Abstract
More research has been conducted on the legalization of some vices than others. One severely under-researched topic is the legalization of prostitution. Activists, politicians, and pundits frequently make grand claims about the dangers or benefits of reforming prostitution laws, but their claims are typically ideologically and politically driven, rather than being evidence-based. This Research Note presents a unique set of studies that employ either (1) a longitudinal design that evaluates key metrics before and after legalization in a particular location or (2) a comparison of legal and illegal sex workers operating in the same location, across a few cities with different policy models (criminalization, decriminalization, legalization), or across cities that vary in the extent to which they abide by national legal norms. All studies in the second category are small-N case studies, which I argue are far superior to superficial large-N designs, including two studies that attempted to assess outcomes of criminalization and legalization across 161 countries. The small-N case studies discussed here identify key outcomes of both criminalization and legalization, and their research designs can serve as a models going forward.
Introduction
What happens after an illegal vice has been decriminalized and legalized? Does legal reform enhance the health and safety of participants? Does it have the opposite effect, rendering actors more vulnerable to exploitation or substance dependence than they were when the activity was criminalized? Or is some vice policy-resistant, with little change on key metrics before and after legalization? The history of alcohol prohibition and its aftermath and the legalization of casino gambling and cannabis suggests that law reform can lay the groundwork for an array of improvements compared to the eras when these vices were criminalized (Leitzel 2008).
We know much less about the aftermath of decriminalizing and legalizing prostitution. Grand claims are commonly advanced in the public square by activists and elected officials who decry liberalization of the law. But those claims tend to be hyperbolic, politically driven, based on selective “evidence,” or grounded in faulty logic (Shrage 2021; Weitzer 2011; Zatz 1997).
This research note explores the following methodological questions: What kinds of data are suitable for measuring the impact of legalization? What is the appropriate research design: Cross-sectional, longitudinal, qualitative? What is the appropriate level of aggregation: multinational, national, local? And what outcomes are most relevant?
Measuring Vice Law Reform’s Impact
Prostitution is legal and state-regulated either nationally or regionally in Australia, Austria, Belgium, Germany, Mexico, the Netherlands, New Zealand, and Switzerland. These are not the only countries with liberal prostitution laws, but I know of no case-study research on those countries. Some nations and states permit only one limited type of prostitution and are thus problematic to include in the “legal” category, since so many other practices are criminalized. In Britain, Denmark, and Tasmania (Australia), only independent escorting is legal. Escort agencies, brothels, and street prostitution are outlawed. Indian law similarly allows only independent operators to work lawfully, despite the abundance of illegal brothels in the country. In the United States, prostitution is permitted only in a few rural counties in Nevada—and only in highly regulated brothels. Fully 85 percent of the state’s population lives in places where all prostitution is illegal.
It is worth noting that most of the countries in the opening sentence above are Western; most have fairly good human rights records; and their human trafficking policies and practices are rated highly. Apart from Mexico, each of them scores in the “best practices” top tier (out of four) in the U.S. State Department’s annual Trafficking in Persons Report and have remained in that position over the two decades that the Department has conducted its assessment. This does not necessarily mean that legalization translates into superior antitrafficking enforcement practices, but it does clash with the claim that liberalization of the law begets more sex trafficking (Dorman 2022; Goldberg 2014; Guinasso 2022).
Mirroring the prohibition of vices such as alcohol, drugs, and organized gambling, research discussed below shows that illegal prostitution is associated with a host of negative outcomes. This does not necessarily mean that decriminalization/legalization is superior. A vice can be policy-resistant, with legal reform having little or no impact on participants’ health and safety. For prostitution, the case studies reviewed below suggest otherwise.
In identifying the most important outcomes of vice law reform generally, a summary measure is the “health and safety” of participants. But other metrics are important as well: entry pathways, working conditions, relations with third parties, job satisfaction, self-esteem, policing practices, and impact on surrounding communities (Leitzel 2008; Weitzer 2012). If trafficking networks are heavily involved in supplying workers where prostitution is criminalized, if working conditions are poor, if third parties tend to coerce and exploit workers, if abuse by customers occurs frequently, and if the authorities tend to compound the harms—can these problems be ameliorated once prostitution is decriminalized and legalized?
