On Our Own Terms: The Working Conditions of Internet-Based Sex Workers in the UK^[1]

Introduction

1.1 It is only recently that discussions around the quality of work and job satisfaction of sex workers have been brought to the research agenda. This tardiness has been for a range of reasons, which in turn reveal broader political dynamics that shape how we come to understand the sex industries, the direction of research in this field, and what knowledge is privileged. It can be said that the often stifling nature of legal rhetoric, reform and policy around commercial sex has drawn the 'research eye' away from asking new questions and applying robust methodological design to answer those queries. A key reason why quality of work questions have not been asked about sex work until recently is because of the ideologies which have dominated debates around the nature of selling sex for exchange.

1.2 The feminist debates on the 'prostitute body' have taken their toll on broader conceptualising around what is going on in the commercial sex industry, as the arguments become reduced down to the 'coercion' versus 'choice' dichotomy (Scoular 2016). Such simplistic binary ideas about the nature of sex work do not reflect any type of work especially the complexities of using sexual labour to make a living (Connelly 2016; O'Connell Davidson 1998). Radical feminists have spent many years claiming that a woman engaged in selling sex cannot be a sex 'worker' because they are treated as a sex 'object' due to patriarchal and structural power relations, leaving no possibility for consensual commercial sex (Barry 1995; Jeffreys 2008). Radical feminist accounts generally focus on street sex work because, while this is a small sector of the industry, it predominantly contains the highest levels of vulnerability (Brown and Sanders 2016). This is done at the cost of discounting other forms of sex work present in the UK, namely indoor markets including brothels, escort agencies and those working independently in privately rented flats. They also exclude male sex work as this goes against the premise that prostitution is male violence against women.

1.3 Radical feminist arguments have been met by the sex positive alliance who point out that selling sex is sexual labour, prioritising the focus on emotional, physical and sexual skills that the sex worker performs (Brewis and Linstead 2000; West and Austrin 2002; Vanwesenbeeck 2005). While the body/sex work literature is now flourishing (for a review see Walkowitz et al. 2012), the bifurcation of 'exploitation' and 'choice' continues to cloud investigation into the condition of sex as work (Weitzer 2012; Lewis et al 2005). This paper in part addresses this lack of inquiry, excavating further into the lived experiences of sex workers and circumnavigating the ideological 'sex wars' (Weitzer 2007) which often represent the political manoeuvring of privileged white female scholars. This paper focuses on the area which we know least about yet is the most significant change – alongside migration (Agustín 2007; Andrijasevic 2010) – to influence the sex industry in recent decades: digital technologies and the sex industry. Using empirical quantitative findings from a survey (n=240) of mainly female independent escorts working in the UK, the paper begins by setting out the context of online sex work and the turn to digital commercial sex.

Selling sex online: escorts operating through the internet

2.1 Research is beginning to demonstrate the everyday realisation that the majority of sex work is now mediated by or provided through digital technologies and that internet-based sex markets are pervasive (Cunningham and Kendall 2011; McClean 2015; Pitcher 2015; Sanders 2005; 2008b; 2009) and is estimated to be a $1 billion dollar industry (Minichiello and Scott 2014). Beginning around the 1990s, commercial sex is now largely marketed through web-based advertising (Pajnik 2015) and facilitated by computer-mediated communication such as email, chat-rooms, social media forums, phone, and webcam (Jones 2015). We are aware of the migration to online platforms and services by witnessing the decline of the two traditionally predominant markets: street-based markets have shrunk significantly in addition to a reduction of managed premises for selling sex in the form of brothels, flats, or saunas (Brooks Gordon et al. 2015; Scoular and O'Neill 2007; Brown and Sanders 2016). Instead the organisational features of cyberlife have dramatically affected the organisation of the sex industries.

