected non-gang members admitted to Canadian federal penitentiar ies between January 1, 2006 and August 31, 2009 were examined. For the purposes of this research, gangs were classified into five broad groups utilized by CSC officers, including: Aboriginal, Asian, Outlaw Motorcycle Groups (OMG), street, and traditional organized crime (TOC).

Overall, street gang members have higher levels of risks and needs than their non-gang counterparts, as well as greater previous involvement in drug and property offences, and greater likelihood of having previously been incarcerated. Taken together, these attributes tend to result in the placement of gang members into facilities or units with higher security levels.

Street gangs had the highest average rates of institutional misconduct.

Aboriginal, Asian, and OMG gangs were closely matched, each with slightly less average rates of misconduct compared to street gangs. Interestingly, TOC members had the lowest average rate of institutional misconduct – a rate that was one third less than that of non-gang members. Anecdotal evidence from corrections officials suggest that these data only show part of the picture – older and more established gang members, such as TOC gang members, may hire other offenders to carry out misconduct on their behalf.

Non-gang members were less involved in institutional misconduct, but were in minor incidents of victimization to about the same extent as gang members.

Aboriginal and street gangs had the highest overall proportion of members involved in at least one major incident (40.1% and 39.9%, respectively), followed by Asian (33%), OMG (21.7%), and TOC (5.2%) members.

Like all research, this study is subject to certain limitations. Some research has shown that gang members are more likely to receive a misconduct compared to non-gang members for committing similar acts, so there is a potential for bias in the data itself. Further, there is likely variation regarding the risk that certain gangs pose between penitentiaries across Canada. Finally, gang membership is likely undercounted, due to its secretive nature.

Ruddell, Rick and Shannon Gottschall. (2011). "Are all gangs equal security risks? An investigation of gang types and prison misconduct. " American Journal of Criminal Justice, vol. 36, 265-279.

Motives and methods for leaving a gang

The easiest way to leave a gang is for marriage, family, or employment reasons. People who left a gang for reasons external to the gang did not experience violence upon leaving the gang.

Using a sample of 84 recently arrested young offenders in Arizona who had left their gang, this research examines the motives and methods for leaving a gang as well as the social ties that remain after an individual is no longer a gang member.

Generally speaking, criminal offending tends to peak in the late teens, and decline thereafter. Previous research has suggested that changing peer relations, such as getting married, can account for leaving a life of crime.

A study of former St. Louis gang members suggests that there are two ways to leave a gang: abruptly or through gradual desistance from the group (Decker and Lauritsen 2002). The gradual changes discussed in the research occurred over time, with gang members slowly developing new ideas and commitments, and identifying less and less with the gang. The abrupt changes documented in that research were often associated with the individual physically leaving the neighbourhood and/or the city. Research in Los Angeles observed that it was not uncommon for members to be "beaten out" of the gang (Virgil 1988), that is, to be physically assaulted by other gang members as an exiting ritual.

On average, respondents to the current research indicated that they had left the gang 20 months ago. 'Leaving the gang' was defined through self- identification. Interestingly, 35% had nonetheless experienced violent victimization in the 30 days prior to their arrest, and 33% had been arrested for a serious offence. The authors posit that "leaving the gang is not necessarily tied to exiting precarious social environments" (421).

The most frequently cited motive and method for leaving the gang was to 'just walk away' without incident in order to leave the gang lifestyle. About two out of three respondents indicated they left the gang to avoid the violence and trouble associated with gang membership; the remaining youth indicated that they left for reasons external to the gang, such as family or employment. "Youth leave for their own motivations and do not have to be coerced or persuaded to exit" (421).

About one out of every five gang exits were hostile – that is, they involved gang members being "beaten out" or the member had to commit a crime before exiting. "The relationship between hostile departure and separation may be a function of leaving gangs at different developmental stages, where non-hostile methods are treated differently at younger ages" (421). Of the 26 respondents who reported that they had left the gang for reasons external to the gang – such as marriage, family or employment reasons – none experienced a hostile departure. On the other hand, nearly one-third of respondents who left the gang because of reasons internal to the gang – such as growing tired of the violence or trouble – experienced a hostile departure. The authors contend that this shows that "the gang "understands" when important life events arise, and the gang does not respond to those life events in a hostile manner" (422).

On average, the respondents indicated at least two social and/or emotional attachments to the gang after the separation. The authors suggest that this implies that "de-identifying as a gang member does not mean disassociating with the gang" (421).

