Illicit drug trade on dark-web markets: Unveiling the causal mechanisms that are of influence on the globalness of drug trade

Institute for Security & Global Affairs

Leiden University – Faculty of Governance & Global Affairs

Illicit drug trade on

dark-web markets

Unveiling the causal mechanisms that are of

influence on the globalness of drug trade

3-3-2019

Master Thesis Crisis and Security Management

I

Master thesis

Program Master Crisis & Security Management Student Joep Berendsen

Student Number S2013001 Date of admission 3-3-2019

Subject Illicit drug trade on dark-web markets

Word count 22560 words excluding bibliography, appendices, transcripts 62863 words including bibliography, appendices, transcripts First reader Dr. J. Matthys Assistant Professor at Leiden University Second reader Dr. G.M. van Buuren Assistant Professor at Leiden University

II Foreword

In this master thesis I researched the characteristics of dark-web illicit drug trade markets to unveil which causal mechanisms are of influence on the globalness of those markets. The globalness of dark-web markets refers to the extent to which these markets are used for local or global trade. I was motivated to research this topic because of the many news articles I have read about these markets and the dark-web in general.

This research adds to the body of knowledge by elaborating how local or global these markets are and by giving an insight into the characteristics and causal mechanisms influencing this variable. Authoring a thesis on dark-web markets can be difficult since the body of knowledge is relatively small and there are just a few experts. However, these challenges are stepping stones to add to the body of knowledge and informing policy makers on how to regulate this new form of drug trade. Moreover, dark-web markets are interesting study-objects since criminals are ‘hiding in the open’. This is facilitated by technology such as a VPN blocker, a Tor browser, and encryption tools. This development could transform international illicit markets and our ability to regulate those markets. This research will hopefully provide both academics and policy makers with more knowledge in order to better understand the complex dynamics of dark-web markets.

I really appreciated researching this phenomenon and I would like to thank my first reader Dr. Joery Matthys for giving me the chance to start this interesting and challenging thesis project. Thank you for allowing me to prove myself by conducting this research and for giving helpful feedback.

Secondly, I would like to thank all interviewees, and the individuals in my network that helped me reach these participants. These interviewees enabled me to answer the research questions and build a causal framework. They really took their time to help me and showed genuine interest in my work.

Finally, I would like to thank my roommates, friends, and family for supporting me and my co-students for reading my paper and providing feedback.

Joep Berendsen 03-03-2019

III Summary

Background: Dark-web markets are illegal online platforms that facilitate drug trade. These

markets have been analysed by scholars that add to a still growing body of knowledge. Previous researchers have started the discussion on the globalness of dark-web markets. Indicating that these markets are potentially very global and that many vendors are willing to ship their drugs internationally. However, it is currently still unknown how global these markets are exactly and what is influencing this globalness. In other words, until now no research has been done into the causal mechanisms influencing the globalness of dark-web markets.

Objective: The main objective of this research is to provide a comprehensive framework that gives

an insight into the globalness of dark-web markets and the influencing characteristics and causal mechanisms.

Methods: This research has been conducted by applying a triangulation of methods, namely digital

field research, directed content analysis, and expert interviews. The digital-field research was used to build knowledge and experience on the subject in order to be able to interpret and analyse the findings of the other two methods. The directed content analysis has been used as a qualitative research method to systematically collect the results of academic articles and governmental report within their context. Moreover, seven dark-web market experts from different countries and academic backgrounds have been interviewed.

Results: Dark-web illicit drugs markets have a significant global potential and enable vendors and

purchasers from all over the world to come in contact with each other. There are only two conditions for participating, namely an internet connection and a postal address. Nevertheless, these signs of globalness do not seem to fully affect the geographical trade structures of dark-web markets. These networks of vendors and purchasers are predominantly local clusters, entailing that the borders of countries, and especially continents shape trade on dark-web market.

Moreover, the results confirm that law enforcement, shipment, supply and demand, and quality, price and rating are the four influencing characteristics and that they have underlying causal mechanisms influencing the globalness of dark-web markets.

Conclusion: The identified characteristics of dark-web markets influence the globalness through

twelve causal mechanisms. Of these twelve causal mechanisms four are contributing to the globalness of dark-web markets and eight are constraining it. This research resulted in a substantiated framework of independent variables, causal mechanisms and a dependent variable.

IV

Table of content

Foreword ... II Summary ... III List of tables and figures ... V

Chapter 1: Introduction ... 1 1.1 Academic relevance ... 2 1.2 Societal relevance ... 2 1.3 Research objective ... 3 1.4 Research question ... 4 1.5 Reading guide ... 4

Chapter 2: Theoretical framework ... 6

2.1 Position in the body of knowledge ... 6

2.2 Analytical framework ... 11

2.3 Conceptualisation ... 13

Chapter 3: Research Design & Methods ... 18

3.1 Methodological justification ... 18

3.2 Operationalisation ... 22

3.3 Data analysis ... 24

3.4 Limitations ... 26

Chapter 4: Results and analysis ... 28

4.1 Current globalness of dark-web illicit drug markets ... 28

4.2 Characteristics and causal mechanism of dark-web drug markets ... 39

Chapter 5: Conclusion ... 66

5.1 Conclusion: Research question ... 66

5.2 Relevance and limitations ... 71

5.3 Recommendations ... 72

Bibliography ... 73

Appendices ... 82

Appendix one: Content analysis preparation ... 82

Appendix two: Interview scheme ... 93

V List of tables and figures

Table 1: Operationalisation of concepts……… Page 22

Figure 1: Schematic overview characteristics and possible causal mechanism……. Page 12

Figure 2: Geographic source of diverse drug-types on the Agora market………….. Page 31

Figure 3: Network of market trades between purchasers and vendors…………...… Page 33

Figure 4: Likelihood of a single purchasers using multiple sellers from

same and different countries. ………. Page 34

Figure 5: Dark-web market opioid drug distribution networks………. Page 36

Figure 6: Transactions by destination in percentages ……… Page 37

Figure 7: Causal framework; Law enforcement………. Page 41

Figure 8: Causal framework; Shipment ……….. Page 49

Figure 9: Causal framework; Supply and Demand………. Page 54

Figure 10: Causal framework; Price, Quality, and Rating……….. Page 60

Figure 11: Final framework of the characteristics, causal mechanisms, and the

globalness of dark-web illicit drugs markets. ………. Page 65

Figure 12: Causal framework; Law enforcement………. Page 67

Figure 13: Causal framework; Shipment ……….. Page 68

Figure 14: Causal framework; Supply and Demand……… Page 69

1

Chapter 1: Introduction

Dark-web markets are illegal online platforms where criminals and offenders from all over the world can virtually come in contact to trade with each other. These relatively new marketplaces challenge policy makers and law enforcement agencies since these provide offenders the opportunity to anonymously transfer part of their criminal activities into the virtual world, while giving them access to a global community (Dolliver D. , 2015). The main product traded on dark-web markets is illicit drugs (Christin, 2013). Trade of these substances on dark-dark-web markets have seen an enormous growth that has not stagnated yet (Buxton & Bingham, 2015). Dark-web markets are often compared to online shops, such as eBay, since they provide sellers and buyers a global platform where trade can be initiated (Mikhaylov & Frank, 2018). This enables international drug trade by virtually connecting drug dealers and users from variating geographical areas. When entering these markets, a drug user can choose from a variety of products from vendors all over the world. They select the drugs they want to buy, pay with a cryptocurrency, and the drugs will be delivered in their mail box (Bartlett, 2014).