2.2 Yet, as a result of policy and discourse focusing largely on female street sex workers, contemporary research has lagged behind the online developments, leaving much research enterprise to be done to unearth the complexities and creativities of internet-based sex work. In outlining the state of the commercial sex terrain, Jones (2015: 558) explained that 'we still do not know enough about how the Internet has affected the work experiences, wages and working conditions of many sex workers', calling for a comprehensive research agenda to address this lack of knowledge. Relatively little is known about internet-based sex workers as initial research in this area has focussed on the health of male sex workers as a specific group (Koken et al. 2004; Walby 2012; Minichiello and Scott 2014), as well as how male customers use the internet to locate and review sex workers (Soothill and Sanders 2005).

2.3 We do know that despite the rise of online communities and forum spaces (Ashford 2009), many internet-based sex workers in the UK operate in total isolation, because if they work collectively they are exposed to the laws relating to brothel management and controlling for gain (Pitcher and Wijers 2014). We also know that there can be detrimental economic effect of the online markets as competition is high and it has been established through analysis of profile sites that there is a wage penalty for condom use (Adriaenssens and Hendrickx 2012). Yet still little research has focused on experiences of violence and crime in the course of online sex work, apart from a small number of studies showing that negotiating risk and safety issues are an everyday reality and an often unmet concern (Davies and Evans 2007; Laing and Pitcher 2013) with different issues presenting for male escorts (Bryce et al 2015). With this state of knowledge as the background to establishing the nature and characteristics of online sex work, our questions were motivated by the search for more information particularly around working patterns, job satisfaction, and experiences of crimes and exploitation.

Measuring Quality of Work: Exposing Working Conditions

3.1 This piece of research was funded by the Wellcome Trust (Grant number 105446) in 2015 and was carried out in partnership with the organisation National Ugly Mugs (hereafter NUM). NUM provides a service which enables confidential reporting of dangerous incidents and assaults for sex workers. Members of the scheme consist of sex work support projects, individual sex workers (predominantly female independent escorts), and sex businesses, such as brothels or escort agencies. NUM takes reports of incidents from its members, which go to one centrally-coordinated web-based hub and members receive legally compliant warnings via email and a mobile phone app. NUM work closely with the police, acting as third party facilitators where sex workers are worried about reporting their experiences.

Recruitment and Sample

3.2 Through a number of sex work research forums between academics and sex work support organisations, the researchers had an existing relationship with NUM as the leading safety organisation for sex workers in the UK. The Principle Investigator and the CEO of NUM are engaged in policy forums and policing advisory groups, where the relationship between research and partners can be established and enhanced in the broader production of evidence based policy. NUM facilitated access to their individual membership which at the time was approximately 2,500 people. This was highly beneficial to the funded pilot study which had to be completed within six months, including recruitment of staff through to the dissemination of findings. The intense time constraint limited the researchers' ability to make relationships with other online platforms in order to recruit and disseminate the survey. NUM representatives were subsequently involved in the construction of the survey, which aided the design of questions and reflected the issues that were reported to NUM on a daily basis.

Sample limitations

3.3 The survey only recruited from NUM's membership – that is, respondents had to be a member of NUM in order to complete the survey. This meant that our sample was restricted to those who had signed up for this safety service (which is free and open to all working in the sex industry). Hence there were limitations to the sample in using this population as the majority of NUM members are White British females and there were large numbers of highly educated women. As such, there is an underrepresentation of other ethnicities as well as those working in street markets. Further a key limitation to the sample is that because there were no funds to translate the survey into other languages, the survey is biased against those who cannot read/write proficient English, migrants who do not have access to signing up to the NUM services, and of course those who are experiencing strictly coercive circumstances. We surmise that members of NUM could also potentially have greater control of their working lives and have greater freedom to access safety support such as that offered by NUM. Further, as the survey was administered through email, the sample required a level of access to and literacy of online technologies.

