Many anti-sex trafficking analyses use the term institution in a narrow meaning, comprising mainly formal-legal political structures (public laws and governmental organizations). However, by bringing in the new institutionalism approach, it is argued that an anti-sex trafficking institution should refer to a relatively enduring collection of rules – including also informal rules such as norms and routines – and organized practices that prescribe appropriate behaviour for any actor, public or private, combating sex trafficking. Based on a review of current research it is concluded that anti-sex trafficking institutions in the early 21st century tend to focus on behaviour that aims at detection, prevention, protection, crisis management, consequence management, and response. Finally, reflecting different strands of the new institutionalism approach, it is argued that the design of anti-sex trafficking institutions depends on path-dependencies, social constructions, international institutions, and domestic politics.
The term institution has entered into anti-sex trafficking analyses. For example, references have recently been made to “institutional responses” (Dewey, 2008), “institutional measures” (Jayagupta, 2009: 243), “institutional mechanisms” (Samarasinghe, 2008: 160), “institutionalized support processes” (Segrave et al., 2009: 176), and the “institutionalization” of a moral crusade (Weitzer, 2007: 447). However, despite the increasing use of the term, these scholars do not refer to the new institutionalism approach, which otherwise is frequently used in the modern social sciences (Rhodes et al., 2006). One consequence of this is that the term is used in a narrow meaning, comprising mainly formal-legal political structures (public laws and governmental organisations). Another consequence is an underestimation of some of the dynamics that presumably affect the design of anti-sex trafficking institutions. Taken together, this means that the potential of the new institutionalism approach still remains to be discovered.
In order to fill this gap, this article will offer a systematic introduction of the new institutionalism approach to the study of anti-sex trafficking activities. The presentation will centre on two key aspects of the approach. First, I will specify the meaning of the dependent variable “anti-sex trafficking institutions”. According to the general understanding, institutions constitute “the rules of the game” (Rothstein, 1996: 145). However, because there are different views about exactly what this includes, it remains to be specified what rules are and in what respects institutions vary. For example, should the term anti-sex trafficking institutions be reserved merely for laws against sex trafficking or should we also include norms against sexual exploitation and routines for protection of victims?
Second, I will address the conditions for institutional design. That is, how are we to understand the composition of anti-sex trafficking institutions and possible variations between countries? As regards explanations, the new institutionalism approach does not offer a single answer. Instead, we find a debate between several schools of thought (Peters, 2005; Rhodes et al., 2006). Seen from a rationalistic perspective, we expect institutions to mirror the intentions and capabilities of the designers. However, this perspective has been criticized for ignoring important contexts as well as the dynamics between actors and contexts. Therefore, I will focus on four strands of the new institutionalism approach, each highlighting a particular context in which designers operate. The four strands are: historical institutionalism, constructivist institutionalism, international institutionalism, and domestic institutionalism. In short, the four strands suggest that institutional design is a result of path-dependencies, social constructions, international institutions, and domestic politics.
What is an Anti-Sex Trafficking Institution?
The founders of the new institutionalism approach, James G. March and Johan P. Olsen, write: “An institution is a relatively enduring collection of rules and organized practices, embedded in structures of meaning and resources that are relatively invariant in the face of turnover of individuals and relatively resilient to the idiosyncratic preferences and expectations of individuals and changing external circumstances” (March and Olsen, 2006: 3). In other words, they continue, institutions are “constitutive rules and practices prescribing appropriate behaviour for specific actors in specific situations”. On the basis of this definition, it is suggested that an anti-sex trafficking institution should refer to a relatively enduring collection of rules and practices that prescribe appropriate behaviour for any actor, public or private, combating sex trafficking. The definition highlights four aspects of an institution.
First, an institution defines a domain or an area of activity. As regards anti-sex trafficking, we have witnessed an increased clarity of the domain. The credit for this can be given to the General Assembly of the United Nations (UN) that on 15 November 2000 adopted the Protocol to Prevent, Suppress and Punish Trafficking in Persons, Especially Women and Children, which is the first international document that contains a commonly agreed definition of trafficking in persons. The Trafficking Protocol, which entered into force on 25 December 2003, establishes (UN, 2000: article 3a):
‘Trafficking in persons’ shall mean the recruitment, transportation, transfer, harbouring or receipt of persons, by means of the threat or use of force or other forms of coercion, of abduction, of fraud, of deception, of the abuse of power or of a position of vulnerability or of the giving or receiving of payments or benefits to achieve the consent of a person having control over another person, for the purpose of exploitation. Exploitation shall include, at a minimum, the exploitation of the prostitution of others or other forms of sexual exploitation, forced labour or services, slavery or practices similar to slavery, servitude or the removal of organs.