Nevertheless, it is still unclear how global or local these markets in practice are. In other words, are these drug users really buying their drugs from vendors all over the world? Or are these markets only connecting dealers and users in the same country or direct area?

This research clarifies these questions by elaborating on how global or local these markets are in practice, and by exploring the causal mechanisms that influence this globalness of dark-web markets. Thus, the gap in the literature that is addressed in this research concerns the causal mechanisms that have an influence on whether dark-web markets are either used for global or local trade.

The current body of knowledge elaborates on a variety of subjects that are conceptualised in this research as ‘law enforcement’, ‘shipment’, ‘supply and demand’ and ‘quality, price, and rating’. These characteristics of dark-web markets are used as the starting point for finding the causal mechanisms. Subsequently, these causal mechanisms give an insight into what is constraining and what is contributing to the globalness of dark-web illicit drug trade markets.

2 1.1 Academic relevance

The discussion on the globalness of dark-web illicit drug markets is part of a larger trend of the globalisation of crime, where criminals and their victims/customers do no longer need to be in a close geographical proximity to each other (Aas, 2007). Additionally, Andreas (2011) states that there are some misconceptions about the current globalness of crime and that it will be beneficial to devote more attention to this subject by researching and debating the dynamics. This research focussed on dark-web markets because current academics do not yet fully understand the facets of globalisation of these markets (Romeo, 2016).

More specifically, Norbutas (2018) concluded that additional research is needed to understand the dynamics that drive vendors and buyers to trade either globally or locally. Giving more insight into the characteristics of dark-web markets and exploring the causal mechanisms that influences the globalness of dark-web markets has not been done before. The resulting framework of characteristics and causal mechanisms will therefore contribute to the body of knowledge.

The field of research on dark-web markets lies between the realms of criminology, international relations, and international economy and has not yet been properly addressed by either one of them (Andreas, 2011). This research has a multidisciplinary approach and will provide a framework that addresses above-mentioned realms by focussing on the characteristics ‘law enforcement’, ‘shipment’, ‘supply and demand’ and ‘quality, price, and rating’. Hence, this research contributes to the academic debate on dark-web illicit drug markets

1.2 Societal relevance

When an innovative technology arises, governments have to decide whether or not they have a role in regulating it and what the appropriate tactics for these possible government interventions are (Chertoff, 2017). Deciding on these interventions is particularly complex when the cyber domain creates globalisations, since international cooperation is required (Passas, 2000). The insights provided by this research will help to better understand the overall dynamics and complexities of dark-web illicit drug trade markets and provides more insight for policymakers and law enforcement agencies on how to regulate such markets.

3 Obtaining more insights into this phenomenon is crucial because of the increasing importance of dark-web markets for the coming years (Soska & Christin, 2015). Moreover, it is important since these dark-web markets pose a serious risk to the international (cyber-)security (Romeo, 2016).

Researching the structure of dark-web markets and how they operate will provide a better insight into their possible effect on the overall trade and consumption of illicit drugs (Demant, Munksgaard, & Houborg, 2016). This will, in turn, benefit the ability of policy makers and law enforcement agencies to understand this complex phenomenon and will help to enforce the right policy measures or law enforcement actions (Broséus, Rhmorbarbe, Morelato, Staehli, & Rossy, 2017).

That more knowledge on this subject is essential has been proven in the past when dark-web markets have been shut down by law enforcement agencies. This has effectively caused turmoil and uncertainty within these markets (Afilipoaie & Shortis, 2015). However, these actions have also sparked innovation in dark-web markets, and made them more resilient and complex (Soska & Christin, 2015) (Aldridge & Askew, 2017). Or as R.W. Ulbricht, the convicted creator of the first successful and most impacting dark-web market, triumphantly stated:

“Silk Road has already made an impact on the war on drugs. The effect of the war is to limit people’s access to controlled substances. Silk Road has expanded people’s access. The great thing about agorism [libertarian philosophy] is that it is a victory from a thousand battles. Every

single transaction that takes place outside the nexus of state control is a victory for those individuals taking part in the transaction. So, there are thousands of victories each week and one

makes a difference, strengthens the agora, and weakens the state.” (Ulbricht, R.W., 2012, as found in (Greenberg, 2013))

1.3 Research objective

This research has a positivist empirical explanatory character (sub-question one) and a positivist empirical exploratory character (sub-question two). In other words, the globalness of dark-web illicit drug markets is explained by answering sub-question one. And the influence of the identified characteristics and their causal mechanisms on that globalness is explored by answering

sub-4 question two. The main objective of this research is to provide a comprehensive framework that gives an insight into the globalness of dark-web markets and the influencing characteristics and causal mechanisms. This has been done by applying a triangulation of methods, namely digital field research, directed content analysis, and expert interviews.

1.4 Research question

To come to above mentioned research goal the following research question is answered:

How do the characteristics of dark-web illicit drug trade markets constrain or contribute to the globalness of those markets?

This research focusses on globalness as the dependent variable and the characteristics of dark-web markets as the independent variables. To be able to answer the main question the dependent variable has been elaborated upon, this has been done by answering sub-question one. After this, the relation of the characteristics and their causal mechanisms are explored. This has been done by answering sub-question two.

1. Are current dark-web illicit drug markets predominantly local or global networks? 2. How do the characteristics and their underlying causal mechanisms influence the

globalness of dark-web illicit drug markets?

1.5 Reading guide

The second chapter of this research clarifies the positioning of this research in the current body of knowledge. This has been done by providing insight into the topics; the internet and dark-web markets, online and offline illicit drug markets, and global and local illicit drug markets. Subsequently, an analytical framework is presented and the for this research important characteristics and terms are conceptualised.

The third chapter describes the triangulation of methods and the procedures that are followed to provide a valid answer to the research question. Moreover, the scope is explained, the concepts are

5 operationalised, and the data analysis is elaborated. Lastly, the reliability and validity limitations are outlined.