The Survey

3.4 This project was designed as a pilot study to capture baseline data in order to develop a much larger project on the internet and sex work (which was funded by the Economic and Social Research Council 2015-2018: Beyond the Gaze: Working practice, safety and regulation of internet-based sex work – see Sanders, Scoular et al 2015). The pilot survey aimed to reveal some core demographic information about online sex workers, as well as information about issues such as: work satisfaction, work histories, stigma, isolation, safety, contact with support services, and crimes experienced. The survey, hosted by Bristol Online Surveys, ran for 12 weeks from November 2014 to the end of January 2015. Data from the 240 responses was imputted into SPSS, where it was coded and analysed. The use of online methodology was only suitable because we were reaching a sample who used the Internet to organise their work, but also because it reflected how work more broadly is bound up with the online experience (Pettinger & Lyon 2012). Overall we received approximately a 10% response rate which is a robust return for an online survey. Of the 240 completed surveys there was 100% response rate to all questions though note that within this there were responses such as 'rather not say/prefer not to say'.

3.5 The survey consisted of 63 questions and took approximately 15 minutes to complete. We administered a pilot survey before it was launched to the NUM membership to time the survey, in addition to test the questions and identify any errors. There were a range of open, scale and multiple choice questions for respondents to answer. Questions were organised by theme, including social demographics, work histories, working arrangements, job security and satisfaction, control at work, relationships with clients, stigma, and safety and crime. There are a number of areas worthy of expansion here. Firstly, when asked about working arrangements, respondents were asked to consider their time at work and how it is spent, income and methods of payment, costs of living expenses and overheads associated with the work, numbers of clients, advertising methods, and the level of respect received in communication with clients and other sex workers. We measured job security by the length of time expected to remain working in the sex industry, and the level of control they felt in deciding this, including their perceived ability to stop sex working if desired. Job satisfaction was measured by asking sex workers to choose from a large selection of positive, negative and neutral terms associated with work. These included asking sex workers whether they associated terms such as 'flexible', 'rewarding', 'stimulating', 'demanding', 'repetitive', 'time-pressured', or 'dehumanising' with their work (in addition to other descriptors). We also asked about the levels of support they received from colleagues, managers, friends and family, online services, and specialist support projects. Control at work was measured by asking respondents to rate the extent they "can make their own decisions at work', are 'able to refuse clients by their own judgement', feel 'under pressure from managers at work', whether they 'can decide how often and how long they can work' and 'when to take time off work'. Respondents were also asked whether they 'set the prices for their services' and if they 'decided what to wear and how to present themselves'.

Ethical Considerations

3.6 A number of ethical issues were considered in the course of this research. Respondents were asked to read an information page and consent form at the beginning of the survey and had the option to agree to the research or to leave before any questions were asked. Respondents were also advised that they could withdraw from the survey and retract their answers after they had been submitted within a designated timeframe. Contact details of the Principle Investigator were shared for respondents to use if they required further information, to discuss the research, or to withdraw from the research. This served the aim of helping respondents feel comfortable with their involvement in the research and aid their understanding of how to withdraw at various points if they wished to. We also worked to reduce the potential for ethical complications in our partnership with NUM, in order to avoid misinterpretation that the services of the organisation were somehow dependent on their completion of the survey. To reduce such implications, the information provided at the beginning of the survey made clear that the research was being undertaken by the University of Leeds, and while we felt that the results would be insightful for sex work researchers more generally, it was in no way linked, or dependent upon, the delivery of NUM's usual safety service. It was also explained to respondents that NUM were hosting the survey in order to access those working in the sex industry. Further, we involved NUM in the development of the survey questions so those within the organisation felt comfortable that the survey would not pose any challenges to the existing relationships and levels of trust they had developed with their service users. The information page and consent form documented the purpose, aims and details of the research project to ensure that the research, its proposed benefits and the research relationship between the researchers and NUM were made clear. The information page and consent form also detailed the arrangements for confidentiality upheld throughout the research. The survey did not ask for identifiers, such as name or precise living location, confirming that the data is only presented in its combined form in order to protect the identities of individual respondents. We did not collect any IP addresses so there could be no traces and this was stated in the opening statement for prospective respondents.