However, despite this conceptual progress, analysts should be aware of three circumstances that complicate the demarcation of anti-sex trafficking institutions. These can be summarized as the problems of diverse interpretations (i.e. sex trafficking often means different things in different locations; Segrave et al., 2009: 18), horizontal confusion (e.g. the domains of anti-sex trafficking and anti-prostitution are not always clearly separated; Weitzer, 2007: 464–466), and vertical embeddedness (i.e. the fight against sex-trafficking is also guided by the general rules and practices of the welfare state). Because of these circumstances, we should still expect to find national differences in institutional design.
Second, institutions comprise public laws as well as informal rules. The latter implies that people are guided also by norms, routines, customs, and so on. Mohammed Mattar at Johns Hopkins University has stressed the importance of informal rules in the combat of sex trafficking (Mattar, 2004: 105): “It is not enough that the law considers illegal the behaviour of the customer of sexual services. The behaviour of the customer who obtains sexual services, the functional equivalent of the law must also recognise such behaviour as unacceptable. By ‘functional equivalent of the law,’ I mean the traditions, the customs, the acceptable behaviour of the people.” This reasoning can be exemplified by the Swedish criminalization of sex buyers. That is, if sex buyers do not adhere to the norm that it is unacceptable to buy human beings for sexual exploitation, then this might take away the effect of the criminalisation because otherwise buyers will just seek places where there is a low risk of detection or travel to countries where prostitution is legal (cf. Samarasinghe, 2008: 179).
Third, because institutions guide the behaviour of specific actors, it is worth noting that much anti-sex trafficking work presuppose effective cooperation between different actors (Samarasinghe and Burton, 2007; Samarasinghe, 2008: 187–188). Thailand is a case in point as the Thai institution since 1997 includes rules and practices for inter-departmental and inter-agency cooperation, cooperation between government agencies and Non-Governmental Organizations (NGOs), and cooperation within the NGO community (Jayagupta, 2009: 232–233). However, although these types of national mobilizations often are regarded as desirable, they are not without problems. For example, as far as NGOs are concerned, there is a risk that the distinction between state and non-state actors becomes blurred so that the roles of assisting victims and the police appear as one and the same (Segrave et al., 2009: 41, 78). But cooperation dilemmas can also appear within the state apparatus. For instance, because the social services and the police have a long tradition of separate rules and practices, much of the current anti-sex trafficking debate focuses on governance reforms such as clarifying roles and improving administrative cooperation both horizontally and vertically (Jayagupta, 2009: 248).
Finally, it is not given exactly what type of behaviour that should belong to an anti-sex trafficking institution. For long, the combat of sex trafficking has been concentrating on the so-called three P: s: prevention, protection of victims, and prosecution of perpetrators (Samarasinghe, 2008: 186–188). Although these behaviours are seen as necessary, there are also other types of behaviour involved. This is seen from some current research on trafficking in persons, which also pay attention to behaviour that aims at detection (e.g. Bales and Soodalter, 2009), crisis management (Hotaling, 2004: 96), and consequence management (e.g. Shelley, 2010: 60–65). Interestingly, the six types of behaviour are not unique to the anti-sex trafficking domain, but could also be found in, for example, the combat of global terrorism (Carter, 2001: 15–17; cf. Karlsson, 2009, 2012).
After addressing the individual components of anti-sex trafficking institutions, we should now try to put them together. Figure 1 summarizes the relationship between behaviour, actors, and rules. The general idea is that rules clarify what should be done (behaviour) and by whom (actors).
The first type of anti-sex trafficking behaviour focuses on the ability to detect victims. Ideally, an institution should encompass routines for discovering the victims before the sexual exploitation starts, which according to the UN Trafficking Protocol would mean during “recruitment, transportation, transfer, harbouring or receipt of persons” (UN, 2000: article 3a). However, in practice, if victims of trafficking are at all detected, then it is usually at a much later stage when they are already in the sex industry (Segrave et al., 2009: 34). In that case, other measures such as consequence management and legal response become more relevant.