In the fourth chapter, the analysed results are reported, and the two sub-questions are answered. This chapter provides an insight into the globalness of current dark-web markets (sub-question one) and provides an elaborated framework of characteristics and causal mechanisms.

In the concluding chapter, the findings and their implications are critically discussed considering the current body of knowledge. Furthermore, the main research question is answered, the limitations of this research and the presented framework are critically discussed, and recommendations are made.

6

Chapter 2: Theoretical framework

In section 2.1 this research is positioned in the current body of knowledge by identifying three superordinate subjects. These are ‘the internet and the dark-web’, ‘online versus offline drugs markets’, and ‘global versus local drug markets’. In section 2.2 the analytical framework used for this research is presented. In the section below, important keywords are emphasised in italics, these keywords are further elaborated in section 2.3 Conceptualisation and are the starting point of identifying the characteristics.

2.1 Position in the body of knowledge

2.1.1 The internet and the dark-web

The internet underwent an explosive growth in the last decade and has created vast intertwined networks that touches upon every part of our daily life (e.g. communication, consuming, entertainment, work, etc.) (Hurlburt, 2017). This growth of networks and functions have made the internet difficult to understand and specially to regulate (Chen, 2011). Estimations indicate that the internet as ‘normal consumers’ use it (via a web browser and by using regular search engines) only holds six till ten percent of the whole internet (Hurlburt, 2017).

The other 90-94 percent of the internet is not yet indexed by search engines or cannot be indexed. This part of the internet is called the deep web and contains private data, corporate data, email-databases, internal networks and other information that is not freely accessible (Chen, 2011) (Hurlburt, 2017). Below this deep web, there is also the so-called ‘dark-web’ with online marketplaces where one could find child pornography, drug-trade, weapons, human traffickers, terrorists spreading propaganda, media pirates, cyber mercenaries, hacking tools, malware, ransomware, and many more illegal goods and services (Hurlburt, 2017).

These marketplaces are still growing, and it is likely that other illegal trade industries will also shift to dark-web markets because it can bring demand and supply together (Harrison, Roberts, & Hernandez-Castro, 2016). Because of these trends, the dark-web poses a serious risk to our international and domestic (cyber-)security (Romeo, 2016). However not all online illicit markets are considered to be dark-web markets. An important distinction is that with ‘open-web illicit markets’, these markets exist and are hosted on the public web, deliver the product to purchasers

7 through prearranged stashes instead of mail and shipping, and exploit mainstream currencies instead of utilizing cryptocurrencies (Mikhaylov & Frank, 2018).

In the last decade, efforts have been made by law enforcement to shut down dark-web markets - sometimes with success - however, this has also helped other dark-web markets to gain more revenue (Buxton & Bingham, 2015). These efforts to stop such markets have not yet worked, on the contrary, they have sparked new innovation in technologies and encryption methods (Buxton & Bingham, 2015). Moreover, Chertoff and Simon (2015) argue that ‘only’ taking down these websites without fully understanding the impact of this could lead to a more fragmented alternative dark web that will complicate the situation even further.

If we zoom in deeper into these black-markets, it becomes clear that there is a variety of products being offered, with illicit drugs being the most common product (Christin, 2013). According to Buxton and Bingham (2015), there has been a dramatic growth in the sale of a variety of illicit drugs on dark-web markets. This growth is made possible by multiple factors, such as expanding internet access; the emergence of a more tech-knowledgeable generation; the evolution of crypto-currencies; the development of new security tools, and the sustained international demand for illicit drugs (Buxton & Bingham, 2015).

Another factor that is assumed to be increasing the popularity of online drugs markets is the more efficient price/quality ratio (Buxton & Bingham, 2015). However, van der Gouwe, et al. (2016) and Bancroft & Reid (2016) argue that currently, the quality-price ratio seems relatively similar to conventional offline markets and argue that only the perception of it differs. Moreover, a single user case study found out that not only the perception of a high quality/price ratio draws users to the online drugs markets, but also the diversity of choice (supply) and the simplicity and speed of the delivering of products (shipment) (Van Hout & Bingham, 2013).

Van Hout & Bingham (2013) argue that this single case study points out that users compare the online drugs markets with eBay or Amazon, where vendors are considered trusted individuals and buying is simple and safe because of the rating systems and the delivery by regular postal services. Vendors that sell illicit drugs are mostly active on several online markets at the same time and may use different usernames. Moreover, they seem to be well organised and promise fast global delivery times (Broséus, et al., 2016). The online distribution networks of illicit drug trade are well

8 structured and organised and many vendors claim global efficient trafficking (Broséus, et al., 2016). Van der Gouwe et. al. (2016), van Hout & Bingham (2013), and Aldridge, Stevens, and Barratt (2017) press the fact that there is still a need for more research in order to understand the full scope of these relatively new online drug markets.

2.1.2 Online versus offline illicit drug markets

Above-mentioned developments –such as easy delivery and the perception of a superior quality/price ratio- are supposedly making the dark-web drug markets more interesting in comparison to the offline drug market for both suppliers and customers. Research into finding deviations between these online and offline drugs markets concludes that there are three main differences (Tzanetakis, Kamphausen, Werse, & von Laufenberg, 2016). First, the concept trust has a different meaning. For traditional drugs markets, this means an interpersonal relationship between the vendor and the user. For online drug markets this concept is promoted by administrators and vendors who use market structures, such as rating systems. Secondly, online drugs markets provide conflict-resolution by third parties (administrators). Whereas, the traditional offline market does not have this service. Finally, the logistics differ; where both markets rely on hiding and concealing drugs, drugs distribution with online markets fully relies on legal postal companies for all shipments.

On a more abstract level, offline drug markets can be considered a network economy where relatively stable networks of dealers and customers use known supply chains (Hull, 2018). The cost for new competitors to enter these markets is high and they need to develop trustworthy sources, customer networks and face retribution from other competitors. Dealing on dark-web markets is part of a market economy where better information is available for customers, new competitors gain trust by successfully completing transactions and earning good ratings, and where competition is based on price, quality and service (Hull, 2018). It is not difficult to imagine that traditional drug organisations will try to exploit the offline market and make use of these new features of drugs trade (Hull, 2018).

Moreover, offline dealers and organisations that sell illicit drugs are often described as small, short-lived, and local (Décary-Hétu & Quesy-Doré, 2017). These dealers and organisations are limited due to their inability to build large pools of customers without being detected by law enforcement agencies (Décary-Hétu & Quesy-Doré, 2017). Staying small and local has in general

9 been the modus operandi for actors on illicit markets to not be detected (Eck & Gersh, 2000). Illicit drug markets are therefore best depicted as a cottage industry where illicit drugs are supplied by a relatively large number of free-lance dealers and traffickers, instead of big conglomerates (Eck & Gersh, 2000). This local orientated cottage industry could, with help of dark-web markets, become more global. However, the question remains if dark-web markets themselves are globally or locally orientated and hence whether they can be compared to this ‘cottage industry’ description.