Socio-Demographics of Sample

Table 1. Descriptive statistics for socio-demographic variables

3.7 Reflecting the membership demographics of NUM, respondents were mainly female (82%; n=196), without dependent children (77%; n=184). Almost half (46%; n=111) defined their sexuality as 'bi-sexual'. In relation to ethnicity, the sample was mainly comprised of white respondents, with the vast majority self-defining as 'White British' (73%; n=175) or 'Other White' (13%; n=31). As highlighted in Table 1, the age of respondents was varied, with the youngest respondents being 18 years old and the oldest being 63 years old. Reflecting the possible trend that those who work indoors are slightly older than street workers (Sanders 2006), the mean average age of respondents was 36 years old. The age when respondents reported having started working in the sex industry varied considerably from 12 years old to 56 years old. The mean average age of entry was 29 years old, again signifying that those who consider independent escorting are often older and have had previous work experiences in the mainstream economy (Sanders 2005).

3.8 Many respondents had also worked in other sex markets before independent escorting: 35% (n=84) had engaged in webcamming; 30% (n=71) had worked for an escorting agency; 24% (n=58) in brothels, parlours or saunas; and 8% of respondents (n=20) had worked in the street sex market prior to independent escorting. This points to independent escorting being part of a longer trajectory within the sex industry, with sex workers moving between, and overlapping, different sex markets over time.

3.9 In relation to respondents' current housing situation, the majority (57%; n=137) were privately renting accommodation at the time of completing the survey. A further 47 respondents (20%) stated that they were living in a home that they owned, while 27 respondents (11%) were living in social housing. A small number were living with family (5%; n=12) or in their partner's home (3%; n=6). In terms of education level, 38% of respondents (n=90) hold an undergraduate degree, which is similar to the general population (40%), according to 2011 census data (Office for National Statistics, 2014). Postgraduate qualifications were held by 17% (n=40) of respondents.

3.10 Respondents reported working across a range of sex markets (see Table 1). The most common sex market that respondent reported working in was 'independent escorting', with 214 (89%) participating in this type of sex work. The second most common was 'webcams' (n=44), although this represents only 18% of respondents, followed by 'phonelines' (13%; n=30). It is clear that respondents commonly work across two or more sex markets. The length of time spent sex working varied greatly between 1 year and 35 years, with the mean average sex work career being 6 years among this sample.

Job Satisfaction

4.1 There have been concerns raised about the relationship between the rise of new media as a tool that perpetuates unequal gender relations for sex workers in the digital environment (Pajnik et al. 2016). The evidence below taken from our survey explores job satisfaction and work experiences of sex workers who operate online, how they view their work and the daily experiences of autonomy.

Describing sex work in their own terms

4.2 The question over autonomy and control are often those which are raised in the broader feminist debates around the nature of sexual labour (Chapkis 1997). Indeed the binary positions around 'exploitation' and 'choice' regarding sex work use polarised ideas around 'force' and 'freedom' to conceptually understand experiences (Phoenix 2009). Taking our cue from other quality of work/life measurements, and in order to move away from these pervasive ideological positions, we were keen to ask about job satisfaction. Sanders had previously asked about levels of job satisfaction amongst women who worked as strippers (Sanders and Hardy 2014), demonstrating the range of qualitative characteristics which make up levels of job satisfaction. Like in the strippers survey (ibid) levels of job satisfaction were high amongst the survey sample of internet-based sex workers. When asked to describe their work, respondents typically selected positive or neutral words (see Figure 1).

Figure 1. Descriptors of Sex Work

4.3 Out of the top seven characteristics chosen to describe their work, six of the characteristics were positive: 91% of sex workers described their work as 'flexible' and 66% described it as 'fun'; over half of respondents (56%) found their job 'rewarding', 'skilful', 'sociable' and 'empowering'. The main negative descriptor (the second most popular) was 'unpredictable' stated by 76% of respondents which continues to establish the precarious nature of sex work as an income generator. On the other hand, comparatively few selected the more negative descriptors for sex work. Indeed, only 15% believe their job is dehumanising and 14% believe it to be exploitative.