Although the detection of victims has only recently received scholarly attention, two conclusions seem to recur (Bales and Soodalter, 2009: 254–258, 265–266; Derluyn et al., 2010: 176; Segrave et al., 2009: 34–64). First, because victims of trafficking are often afraid to contact the authorities, detection in practice has become a matter of establishing routines for the search of victims. In theory, there are several critical moments of the trafficking process during which victims can be detected. For example, identification can take place in connection to recruitment, border crossings, and at the destination. However, in practice, the search is usually restricted to the final stage and to the discovery of victims at brothels and prostitution streets (Segrave et al., 2009: 45–58). This means that much of the search relies on routines for rescue missions and control of illegal immigrants. As regards the other moments of the trafficking process, search is assumed to be less effective because much recruitment takes place under cover (e.g. through agencies supplying models, dancers, brides, or domestic workers). In the case of border crossings, the UN Trafficking Protocol emphasises border measures and control of travel documents as means to detect victims of trafficking (UN, 2000: articles 11–13). However, because most victims are likely to pass border crossings unnoticed on valid documents, this type of practice appears to be of secondary importance (Segrave et al., 2009: 41–45).
Second, beside public actors, it has been argued that private actors can also play an important role with respect to detection (Bales and Soodalter, 2009: 254–258). For example, some NGOs have routines for receiving reports of suspected trafficking. Certainly, NGOs are likely to face the same problems as described earlier, but because of their status as nonstate actors it might be easier for some witnesses to give a tip. Moreover, it has been argued that more attention should be paid to private citizens and to the indirect profiteers from prostitution (e.g. hotels, airlines, cab drivers). The idea is that these types of actors can help to detect trafficking by reporting suspicious behaviour. On the other hand, to ask private actors to play a role is not unproblematic. There is a risk of reporting innocent people and the fact that much trafficking is controlled by organized crime raises questions about personal safety.
The second type of anti-sex trafficking behaviour is probably the one that receives most attention (Samarasinghe and Burton, 2007). The aim of prevention is to make it more difficult for traffickers from carrying out their plans. This could be achieved by establishing institutions that focus directly on the trafficking acts (e.g. recruitment and transportation) as well as rely on the indirect strategy of changing the demand and supply conditions for sex trafficking. Three preventive measures seem to attract particular interest, namely criminalization, norm-creation, and awareness-rising.
First, the UN Trafficking Protocol urges the signatory states to criminalize trafficking in persons (UN, 2000: article 5.1). This step has been rather easy to take. For example, in 2009, Estonia was the only European country without a specific offence of trafficking in persons (UNODC, 2009b: 5, fn. 3). However, despite this, there are still significant differences as regards the formal design of states' anti-sex trafficking institutions. This has partly to do with different governmental views on prostitution (Hughes, 2004: 73–83). For instance, there is a debate on whether prevention of sex trafficking is promoted by the criminalization of sex buyers (Sweden) or by the legalization of prostitution (e.g. the Netherlands). The expectation of the former is that criminalization will help reduce the demand while the latter believes that legalization will take away the reasons for smuggling.
Second, recalling the importance of informal rules, it has been argued that the success of anti-sex trafficking work rests partly on the attitudes of sex buyers (Kara, 2009: 33; Samarasinghe, 2008: 163–183; Torrey, 2004). That is, as long as there is a demand to purchase sex, it is believed that women will continue to be trafficked into the sex markets. Therefore, following this argument, it is concluded that norm-creation should play an important role in the combat of sex trafficking. The UN General Assembly, which partly has a normative function in world politics, took an important step in this process when it adopted the Trafficking Protocol in November 2000. By expressing norms against sexual exploitation, the Assembly laid a basis for discouraging the demand side of sex trafficking. As this requires norm-diffusion, states were also encouraged to undertake information campaigns and to cooperate with NGOs and civil society (UN, 2000: article 9). However, although there is a strong belief in norm-creation, this strategy has been criticized on several grounds (Doezema, 2010: 124–130; Segrave et al., 2009: 3–7, 16–17). For example, the effectiveness of the strategy is uncertain and, as argued by “pro-sex work” organizations, it violates sex workers' rights to make own choices. Another critique is that the strategy forgets about the supply side.