2.1.3 Global versus local illicit drug markets

As stated above, dark-web markets can be compared to markets such as eBay. Similar to dark-web markets these legal markets are open 24 hours a day, seven days a week. Hence, customers do not depend on offline dealers but have the option to go to online markets that offer more products often for a lower price (Sinclair, 2007). The difference in price is possible since global markets are designed more efficiently and bring supply and demand together. This efficiency is facilitated by the fact that these global market places promote disintermediation. Brining the production of the goods and the end users closer together by eliminating an intermediary (Sinclair, 2007). This efficiency of global markets is combined with low barriers for suppliers and customers to enter and the potential to reach vast international markets (Organisation for Economic Co-operation and Development, 1998). When dark-web markets first appeared, they were small. However, even then the Organisation for Economic Co-operation and Development (1998) predicted an enormous potential for growth with an expansion to international markets and facilitation of global trade. This prediction has become reality to a certain extent and these eBay-like markets have become increasingly popular in many different nations (Haucap & Heimeshoff, 2013).

Variating scholars from diverse backgrounds1 argue that dark-web markets enable purchasers and vendors to buy and sell drugs globally. However, the fact that it is possible to contact vendors from all over the world does not mean that this is the new norm. It could, for example, be argued that the path of least resistance is ordering from a dealer within the same country because shipping would be cheaper, and the illicit drugs do not have to pass border control agencies. This raises the

1 _{(Aldridge & Askew, 2017) (Barratt & Aldrigde, 2016) (Barratt, Lenton, & Allen, 2013) (Décary-Hétu,}

Paquet-Clouston, & Aldrigde, 2016) (Dolliver, Ericson, & Love, 2018) (Tzanetakis, Kamphausen, Werse, & von Laufenberg, 2016)

10 following question: ‘Can dark-web illicit drugs trade markets be considered global markets, or are they mainly oriented on local trade?’. Norbutas (2018) also recognised this gap in the literature and stated that previous research found that dark-web market vendors are often willing to ship illegal drugs internationally. However, then the question remains whether this willingness results in an actual global dark-web drug trade market or only a potential global drug trade system. In order to partially answer this question, Norbutas (2018) analysed a single dark-web market and concluded that online drug trade networks might still be heavily shaped by geographic constraints, despite their claim to provide access for purchasers to an international network of drug trade. In other words, purchasers and vendors seem to prefer to trade with users that are geographically close to them.

2.1.4 Position and gap in the body of knowledge

The three sub-sections above indicate the position of this research in the current body of knowledge. First it is part of a broader discussion on the phenomenon of dark-web markets in general and the societal implications these modern technologies have. Furthermore, it is part of a discussion of criminal trade moving from offline structures towards online structures. Finally, if we zoom in deeper, this research addresses part of the ongoing discussion on how global or local these online criminal trade structures in practice are. The gap in this literature that this research addressed focusses on the last topic, namely the globalness of dark-web markets and the causal mechanisms influencing this dependent variable. As stated above, the keywords in italics in each sub-section illustrate what the central concepts of the body of knowledge are. These concepts are used as the starting point for the characteristics that are further elaborated in the next section and in conceptualised in section 2.3.

2.1.5 Characteristics of dark-web illicit drug markets

The environment of dark-web markets shapes the decision-process for vendors and purchasers to sell or buy drugs internationally (Décary-Hétu, Paquet-Clouston, & Aldrigde, 2016). The characteristics presented in this section form this environment of dark-web markets.

First of all, dark-web markets are constantly monitored by law enforcement. And if possible, packages are intercepted, and markets disrupted. These actions result in the risk for the vendors and purchasers to be apprehended (Décary-Hétu & Giommoni, 2017).

11 Secondly dark-web drug markets exploit the professionalism of postal companies and thus benefit of their ability to distribute packages over the whole world with low costs and fast delivery times (Tzanetakis, Kamphausen, Werse, & von Laufenberg, 2016). Tzanetakis, et. al. (2016), continue by stating that these online dark-web markets increase the availability of drugs by being able to distribute packages seven days a week to any given location with a postal address (shipment).

Thirdly, these dark-web markets have the same characteristics as licit online markets and gives purchasers and vendors the opportunity to access a global market with a high supply and demand for illicit drugs (Dolliver & Love, 2015).

Finally, an important reason for going on dark-web markets in the first place is due to the perception of access to better-quality drugs, better prices, and access to a rating system to increase confidence and compare quality and price (Barrat, Ferris, & Winstock, 2013).

Above-elaborated characteristics are distilled from the current literature and are further elaborated as independent variables in section 2.3 Conceptualisation. Moreover, the results of what these characteristics exactly entail in light of this research will be elaborated in section 4.2. Gaining more insight into these characteristics and unveiling their constraining and contributing causal mechanisms in relation to the globalness of dark-web markets, will help better understand the dynamics behind the discussion if dark-web illicit drugs markets are predominantly global or local oriented networks.

2.2 Analytical framework

An analytical framework is needed in order to analyse how the characteristics derived from the literature affect the globalness of dark-web markets. This analytical framework is established by combining multiple academic sources, because there is currently no such framework. In other words, literature is used to structure the data collection and analysis. The purpose of such an analytical framework is linking the theoretical question to the empirical analysis by summarising what is theoretically known about relevant empirical phenomena (Glaser, 2015).

12 The factors constraining or contributing to the globalness of markets can also be referred to as causal mechanisms. This research aims at exploring and explaining these causal mechanisms. In other words, the four characteristics and their possible causal mechanisms in relation to the globalness of dark-web illicit drug markets are researched. A causal mechanism is defined as “a sequence of events, conditions, and processes leading X to Y” (Little, 1991, p. 15). This relation between X and Y only exists if there are a set of causal mechanisms that connects X to Y. In this research X represents the characteristics of dark-web illicit drug markets and Y represents the globalness of dark-web illicit drug trade markets. X can only influence Y if there is a set of causal mechanisms that connects them. In the literature review above multiple characteristics have been identified. These characteristics can contain two different causal mechanisms: contributing and constraining mechanisms. In other words, the causal mechanisms have a positive or negative influence on the dependent variable. It is also possible that a characteristic does not lead to a causal mechanism. See Figure 1 for a schematic overview of the analytical framework. This analytical framework is the basis for answering the research question and building the final framework of characteristics and causal mechanisms influencing the globalness of illicit dark-web drug markets.