4.4 To ensure that responses were not delineated too greatly by pre-assigned categories and to give respondents some space to write freely about how they considered their work experiences, we asked them to comment in the form of an open response on the three aspects they enjoy most about sex working. The most common responses were 1) financial rewards; 2) flexible working hours; 3) the choice and freedom they have in their decisions around when, where and how to sell sex. These three responses are powerful indicators about the levels of job satisfaction amongst the sample, particularly in relation to the levels of control they experience in their everyday work. Conversely we asked about the three aspects respondents dislike most about their job: 1) time-wasting clients; 2) stigma and poor attitudes towards sex work; 3) the dangers of sex working and violence experienced. Potential customers wasting the time of sex workers, either through enquiries by email or in person, or arranging bookings and not showing have been well documented as a major irritation of sex workers (Sanders 2005), which ultimately costs them time and money. The other two issues regarding stigma and negative attitudes towards sex work and the fear of crime in relation to their work are unsurprising as they have been documented over many years (Day 2007; Kinnell 2008). The complex issues around the dangers and experiences of violence for internet-based sex workers will be discussed below.

Control in their work

4.5 Overall responses indicate that respondents revealed a great deal of choice in their decision making. We were not particularly surprised by this finding given the composition of the sample and the reality that those who are working under coercive conditions are most likely to not be NUM members, or have access to the survey. Using a series of likert scale questions (see Table 2) we asked respondents to tell us about their everyday labour relations with clients, third party organisers and co-workers in order to gather a sense of the levels of control and decision making power individuals had over their sex work activities.

Table 2. Descriptive statistics for likert scales (%)

4.6 These results indicate that for this sample of independent escorts there was a high level of control and autonomy over their working patterns and daily decisions in their sex work. What we can assert is that there were very few respondents who had their work patterns dictated to them, suggesting that there were low levels of third party control experienced by the sample. Indeed further, we found that 79% of respondents (n = 190) are able to refuse clients all of the time, compared to 0.4% (n = 1) who could never refuse a client. This suggests that from the sample of 240 there was very low levels of control evident, confirming that the sample were largely independent, self-employed sex workers controlling many aspects of their work, time and client encounters.

Relationships with clients

4.7 While there is now a robust literature on clients (although mainly men) who purchase sex (see Sanders 2008a; Birch 2015; Hammond 2015), within this, sex workers opinions of clients are seldom sought. We were keen to ask respondents about their relationships with clients given this is a significant part of their work. The majority of respondents appear to have good relationships with their clients. Indeed, when asked to rate their relationship on a scale of 1 (positive) to 10 (negative), 90% of respondents fall on the 'positive' end of the scale (see Table 2). Respondents were also asked whether or not the online communication they receive from clients and potential clients is respectful. By far the most common response was that communication is respectful 'most of the time' (73%; n = 176). The question about helping clients with their emotions and relationships elicited the most variable responses. That said, most fell towards the end of the scale signifying that they do help with emotions and relationships. This supports existing research which elucidates that emotion and intimacy can be a key part of commercial sexual relationships (Sanders 2008b; Milrod and Monto 2012).

4.8 We were keen to establish the level of control sex workers had over the sex they sold: 56% of respondents (n = 138) have control over the sex acts they perform 'all of the time', 24% (n=58) of respondents said 'most of the time', compared to 2% of respondents (n = 4) who 'never' have control over their sex work. Interestingly, given what we know about the high rates of violence experienced by sex workers, most respondents positioned themselves closer to the 'never' end of the scale when asked about client's aggressiveness. That said, the majority also indicated that if clients do become aggressive, individuals are able to manage the situation. Indeed, while 25% (n = 61) said they can manage this 'all of the time', only 2% (n = 4) noted that they could 'never' manage client's aggression suggesting a small number of very vulnerable sex workers. While response to this question varied along the spectrum between 'all the time' and 'never', there was an observable trend towards sex workers having more, rather than less, control over aggressive clients.