Third, awareness-rising has been described as a cornerstone of prevention (Samarasinghe, 2008: 189). By making potential victims and buyers aware of the phenomenon of sex trafficking, awareness-rising-campaigns can be used to attack both the supply side and the demand side of trafficking. The campaigns are often aimed at specific groups. For example, the purpose may be to raise awareness among young girls in a specific region or among foreign tourists.
Although the fight against sex trafficking largely depends on long-term preventive measures, a well-developed institution should include rules and practices for dealing with acute situations as well. One type of behaviour that fits into this context is protection (Crowhurst, 2006; Scarpa, 2008; Segrave et al., 2009: 176–180). That is, if there is a direct and imminent threat against a victim of sex trafficking, we expect states and NGOs to be guided by some established set of rules and practices for giving urgent protection. Because the goal is to provide for the physical safety of victims, protective measures usually focus on the establishment of safe environments. To accomplish this, victims are, for example, offered safe shelters and protected identities.
The protective element of the anti-sex trafficking institution has in recent years been criticised mainly on two grounds. One critique, directed at the UN General Assembly, claims that the Trafficking Protocol shows limited interest in protection and that the relevant provisions are not binding (Scarpa, 2008; Segrave et al., 2009: 18–19). Because of this, states are only committed to protection “in appropriate cases and to the extent possible under its domestic law” (UN, 2000: article 6.1). The second critique calls attention to a widespread neglect of the wishes and needs of the victims (Segrave et al., 2009: 119–120). If victims are excluded from the design of this part of the institution, there is a risk that they will develop resistance strategies (e.g. run away and maintain contacts with their traffickers).
“We are working in ‘crisis mode’ rather than on prevention” (Hotaling, 2004: 96). This remark, which was made by Norma Hotaling at a conference on sex trafficking held at DePaul University in 2003, expresses a common experience of anti-sex trafficking work. While there is considerable awareness of the need to focus on long-term preventive strategies, much of the actual work tends to consist of acute crisis management. Crises usually arise unexpectedly, put considerable values at stake, and require immediate actions. In the case of sex trafficking, a crisis describes a situation that (a) arises from an unexpected detection of trafficking, (b) put the victims' human rights at stake, and (c) requires immediate intervention. For example, in 2005 and 2006 the Netherlands experienced several cases in which young Nigerian women suddenly disappeared from refugee centres and a few days later turned up as street prostitutes in other West European countries (Radio Netherlands Worldwide, 2009). Considering that it was only a few hours before the victims were outside the jurisdiction of Dutch intervention, this type of disappearances could very well be interpreted in terms of crises.
The role of crisis management is not directly addressed in the UN Trafficking Protocol. As regards the academic literature, despite the crisis mode noted by Hotaling, scholars rarely use the term. On the other hand, it is quite common for empirical studies to pay attention to the separate criteria of a crisis (e.g. McCabe and Manian, 2010; Wylie and McRedmond, 2010). Moreover, it seems that crisis management can be of importance during all stages of the trafficking process. In other words, intervention could be made during recruitment and transportation as well as on arrival at the destination. By developing rules and practices for crisis management, authorities and NGOs can create opportunities for quick interventions that prevent crises from becoming even worse. However, interventions by public and private actors are not always unproblematic (Crowhurst, 2006: 222–223; Doezema, 2010: 140; Segrave et al., 2009: 108–114). To be effective, victims must be prepared to get out of the sex business and to cooperate with the authorities. Moreover, the literature contains many examples of “rescued” women who have been put into so-called “safe havens”, where they have been exposed to violence, racism, and sexism.
Sex trafficking has serious effects on victims' lives and bodies (Segrave et al., 2009: 97–98; Shelley, 2010: 60–65). The most common health consequences relates to three areas: (a) increased exposure to sexually transmitted diseases, (b) threats to mental health, and (c) difficulties with health care access (Samarasinghe, 2008: 31–32). Because of this, a well-developed institution should also have rules and practices for managing the consequences of sex trafficking. Rules can in this context aim both at taking care of victims (e.g. by guaranteeing access to health care services) and at limiting further consequences (e.g. by taking compulsory measures to stop the spread of infectious diseases). Practices are used in a broad sense, including practices for treatment as well as routines for meeting victims' special needs (e.g. privacy).