Figure 1: Schematic overview characteristics and possible causal mechanism

Causal mechanisms Y Globalness of dark-web illicit _{drug markets} Constraining

Contributing X1 Law Enforcement

X2 Shipment

X3 Supply and demand X4 Price, quality, and rating

13 2.3 Conceptualisation

2.3.1 Independent and dependent variables

The body of knowledge indicated that there are several characteristics that are important for the globalness of dark-web illicit drug trade markets. The theoretical framework helped to select the relevant characteristics that are further researched in this study. During this research the possibility was kept open to add or remove characteristics if the results indicate that this was necessary. However, both the interviews and the content analysis did not indicate that the analytical framework should have been modified.

As concluded, the four characteristics structuring the analytical framework are; law enforcement, shipment, demand and supply, and quality, price and rating. These characteristics will be used as an analytical framework for exploring the possible causal mechanisms. The characteristics are elaborated on below to illustrate their relevance to the globalness of dark-web market and the research question.

Independent variables

Law enforcement

Some dark-web markets have been shut down by law enforcement. This caused turmoil in other marketplaces and has pushed vendors to become more creative and markets to become more complex (Afilipoaie & Shortis, 2015). These law enforcement actions are the main risks for vendors and purchasers and could, therefore, affect geographical dark-web markets trade structures (Cunliffe, Martin, Décary-Hétu , & Aldridge, 2017). Moreover, differences in the use of dark-web markets per country appear to be driven by domestic drug policy (Van Buskirk J. , Naicker, Roxburgh, & Bruno, 2016). Another important factor is that dark-web markets users are adopting new law enforcement mitigating technologies that enable them to buy and sell substances across the world (Décary-Hétu & Giommoni, 2017). This could disrupt the ability of law enforcement to regulate illicit markets and thus holds the potential to change the distribution and sales of illicit drugs.

Shipment

Dark-web market substances are unwillingly and unknowingly delivered by regular postal services. This manner of distributing drugs affects international and domestics mail centres. These centres therefore play a key role in the distribution of illicit drugs (Afilipoaie & Shortis, 2015).

14 International and local postal services are becoming more efficient and their infrastructure is used to ship illicit drugs locally and globally. This is an important feature of dark-web markets because it enables dark-web illicit drugs trade. The duration of the shipment is mainly determined by the location of the vendor and the purchaser. It could therefore take days of even weeks for the illicit drugs to arrive at the right geographical destination (Aldridge & Askew, 2017). Moreover, this usage of postal services decreased the geographic separation of drugs trade. When a drug is purchased on a dark-web market it can be delivered to any place registered with a postal address. This use of postal services circumvents the need for the user to seek a dealer with specific drugs in their geographic location (Dolliver, Ericson, & Love, 2018).

Demand and supply

Dark-web markets potentially affect the overall market structure of illicit drug trade by shortening the lines of supply chains (Martin, 2014). Vendors often sell directly to end-users, or to wholesalers who may resell the illicit drugs in the offline or online markets. The vendors themselves can fill their supply by sourcing from dark-web markets and offline markets, the same way as offline dealers do (Aldridge & Askew, 2017). Hence, dark-web markets are online environments where global demand and supply converge. One of the biggest dark-web markets so far (Silk Road) was, amongst others, used to sell highly demanded drugs like MDMA, 2-C, cannabis, cocaine, opioids, and LSD (Barratt, Ferris, & Winstock, 2013). In other words, dark-web markets can function as an online broker system, connecting wholesalers to vendors, wholesalers to street dealers, or vendors to end-users. These dark-web markets may, therefore, be able to kick-start changes in patterns and locations of trade (Aldridge & Décary-Hétu, 2016).

Quality, pricing and rating

The main question when it comes to quality and pricing and the globalness of markets is whether purchasers will try to take advantage of online illicit drugs markets by seeking the best price-quality ratio no matter the location (Aldridge & Décary-Hétu, 2014). Quality is one of the reasons drug users and vendors prefer dark-web markets instead of offline markets (Van Hout & Bingham, 2013). So, it could be that users are enabled to choose the product they desire by comparing quality and price, no matter the location. Moreover, dark-web markets enable a setting where drug users can discuss and compare international value and quality of drugs (Bancroft & Reid, 2016). These discussions on quality and price take place in online dark-web forums and by embedded rating systems (Demant, Munksgaard, & Houborg, 2016). It is suggested that these systems result in

15 availability of high quality drugs (Christin, 2013) (Martin, 2014) and that this makes quality more transparent for purchasers (Bancroft & Reid, 2016).

Dependent variable

Globalness

Dark-web markets are a relatively new technological innovation that has enabled the distribution of illicit drugs by limiting geographical restrictions (Tzanetakis, Kamphausen, Werse, & von Laufenberg, 2016). These markets are stretched across time, virtual and geographical space and the globalness variates depending on the location of vendors and purchasers that interact and trade with each other (Aldridge & Askew, 2017). This new ability of shipping drugs across national and continental borders by using regular postal companies can be utilised by vendors when a purchase is made. However, offering this global service does not mean that it is also used because of possible restrictions that could lead to geographical clustering of customers and purchasers (Norbutas, 2018). Ordering from local dealers on dark-web markets could be preferred by purchasers for multiple reasons (Van der Gouwe, Brunt, van Laar, & van der Pol, 2016). Because of this, drug trade networks on dark-web markets may be more locally oriented (restricted by national and continental borders) than often assumed (Norbutas, 2018).

2.3.2 Definitions of remaining terms

In this section the important concepts of this research are defined. See chapter 3.2 for the operationalisation of the characteristics (independent variables) and globalness (dependent variable).

Dark-web illicit drug market: Online marketplaces where vendors and purchasers come together

to offer or purchase drugs. These online places must be accessed by using a Tor browser and is thus part of the internet that is not accessible through normal web-browsers. It is therefore not indexed by regular search engines such as Google or Yahoo (Hurlburt, 2017). Illicit drug markets that can be accessed by using regular web-browser are not part of this research.

Offline drug market: Traditional offline drugs markets where drugs dealers develop face-to-face

relationships with customers at a specific geographical location. These transactions are out of sight for law enforcement and other dealers and users (Aldridge & Askew, 2017).

16

Illicit drugs: All drug types that are forbidden to sell and buy online, such as cannabis, stimulants,

ecstasy-types, prescription drugs, opioids, and psychedelics including New Psychoactive Substances (NPS) that are not yet registered by some countries (Rand Europe, 2016).

Vendor: A person (or group) that offers illicit drugs on dark-web markets.