Experience of Crimes

5.1 There is significant evidence of the burden of violence against sex workers as a global issue (Deering et al. 2014), and it remains that in the UK sex workers are 12 times more likely to be murdered than their non-sex work female peers (Ward et al. 1999). Safety is a significant concern for all sex workers. The types of crimes that are feared range from hate crimes motivated by whorephobic/homophobic/racist attitudes as well as other crimes such as robbery and men posing as clients who then commit serious violence and rape (see Campbell 2014). While there has been evidence that indoor markets are safer (Kinnell 2008; Sanders and Campbell 2007), indoor sex workers are increasingly experiencing violence (Boff 2012; O'Doherty 2011) with a trajectory of migrant sex workers being targeted as the most vulnerable group of sex workers across markets (Brown and Sanders 2016). Our findings (see Figure 2) suggest that the online world of sexual services may well pose new forms of exploitation and risk for sex workers specifically in relation to crimes that originate through digital communication (such as harassment by persistent email or phone calls) and then move offline into physical crimes such as assaults or physical stalking. About half of respondents (47%) had been victims of crime during the course of their sex working. Most commonly this manifests in the form of threatening or harassing texts, calls and emails and verbal abuse, with 36% and 30% of respondents experiencing these forms of victimisation, respectively. Respondents had also experienced more serious forms of violence such as robbery, rape, physical assault and attempted abduction (see Figure 2) – but these forms were less common.

Figure 2. Experiences of Crimes

5.2 While established research evidence suggests that indoor sex workers generally experience less violence compared to street-based sex workers (Sanders and Campbell 2007), the rise of online sex work and the migration to this mode of working may be prompting changes in crime patterns. Crimes experienced by sex workers can particularly focus on issues around privacy, online identity and ensuring images are only posted with the sex worker's consent.

5.3 We asked our respondents if they feel confident in their own ability to maintain their online profile in a safe way. The results are recorded in Table 3. The majority of respondents (57%) indicated that they do have the skills and technology knowledge to keep their work safe online.

Table 3. Descriptive statistics for likert scales (%)

5.4 A further 50 respondents (21%) indicate that they sometimes feel confident, while 27 respondents (11%) do not feel confident in maintaining a safe online profile. Fewer respondents indicated that they feel confident in their ability to protect their identity online. That said, 96 respondents (40%) still indicated that they feel confident, and a further 67 respondents (28%) believe that they sometimes feel confident in protecting their online identity. 42 respondents (18%), however, do not feel confident in protecting their identity online and 24 (10%) were not sure. Respondents were asked to indicate their level of confidence in their own ability to remove their online profile and traces of personal information held online should it be necessary. Responses to this question were very varied. The most common response was still the 'yes' option, with 80 respondents (33%) selecting this. However, a similar number of respondent (32%) selected the 'no' option, indicating that they do not feel confident in removing online personal information. When we interpret these findings with those in Figure 2 which highlights the levels of digitally facilitated crimes, we can see that this is a new area where sex workers are possibly vulnerable to perpetrators and those who target sex workers driven by a range of whorephobic attitudes.

5.5 It appears that what we have traditionally understood as processes whereby sex workers can be victimized and exploited (Prior et al. 2013) may well be taking a digital turn as technology is used against them as they are exposed to computer related privacy issues which threaten their ability to keep their sex work hidden and their identity secret. These findings support initial suggestions from a study of female British escorts who reported harassment and crimes on a forum website (Davies and Evans 2007) where it was concluded that the Internet had introduced a new threat to business through various forms of electronic abuse. While sex workers have a myriad of strategies to protect their identity and are often aware of strategies to protect themselves, the risks from the digital world are complex and ever changing.

5.6 While the types of crimes that sex workers are experiencing adds a nuanced understanding to the literature about the victimization of sex workers in the digital age, it was not particularly surprising that levels of confidence in the police are low given difficult relationships with the police have been documented in the UK (Boff 2012). Almost half of respondents (49%) were either 'unconfident' or 'very unconfident' that police would take their reports of crime seriously. Only 19% were either confident or very confident that the police would treat their crimes as like any other. Essentially, we can conclude then where sex workers are experiencing online harassment, or crimes which start online then move to off-line physical attacks and violence, sex workers are most likely not to report to the police. This emphasises the importance of NUM, which acts as a third-party mediator between the police and sex workers in cases where the sex worker would like to anonymously report crimes to the police.