The generalizability of this research is limited insofar as it is based on the particular experiences of young people in Arizona.

Pyooz, David C. and Scott H. Decker. (2011). "Motives and methods for leaving the gang: Understanding the process of gang desistance. " Journal of Criminal Justice 39, 417-425.

Who stops dealing hard drugs?

Reducing a drug dealer's personal level of addiction or their capacity to obtain supplies may encourage them to stop dealing.

"Street-based drug dealing involves a high risk of violence, including gun-related violence and homicide, because of a lack of formal dispute resolution mechanisms and the substantial profits associated with this trade" (459). The lack of formal dispute resolution invites organized crime groups to fulfill this function. In addition, high profits and strong demand attract organized criminal offending to coordinate supplying illicit commodities.

The analysis for this study was conducted with data obtained from a prospective cohort of injection drug users and street-involved youth in Vancouver between 2005 and 2009. The study collected data using questionnaires at six-month intervals.

At the start of the study period, 381 drug users reported dealing drugs, 194 (51%) of whom ceased dealing drugs during the two and a half year study. Dealers primarily reported selling crack cocaine (272, 71%), heroin and/or methadone (183, 48%), and cocaine (163, 43%). The median hours per week spent dealing was 15.

Being willing to stop selling drugs was associated with older age, recent crack use, recent public injection of drugs, and the impression that police presence affects the location of drug purchases. However, these findings were only true where the income from drug dealing was not needed to fund personal drug use. Recently injecting crystal methamphetamine was associated with not being willing to stop dealing drugs.

Interestingly, being stopped or detained by police, committing or being the victim of physical assaults, being involved in the commercial sex trade, or residency in Vancouver's downtown eastside neighborhood were not associated with people being more or less willing to stop dealing drugs.

The "cessation of drug dealing was associated with spending less than $50 per day on drug use and not having a regular source for drug purchase" (461). While dealers might report that police presence at drug purchase locations increases their desire to stop drug dealing, the authors identify the intensity of drug dependence as the over-riding factor in people actually giving up street-level dealing. A person who is drug addicted and well-connected to suppliers is not very likely to leave street-level drug dealing.

The authors "conclude that drug policies targeting street - based drug dealing should focus on the provision of addiction treatment in tandem with structural interventions" (462). It is implied that addiction treatment would focus on individual dealer-users. Presumably, 'structural interventions' could include the periodic restriction of the drug supply through wholesale-level seizures and/or the disruption of wholesale drug distribution networks by removing brokers to street-level dealers, intending to have an adverse impact on street- level dealers' access to regular sources of supply.

The study was not structured to address the question of the overall supply of drug dealers in the illicit marketplace or the volume of drugs sold by dealers. Although there is a suggested intervention to increase the rate at which street-level dealers cease selling drugs, the authors do not comment on how to reduce the overall number of street-level dealers or the policies best-suited to restricting the overall supply of street-level drugs available for consumption.

Werb. Dan, Martin Bouchard, Thomas Kerr, Jean Shoveller, Jiezhi Qi, Julio Montaner and Evan Wood. (2011) "Drug dealing cessation among a cohort of drug users in Vancouver, Canada." Drug and Alcohol Dependence, no. 118, 459-463.

Using wiretap data to understand criminal organizations

Wiretapped phone conversations gathered in law enforcement investigations can offer a wealth of data for understanding criminal networks.

The use of social network analysis in the field of criminology has rapidly expanded over the past few years, being used specifically in the context of organized crime research to gain a deeper understanding of the structures of large criminal networks. Like conventional statistics, social network analysis is a quantitative technique, and as such certain constraints are imposed upon the data used to ensure its quality and suitability for generating valid findings.

The authors lay out a set of formal criteria for the use of wiretap data in social network analysis for the study or investigation of organized crime.

While wiretapped telephone conversations, as a record of who has spoken with whom, and about what, can be immensely useful for understanding the workings of criminal organizations, the authors note that certain minimum standards must be met in order to use this type of data for quantitative analysis. They outline three criteria which they believe should guide the use of wiretap data in analytical studies.

First, there should be no self-censorship in the recorded conversations. Specifically, this means that those speaking with each other do talk about the criminal activities in which they are engaged. Lack of self- censorship is not the same as speaking in code, and even coded conversations can be used as long as the code has been decrypted. However, there should be no crimes relevant to the organization omitted from the conversations between its members.