I distinguish decriminalization (i.e., removal of criminal penalties) and legalization (regulations imposed by the state postdecriminalization). Most of the countries that have liberalized the law have opted for legalization in order to exert control over the enterprise. Legalization may include formal intake interviews and registration of workers, mandatory health examinations, official screening and licensing of business managers, “sin taxes” not applied to other businesses, zoning, advertising restrictions, or periodic site visits by the authorities. Maximalist regimes impose all or most of these regulations (e.g., Nevada, Austria, Germany) while some others opt for a minimalist form of legalization (e.g., New Zealand, Switzerland, Victoria and New South Wales in Australia). Most of the countries or states that have liberalized the law continue to outlaw or discourage street prostitution, because of its adverse externalities in host communities (Weitzer 2012).
I argue that cross-sectional research, especially on a large number of countries at a single point in time (Cho et al. 2013; Jakobsson and Kotsadam 2013) is incapable of measuring and explaining outcomes traceable to legal reform. These two studies attempted to measure the impact of legal versus criminalized prostitution on rates of human trafficking across 161 countries. The research designs were bedeviled by reliance on unstandardized data sources across countries, flawed categorization of legal regimes, or other serious deficiencies (Hedlin 2017; Weitzer 2025; see “Conclusion”). If a cross-sectional comparative study is limited, however, to two or three cases where full information is available and the comparison is carefully conducted, this can be a worthwhile approach—as reflected in some of the studies discussed below.
A preferable research design is longitudinal: measuring outcomes before and after legalization in a particular locale. A longitudinal design allows the analyst to directly measure the impact of law reform. The challenge: The success of a longitudinal design depends on the availability of solid data both prior to and after a vice is legally liberalized. Such studies are therefore rare but include some of those described below.
Scoping Reviews
A sizeable amount of research indicates that criminalization adversely affects sex workers’ health, rights, and safety by stigmatizing and marginalizing them and alienating them from the authorities and social services (Levy 2014; McCann, Crawford, and Hallett 2021; McCarthy et al. 2012; Östergren 2020; Platt et al. 2018). Whether exploitation and victimization are rife or rare situationally or structurally (e.g., varying by type of prostitution), abuse remains an occupational hazard. The illegality of the work makes sellers more vulnerable to bad actors and organized crime, which thrive under conditions where a desired service or commodity has been criminalized.
Some studies focus on criminalization’s impact on health outcomes. A recent scoping review concluded that criminalization adversely affects the health of sex workers, their customers, and public health more generally (Platt et al. 2018). Another analysis found that those who work in decriminalized or legalized systems demonstrated a range of positive health practices and outcomes (McCann et al. 2021). Other scoping reviews reach similar conclusions (Deering et al. 2014; McCarthy et al. 2012; Vanwesenbeeck 2017). While such reviews are geared to producing generalizable knowledge, case studies are uniquely well suited to identifying the factors responsible for negative and positive outcomes after legal reform. Below, I address this issue with an exploration of case studies conducted in Indonesia, Switzerland, Belgium, Australia, Mexico, and Germany. The cases were selected based on the availability of either small-N comparative or longitudinal studies relevant to the impact of legal reform.
Case Studies
A natural experiment was conducted after brothels in one district in East Java, Indonesia were outlawed in 2014. Police conducted raids on the brothels, but illicit transactions continued surreptitiously at many of them. Health workers had previously conducted exams and provided free condoms at the brothels, outreach that ceased. To measure the impact of criminalization, researchers conducted a longitudinal and ecological analysis: before and after the law’s change in the targeted district in comparison with a bordering district where prostitution remained legal. In contrast to the latter, criminalization in the other district was associated with (1) a surge in sexually transmitted infections due to the lack of condoms (STI prevalence was stable in the other district); (2) declining weekly earnings, inadequate to meet workers’ needs; (3) an increase in the number of children working to support their families; (4) a decline in sex workers’ happiness quotient; and (5) increased street prostitution (Cameron, Seager, and Shaw 2021). This study’s research design—both longitudinal and geographically comparative—highlights multiple ways in which criminalization can produce or intensify risks and harm to both participants and the wider public.
Since 1992, federal law in Switzerland has allowed each of the country’s cantons to determine the kinds of prostitution that they will permit and outlaw, as long as the work is voluntary and restricted to adults. Zurich is known for street prostitution and “drive-in” prostitution (in a car park), whereas Geneva hosts cabaret-type bars, erotic massage parlors, and some window prostitution. Sex workers in Geneva must register with the police and then attend an information session run by the sex workers’ organization Aspasie. Importantly, local authorities periodically monitor erotic businesses by conducting site visits to ensure compliance with municipal regulations. Due to this oversight and collaboration between government officials and the managers of sex businesses throughout Switzerland since 1992, there has been “a low level of criminality in the sex industry” (Chimienti and Bugnon 2018:148).