Concluding Comments: Autonomy, Rights and Protection

6.1 There is evidence from this specific survey of mainly female internet based sex workers in the UK that job satisfaction is high with decision making powers evident across aspects of sex work. Relationships with clients are generally positive, although reports of experiences of crimes were significant (n = 47%) with limited confidence reporting to the police. Whilst these findings are specific to this sample (of which the limitations are discussed above), we can place them in the context of the sex work debates which frame the discussions around the coerced/choice continuum, much of which makes little sense to either the complexities of research findings or individuals' experiences of their situation. This research shows that for independent sex workers with these characteristics there is clear agency and decision making power in most work activities, particularly over crucial issues such as when to work and which clients to accept. Yet at the same time for this group there are consistent issues around safety and protection.

6.2 These findings are important as although they speak from a specific sample there is evidence to demonstrate that there are groups of sex workers who are not experiencing coercion to sell sex, remain in sex work, or exercise minimal control over the use of their own bodies. Instead their daily practices of doing sex work are governed by their own autonomy. Issues relating to exploitation, controlling behaviours, limited access to choices around working conditions or patterns do exist in very small numbers in this sample suggesting that this experience is in the minority. There is evidence that in general this sample of sex workers have significant decision making power, something which is usually denied by arguments that reduce sex workers to the status of only victims, or refuse to engage in the context in which sex workers experience their work. Arguments that prefer to remain at an ideological level of whether it is 'right' for individuals to engage in sexual labour, or prefer the discourse of false consciousness whenever sex workers express their freedoms are presented here with a set of circumstances which refute such claims.

6.3 Yet these points we make about the autonomy in sex work should not trivialise the concerns that this research has flagged up in relation to issues around the law, regulation and policing practices/culture that compromise the safety of this group and leave serious questions regarding the protection of sex workers. Some researchers have argued that although the workplace may have changed for online sex workers, their experiences of work or the services that are being sold may not have changed (Walby 2012). It is emerging from this data that there are new dangers apparent which affect the everyday experiences of working online for sex workers. Here we suggest new trends through crimes mediated through phones, texting and email. As ongoing research by Sanders and Scoular et al (2015) notes, the rise of online platforms where sex workers can upload profiles at minimum costs, present as minimal organisational risk yet masks the dangers such as the misuse of information, 'doxing' of personal information without consent (also see Jones 2015), stalking and harassment through digital technologies. It is these areas where sex workers experience infringements of privacy and become victims of crimes, usually with limited power to speak out to the police because of fear of what will happen and the stigma that will result. Digitally facilitated crimes (which usually takes the shape of relentless harassment and verbal abuse) usually remain un-reported to the police, hidden from detection figures and rarely acted upon in terms of any form of investigation of abusive behaviour. We know this because the data indicates the very low levels of confidence in reporting crimes to the police, often because of previous bad experiences or because of a fear that they will be arrested for selling sex. These new types of crimes experienced online can be understood in the continued context of the criminalisation of sex work in the UK, which significantly affects the ways in which sex workers can access ordinary routes to protection and justice.

6.4 Within this changing sexual commerce landscape the question around the quality of work of those involved in sex work becomes ever pressing, as the need for accurate data to inject into policy debates around the legal status of sex work become a permanent political issue in many European states and beyond. There is a need to make further comparisons of sex work with other types of work in order to establish how sex workers job satisfaction and working patterns compare to other, similar, types of workers as well as placing the crimes experienced by sex workers online into the context of rights to justice and protection.

---

Acknowledgements

This article was supported by grant number 105446 from the Wellcome Trust. Thanks to Alex Feis-Bryce and Rosie Campbell for their guidance and contribution throughout the project.

This work is licensed under a Creative Commons Attribution 4.0 International License.

---

References