The UN Trafficking Protocol addresses consequence management, even though the term is not explicitly used. For example, article 6.3 refers to “measures to provide for the physical, psychological and social recovery of victims” (UN, 2000). However, despite this, we should expect to find considerable differences between national institutions for consequence management. One reason for this is the lack of precision, i.e. the vague wording of the article is open to different interpretations. Another problem is that this part of the protocol is not legally binding. Each state party “shall consider” implementing this type of measure, but there is no obligation to establish them.
Response is used as a generic term for all measures that, from the victims' point of view, mark the transition to the post-trafficking stage. By stressing transition, it is understood that response includes measures that put an end to the sex trafficking as well as help victims to start a new life. After reviewing the UN Trafficking Protocol (UN, 2000: articles 6–8) and current research (Bernat, 2011; Jayagupta, 2009; Jobe, 2010; Segrave et al., 2009), it seems reasonable to make a distinction between four types of responses. First, legal response refers to the investigation and prosecution of the offence of trafficking in persons. To make sure that victims' views and concerns are considered, states are recommended to provide legal assistance. Second, social response aims at helping victims of trafficking to start a new life. This means that victims are provided with, for example, appropriate housing, education, vocational training, and employment. The third type of response concerns residence permits. Many victims of trafficking become subject to deportation because they lack valid travel documents and awareness of their legal rights. Therefore, following the Trafficking Protocol, each state shall consider measures “that permit victims of trafficking in persons to remain in its territory, temporarily or permanently” (UN, 2000: article 7.1). The last type of response focuses on repatriation. Victims of trafficking have a right to return to their country of nationality. This right is recognized by the protocol, which also establishes that the receiving countries shall accept and facilitate the return (article 8). Some sending countries and international organizations have also designed special reintegration assistance programs.
Explaining the Design of Anti-Sex Trafficking Institutions
For a long time, it was common for scholars to assume that the formation and operation of institutions reflected rational choice (Peters, 2005: 12–15; Shepsle, 2006: 25–26). In other words, rules and practices were explained by making references to the intentions and capabilities of the designers. However, following the entry of the new institutionalism approach, this perspective has been criticized for ignoring important contexts as well as the dynamics between actors and contexts (Peters, 2005: 16–21; Rothstein, 1996: 153–156). In this section, I will therefore present four strands of the new institutionalism approach that each stress a particular context in which designers operate. The four “schools of thought” are labelled historical institutionalism, constructivist institutionalism, international institutionalism, and domestic institutionalism.
According to this school of thought, the design of institutions could not be understood without paying attention to previous rules and practices (Greener, 2005; Pierson, 2000a). The essence of this effect, which historical institutionalists usually refer to as “path-dependence”, is that institutions have a constraining effect on actors' preferences and future choices. That is, once rules and practices are in place they embed actors and forge particular paths of behaviour. The reason for this is that institutions tend to develop a stabilizing dynamic that reduce the effect of external pressures for change. As a consequence of this, it is common to assume that only external shocks will break down the stability or inertia of established institutions and path-dependencies. Interestingly, although the term shock is rarely used in the anti-sex trafficking literature, Ronald Weitzer notes that the frequent claim that trafficking has reached an epidemic level has a “shock value” (Weitzer, 2007: 455).
As regards the combat of sex trafficking, scholars often come back to two phenomena that reveal the presence of path-dependencies. In both cases, the paths weaken or even undermine the fight. The first type of path-dependence is associated with the organization of the state (Gallagher, 2006: 146–148; Jayagupta, 2009: 232–233; Samarasinghe, 2008: 116). One typical pattern for the last one and a half century, especially among Western countries, is that states embrace both a welfare state and a security state (e.g. the Police, the Armed Forces, the Intelligence Community, and the State Border Guard Service). Because these are based on separate sets of rules and practices, it has been common to define anti-sex trafficking as either a welfare task or a police matter. However, in order to achieve effective uses of national resources, some countries have implemented anti-trafficking reforms that aim at strengthened coordination across the civil-police divide.