Purchasers: A person (or group) that buys illicit drugs on dark-web markets for own use or to

further distribute.

Administrator: The person (or group) that is responsible for building a dark-web market,

maintaining it, overseeing the trade deals, and providing conflict resolution if necessary (Afilipoaie & Shortis, 2015).

Trade network: Completed transactions between vendors and purchasers in certain geographical

locations. These trade networks can be national, international, and intercontinental (Norbutas, 2018).

Local trade: Trade within the borders of a single country. Both vendor and purchaser are located

in the same country (Norbutas, 2018).

Global trade: International and intercontinental trade. Vendor and purchaser are not located in

the same or neighbouring countries (Norbutas, 2018).

Globalness: The degree to which a network of trading in illicit drugs is either locally or globally

oriented. These markets are stretched across time, virtual and geographical space. These variables of globalness variate depending on the location of vendors and purchasers that interact and trade with each other (Aldridge & Askew, 2017) Because of this the network structure of dark-web markets can be either local or global oriented.

Constraining causal mechanisms: Causal mechanisms distilled from the for this research

important characteristics of dark-web drugs markets that negatively influence the globalness of those dark-web illicit drugs trade markets. In other words, events, conditions, or processes that localise dark-web illicit drug markets.

Contributing causal mechanism: Causal mechanisms distilled from the for this research

17 those dark-web illicit drugs trade markets. In other words, events, conditions, or processes that globalise dark-web illicit drug markets.

Law enforcement: Collaboration of organisations and agencies that investigate or try to hinder

illicit drugs trade on dark-web markets. This can be done by either online activity or by interfering in offline parts of the market (Aldridge & Askew, 2017).

Shipping: International and domestic postal services that are being used by vendors on dark-web

markets to ship their products to the purchasers (Afilipoaie & Shortis, 2015).

Demand and supply: The amount and variety of illicit drugs that is being offered on dark-web

illicit drugs markets and the number of individual customers and vendors that use those markets. Because of the dark-web as technological innovation vendors and purchasers are brought together in one single location (Décary-Hétu & Quesy-Doré, 2017)

Quality, pricing and rating: The (perceived) quality of the drugs in terms of substances and effect

(Van der Gouwe, Brunt, van Laar, & van der Pol, 2016) and the sum of cryptocurrency that is being asked and offered for those illicit drugs. These variables can be compared by customers by using the build in rating systems (Hardy & Norgaard, 2016). These dark-web rating systems help customers compare the quality and price of different products and vendors.

18

Chapter 3: Research Design & Methods

This chapter elaborates the methodological justification and the procedures that have been followed to reach a valid and reliable answer to the following research question:

How do the characteristics of dark-web illicit drug trade markets constrain or contribute to the globalness of those markets?

This research question is divided into the following sub-questions:

1. Are current dark-web illicit drug markets predominantly local or global networks? 2. How do the characteristics and their underlying causal mechanisms influence the

globalness of dark-web illicit drug markets?

3.1 Methodological justification

This research is designed to explore and explain the causal relation between the characteristics of dark-web markets and its globalness. This research is divided into two sub-questions. The first sub-question provides an insight into what is known about the current globalness of dark-web illicit drug markets. To answer this question existing academic literature and official reports were analysed. The second sub-question explains the relation of the characteristics to the globalness of dark-web markets and explores the possible causal mechanisms. This is done by combining the data collected by interviewing experts and the data collected with the directed content analysis. To improve the ability to produce a valid answer on the research-question a digital field research was conducted to gain more knowledge on dark-web markets.

3.1.1 Scope

This research only focussed on dark-web markets and not on other online markets that can be accessed by using normal web browsers. Furthermore, it did not take in account trade in other illicit products such as weapons, legal supplements, or hardware. Moreover, the effect of dark-web markets on the entire illicit drug trade markets is not addressed. Finally, it is important to state that this research does not prove if specific dark-web markets are active on a global or local scale. Instead, it explains the current globalness of markets in general, the influencing characteristics, and the underlying causal mechanisms.

19

3.1.2 Data collection methods

To produce a reliable and valid answer to the research question, a triangulation of methods has been used. The first method that has been used is a digital field research, which has been executed by entering and observing dark-web illicit drugs markets. The second data collection method was a qualitative directed content analysis of current literature and reports. And the final method, was interviewing experts. These three data collection methods are discussed more extensively below.

Digital field-research

The main reason for including this data collection method was to gain the necessary experience and knowledge to be able to understand the data that was collected with the other two methods. Doing field-research is important when it comes to understanding dark-web markets, because it provides necessary insights and understanding in this complex online world (Bakken, 2019). This method was not used to collect empirical data but only to gain essential experience and knowledge on the topic of dark-web markets. It therefore helped with preparing for the interviews, analysing the data, and constructing valid and reliable conclusions. This digital field-research started in tandem with the start of writing the proposal for this research. Approximately 20 hours were spent on dark-web markets and their forums to gain knowledge. During this time, different market places were observed, and forum discussions were read to find information related to the globalness of dark-web markets. During this research method, no laws were violated and necessary security measures such as using a VPN were taken.

Directed content analysis

To systematically analyse the current knowledge on the globalness of dark-web markets a directed content analysis was conducted. This is a qualitative content analysis method that focusses on the characteristics of language as communication with attention to the content and contextual meaning of the text (Budd, Thorp, & Donohew, 1967; as seen in Hsieh & Shannon, 2005). This method is useful in case of prior but still incomplete research on a certain phenomenon (Hsieh & Shannon, 2005). This method is used to conceptually extend the already existing knowledge on this topic and explore the possible causal mechanisms. “The goal of content analysis is to provide knowledge and understanding of the phenomenon under study” (Downe-Wamboldt, 1993, p. 314). In this case the phenomenon under study are the dependent and independent variables.

20 This analysis went beyond counting words to classify amounts but focussed on content and context. In total 45 academic articles and 10 governmental reports have been analysed. Moreover, the information gathered by interviewing experts has been coded and analysed by using the same categories, indicators and rules. This has been done, because a directed content analysis is also useful to structure interview data that has been collected by open-ended questions (Hsieh & Shannon, 2005). This content analysis resulted in three Excel sheets with the essential information for answering the two sub-questions. The categories used for this directed content analysis are the independent and depended variables.

In this research, only the latent content of the collected data has been used and not the manifest content. This means that only the context and meaning of the data has been analysed and not the number of times it comes back in the literature. This resulted in a clear overview of the currently available information and its context for answering both sub-questions.

The documents used for this analysis were selected by using Google scholar, the online library of the University of Leiden, and by using the so-called ‘snowball’ technique. This entails looking at sources used in one document and adding them if they are found relevant. The sources used for this content analysis were critically reviewed, assessed by looking at who the author is, why the author conducted that research, how many and what kind of sources they used, what the research goal was, the date of publication, and whether it is published by a relevant journal (Schreuder Peters, 2012).

