Evidence assessment of the impacts of the criminalisation of the purchase of sex: a review

Review of evidence on the impact of the criminalisation of the purchase of sex.


1 The term 'prostitution' has been used throughout this report in line with the Scottish Government remit, however the contested nature of language in this area is acknowledged (see p7).

2 We are grateful to the Group for their contributions to the development of this report.

3 The limitations of this data should be borne in mind as should the limitations imposed by the time-frame and remit of this review.

4 Evidence from countries where legislation has recently been introduced, including France (in 2016) is largely unavailable at this time.

5 Although we are aware that this results in a context surrounding the presentation of evidence that may result in it being 'read' in a particular way. Nevertheless, the context is necessary in order to frame consideration of existing evidence.

6 This refers predominantly to women, who account for around 95% of those involved in prostitution, according to the English Collective of Prostitutes (see All-Party Parliamentary Group on Prostitution and the Global Sex Trade, 2014).

7 For a comparison of prostitution regimes across different countries see for example Kelly et al. (2009 and 2014); Barnett and Casavan (2011 revised 2014); Skilbrei and Holmstrom (2013).

8 For example, the complexities of law and ethics relating to the principle of harm and subsequent debates surrounding criminalisation in 'voluntary' prostitution are outlined by Persak (2012); while O'Connell Davidson (2002) notes the challenges of supporting the rights of those who work in prostitution as workers, remaining critical of the social and political inequalities that underpin market relations in general, and prostitution in particular. Scoular (2010) highlights the significance of law in the regulation of prostitution. Conversely, prostitution as exploitation and/or gender inequality and the associated harms within this context are presented by others such as Ekberg (2004 and 2013); Kelly et al. (2009; 2014); Waltman (2011a and 2011b); Coy (2012).

9 Cho et al. (2012) consider the potential effects of legalised prostitution on human trafficking in terms of two potential effects: substitution effect (away from trafficking) and scale effect (increasing trafficking). Their analysis of global data indicates that scale effect dominates. They note that the clandestine nature of both prostitution and trafficking makes it difficult to obtain hard data and therefore their finding is posited as based 'on the most reliable existing data, but needs to be subjected to future scrutiny' (p76).

10 Post 1990 English language studies conducted in Australia, Finland, the Republic of Ireland, the Netherlands, North America, Sweden and the UK.

11 Relatedly, Carson and Edwards (2011) in a discursive analysis of law and policy in Sweden and Victoria, Australia note the fewer internal contradictions within Swedish legislation as a result of its relatively clear philosophical basis.

12 The Swedish Government evaluation was primarily based on already published material supplemented with interviews with key service representatives. The English translation of excerpts of the report (Swedish Institute, 2010) provides details of the key reports, studies and information that form the basis of the evaluation conclusions.

13 "Prostitution represents an insidious form of abuse of women and men. Those involved are not only degraded by the act itself but are exposed to a significant risk of non-consensual sexual and physical abuse" (Lord Advocate's Guidelines, 2012: para 3).

14 The Equality Duty was introduced by the Equality Act 2006 and requires Scottish Ministers to set out priority areas across functions and activities relevant to Scottish public authorities. They must provide an update report every three years. This report, along with a report on occupational segregation, was produced in furtherance of this duty.

15 Scottish Government (2010) para. 3.11.

16 The Crown Prosecution Service of England and Wales, have expressly encompassed women involved in prostitution in their policy on violence against women www.cps.gov.uk.

17 Available figures are for the time period 2003-2013: see Freedom of Information Request published 01 July 2014 available at http://www.crownoffice.gov.uk/foi/responses-we-have-made-to-foi-requests/41-responses2013/472-prosecutions-prostitution.

18 While a number of studies have been conducted on the purchasers of sexual services, this is discussed here in relation to potential deterrent impacts. A number of problems (methodological and ethical) have been associated with such studies - they are referred to here by way of context rather than as central to the evidence review.

19 This study is based on a small number of respondents accessed via Snowball and Respondent Driven sampling and limited by only 4 available options to the question on deterrence. Other responses available to respondents were: payment of fine, public exposure (both responses were selected by 49% of respondents).

20 To which 86 academics and researchers responded, writing to Members of the European Parliament to voice their concerns about the report on the basis of inaccuracies and misrepresentations which they claimed meant that the report was an inadequate basis on which to hold a vote in the European Parliament. 79 academics and researchers wrote in support of the resolution: http://www.troubleandstrife.org/wp-content/uploads/Honeyball_support_letter_FINAL.pdf

21 Debates and consultations are ongoing in a number of other countries.

22 Generally referred to as prevalence in relevant literature.

23 As noted above, terminology is problematic here - while this could also be seen to include third parties involved in selling sex (pimps, brothel owners etc) it is used here to refer to individuals who sell sexual services.

24 Rasmussen et al (2014) note an international increase in numbers involved in prostitution in 2008 which they attribute to the onset of the global financial crisis.