The second type of path-dependence that poses a threat to anti-sex trafficking is corrupt practices (Dewey, 2008; Jayagupta, 2009: 231–232; Samarasinghe, 2008: 117–119, 156–157; Smith and Mattar, 2004: 172–173). The fact that sex trafficking generates enormous profits has created a breeding ground for corruption, which is especially noticeable in less developed countries and in war-torn societies. For example, studies of sex trafficking in South-East Asia and in former Yugoslavia report that local officials sometimes allow and even help the trade to continue (Samarasinghe, 2008; Segrave et al., 2009).
Following constructivist institutionalism, “institutions are socially constructed by the perceptions and cognitions of their members rather than being objective entities” (Peters, 2005: 117). This means that anti-sex trafficking institutions have no obvious design. Instead, they are subjected to cultural and cognitive variations, which might occur along national and sub-national lines as well as along organizational and professional lines. In other words, the design depends largely on how sex trafficking is socially constructed by countries, citizens, authorities, and occupational groups. Judging from a recent review of literature on the female sex trafficking discourse (Samarasinghe, 2008: 25–32), sex trafficking has been constructed as a problem that concerns gender rights, development, criminal networks, labour, abolitionist/anti-legalization, migration, and health.
The list of social constructions does not make any claims of being exhaustive, nor are the single problems necessarily mutually exclusive. For example, considering that human trafficking has become increasingly “presented as an organised crime threat to the state as well as a threat to state borders” (Lobasz, 2009: 323), suggests that security should be added to the list or be brought up in the analysis to highlight the link between criminal networks and migration (cf. Aradau, 2008; Friman and Reich, 2007; Jonsson, 2008).
The construction of the problem has implications for the choice of strategies and means. For example, if sex trafficking is seen as a threat to the state there will most likely be more focus on border control, migration control, and international law enforcement cooperation (Lobasz, 2009: 326–329). Clearly, the multiplicity of social constructions opens the door for several strategies and means for combating sex trafficking. On the other hand, some studies claim that a discursive shift is currently taking place, which might result in more focus on certain counter-measures (Penttinen, 2008: 15; Segrave et al., 2009: 5–6; Zheng, 2010). The meaning of this shift is that sex trafficking is increasingly presented as a human rights problem. As a consequence of this, we should expect that the promotion of human rights norms would play a more prominent role in the future. However, recalling that human trafficking is also increasingly seen as a security threat, suggests that institutional designers will have to pay more attention to the inherent conflict between human rights and security (cf. Lund Petersen, 2001).
International institutions provide a third context for the design of national anti-sex trafficking institutions. Resembling the definition given earlier, international institutions are established sets of rules and practices that prescribe appropriate behaviour for states and other actors in the international system (cf. Peters, 2005: 143–146). International rules and practices can have an informal status, but also be formalized by international conventions and include international organizations. As far as the anti-sex trafficking domain is concerned, there are four types of formal international institutions (Figure 2).
International institutions put pressure on the design of national institutions (Peters, 2005: 148). Exactly how this process works and with what effects is less clear. Leaving unlikely scenarios aside (e.g. in which external pressure automatically translates into national rules and practices), the research debate centres on mainly two assumptions (Kelley, 2004: 428). The influence of international institutions is supposed to be a result of either membership conditionality (the use of incentives and sanctions) or of socialization-based methods (the use of norms to persuade, shame, and praise certain behaviour). In the case of anti-sex trafficking, influence has to be explained mainly as a result of socialization.
Finally, the outcome of the international pressure might be difficult to establish. One indication of a significant effect is the fact that “the number of countries having anti-trafficking legislation more than doubled between 2003 and 2008 in response to the passage of the protocol” (UNODC, 2009a: 8). As of November 2008, nearly 100 countries reported that they had passed laws against trafficking in persons. On the other hand, this is not a guarantee for a more effective fight against trafficking. As is clear from several studies, many states lack the will or necessary resources to fully implement the UN Trafficking Protocol and other international agreements (Samarasinghe, 2008: 178–179; Segrave et al., 2009: 24–26; Smith and Mattar, 2004: 167–173). It is therefore not surprising to find references to “a general pattern of discrepancy between policy and practice” (Munro, 2006: 322; cf. Friman and Reich, 2007) or “that gaps between policies and practices have often been the norm” (Banerjee, 2006: 193).