Only sources of the year 2008 and later were used in this analysis. This choice has been made to ensure relevancy of the data since internet related phenomena change rapidly. Another reason for this time frame is that in the past decade, the marketing and sales of illicit drugs via dark-web markets has grown rapidly (Van der Gouwe, Brunt, van Laar, & van der Pol, 2016). Moreover, the noticeable use of dark-web markets by drug dealers and users started approximately in 2007-2008 (Barratt, Lenton, & Allen, 2013).

The sources have been read carefully. Important parts related to the categories and indicators were marked, and these parts with their necessary context were copied into the assigned sheet and column. To prevent losing any important context, the unit of analysis that has been used is one or multiple sentences. All marked documents and the Excel sheets are stored and can be obtained by

21 contacting the author. For more information on how the content analysis is performed, see appendix one.

Expert interviews

The third method is interviewing experts. The interviews were held when the other two methods were almost completed. This allowed for acquiring the necessary knowledge on the subject first, which could then be used during the interviews. This way, more directed and relevant questions could be asked.

The experts were selected on the basis of their experience and knowledge on the subject of drugs trade on dark-web markets and the perceived local and global structures of this illicit trade. Since this research aims to explain a global phenomenon with global implications, experts from different countries and continents were contacted. Respondents that have been approached to participate in this research are from the following organisations; Europol, Interpol, the Dutch Police force, Trimbos Institute, Global Drug Survey, National Cyber Security Centre, Netherlands Organisation for Scientific research, Rand Corporation and multiple scholars from variating universities. Many of these organisations and individuals responded. However, some did not have time to participate and some deemed others more knowledgeable to answer the interview questions.

Because of the explorative nature of the topic and research question, a semi-structured interview scheme was applied (Schreuder Peters, 2012). The four independent variables where the guiding topics and the interviewees were able to determine whether they saw a relation with the dependent variable and if so, how this causal mechanism works. By using a semi-structured interview, the experts were able to choose any causal mechanism they considered relevant. In order to find out if there were any independent variables missing, each interview ended with the following question: ‘Are there any independent variables or causal mechanisms that you deem important for this discussion that have not yet been discussed?’ See appendix two for the interview scheme.

22 3.2 Operationalisation

In chapter 2.3, the concepts have been conceptualised. The following step is to operationalise them by defining observable indicators that serve to identify important information for the directed content analysis. The concepts have been coded by assigning them the letter that interlinks with the relevant column in the Excel sheet. See Table 1.

Because these indicators signal every relevant word in the sources regardless of their context a second step was added in order to only collect relevant data. This second step consisted of carefully reading the surrounding text and analysing whether this had something to do with the concept itself and the dependent variable. In some cases, this was challenging to assess. However, when it was not clear whether the information could be relevant, it was included in the Excel sheet and marked red. This choice has been made because some data can be more valuable when compared to other findings of other authors. If these units were deemed unnecessary after analysing all documents, they were removed. By doing this assessment each time when an indicator was found, relevant information was collected that could be compared in a later stadium. How the relevant information was processed is further elaborated in the next section.

Table 1: Operationalisation of concepts

Code Concept Definition Indicators

D Law

enforcement

Collaboration of organisations and agencies that investigate or try to hinder illicit drugs trade on dark-web markets. This can be done by either online activity or by interfering in offline parts of the market (Aldridge & Askew, 2017).

Crime, criminal, forensic, crime fighting, law, policy, law enforcement, rule of law, regulation, agency, police, police force, investigate, interfere, customs, intercept, mitigate, arrest,

intervention, hinder, deter, detection,

E Shipment International and domestic postal services that are being used by vendors on dark-web markets to ship their

Postal service, delivery companies, posting, delivery, distribution,

23 products to the purchasers (Afilipoaie &

Shortis, 2015).

trafficking, destination, transferring, mail, mail centre, send, shipment, concealment, packaging wrapping, package, batch, load,

F Demand and supply

The amount and variety of illicit drugs that is being offered on dark-web illicit drugs markets and the number of individual customers and vendors that use those markets. Because of the dark-web as technological innovation vendors and purchasers are brought together in one single location (Décary-Hétu & Quesy-Doré, 2017)

Demand, user, buyer, customer, purchaser, supply, vendor, supplier, seller, dealer, retailer, amount, quantity, variety, diversity, assortment, array, variation, revenue, profit, collection, offer, availability, capacity, sourcing stock, wholesale, purchases,

G Quality, pricing and rating

The (perceived) quality of the drugs in terms of substances and effect (Van der Gouwe, Brunt, van Laar, & van der Pol, 2016) and the sum of currency that is being asked and offered for those illicit drugs. These variables can be compared by customers by using the build in rating systems (Hardy & Norgaard, 2016). These dark-web rating systems help customers compare the quality and price of different products and vendors.

Quality, purity, adulterant, concentration, chemical substance, substance, ingredients, predictability, expectedness, reliability, price, worth, charge, value, money, currency, rating, evaluate, comparing, experience, grading, review, valuation, judgement I Globalness (geographical

The degree to which a network of trading in illicit drugs is either locally or globally oriented. These markets are stretched

Globalness, global, globalization, worldwide, universal, international,

24 trade

structure)

across time, virtual and geographical space. These variables of globalness variate depending on the location of vendors and purchasers that interact and trade with each other (Aldridge & Askew, 2017) Because of this the network structure of dark-web markets can be either local and global oriented.

internationalisation, continental, transnational, intercontinental, local, localisation, regional, domestic, origin, location, geographical,

topographical, time, space, borders, country, nation,

3.3 Data analysis

In section 3.1.2 it is explained which methods were used to answer the sub-questions of this research. In this section, these three methods are brought together, and an elaboration is provided on how the research has been conducted step by step. This has been done to create transparency and thus repeatability.

First, the digital field-research. As mentioned before, this part of the method is used to create an insight into the dynamics of dark-web markets. Although this method has not directly contributed to answering the sub-questions it is a vital part of understanding the outcomes of both the content analysis and the interviews. Without first-hand experience in what these markets are the results could easily be misunderstood or misrepresented.

Secondly, the directed content analysis. During this method, the four operationalised independent variables and the dependent variable are used to analyse the content of 45 academic articles, 10 governmental reports and 7 interviews. The collected data on the dependent variable is used to answer question one and the data collected on the independent variables is used to answer sub-question two. The analysed documents and interviews are marked and coded by using the colours related to the variable, see section 3.2 Operationalisation.