Research offers another revealing finding, both comparatively and longitudinally. Many of the providers who work in Geneva are migrants from Spain. According to researchers, the Spanish women consistently report that the working conditions, earnings, and relations with customers are far superior in Geneva than they were in Spain—a longitudinal measure. Compared to Spain, sex businesses in Geneva take a lower commission fee from workers; condoms are provided free; and, unlike Spain, customers in Geneva were “more polite, did not haggle over prices, were cleaner, and they complied with the established negotiation” (Meneses-Falcón 2026:240). Compared to Spain, which has become less sex worker-friendly recently, Spanish sex workers experienced Geneva as “safer, less exploitative, and more financially beneficial”; the in-take interviews and official site visits stymied the development of trafficking networks; and the study concluded that “decriminalization and regulation of sex work in Switzerland, and especially in Geneva, is a very attractive aspect for Spanish sex workers” (Meneses-Falcón 2026:242–246). This study had the advantage of being both comparative and longitudinal in the sense that it tracked workers’ perceptions and experiences over time from one national context to another.
Belgium decriminalized prostitution in 2022 and legalized it in 2024 by enacting a package of rights and regulations: social security and unemployment benefits, maternity leave, screening of prospective business owners, requiring on-site alarm buttons, and formalizing the right to refuse a client or a particular sex act. A municipality may impose other rules. In Antwerp, novice sex workers are required to undergo an intake interview with city officials. Some of these regulations were in place prior to 2022, when third-party involvement was illegal but de facto allowed and unofficially regulated in Antwerp, Brussels, Ghent, and a few smaller cities (Weitzer 2014). Regarding the question of whether legalization improves participants’ health and safety, the evidence to date in Antwerp and Ghent points in a positive direction. Both city’s window-prostitution red-light districts feature a city-funded health center that is exclusive to local sex workers, most of whom access its free counseling and health services. (The latter are also available to, and accessed by, sex workers who work independently outside the red-light zone; these freelance workers are also subject to an intake interview and are periodically visited by city officials.) Earlier research in both Antwerp and Ghent reported a range of benefits stemming from the authorities’ de facto regulation and monitoring of managers and workers in their red-light districts (Fuchs 2013; Loopmans and van den Broeck 2011; Weitzer 2014; Weitzer and Boels 2015), oversight now reinforced de jure by national law.
Brussels profiles quite differently. City officials and the police have largely opted not to impose regulations in the window-prostitution district, jeopardizing the health and safety of participants. In the past as well as today, Brussels’ authorities have not routinely patrolled the area or screened the workers or their managers, many of whom are connected to organized-crime networks based in Eastern Europe or West Africa. Unlike Antwerp and Ghent, the city’s red-light district ranks high on both physical disorder (trash, graffiti, decaying buildings) and social disorder (normlessness reflected in episodic microaggression toward window workers, theft from and assault of some visitors and clients by local predators, lack of security cameras) (Weitzer 2014; Weitzer and Boels 2015). Brussels’ laissez-faire approach underscores the importance of implementing and enforcing national-level regulations, just as it illustrates the need to conduct comparative research at the local, city level.
An impressive test of policy models was conducted in Australia, where it is possible to compare very different legal regimes across states. Researchers conducted site visits at brothels and interviewed 604 brothel workers in a criminalized system (in Perth, Western Australia), a decriminalized one (in Sydney, New South Wales), and a legalized regime (in Melbourne, Victoria). The latter, legalization, was found to be superior to both criminalization and decriminalization on measures of brothel cleanliness, lighting, health practices, supportive staff, and security measures (Harcourt 2010). Field researchers also gave each regime a 1- to 5-star rating based on an overall assessment of the brothels as worker-friendly: 20 percent of the brothels in the legalized system were awarded five stars, compared to 3 percent in the decriminalized case and 0 percent in the criminalized system. Perhaps surprisingly, brothels in the decriminalized system in Sydney were most likely to be ranked low: 33 percent of them received a 1-star rating. Recall that decriminalization means virtually regulation-free, unlike legalization.