25 Waltman (2011a and 2011b) cites these figures and others from the SOU (2010) report, leading Skilbrei and Holmstrom (2013) to note that while they support his position, his uncritical acceptance of official figures may benefit from further questioning.

26 Providing English extracts from the official report SOU, 2010).

27 Ostergren in particular, has been criticised for her methodological approach (see Waltman, 2011).

28 This is reflected elsewhere (for example in Scotland) where local priorities can determine the implementation and enforcement of legislative priorities.

29 Australia, Finland, Germany, Ireland, Netherlands, New Zealand, Spain, South Africa, Sweden.

30 For example, Rasmussen et al (2014) note that timing of observations may be a potential source of error, however even when this is taken into account, the same trends are evident.

31 This report was commissioned by the Norwegian Ministry of Justice and Public Security in 2013 to evaluate the legislation. We have accessed the English summary and are unable to consider the detailed information contained in the main report.

32 Although, this evaluation has been criticised by organisations including the Ombudsman for Discrimination, the National Board of Health and Welfare, and the Swedish Agency for Public Management ( SOU, 2010; Dodillet and Östergren, 2011; Jordan, 2012).

33 Fredlund et al. (2013) compared their survey results with a similar survey conducted by Svedin and Priebe's (2007) in 2004. Fredlund et al. used a modified version of the survey conducted by Svedin and Priebe. Based on a representative sample of 3,498 young people, they suggest a significant shift (among those young people who reported ever having sold sexual services) (n=51) to contacting 'buyers' via the internet (57% of those young people who reported ever having sold sex in 2009 had made contact with the 'buyer' over the internet compared to 17% in 2004). Numbers involved are small.

34 Ekberg (2004) notes that the National Organization for Women's Shelters and Young Women's Shelters in Sweden made calls for the criminalisation of men who purchase sex part of their yearly Plan of Action in 1987.

35 In the 2002 survey, it was asked whether it should be illegal to buy sex whereas in the 2010 survey it was asked 'should we retain the law prohibiting the purchase of sex?'

36 Conducted by a research organisation and commissioned by the tabloid press this survey involved telephone interviews with 1000 individuals.

37 This considers the average difference over time in attitudes between Sweden and Norway.

38 Buying sex became a criminal offence in January 2009 in Norway so the survey collects data on attitudes before and after the law.

39 Finland has criminalised clients of trafficking victims; however, not knowing that the person was trafficked can be a defence.

40 It is clearly a challenge to evidence 'normative' impact (see Scoular, 2015 for further discussion).

41 Buyers of sexual services were asked what they would do if 'paying for sex was a crime'. Only 16% of respondents said they would stop paying for sex altogether (and 20% of respondents from Northern Ireland) whilst the majority response (If I was in a relationship) was selected by 35% of respondents (33% of Northern Ireland respondents). Other responses included: 'If I could have sex without paying for it' (27%) 'If my partner found out and wanted me to stop' (19%) 'If I could express my sexuality/sexual preferences without paying for it' (14%) 'If there was less stigma around my sexual preferences' (5%) and 'none of the above' (12%).

42 We have been unable to access the original report by the Nordic Gender Institute.

43 Some of the data can be challenged where trafficking for labour and that for sexual exploitation are not sufficiently distinct.

44 Cho et al. (2013) investigated the impact of legalised prostitution on trafficking flows from 150 countries and despite the limitations of the data, which they acknowledge and outline in their paper, conclude that: countries where prostitution is legalised report a larger incidence of trafficking flows; and that legalised prostitution results in increased demand.

45 It is easier to produce evidence of criminal guilt when the police wait till sex has taken place; but as noted by the Norwegian Ministry for Justice and Police Affairs (2004) and Swedish Institute (2010) this is in conflict with the police duty to prevent criminal actions.

46 It has been pointed out that incidents of serious violence (rape and assaults) actually fell, however, we have only been able to access the summary version of this report.

47 Their study does not draw conclusions specifically related to the impacts of legislation criminalising purchasing of sex as this was a policy rather than legislative approach but is considered worth referring to here.

48 Specifically, in the 8-month period post-policy implementation, 24.6% (58/236) of sex workers experienced work-related physical and sexual violence (as compared with 23.7% (65/275) interviewed in the 8 months pre-policy in 2012). However, it should be noted this refers to a policy by the Vancouver police rather than a legislative change.

49 However, as Florin notes, this is based on a non-random selection by the prostitution units with the majority of those selected claiming to have stopped selling sex prior to the intervention (p275).

50 It is not clear what proportion of this 181 permit holders included 'buyers' as opposed to 'vendors'.

51 Ekberg notes this information is based on personal conversation in 2001.

52 As outlined in Scottish policy documents.

53 Similar challenges were experienced by the House of Commons Home Affairs Committee (2016) who examined a range of legislative approaches to prostitution but were unable to draw clear conclusions on the basis of available evidence.


Email: Justice Analytical Services

Back to top