The interviews where the last documents added to the content analysis. The interviews were recorded, transcribed, and finally coded and added to the coding sheet. One coded part of interview text equals one unit of analysis. This is done in order to preserve the context of the statement.

25 The final selection of respondents that participated is:

1. University of Copenhagen: PhD fellow S. Bakken 2. University of Montréal: Prof. D. Décary-Hétu

1. University of Alabama: Assistant professor D. Dolliver 2. Trimbos Institute: Dr. D.Z. van der Gouwe

3. Rand Corporation: Associate Director S. Hoorens 4. University of Montréal: PhD candidate R. Munksgaard 5. University of Utrecht: PhD candidate L. Norbutas

All participants have been given the chance to comment on the transcript and remove or change any given information if deemed necessary. Moreover, none of the participants desired to be anonymous. See appendix three for the transcribed and coded interviews.

This approach resulted in an Excel document with 937 units of analysis, divided over three sheets with each five columns. In case a unit of analysis relates to multiple variables, it is also marked in multiple columns. Each sheet also contains two columns with information that was considered important for creating context. Furthermore, these extra columns were used to find possible other characteristics that could be important for this research. However, this has not resulted in an outcome. Moreover, the interviewees did not suggest more independent variables. Because of this the four independent variables that were found in the theoretical framework have not been modified.

The 937 units of analysis were compared to each other by filtering the columns and clustering the information. When the context of a unit of analysis was unclear, the original document was reopened to add context. When a column is filtered, the data of one independent variable becomes visible. These were then compared, added to the chapter results, and marked with a light green colour to indicate that it has been used. See the Excel document for the coding sheets.

26 3.4 Limitations

The current body of knowledge was deemed extensive enough to start a feasible research and has presented a clear gap, namely: The unknown characteristics and causal mechanisms influencing the globalness of dark-web markets. By researching the relevant characteristics and their possible causal mechanisms that could influence this globalness an insight has been provided into the important events, conditions or processes.

To my knowledge, this is the first research that is devoted to this framework of characteristics and the first research that qualitatively addresses the causal mechanisms in relation to the globalness of dark-web markets. The main goal of this research is to explore and elaborate the possible causal mechanisms. It is certain that more research would be beneficial to strengthen any conclusion. Nonetheless, this research is valuable since it combines disciplines2, explains the characteristics and provides insight into the causal mechanisms with.

However, there are some challenges concerning the reliability and validity to consider.

3.4.1 Reliability

The elaborative and exploratory character of this research gives rise to some reliability issues. The test-retest reliability of this research is under pressure due to the complex dynamic and fast developing topic of dark-web markets. Another reason for the limited test-retest reliability is the nature of such a qualitative and explorative research. The results of the three methods have all been interpreted, meaning that other students or scholars could end up with slightly different results.

Another possible reliability issue for this research is the so-called inter-rater reliability. This issue appears because only a relatively small number of experts can be interviewed, and they could interpret or evaluate the concepts different than other experts (Phelan & Wren, 2005). It is possible that other researchers find more or different causal mechanisms that influence the dependent variable. This limitation exists even though all experts have been asked if they missed any important characteristics or causal mechanisms. It could be that other experts have another opinion or that the same question will be answered differently in the future because of the changing

27 dynamics of dark-web markets. Moreover, it could be that other experts see the influence of the causal mechanisms in a different way than the ones interviewed for this research. These reliability issues are counterbalanced by using a triangulation of methods and by interviewing experts from different countries with different specializations regarding dark-web markets.

Nonetheless, the conclusions of this research should be retested or further researched to strengthen any conclusions. Recommendations for this are made in chapter 5.3. To provide future students or scholars the chance to test or add to this research, all data that is collected can be obtained by contacting me via LinkedIn. To add to the transparency, all respondents of this research have approved for the transcripts to be published.

3.4.2Validity

It is important to make sure that what is being measured truly answers the research question. This is called the construct validity (Phelan & Wren, 2005). To make sure this construct validity is high, the important concepts are defined as specific as possible and these are operationalised by reviewing the current literature on these concepts in relation to the main research question. Concerning this topic, it is important to keep in mind that this research is partly explorative and that the second sub-question does not try to prove the extent to which a causal mechanism is influencing the dependent variable. However, the results do show which causal mechanisms are of influence according to the current literature and experts and how this works. A future researcher could use this information to try to prove to which extent these causal mechanisms are of an influence.

Another challenge is the internal validity. This raises the following question: ‘are the independent variables influencing the dependent variable or is there more mechanism influencing the dependent variable?’ (Shuttleworth, 2008). For this, the same factors as for the inter-rater reliability are important.

Finally, the sampling validity. This regards the content analysis and the interviewing of experts. Not all experts can be interviewed and not all academic sources and important reports can be analysed. This issue remains because the research needs to be feasible within a certain period. However, because of the mixed methods and variety of experts this problem is deemed acceptable.

28

Chapter 4: Results and analysis

In this chapter the results are presented by answering the two sub-questions. Both questions are answered by analysing the results of the triangulation of methods. The first section of this chapter answers sub-question one ‘Are current dark-web illicit drug markets predominantly local or global networks?’ This section provides an elaborative insight into the current knowledge and features of globalness and illustrates what the dependent variable exactly entails. The second section answers sub-question two ‘How do the characteristics and their underlying causal mechanisms influence the globalness of dark-web illicit drug markets?’ This section starts by elaborating on what the independent variable entails in relation to the globalness of dark-web markets and their causal mechanisms are substantiated. This resulted in a comprehensive framework that is depicted in section 4.2.5 conclusion. These results, their implications and limitations will be further discussed in chapter 5 Conclusion.

4.1 Current globalness of dark-web illicit drug markets

In this section the first sub-question ‘Are current dark-web illicit drug markets predominantly local or global networks?’ is answered. This section is divided into four sub-sections. The first sub-section zooms in on the potential globalness of dark-web markets and the features that enables this potential. The second sub-section will elaborate on the currently involved countries. The third sub-section elaborates on the willingness of vendors to ship internationally. And the final sub-section elaborates on the results that indicate that dark-web markets are predominantly used to trade locally and the current trend of globalness. After these sub-sections, a conclusion is given.

4.1.1 Global possibilities

Current scholars agree that dark-web markets have the potential to become global drug trading markets by circumventing current offline drug trade networks (Demant, Munksgaard, Décary-hétu, & Aldrigde, 2018). Dark-web markets enable purchasers and vendors from all over the world to come together in an online environment and trade with each other. On these dark-web markets purchasers can choose from which vendor, and thus from which country (or region), they want to buy their illicit drugs. When a purchaser has made his/her decision, an order is placed via this online platform. Depending on the location of both vendor and purchaser, the package will be