The essence of the final explanation is that institutional design cannot be seen in isolation from the domestic political context (Peters et al., 2005; Pierson 2000b: 478–483). That is, institutional designers are affected by domestic politics and the competition between various interests for power and influence. Sometimes, the designers might even themselves represent a political or societal interest. The domestic political explanation presupposes that the domain is of significance for power and influence. If this is the case, political parties and interest groups might be lured into politicizing the issue, which lead to a risk that the design process will be drawn-out or subjected to frequent changes.
According to a widely spread stereotype, the combat of sex trafficking is not a political issue and the people involved are without selfish or parochial interests. This view is convincingly challenged by current research, which notes that the issue of sex trafficking on the contrary is “a highly political topic” (Penttinen, 2008: 15; cf. Aradau, 2008; Segrave et al., 2009) and that it has become “increasingly politicised in recent years” (Weitzer, 2007: 447). One explanation for the politicization is that trafficking is often coupled together with the issue of prostitution (Penttinen, 2008: 145, fn. 1; Samarasinghe, 2008: 4–6; Segrave et al., 2009: 2–3). In other words, the anti-sex trafficking domain has become yet another battleground for groups that are for or against prostitution. Regardless of which group that becomes predominant, we expect that any shift of balance between them will leave significant marks on the institutional design.
By using the term institution, recent studies of the combat of sex trafficking have opened the door to the new institutionalism approach. After elucidating the potential of this approach to map and explain national efforts to counter sex trafficking, I will conclude this paper by making three remarks that should accompany applications of institutional theory to this type of issues. The remarks bring attention to three research problems. The first problem concerns definitions. Because trafficking is sometimes mixed up with migration or people smuggling, analysts should be vigilant to if the institution has been designed specifically for the combat of sex trafficking or not. For example, if victims of sex trafficking are handled according to rules and practices for dealing with illegal immigrants, it means that states apply institutions that in reality have been designed for other domains. In these cases, it might be more relevant to talk about extra-institutional guidance or overlapping institutions.
The second problem is analytical. That is, how are we to understand the design of anti-sex trafficking institutions? Considering that the design takes place within four different contexts, it is expected that the outcome will vary considerably. As a matter of fact, only one context (international institutions) seems to be working unambiguously for convergence of national institutions. To find out what accounts for the design of anti-sex trafficking institutions, it is important to carry out systematic case studies across both space and time. The former type of case study aims at comparing national institutions at a specific point in time while the latter highlights their development, for example, since the adoption of the UN Trafficking Protocol in 2000.
The final problem concerns the possibility for scholars of recommending a policy or a specific institutional design. Following previous research, there is reason to be cautious about appointing best practices when it comes to the combat of sex trafficking. As stressed by Samarasinghe and Burton (2007), many projects are poorly funded, quickly set-up, and not rigorously monitored or systematically evaluated. Therefore, lessons can rarely be duplicated or simply transferred from one country to another. Moreover, because of a lack of accurate data on sex trafficking (Samarasinghe, 2008: 15–19; Segrave et al., 2009: 12–13; UNODC, 2009a: 18–19), it is impossible to establish how many victims are actually detected or enjoying protection and consequence management. Considering the hidden nature of sex trafficking, it is to be feared that a large share of victims are left to their fate. If this is the case, it might suggest that many victims practise some sort of self-help to manage the consequences (incl. the use of alcohol and drugs). On the other hand, this is not to say that policy recommendations are ruled out. For example, much research suggests that the effectiveness of anti-sex trafficking institutions depend on the inclusion of victims' perspectives (Segrave et al., 2009; Zheng, 2010), the attention to local rules and practices (Bernat, 2011; Jayagupta, 2009: 248–250), and the efforts to fight the demand and supply side of trafficking (Di Nicola, 2009; Kara, 2009: 33; Samarasinghe, 2008: 163–183; Torrey, 2004).
For helpful comments on previous drafts, I thank participants at the 2010 Stockholm Criminology Symposium, the Political Science Seminar at Södertörn University, and the anonymous reviewers. Thanks also to participants in the Masters Seminar in Gender Studies at Södertörn University for valuable comments and conversations. This research was supported by a grant from the Swedish Crime Victim Compensation and Support Authority (Brottsoffermyndigheten). The usual disclaimer applies.