Another Australian state—Queensland—legalized independent escorting (in 1992) and brothels (in 1999); other types of prostitution remain unlawful. Two studies compared those who work legally with their criminalized counterparts: The latter profiled much worse on mental health indicators, intravenous drug use, and victimization (assault, rape) by a customer (Seib, Fischer, and Najman 2009). Regarding organized crime in the legal sector, a government inquiry reported that both were negligible. More generally,
There is no doubt that licensed brothels provide the safest working environment for sex workers in Queensland. Legal brothels now operating in Queensland provide a sustainable model for a healthy, crime-free, and safe legal licensed brothel industry [and are a] state-of-the-art model for the sex industry in Australia. (Crime and Misconduct Commission 2004:75, 89)
Escorting also profiled quite positively compared to the illegal market.
At least one type of prostitution is legal in about half of Mexico’s states. Tijuana allows us to compare legal and illegal sex work in the same location. Legal sex work is restricted to certain areas of the city, where sex businesses must be licensed and the authorities conduct periodic site visits. Legal workers must register and carry health cards indicating they have tested negative for STIs. Like Queensland, research by Katsulis (2009) shows that the legal sector offers several advantages over the illegal market. First, legal workers had superior health profiles. Second, they were half as likely as their criminalized counterparts to have been assaulted or robbed and three times less likely to have experienced extortion or violence by police officers. Third, legality was associated with intangible benefits. Compared to those who work illegally, the legal workers
enjoy better working conditions and job satisfaction, less fear about the nature of their work, a higher degree of sophistication and confidence with regard to their speaking skills, appearance, and demeanor. Registration and monthly checkups appear to encourage behaviors that are protective of health as well as provide a barrier against police harassment. Registration increases the sense of legitimacy and community and is correlated with much lower levels of depression and mental stress. (Katsulis 2009:77)
The intangible dividends of working legally—legitimacy, self-confidence, a sense of community, job satisfaction, self-esteem—are important but under-researched aspects of liberal law reform.
The Tijuana findings are replicated elsewhere, at the most micro level: a study of a legal mega-brothel that has operated in the Mexican state of Chiapas since 1991. Like Tijuana, some of the regulations—for example, mandatory 3-month STI exams at a nearby clinic—are intrusive. Otherwise, the legal workers have a “great deal of freedom and exercise control over their work” (Kelly 2008:79). They alone decide their rates, when to work and for how long, and who they will serve; they come and go as they please and take extended leaves to visit family members. Almost all of the 140 women at the brothel work independently, with only a dozen connected to a pimp. On a good day, they can earn ten times the daily minimum wage in Chiapas. The women use their earnings to buy items for their children in addition to consumer goods that they otherwise could not afford, such as nice clothing, cell phones, and jewelry. This bolsters their self-esteem.
The brothel is located outside the city of Tuxtla, where street, bar, and home prostitution are illegal. The police frequently raid these places and arrest the workers, many of whom are minors (Kelly 2008). Like Tijuana and Queensland, the findings highlight sharp contrasts between parallel legal and illegal systems.
A law passed in Germany in 2002 defined sex work as an occupation, tightened the definition of third-party offenses, and gave workers the right to social security and unemployment benefits. Otherwise, the law granted each state and municipality the power to implement their own regulations. This led to an assortment of local norms across the country. A 2017 federal law negated some of these state and municipal regulations by substituting a set of nationwide rights, obligations, and regulations.
At the national level, the number of criminal charges involving third parties declined significantly after the 2002 law was enacted. In 2000, there were 1,888 state-certified victims of third-party “exploitation of prostitutes” (Criminal Code §180a), a figure that steadily declined after legalization: to 36 in 2019. Over the same time span, the number of officially certified victims of pimping/procuring/promotion (§181a) dropped from 1,304 to 149 (Henning, Hunecke, and Walentowitz 2021).
These figures may be interpreted in different ways, but they are at least suggestive of a positive trend since legalization. The 2002 law redefined third-party offenses. Unlike the previously broad offenses of pimping or promoting prostitution, the law now criminalizes specific acts, such as preventing someone from leaving prostitution, dictating the type of services to be provided, coerced servicing of a particular client, or restricting one’s freedom of movement (§181a). Similarly, “exploitation” is now defined as keeping an employee “personally or financially dependent” on one’s business operation, such as debt bondage that becomes more onerous over time (§180a). In short, statutory changes—specifying particular harms—explain the steadily declining number of individuals implicated in these crimes. While these national-level data differ from the others described in the article, the figures have the advantage of being longitudinal.