Second, there should be reasonably wide coverage of the group, with all key members having been put under surveillance. Observing the entire network, and including all of the key members ensures that social network statistics do not give misleading results.

Third, the data should be made up of a large number of conversations over a reasonably long period of time. This helps to ensure that all key members of the group would be included, as well as allowing for the extraction of random samples of conversations which can be used to discover actor attributes such as their specific role in the network.

If these criteria are met, a number of useful techniques can be applied to the data from wiretap investigations collected during the course of law enforcement investigations, including social network analysis, content analysis, correspondence analysis and, in certain cases, longitudinal stochastic actor-oriented modeling. Each of these techniques can give us important information on the structure and composition of the network.

Using content analysis, which identifies important aspects of written communications, the key activities and their relative importance to the network can be discovered, as well as features of network members and who is engaged in which activities. For example, using content analysis, it is possible to determine which members of a network are responsible for issuing orders or giving advice, possibly indicating the rank of members of the network.

Correspondence analysis, which describes the similarities between individuals on a number of characteristics, can determine whether members of the network are similar based on their roles or activities, and if so, whether the network has an internal division of labour. For instance, using correspondence analysis, one could determine whether there is a difference in offending patterns between higher or lower ranked members of a criminal network.

Finally, longitudinal stochastic actor-oriented modeling can be used to answer questions about the effects of individual characteristics, such as age or prior offence history, on the likelihood of contacts between individuals, and can also be used to assess the evolution of the network over time. This technique, which looks at how networks change over time, could be used to examined changes in a criminal network when individuals are removed, such as through arrest as a result of a law enforcement investigation.

Campana, Paolo and Federico Varese. (2011) "Listening to the wire: criteria and techniques for the quantitative analysis of phone intercepts " Trends in Organized Crime, published online 26 April 2011.

Effectively dismantling a drug ring

It is more effective to disrupt a drug ring by eliminating well-connected members of the network than by focusing on the jobs of the criminals in the operation.

This study used a number of innovative, quantitative methods for testing the effectiveness of different law enforcement strategies aimed at dismantling criminal networks.

Policing resources are limited and prosecuting criminal organization cases is difficult, lengthy and expensive. Focusing on the most critical people to remove from a criminal network to maximize the disruption of their illegal activities is important.

In this study, sentencing records were used to construct a social network map of a criminal group that produced and trafficked methamphetamines in Australia during the 1990s. The researchers identified the role that each individual played within the syndicate, as well as the nature of the connections between people.

Four enforcement scenarios on the mapped criminal network were examined, which targeted individuals for removal based on: 1) the connectedness of people to others in the network (degree attack); 2) the role or job of the person in the criminal enterprise (weight attack); 3) a combination of first two (mixed attack); and 4) randomly choosing people in the network (random attack).

Most previous work on disrupting criminal networks using social network analysis (SNA) has focused on the position of individuals in the structure of the social network. Previous analyses of degree have found that although both attacks on "hubs" (people connected to large numbers of others) and "bridges" (people that connect groups of unconnected people) were effective in dismantling networks, attacks on 'bridges' were the most effective.

This research also examined the role or job of individuals, since certain people may be critical to a criminal enterprise without knowing many people in the network. For instance, a single producer might only be connected to many traffickers through a manager or a single trafficker, but still be the sole supply of the drug for the entire network. An innovation of the method used in this study was the creation of a numerical weight measuring the replacability of each person in the criminal enterprise based on the number of people fulfilling the same role. For instance, two out of the 40 people in the studied network were managers, while 20 of the 40 people were workers or labours. Managers were assigned a weight of 1/2 and workers 1/10, since managers were five times harder to replace from existing people in the network. "The weight measures how difficult it might be to replace that individual, were they to be removed from the network." (11)

Each strategy was applied by sequentially eliminating the most attractive target under the given attack scenario, measuring the disruption that the action had on the network. Disruption was measured by examining the number of people remaining in the largest connected component of the current network, as well as by a dding up the role weights of the remaining people in the largest connected component of the network.

As a result of the analysis, the researchers found that law enforcement strategies that target criminals in a network based on how they are connected to other people in the network is a very effective disruption strategy. Marginally more effective is applying interventions that also consider the role of people in the criminal network in addition to connectivity. Interestingly, targeting people for removal from the network just based on their role in the criminal network was not effective.