No one would claim that legalization is a panacea. In any given context, bad actors may find ways to involve themselves in sexual commerce post law reform. A good example is Staiger’s (2022) ethnographic study of a legal brothel in Germany. Staiger found that some pimps, posing as customers, had infiltrated the brothel. These men seek to develop friendships or emotional ties with specific workers in order to later control and exploit them financially. Most of the women viewed the pimps with contempt and denied that they were associated with them, and Staiger ultimately concludes that only a minority were so connected. The study’s contribution is to note that sidelining bad actors is difficult when they can assume the persona of an ordinary customer. It is unknown whether or to what degree nefarious third parties are involved in other German sex businesses, but they would be liable for arrest if their behavior contravened §181a of the 2002 law. Regarding the brothel workers, they were generally satisfied with their interactions with other customers, had full autonomy in deciding who to flirt with and service, and cherished the income, gifts, and opportunity to work legally (Staiger 2022).
Research at the city level confirms the importance of a micro-level research design. As in Belgium, German cities differed in the way local authorities responded to the passage of the 2002 law. In Leipzig, for example, officials continued to pathologize sex workers by requiring them to register with the authorities and via a carceral approach to street prostitution. Dortmund took a softer, inclusive approach. City officials engaged in collaborative outreach and incorporation of sex worker representatives and erotic business owners, treating them as professionals whose interests were generally respected by the authorities (Dölemeyer, Pates, and Schmidt 2010). The 2017 federal law mentioned above was an attempt to standardize local authorities’ regulation and engagement with legal workers.
Conclusion
When it comes to something as controversial as prostitution policies, a research design and associated findings matter not just academically but potentially more broadly. Earlier in the article, I mentioned two multinational cross-sectional studies that attempted to assess the performance of 161 countries that varied in whether prostitution was legal or outlawed, then compared these laws to each country’s performance in combatting human trafficking (Cho et al. 2013; Jakobsson and Kotsadam 2013). Both studies reported that human trafficking writ large (not the proper measure: sex trafficking) was more prevalent in countries where prostitution was legal than where it was criminalized. In a recent article, I offered a comprehensive critique of these authors’ assumptions, research designs, and conclusions. I also documented how their flawed conclusions had been embraced by major news outlets, in some legislative debates and bills domestically and internationally, by the Canadian government during a recent constitutional challenge to its national prostitution law, and in Google’s topline response when asked if legal prostitution was related to sex trafficking (Weitzer 2025). Whatever other problems such large-N studies may have when applied to a relationship as complex as prostitution law and (widely problematic) country-level figures on human trafficking, the fact that these deeply flawed studies have had an outsized impact in the public square and have now become the conventional wisdom highlights the importance of an alternative research design: carefully conducted small-N case studies that are either comparative or longitudinal.
It is easier to document the adverse effects of criminalization—as the scholarly literature consistently does—than to identify outcomes that are directly traceable to legalization. My review of case studies, most of which are little known, suggests that we can stipulate certain advantages of legal prostitution, but much hinges on the kinds of regulations governing any given sector or locale as well as how well these norms are implemented and enforced at the local level (Weitzer 2012). The Brussels case illustrates how national law in Belgium can be disregarded—in terms of deficient regulation and oversight—at the city level, whereas the authorities in Antwerp and Ghent have shown much more commitment to implementing national norms and monitoring the operations of sex business owners and managers. The case of East Java shows how criminalization of previously legal commerce can impair the quality of life of sex workers and their children. And the systematic comparison of diverse policy models in three Australian cities shows that legalization can be superior to both criminalization and laissez-faire decriminalization.
Future research has the greatest potential if based on a research design that is comparative or longitudinal. A comparison of a few settings at a single point in time can pay dividends—demonstrated in some of the studies described above—but the N needs to be small in order to control for extraneous factors and properly investigate what is happening on the ground. Even more promising are longitudinal studies of a single setting, a design suited to identifying both changes over time and their causes. Again, this means that the scope should be limited: preferably to the state or city level. Larger levels of aggregation—national, multinational—are detached from key micro processes and outcomes. As noted above, many of the countries that have legalized prostitution at the national-level delegate implementation of the law to states or municipalities that often impose additional regulations of their own. But even where national law is determinative, the most meaningful studies will be conducted at the local level.