"To make cost-effective arrests, law enforcement agencies" need to invest more resources in gathering "quality intelligence, sophisticated SNA, and interventions targeted at vulnerable areas" of criminal networks (23).

The researchers commented that the law enforcement of organized drug crimes could become more cost -effective if their analytic methods were applied. Further research will need to be undertaken to assess whet her the application of these methods are more cost-effective than alternative strategies.

This study contributes to possible ways to evaluate law enforcement operations and network disruption. As the researchers note, "there may be political pressure on law enforcement agencies to seize drugs and money, make arrests, etc., as indicators of success, rather than to engage in prolonged intelligence gathering, costly investigations, and interventions designed to [effectively] dismantle a criminal network." (23) The researchers note that, despite recent work by groups such as the RCMP on developing a Disruption Attribute Tool (DAT), "there are no clear and accepted [publicly published] methodologies which measure the effectiveness of law enforcement interventions aimed to dismantle criminal networks." (24) This study contributes to on-going efforts to develop "alternative performance indicators which can accurately reflect … law enforcement goals such as the dismantling of criminal networks." (24)

This research was only designed to address the goal of disrupting the ability of an observed criminal network from being effective in its objective through removing actors from the network. Other possible goals of law enforcement intervention in a criminal network exist. "For example, the aim may be to incapacitate the network so that the groups can no longer act illicitly, or to breach trust within the network such that the network disintegrates via internal distrust and conflict." (25)

Bright, David, Catherine Greenhill and Natalya Levenkova. (2011) "Dismantling criminal networks: Can node attributes play a role?" Presented at the Illicit Networks Conference 2011, Montreal, Canada.

Organized crime involvement in the illicit antiquities trade

"The antiquities trade involves local, small-scale thieves, larger groups of looters, and internationalconnections with auction houses, galleries, museums, dealers and collections." (4)

The trade in antiquities is a gray market, involving both licit and illicit activities. There are legal and cultural disputes about who has the right to remove items from archaeological sites; further, objects are removed from known as well as unknown sites – often items will not yet be documented, compounding the difficulty in regulating the trade. The drivers of looting are poverty and a lack of legitimate economic opportunities.

How involved are criminal organizations in the illicit antiquities trade? This research focuses on the supply side of the trade, examining "the problem of theft from archaeological sites from the perspective of archaeologists" (6). This study involved the electronic distribution of a structured questionnaire to nearly 15,000 archaeologists from around the world, and yielded a (good) response rate of 16% (N=2,358). Respondents were asked if they would participate in a follow-up survey, 28% (N=662) of whom agreed, and took an additional survey or participated in a telephone interview.

Most respondents (73.4%) indicated that they had observed looting during their time in the field; 68.2% had seen holes or other signs of damage to the site, 42% had objects stolen from their sites, and 24.1% had encountered looters while the looters were looting. Interestingly, most respondents (87.1%) also reported off-site encounters with admitted or suspected looters. "Still other archaeologists reported having been brought looted objects by the looters themselves, who appeared to expect to have their collections inspected, admired, or even appraised." (13)

Archaeologists from outside North America were less likely to think that criminal organizations were involved in looting and trafficking where they personally work, compared to their counterparts working within North America.

Archaeologists were asked about their general impressions of organized crime involvement in the global antiquities market, the "overwhelming majority" of whom "perceive such [criminal] groups to be involved in not only the movement of looted material but the actual looting of archaeological sites" (14). At the same time, paradoxically, only a small majority of respondents believed organized crime groups were involved with the looting and trafficking of antiquities in their locality. The author attributed this disconnect to stereotypical conceptions of what constitutes an organized crime group. Looting and trafficking in antiquities involves loosely-networked individuals, groups that would constitute a criminal organization by recent accepted definitions of the term.

The study did not specifically discuss the looting or trafficking of artifacts within the Canadian context.

Proulx, Blythe Bowman. "Organized Criminal Involvement in the illicit antiquities trade." Trends in Organized Crime 14 (2011):1-29.

For more information on organized crime research at Public Safety Canada, please contact the Organized Crime Research Unit at

Organized Crime Research Highlights is produced for Public Safety Canada. Organized Crime Research Highlights supports the objectives of the National Research Agenda on Organized Crime by highlighting topical, evidence-based information found in scholarly organized crime articles. The summaries herein reflect interpretations of the report authors' findings and do not necessarily reflect those of the Department of Public Safety Canada.

Date modified: