Sage Journals: Discover world-class research

Abstract

Sex tourism is a persistent and complex dimension of global mobility, often operating in legal and enforcement grey zones. Using a two-stage mixed-effects gravity model on a 1995–2018 dyadic panel (150+ countries), we estimate residual tourism flows unexplained by standard economic, geographic, and cultural factors. We find that destinations identified as sex tourism hubs attract significantly more inbound tourists than predicted. This effect is strongest where sex work is legal but weakly enforced. Economic inequality between origin and destination countries amplifies these dynamics. Interactions show that stronger rule of law attenuates excess flows, while permissive legality combined with low enforcement magnifies them. By isolating “excess” arrivals beyond structural determinants, we provide empirical evidence that sex tourism is shaped by institutional ambiguity and global income gaps. Policy implications point to coordinated, harm-reduction regulation that targets governance capacity and cross-border enforcement rather than legality alone.

Keywords

sex tourism tourism commercial sex institutions prostitution informal economy

Introduction

Tourism is one of the most dynamic sectors of the global economy, shaped by economic opportunity, cultural exchange, mobility technologies, and the institutional frameworks that govern cross-border movement. Over the past three decades, trade liberalisation, cheaper air travel, and the rapid diffusion of digital platforms have expanded international mobility and reduced the frictions associated with searching, booking, and coordinating travel. Tourism now accounts for roughly 10% of global GDP and about one in 10 jobs worldwide (World Tourism Organization, 2024). This expansion is not simply a market outcome; it is also a product of institutional arrangements that enable movement, such as visa waiver agreements, bilateral air service treaties, and regional blocs. Visa facilitation alone can raise arrivals by 5–25%, particularly where latent demand is high (Neumayer, 2010; World Tourism Organisation & World Travel and Tourism Council, 2012).

A growing literature shows that institutions shape tourism performance by affecting perceived safety, predictability, transaction costs, and the credibility of regulatory environments. Institutional quality – often proxied by governance, rule of law, and control of corruption – is consistently associated with stronger tourism outcomes (Alvarez and Campo, 2014; Arezki et al., 2009). Importantly, institutions also condition tourism’s resilience to shocks: when uncertainty rises, destinations with stronger governance frameworks appear better able to sustain arrivals. Recent evidence shows that good governance can moderate the negative effect of global uncertainty on tourism demand (Koçak and Cobanoglu, 2025), and that democratic institutions similarly mitigate the adverse effects of pandemic-induced uncertainty on tourism demand (Kizilkaya et al., 2024). Together, this work motivates a core theoretical premise that guides our empirical design: institutional quality influences tourism not merely by shifting demand up or down, but by structuring the conditions under which mobility and market exchange occur, including risk, trust, information, and enforcement credibility. These mechanisms are likely to be most consequential for tourism segments that rely on discretion and regulatory ambiguity, implying that institutional environments should predict systematic deviations from gravity-based ‘expected’ flows rather than only average differences in arrivals.

This perspective also implies that institutions shape not only the volume of tourism but also its composition, influencing whether travel is oriented toward leisure and business tourism or toward more contested and informal niches (Richter, 1989; Williams and Hall, 2000). Consistent with this view, recent research indicates that tourism development may alter the balance between formal and informal economic activity and produce heterogeneous distributional outcomes across development contexts, underscoring that tourism’s broader consequences are conditioned by the institutional and developmental settings in which it is embedded (Kahyalar et al., 2024; Xu and Lv, 2023). Many tourism-related activities operate partly outside formal regulation, and the gap between “rules on the books” and “rules in practice” can change the visibility, organisation, and profitability of activities that sit in regulatory grey zones. In these settings, formal legality is not sufficient to characterise market conditions: the same legal code can generate very different market realities depending on enforcement intensity, corruption, and state capacity (Helmke and Levitsky, 2004). This is especially relevant for niches where discretion and selective enforcement may shape both perceived risk and expected costs for visitors and intermediaries.

Sex tourism is a clear case where legality, enforcement discretion, and governance capacity can jointly structure market conditions. Tourism economies often coexist with informal and illicit activities operating in regulatory grey zones ranging from unlicensed guiding and street vending to drug markets, commercial sex, and trafficking (Crick, 2012). In such contexts, institutions do not simply permit or prohibit activity; they determine the degree of ambiguity and risk faced by prospective buyers and sellers. Where enforcement is inconsistent, institutions can inadvertently create conditions that increase discretion, reduce perceived risk, and lower expected costs, particularly when cross-country income differences increase purchasing power abroad and expand the economic incentives for transactional exchanges (Helmke and Levitsky, 2004; Oppermann, 1999). This is particularly plausible in ‘grey zone’ settings where formal rules and on-the-ground enforcement diverge. For example, some jurisdictions prohibit commercial sex on paper but tolerate it in practice through selective enforcement; others permit parts of the market while restricting solicitation, advertising, third-party facilitation, or venue licensing. In such environments, discretion and uncertainty can create informality rents, shaping where and how visitors perceive access and risk. These features should matter for travel niches that rely on ambiguity, weak oversight, and intermediated access.

We use the Centres for Disease Control and Prevention (CDC) definition of sex tourism as travel planned primarily for sexual services, often to destinations characterised by permissive legality or permissive enforcement (Lu et al., 2020). Sex tourism spans male and female markets, heterosexual and same-sex markets, and includes child sex tourism in its most exploitative forms (Brents and Hausbeck, 2007; Cabezas, 2008; Graburn, 1983; Richter, 1989; Rivera, 2015; Taylor, 2006). While often framed as a contemporary phenomenon, sex tourism is historically and structurally linked to commercial sex markets that provide the infrastructure for transactional encounters (Salom, 2008). Digital platforms and social media can further reduce search and coordination frictions by connecting travellers to venues and service providers and embedding commercial sex within broader mobility networks (Lu et al., 2020). These dynamics raise important policy and public health concerns and intensify ethical dilemmas where tourism demand intersects with coercion, exploitation, and harm affecting vulnerable groups. Although this study does not directly measure individual harms, the institutional environments that enable discretionary enforcement and low accountability are frequently those in which worker protections and safeguarding capacity are most limited.

Despite an extensive qualitative literature on motivations, identities, and local consequences, important theoretical and empirical limitations remain. First, existing research is stronger at describing demand-side motives, participant profiles, and case-specific dynamics than at explaining why sex tourism systematically concentrates in particular destinations and not others. Second, prior work often treats legal status as a key explanatory feature, yet formal legality alone may not adequately characterise market conditions where enforcement is selective, informal, or weak. In such contexts, the gap between rules on the books and their implementation in practice may be more consequential than legality itself. Third, because sex tourism is partially hidden and difficult to observe directly, the literature has lacked a tractable macro-comparative strategy for identifying its structural correlates across countries. A complementary macro-institutional account suggests that cross-border transactional sex is more likely to concentrate where purchasing power gaps between origin and destination are large, where commercial sex market infrastructure is present, and where legal status interacts with enforcement capacity to create permissive or ambiguous regulatory environments that reduce expected costs and increase perceived access.

This study addresses these gaps by providing internationally comparative evidence on how institutions, legality, enforcement capacity, and global economic asymmetries shape sex tourism patterns embedded within broader tourism flows. Using a dyadic panel of bilateral tourist arrivals and a two-stage approach, we first estimate expected tourism flows using a structural gravity framework and then analyse the residual (“excess”) of bilateral tourist arrivals that remains after accounting for standard gravity determinants such as economic size, geography, and cultural proximity. We then test whether these excess inflows are systematically associated with sex-tourism market indicators and whether these associations are conditioned by legality, enforcement capacity, and cross-border inequality. In doing so, the study contributes to both the tourism and institutional literatures in two ways. First, it reframes sex tourism as an institutionally structured component of tourism composition, shaped not simply by individual demand or formal legality alone, but by the interaction of market infrastructure, regulatory ambiguity, and enforcement discretion. Second, by isolating excess flows beyond structural tourism determinants, it offers a tractable macro-empirical strategy for studying a sector that is difficult to observe directly, while generating testable implications about how governance and legality jointly condition contested tourism niches.

Background

Scholarship over the past three decades has generated important insights into sex tourism’s drivers, consequences, and intersections with gender, class, race, exploitation, public health, and policy (Hillis et al., 2022; Nørfelt and Kock, 2024). However, this literature remains stronger at documenting participation patterns, harms, and case-specific dynamics than at providing an integrated explanation of why sex tourism systematically concentrates in particular destinations. To clarify the mechanisms most relevant to the present study, this section reviews five themes of scholarship: empirical and measurement constraints, demand-side participation patterns, documented harms and ethical dilemmas, institutional and legal grey-zone mechanisms, and the role of global inequality in structuring directional flows. Taken together, these themes motivate an account of sex tourism as embedded in broader tourism systems, institutional conditions, and political economy asymmetries.

Empirical constraints and measurement challenges

Empirical progress has been constrained by the industry’s covert nature. Much of the evidence base relies on ethnographies, case studies, and surveys, with relatively limited cross-country quantitative research on system-level drivers (Lu et al., 2020). Two constraints are central. First, commercial sex markets are illegal or partially illegal in many jurisdictions, restricting observability and increasing reporting bias. Second, where commercial sex is tolerated, enforcement can be discretionary, shifting activity into “grey zones” that are difficult to capture in official data. Consequently, a large share of research has focused on rights, exploitation, and the economic pressures shaping participation by states and individuals (Leheny, 1995), alongside measurement problems inherent in informal and illicit markets (Hall and Brown, 1996). These constraints motivate approaches that infer sex tourism-related patterns from broader mobility data rather than attempting to observe the sector directly.

Demand-side mechanisms and participation patterns

The micro-level literature often describes sex tourists as disproportionately male, frequently travelling for work or long stays (e.g., military or business travel), and sometimes engaging in higher-risk behaviours such as heavy alcohol or drug use (Lu et al., 2020). Importantly, sex tourism is not always the sole purpose of travel. Rivera (2015) notes that some participation is opportunistic and occurs alongside general tourism, implying that sex tourism can be embedded within and amplified by mainstream tourism flows. Consistent with this, sex travel motivations overlap with broader tourism motives such as novelty seeking, unmet preferences at home, emotional longing, and convenience, rather than constituting a completely distinct category of travel (Hakim, 2015; Svensson et al., 2018). This is important for macro analysis: if sex tourism is partially nested within ordinary travel, then destinations associated with sex tourism may exhibit tourism volumes that exceed structural expectations even after controlling for conventional determinants.

Consequences, externalities, and ethical dilemmas

Beyond motivations, the literature emphasises substantial externalities associated with sex tourism. In many destinations, sex tourism is discussed alongside risks of coercion and exploitation of vulnerable cohorts, and in some contexts, sex tourism and trafficking are intertwined, particularly where institutions are weak and protective services are limited (Oppermann, 1999; Thanh-Dam, 1983). Public health concerns such as STI transmission risk, condom negotiation, and access to prevention and care are recurrent themes, especially where stigma and informality reduce health service access (Hillis et al., 2022; Lu et al., 2020). At the community level, the distribution of benefits and harms can be uneven: tourism revenues may accrue to intermediaries and venue owners while risks are borne by sex workers, particularly in informal settings with limited labour protections (Hall and Brown, 1996). These realities generate ethical dilemmas for policy: prohibition can push markets further underground and increase vulnerability, while permissive frameworks without enforcement capacity may enable exploitative segments to persist. This study does not estimate these harms; rather, it identifies the economic and institutional conditions under which sex-tourism-associated excess travel is more likely, contexts that prior work links to elevated health, exploitation, and human rights risks.

Institutions, legality, and enforcement as market-shaping mechanisms

A core theme of the sex tourism literature is that the institutional environment shapes the structure and visibility of commercial sex markets and, by extension, sex tourism. Legal status alone is rarely determinative; what matters is the interaction between legality, regulatory design, and enforcement capacity. Where sex work is legal or tolerated, but enforcement is weak, markets may become more visible and commercially organised; where sex work is illegal but weakly enforced, markets may persist in informal forms with greater discretion. These “grey zone” dynamics help explain why sex tourism can concentrate in locations where legal ambiguity and enforcement discretion reduce perceived risks for visitors and intermediaries while simultaneously limiting protections for workers. This broader emphasis on governance is consistent with Sou and Vinnicombe (2023), who argue that governance influences tourism not only through direct demand-side effects, but also through the wider supply-side environment by shaping investment conditions and destination competitiveness. Cross-national legal asymmetries may also matter, travellers may seek destinations where the expected costs of participation are lower and anonymity is higher, or where markets offer specialised or stigmatised services not available at home (Ying et al., 2022; Ying and Wen, 2019). Evidence from China, for instance, points to convenience and time constraints as drivers for some travellers purchasing sex abroad (Ying et al., 2022; Ying and Wen, 2019). More broadly, institutional arrangements governing mobility, such as visa facilitation, may interact with these markets by reducing access costs and expanding the pool of potential tourists.

Global inequality and the political economy of directional flows

Inequality is a consistently identified structural theme in the sex tourism literature. Travellers from wealthier origins can convert purchasing power advantages into status and access in lower-income destinations, creating conditions conducive to transactional relationships (Brennan, 2001; Lu et al., 2020). This is consistent with broader tourism research showing that tourism’s effects are not uniform, but vary with underlying development conditions, with tourism growth affecting inequality differently across economic contexts (Camacho and Ramos-Herrera, 2025; Wang and Tziamalis, 2023). From a behavioural ecology perspective, Kock (2021) argues that economic and social “fitness” constraints at home can be offset by travelling to poorer markets, increasing access to partners. These mechanisms help explain why many well-known sex tourism destinations are located in developing or lower-income countries and why flows often run from higher-income origins to lower-income destinations.

At the destination level, tourism can be central to development strategies, but local attitudes toward sex tourism are frequently ambivalent and shaped by hardship and opportunity constraints (Lu et al., 2020; Taylor, 2006). In such settings, households may rely on sex work income, often in informal markets with lower pay and higher risk (Lu et al., 2020; Rivera, 2015). The literature also documents relationship pathways linked to economic mobility such as marriage with foreign clients (Jaisuekun and Sunanta, 2021; Sunanta and Angeles, 2013) or transactional romantic relationships involving female tourists and economically marginalised men (Bindel, 2003; Weichselbaumer, 2012). These dynamics reinforce the view that sex tourism is embedded in broader inequalities rather than being driven solely by individual preferences.

Synthesis and implications for the present study

Taken together, the literature suggests that sex tourism is patterned rather than idiosyncratic, arising from the conjunction of economic incentives, market infrastructure, and institutional-legal configurations. In particular, sex-tourism-related activity is more likely where (i) cross-country income differentials generate purchasing power advantages for travellers, (ii) commercial sex markets and related adult-service infrastructure provide the organisational basis for transactional exchange, and (iii) formal legal status interacts with enforcement capacity and governance quality to produce permissive, discretionary, or ambiguous regulatory environments. Yet because sex tourism is frequently embedded within informal or semi-illicit markets and is rarely observed directly in official statistics, macro-comparative evidence on these mechanisms remains limited.

This study advances the literature by providing dyadic cross-national tests of these propositions. Using a two-stage gravity framework, we first benchmark expected bilateral arrivals based on standard structural determinants and then examine whether systematic deviations from this benchmark are associated with sex-tourism indicators and conditioned by sex-work legality, institutional quality as a proxy for enforcement capacity, and global inequality. This synthesis also underscores the policy relevance of an institutional perspective: to the extent that excess inflows are associated with regulatory ambiguity and weak or selective enforcement rather than legality per se, effective policy responses are likely to require improvements in governance capacity, enforcement consistency, and worker protections, complemented by cross-border coordination, rather than reliance on binary legal classifications alone.

Data

To examine the relationship between sex tourism-related characteristics and international tourism flows, we assemble a dyadic (origin – destination - year) panel of bilateral tourist arrivals covering 1995–2018. The dataset includes more than 150 origin and destination countries and is structured to support directional analysis of flows from origin i to destination j in year t. Our dependent variable is the annual number of arrivals from i to j, sourced from the United Nations World Tourism Organization (UNWTO), which reports arrivals by destination and country of origin (World Tourism Organization, 2024). Following standard practice in tourism gravity modelling, arrivals are transformed using the natural logarithm to reduce skewness and heteroskedasticity and to facilitate semi-elasticity interpretations.

As sex tourism is not captured in standard international statistical systems, we construct a sex tourism destination indicator using a transparent triangulation protocol. We first compiled candidate destinations through systematic web-based searches using predefined keywords (e.g., “sex tourism destinations”, “sex tourism hotspot”, “where to go for sex tourism”). We then cross-validated candidates across multiple source types, including travel journalism, travel industry sources, and peer-reviewed research. A destination is coded as a sex tourism destination if it is identified as such in at least three independent sources. When sources conflicted or classifications were ambiguous, destinations were cross-checked against the academic literature and the country’s legal context to ensure consistent classification. This approach recognises that sex tourism status is not mechanically implied by legality: some countries have legal sex work without being widely identified as sex tourism hubs, whereas other destinations remain prominent in both popular and academic accounts despite formal legal restrictions. In robustness checks, we assess sensitivity to classification by varying the inclusion threshold (e.g., ≥2 sources), excluding destinations identified only by non-academic sources, and excluding high-profile destinations one at a time to ensure results are not driven by a small number of cases.

To proxy sex tourism infrastructure, we construct a destination-level measure of adult-oriented venue density using OpenStreetMap’s LoveMap API (Skowron, 2014). We extract counts of adult-oriented venues such as brothels, strip clubs, erotic massage parlours, love hotels, and adult entertainment shops. Venue counts are normalised by population and log-transformed to address skewness. This measure is intended to capture structural differences in the presence of adult-service infrastructure that may support sex tourism demand. We acknowledge two limitations. First, OpenStreetMap coverage varies across countries. Second, the venue data are not consistently time-varying over the study period, and therefore cannot capture openings, closures, or short-run shocks. We interpret the measure as a structural proxy rather than a precise dynamic indicator.

We incorporate the legal status of sex work using cross-country classifications from Forza (2024). Countries are grouped into three categories: legal, limited legality, and illegal. “Legal” denotes regulated sex work with formal monitoring provisions. “Limited legality” captures partial or mixed regimes (e.g., some activities tolerated or decriminalised without full legalisation), reflecting a regulatory grey zone in which enforcement and legal boundaries may be ambiguous. These categories are used both directly and in interaction with institutional quality measures to evaluate whether legality operates differently across varying enforcement capacities. See Table 1 for a complete list of countries by status and type.

Table 1.

Sex tourism regions, status, and type.

Sex tourism region	Sex work status	Type of sex tourism
Brazil	Legal	Traditional
Colombia	Legal	Traditional
Cuba	Illegal	Traditional
Dominican Republic	Limited legality	Traditional
Honduras	Limited legality	Traditional & child sex tourism
Indonesia	Legal (no specific law)	Traditional & child sex tourism
India	Limited legality	Traditional & child sex tourism
Jamaica	Legal	Traditional
Kenya	Limited legality	Traditional & child sex tourism
Malaysia	Illegal	Traditional
Sri Lanka	Illegal	Child sex tourism
Morocco	Illegal	Traditional & child sex tourism
Mexico	Limited legality	Traditional & child sex tourism
The Netherlands	Legal	Traditional
The Philippines	Illegal	Traditional & child sex tourism
Russia	Illegal	Traditional & child sex tourism
Thailand	Limited legality	Traditional
Turkey	Legal	Traditional
Ukraine	Illegal	Traditional & child sex tourism
Vietnam	Illegal	Traditional & child sex tourism
South Africa	Illegal	Traditional & child sex tourism

To isolate sex tourism-related patterns from standard determinants of tourism, we include a comprehensive set of controls commonly used in the gravity literature. Economic conditions are captured via GDP per capita (current USD) for both origin and destination countries, log-transformed, alongside the official exchange rate (local currency per USD), averaged annually and logged (World Bank, 2024, 2025). We additionally include a common currency indicator to reflect reduced transaction costs and a Gini coefficient measure to account for domestic inequality, which may be associated with the structure and informality of service sector markets, including commercial sex (World Bank, 2023). Destination attractiveness is proxied using counts of UNESCO World Heritage Sites (cultural and natural), which capture baseline tourism appeal unrelated to sex tourism (UNESCO, 2025). In robustness checks, we also incorporate inbound international students and migrant stocks (International Migration Portal) to account for non-leisure travel motives such as education and family visitation (International Organization for Migration (IOM), 2025).

Geographic and physical constraints are controlled for using bilateral distance from CEPII’s gravity dataset (Mayer and Zignago, 2011), along with a common border indicator and destination endowments such as land area and coastline length from the CIA’s The World Factbook (2025). Accessibility is proxied by the number of international airports in the destination country (UNWTO). Institutional quality is measured using International Country Risk Guide (ICRG) indicators, including law and order, internal conflict, and government stability (PRS Group, 2025). Bilateral mobility barriers are captured using visa and exit permit requirements from the DEMIG VISA database (International Migration Institute, 2025). Because DEMIG coverage ends in 2013, values for 2014–2018 are carried forward under the assumption that visa regimes change infrequently; we test sensitivity by excluding the post-2013 period. We also include regional trade agreements (World Trade Organization, 2025), legal origins (Porta et al., 2008), and a weighted common language index following Gurevich et al. (2024). Finally, to account for security-related deterrents to travel, we include terrorism measures from the RAND Database of Worldwide Terrorism Incidents (RAND National Security and Research Division, 2025). To improve interpretability, Table 2 summarises the conceptual role of each variable in the empirical framework, distinguishing between the structural gravity covariates used to estimate expected bilateral tourism flows in Stage 1 and the sex-tourism-related indicators, institutional moderators, and inequality measures used to explain residual excess flows in Stage 2. Detailed descriptive statistics for all variables are reported in the Online Appendix.

Table 2.

Variable roles in the empirical framework.

Variable	Operationalisation	Stage	Role in model	Expected relationship
Panel A. Outcome variables
International arrivals (log)	Log of annual arrivals from origin i to destination j	Stage 1	Main dependent variable in the structural gravity model
Gravity residual (“excess” arrivals)	Observed log arrivals minus gravity-predicted log arrivals	Stage 2	Dependent variable capturing unexplained excess tourism
Panel B. Structural gravity variables
GDP per capita (origin/destination)	Logged GDP per capita	Stage 1	Economic size, purchasing capacity, and destination demand conditions	Positive for tourism-generating and destination capacity
Population/total outflow	Logged population or outbound volume	Stage 1	Scale of tourism supply from origin/size effects	Positive
Exchange rate	Logged official exchange rate	Stage 1	Relative travel cost/affordability	Ambiguous
Distance	Logged bilateral distance	Stage 1	Travel friction/resistance cost	Negative
Common border	Binary indicator	Stage 1	Lower travel frictions/geographic proximity	Positive
Common currency	Binary indicator	Stage 1	Lower transaction costs	Positive
Common legal origin	Binary indicator	Stage 1	Institutional affinity/lower informational frictions	Positive
Colonizer relationship	Binary indicator	Stage 1	Historical ties/path dependence	Positive
Language distance	Weighted linguistic similarity measure	Stage 1	Cultural and information friction	Negative as distance rises
Regional trade agreement	Binary indicator	Stage 1	Institutional integration/mobility facilitation	Positive
Visa required	Categorical indicator	Stage 1	Mobility barrier	Negative
Exit permit	Categorical indicator	Stage 1	Outbound mobility barrier	Negative
Airports	Count of destination airports	Stage 1	Accessibility/transport capacity	Positive
Coastline area/land area	Logged geographic endowments	Stage 1	Destination attractiveness/scale	Generally positive
UNESCO cultural/natural sites	Counts	Stage 1	Baseline tourism attractiveness	Positive
Political risk/government stability/law & order/internal conflict	ICRG indicators	Stage 1	Institutional quality/security environment	Better governance positive; conflict negative
Terror attacks	Count over previous 3 years	Stage 1	Security deterrent	Negative
Inbound students/migrants	Counts	Stage 1	Non-leisure travel links/social connectivity	Positive
Panel C. Focal sex-tourism variables
Sex tourism destination dummy	Indicator for destinations identified as sex tourism hubs	Stage 2	Main explanatory variable for excess arrivals	Positive
Number of sex tourism POI	Logged density/count of adult-oriented venues	Stage 2	Proxy for sex-tourism market infrastructure	Positive
Sex work status	Legal/limited legality/illegal	Stage 2	Institutional-legal environment	Ambiguous on its own; depends on enforcement context
GINI coefficient	Origin-side inequality measure	Stage 2	Socioeconomic heterogeneity/demand-side inequality	Ambiguous main effect
Panel D. Moderators/interactions
Sex tourism × sex work status	Interaction terms	Stage 2	Tests whether the sex-tourism effect depends on the legal regime	Stronger in ambiguous/limited settings
Sex tourism × law & order	Interaction terms	Stage 2	Tests whether stronger institutions attenuate excess flows	Negative
Sex tourism × GINI	Interaction term	Stage 2	Tests whether inequality amplifies sex-tourism-related demand	Positive
Sex tourism × destination GDP per capita	Interaction term	Stage 2	Tests whether effects are stronger in poorer destinations	Negative
Origin × destination GDP and triple interactions	Interaction terms	Stage 2	Tests purchasing-power asymmetry across dyads	Stronger for rich-origin/poorer-destination dyads

Together, these data are used to form a multidimensional dyadic panel suitable for estimating a first-stage structural gravity model of expected tourism flows, and for testing whether residual (“excess”) arrivals are systematically associated with sex tourism indicators, sex work legality, enforcement capacity, and cross-country inequality. For transparency and reproducibility, we have uploaded the complete dataset and all analysis files in an OSF repository (https://osf.io/juw87/overview?view_only=77dd02ecb582437bbe8c9903ada30bf7).

Method

This study adopts a two-step structural gravity framework with dyadic tourism data to investigate whether sex tourism-related characteristics and prostitution status are systematically associated with international tourism flows. Because sex tourism is partially clandestine and not directly captured in international statistical systems, the empirical strategy first estimates expected bilateral arrivals using a gravity benchmark and then examines whether systematic deviations from that benchmark align with theoretically motivated measures of sex tourism market presence and the institutional legal environment. Dyadic data capture outcomes generated by pairwise relationships between sampled units, such as tourist flows between origin and destination countries, where both partners jointly shape the observed outcome (Graham, 2020). Drawing on the structural gravity model originally developed in the context of trade, this study adopts a two-stage approach to assess sex tourism and international tourism flows (Rosselló Nadal and Santana Gallego, 2022; Tinbergen, 1963).

Gravity models have been applied extensively to tourism demand due to their strong empirical performance in accounting for bilateral frictions and multilateral influences that shape dyadic flows (Rosselló Nadal and Santana Gallego, 2022). The original model proposed by Tinbergen (1963) posits that bilateral flows are positively related to economic size and negatively related to “resistance” between countries, such as distance and cultural barriers. Gravity models have produced reliable and interpretable estimates across trade, migration, and tourism applications (Frankel et al., 1996; Silva and Tenreyro, 2006; Yotov et al., 2016). Building on this framework, our approach isolates sex-tourism-specific drivers beyond general tourism determinants. In Stage 1, we estimate a gravity model of bilateral tourism arrivals using a linear mixed-effects regression to recover expected flows between origin and destination countries based on standard structural determinants. In Stage 2, we analyse the residual deviations from this benchmark to test whether excess tourism is systematically associated with sex tourism indicators and whether these associations vary with sex work legality and institutional quality.

In Stage 1, tourism demand is modelled using a linear mixed-effects regression. The model captures variation in logged international arrivals from origin i to destination j in year t as a function of observed covariates and common time shocks:

\log ({A r r i v a l s}_{i j t}) = α + X_{i j t} β + γ_{t} + u_{i j} + ε_{i j t}

where

X_{i j t}

is a matrix of dyadic and country-level covariates representing structural determinants of tourism flows,

γ_{t}

are year fixed effects controlling for common shocks across dyads,

u_{i j}

is a dyad-specific random intercept absorbing time-invariant bilateral heterogeneity, and

ε_{i j t}

is the idiosyncratic error term. Linear mixed-effects models enable partial pooling across dyads, improving efficiency and reducing overfitting relative to fully saturated fixed-effects approaches in large dyadic panels (Snijders and Bosker, 2012). The dyad-specific intercept captures latent bilateral affinity, such as persistent migration ties or durable social and economic linkages, that may be correlated with tourism flows but are difficult to measure directly (Snijders and Bosker, 2012). The dependent variable is log-transformed to reduce skewness and stabilise variance, consistent with standard practice in tourism gravity modelling. Robust standard errors are used across models to improve inference under potential heteroskedasticity (Huber, 1967; White, 1980).

The fitted values from the first-stage model represent gravity-predicted tourism flows conditional on observed covariates. Residual (“excess”) arrivals are computed as:

{\hat{ε}}_{i j t} = \log ({A r r i v a l s}_{i j t}) - \log ({\hat{A r r i v a l s}}_{i j t}^{g r a v i t y})

These residuals capture the extent to which observed arrivals deviate from gravity-based expectations: positive residuals indicate more arrivals than predicted, while negative residuals indicate underperformance. In Stage 2, we examine whether these deviations are systematically associated with sex tourism-related indicators and their interactions with the legal and institutional context:

{\hat{ε}}_{i j t} = Z_{j t} δ + η_{i j t}

where

Z_{j t}

is a vector of destination-specific sex-tourism-related indicators and interaction terms, and

δ

is a vector of coefficients capturing their association with excess inbound tourism. In practice,

Z_{j t}

includes the sex tourism destination indicator, the adult venue infrastructure proxy, sex work legal status, and interactions with institutional quality and inequality measures. Standard errors are clustered at the dyad level to account for repeated observations within origin-destination pairs over time. This residual-based strategy provides a practical way to isolate the contribution of sector-specific variables that may explain deviations from expected travel patterns, consistent with approaches used in related trade and migration literatures (Morley et al., 2014).

Results

The results present a series of linear mixed-effects regressions examining whether destinations associated with sex tourism receive systematically higher inbound international tourism flows than would be predicted by conventional determinants. The dependent variable in all specifications is the Stage 1 gravity residual (i.e., the deviation of observed log arrivals from gravity-predicted log arrivals). Descriptive statistics for all variables, along with baseline first-stage gravity estimates used to generate residuals, are reported in the Online Appendix.

Table 3 reports baseline specifications relating excess arrivals to sex-tourism-related indicators and institutional context. Across all models, the sex tourism destination indicator is positive and statistically significant, indicating that destinations identified as sex tourism hubs receive substantially higher tourism inflows than would be predicted by standard structural determinants. In Model 1, the coefficient on the sex tourism indicator is 0.43, implying approximately 54% higher residual arrivals (e^0.43 − 1) in sex tourism destinations relative to otherwise comparable destinations. This association remains stable across alternative specifications, suggesting that sex tourism is a meaningful component of excess tourism demand. We next add an infrastructure proxy to strengthen interpretation and then examine whether legality and enforcement conditions moderate these patterns, before turning to origin-side inequality.

Table 3.

Mixed effects regression of random intercept residuals.

Independent variable	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
Sex tourism dummy	0.43*** (0.01)	0.41*** (0.01)	0.49*** (0.01)	0.51*** (0.01)	0.78*** (0.02)	0.54*** (0.01)	0.59*** (0.01)	0.36*** (0.05)	0.764*** (0.03)
Number of sex tourism POI		0.02*** (0.00)	0.03*** (0.00)	0.03*** (0.00)	0.03*** (0.01)	0.06*** (0.00)	0.03*** (0.00)	0.03*** (0.01)	0.026*** (0.00)
Sex work at origin legal			−0.01 (0.01)	−0.01 (0.01)	−0.02 (0.01)	−0.00 (0.01)	−0.01 (0.02)	−0.03 (0.01)
Sex work at origin limited			0.00 (0.01)	0.01 (0.01)	−0.00 (0.01)	0.01 (0.01)	−0.00 (0.03)	−0.01 (0.01)
Law & order destination						0.02*** (0.00)
Law & order origin							0.00 (0.00)
Sex work at destination legal			−0.71*** (0.03)	−0.71*** (0.03)	−0.43*** (0.03)	−0.64*** (0.03)	−0.64*** (0.03)	−0.70*** (0.03)
Sex work at destination limited			−0.01 (0.02)	−0.01 (0.03)	−0.02 (0.02)	0.05*** (0.02)	0.05 (0.03)	0.01 (0.02)
GINI origin								−0.00*** (0.00)
Sex tourism x origin sex work
Legal				−0.03 (0.06)
Limited				−0.05** (0.02)
Sex tourism x destination sex work
Legal					−0.98*** (0.03)
Limited					−0.20*** (0.02)
Sex tourism x law & order destination						−0.01** (0.00)
Sex tourism x law & order origin							−0.03*** (0.01)
Sex tourism x GINI origin								0.01* (0.00)
Sex tourism x number of sex tourism POI									−0.071*** (0.01)
Observations	131,140	130,908	130,908	130,908	130,908	118,619	101,567	46,332	130,908
Number of groups destination	11,389	11,348	11,348	11,348	11,348	10,255	8852	8905	11,348
Number of groups origin	168	168	168	168	168	168	121	133	168
Wald chi²	4437.46	4455.82	9637.37	10,134.85	32,286.53	15,292.34	7663.95	5821.41	4530.18
Prob > chi²	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00

Note. Robust standard errors in parentheses. *** indicates significance at p < .01, ** indicates significance at p < .05 and * indicates significance at p < .10.

Model 2 adds a proxy for sex tourism infrastructure: the logged density of sex tourism-related points of interest (e.g., brothels, strip clubs, erotic massage parlours, and love hotels). This measure is positive and statistically significant across specifications, with a modest magnitude (approximately 0.02–0.06). The result is consistent with the view that observable adult service infrastructure, reflecting market presence and visibility, contributes to excess arrivals alongside the destination’s broader classification as a sex tourism hub (Lu et al., 2020; Oppermann, 1999).

Models 3–8 introduce sex work legal status and interaction terms to examine whether the association between sex tourism hubs and excess arrivals varies by legal and institutional context. A notable pattern is that fully legal sex work at the destination is associated with lower residual arrivals, and the interaction terms indicate that excess arrivals linked to sex tourism hubs are smaller in fully legal settings than in limited or ambiguous regimes. This finding complicates a simple legalisation-tourism narrative and is consistent with an institutional interpretation in which regulatory design and enforcement discretion shape market conditions. In particular, demand may be strongest in environments characterised by legal ambiguity or partial tolerance, where discretion and informality are more salient. By contrast, fully legal regimes may be associated with more formalised market structures and regulatory oversight, potentially reducing the “grey zone” conditions that underpin some segments of sex tourism-related demand. In contrast to destination legality, origin-country legal status does not show a consistent association with outbound residual tourism, suggesting that destination-specific conditions, market structure, availability, and governance environment are more salient.

Institutional quality further moderates these relationships. In Models 6 and 7, interactions between sex tourism and law and order are negative and statistically significant, indicating that stronger law and order reduces the magnitude of excess arrivals associated with sex tourism destinations. Substantively, excess inflows to sex tourism hubs are attenuated where institutions are more effective and enforcement is more consistent. A comparable moderation pattern is observed for origins: tourists from higher law-and-order contexts exhibit weaker unexplained travel to sex tourism destinations, consistent with the view that origin institutional environments can shape participation in informal or contested travel niches.

Model 8 incorporates origin-country income inequality (GINI) and its interaction with the sex tourism indicator. The standalone GINI coefficient is negative and statistically significant, indicating that higher inequality in the origin is associated with lower residual outbound tourism overall. However, the positive interaction between GINI and the sex tourism indicator suggests that, conditional on travel to sex tourism destinations, origins characterised by greater internal inequality exhibit stronger excess flows. This pattern is consistent with accounts linking inequality to transactional and status-oriented consumption abroad and highlights that demand-side conditions may be shaped not only by cross-country differences but also by within-country socio-economic structure.

Table 4 extends the analysis by incorporating dyadic economic conditions and additional interaction structures that address purchasing power asymmetries and development-tier heterogeneity. Models 10 and 11 indicate that excess arrivals associated with sex tourism are concentrated toward lower-income destinations: the interaction between destination GDP per capita and the sex tourism indicator is negative, and the triple interaction structure in Model 11 is consistent with stronger effects when tourist of wealthier origins travel to poorer destinations. These findings reinforce an inequality mechanism in which purchasing power differentials amplify sex-tourism-related excess travel, consistent with a political economy account in which cross-border income gradients shape the destination set and the intensity of sex-tourism demand.

Table 4.

Mixed effects regression of random intercept residuals.

Independent variable	(10)	(11)	(12)	(13)	(14)
Sex tourism dummy	0.40*** (0.07)	1.74*** (0.09)	3.69*** (0.62)	0.97*** (0.03)	−.70*** (0.02)
Number of sex tourism POI	0.02*** (0.00)	0.01*** (0.00)	0.01*** (0.00)	0.06*** (0.00)	−0.00** (0.00)
Law & order destination				−0.01** (0.00)
Sex work at destination legal				−0.56*** (0.03)
Sex work at destination limited				−0.05 (0.04)
GDP per capita origin	−0.00 (0.01)		0.61*** (0.04)
GDP per capita destination		0.02* (0.01)	0.71*** (0.05)
Low income destination					−0.35*** (0.02)
Lower middle-income destination					0.03** (0.02)
Upper middle-income destination					−0.01 (0.02)
Sex tourism × legal sex work				−1.21*** (0.03)
Sex tourism × limited sex work				−0.29*** (0.03)
Sex tourism × law & order destination				−0.05*** (0.00)
Sex tourism × GDP per cap origin	0.00 (0.01)		−0.04 (0.07)
Sex tourism × GDP per cap destination		−0.15*** (0.01)	−0.60*** (0.07)
GDP per cap origin × GDP per cap destination			−0.07*** (0.00)
GDP per cap origin × GDP per cap destination × sex tourism			0.03*** (0.01)
Sex work legal × law & order				0.05*** (0.00)
Sex work limited × law & order				0.03** (0.00)
Sex tourism × sex work legal × law & order				0.07*** (0.01)
Sex tourism × sex work limited × law & order				0.03*** (0.01)
Sex tourism × low income destination					1.67*** (0.02)
Sex tourism × lower middle-income destination					0.90*** (0.02)
Sex tourism × upper middle-income destination					1.28*** (0.03)
Observations	130,908	130,908	130,908	118,619	123,934
Number of groups destination	11,348	11,348	11,348	10,255	10,691
Number of groups origin	168	168	168	168	168
Wald chi²	4610.89	5918.73	1945.99	3231.09	26,746.78
Prob > chi²	0.00	0.00	0.00	0.00	0.00

Note. Robust standard errors in parentheses. *** indicates significance at p < .01, ** indicates significance at p < .05 and * indicates significance at p < .10.

Model 14 introduces interactions between the sex tourism indicator and World Bank destination income groups, allowing the association between sex tourism hubs and excess arrivals to vary by development tier. With high-income destinations as the omitted reference category, the main-effect coefficient on the sex tourism indicator reflects the association within high-income destinations, while the interaction terms capture how this association differs in low- and middle-income destinations.

The interaction terms are positive and statistically significant for low- and middle-income destinations, indicating that excess arrivals associated with sex tourism hubs are substantially larger outside high-income settings. This heterogeneity pattern complements the GDP interaction results by showing that the concentration of sex-tourism-related excess arrivals is not merely a function of income levels in isolation, but is systematically stronger in development tiers where purchasing power differentials and institutional ambiguity are more likely to coincide.

Finally, Model 12 introduces a three-way interaction between sex tourism, destination sex work legality, and law and order. The interaction structure indicates that enforcement context conditions how legality relates to excess inflows, supporting the broader conclusion that sex tourism is shaped not by legal status alone but by the interplay between legality, perceived risk, and enforcement capacity. This is consistent with the ‘ambiguity/discretion’ mechanism: excess travel is greatest where market access is feasible but governance is discretionary, whereas fully formalised and regulated regimes may reduce informality rents and the demand for under-the-radar intermediated access. Taken together, the results point to a consistent mechanism: sex tourism-related excess tourism is strongest where adult market infrastructure is present and where the institutional and legal environment creates conditions of discretion or selective enforcement, and it is amplified in dyads characterised by sizeable income asymmetries.

Conclusions

Our results indicate that sex-tourism-related excess arrivals are patterned and systematically associated with economic asymmetries, institutional capacity, and the legal environment. Destinations identified as sex tourism hubs receive inbound tourism flows that exceed gravity-based expectations even after controlling for a wide set of economic, geographic, institutional, and cultural determinants. Taken together, the evidence is consistent with sex tourism operating as a “shadow” segment embedded within formal tourism flows, one that is amplified by uneven development, adult market infrastructure, and regulatory conditions characterised by discretion or selective enforcement.

This study advances the literature in three related ways. First, it provides macro-comparative evidence that institutions shape not only the volume of tourism but also its composition, helping to explain why contested or informal niches may manifest as systematic deviations from structurally “expected” tourism flows. Second, it clarifies that legality alone is not sufficient to characterise market conditions: across specifications, fully legal sex work at the destination is not associated with higher excess arrivals and is often associated with lower excess inflows relative to limited or ambiguous regimes. This pattern supports an institutional “grey zone” mechanism in which the interaction of legal status with enforcement capacity and governance practices conditions the expected costs, discretion, and informality rents associated with participation. Third, by applying a two-stage, residual-based structural gravity framework to dyadic arrivals data, the analysis contributes a tractable empirical strategy for studying clandestine or weakly measured sectors linking deviations from predicted flows to theoretically motivated indicators of sex tourism market presence, institutional quality, and inequality.

Our heterogeneity results further strengthen these inferences. Excess arrivals associated with sex tourism hubs are shown to be substantially larger outside high-income destination settings. They are amplified in dyads characterised by purchasing power asymmetries, consistent with political economy accounts in which cross-border income gradients expand the scope for transactional exchange and shape the direction and intensity of demand.

Although the analysis does not directly observe individual harms or motivations, the findings are still practical and ethical. That is, the institutional conditions associated with larger sex tourism-related excess inflows are frequently those shown to be linked in prior work to elevated exploitation risks, weaker worker protections, and public health vulnerabilities. Two clear implications follow. Firstly, policy responses should avoid treating reform as a binary choice between prohibition and legalisation. Instead, the results suggest that regulatory design and enforcement credibility are central: where enforcement is weak, discretionary, or selectively applied, markets may remain both permissive and informal, expanding conditions conducive to sex-tourism-related activity while limiting protections. Strengthening institutional capacity through consistent enforcement, clearer regulatory boundaries, and anti-corruption measures, along with effective oversight of intermediaries, may reduce harmful segments of demand without merely displacing activity into less visible and potentially riskier settings. Secondly, safeguarding-oriented interventions remain important regardless of legal status. Improving access to health and support services, strengthening reporting and protection mechanisms, and coordinating cross-border information sharing and targeted policing of exploitative practices can help mitigate adverse externalities in contexts where demand is shaped by international income asymmetries and enforcement capacity constraints. For the tourism sector, ethical tourism initiatives and due diligence standards for platforms and intermediaries can naturally complement and support public policy where market access and coordination increasingly occur through digital channels.

Several limitations of the current study should be acknowledged. Firstly, residual-based outcomes capture deviations from predicted flows and may reflect unobserved drivers beyond sex tourism. Secondly, to state the obvious, sex tourism, like all illicit behaviours, is difficult to measure directly. The destination classification and venue-based infrastructure proxy necessarily approximate a clandestine phenomenon, and as such may not capture temporal changes in market size or organisation. Thirdly, cross-national legality classifications simplify complex and sometimes subnational legal environments, and general institutional indicators may not fully capture sector-specific enforcement relevant to commercial sex markets. And finally, macro dyadic data cannot identify traveller motivations or distinguish purposive from opportunistic participation. Future research could extend this framework by integrating micro-level evidence on travel motivations and behaviour, incorporating qualitative fieldwork in key destinations, and developing improved time-varying measures of sex tourism market presence. Further work could also distinguish more explicitly between legalisation and decriminalisation regimes and evaluate whether institutional reforms or enforcement shocks alter the magnitude or composition of excess inflows over time.

Overall, our findings suggest that sex tourism is not random or purely individual-level; it is systematically shaped by economic gradients and institutional-legal configurations. By providing cross-country macro evidence on these mechanisms, our study contributes to tourism economics and the broader literature’s understanding of informal economies by showing how regulatory ambiguity, enforcement capacity, and inequality jointly structure contested tourism niches within global mobility.

Supplemental material

Supplemental Material - Economics and legal influences on sex tourism

Supplemental Material for Economics and legal influences on sex tourism by Ella Hugo, Steve Bickley, Ho Fai Chan, Benno Torgler and Stephen Whyte in Tourism Economics.

Footnotes

ORCID iD

Ella Hugo

Author contributions

E.H: Conceptualization; Methodology; Formal analysis; Investigation; Data curation; Writing - original draft & editing. S.B, H.F. C, B.T, & S.W: Conceptualization; Supervision; Writing – review & editing.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.*

Supplemental material

Supplemental material for this article is available online.

Author biographies

Ella Hugo is a PhD candidate in Economics at Queensland University of Technology. She holds an MPhil in Economics and a Bachelor of Commerce in Finance and Economics from the University of Newcastle. Her research spans welfare economics, inequality, macroeconomics, and behavioural economics, with a strong policy orientation. Ella examines hidden economic domains, including human trafficking, corruption, and informal markets, using quantitative and mixed methods. She aims to generate evidence that supports social equity, institutional accountability, and improved economic outcomes across diverse populations in Australia and internationally, through rigorous research.

Steve Bickley’s research sits at the intersection of behavioural economics and complexity science, drawing insight from real-world social systems. His interdisciplinary work connects micro, meso, and macro socio-economic dynamics to explain decision-making in complex environments. Topics span stock markets, artificial intelligence, quantum technologies, public health, urban design, online communities, and sport. By integrating theoretical and empirical perspectives, Steve develops practical strategies that inform technology design, policy reform, and organisational management. His work demonstrates how cross-disciplinary thinking can address contemporary challenges across science, society, and the economy in policy and practice.

Ho Fai (Ben) Chan is a Senior Research Fellow at QUT whose research integrates economics, data science, and psychology. His work ranges from laboratory and field experiments on individual decision-making to analyses of large-scale bibliometric, mobility, and patent datasets, revealing how incentives and information shape behaviour in markets, science, and society. Dr Chan has authored more than seventy peer-reviewed articles and has been cited over two thousand times. Publications include Nature, Journal of Economic Behavior & Organization, Psychological Science, Labour Economics, and Journal of Business Ethics, with broad policy relevance.

Benno Torgler is Professor of Economics in the School of Economics and Finance at QUT. A former ARC Future Fellow and Adjunct Professor at EBS Universität, he has held research and visiting posts at Yale University, UC Berkeley, and Georgia State University. His interdisciplinary scholarship spans behavioural and experimental economics, public finance, political economy, and the economics of illegality, including corruption and tax evasion. He has published widely across economics, political science, psychology, and law, advancing understanding of informal economies and informing evidence-based public policy in Australia and abroad today.

Stephen Whyte is a Postdoctoral Research Fellow in Behavioural Economics at the School of Economics and Finance, QUT. His research investigates large-scale decision-making in mate choice and human behaviour, bridging applied microeconomics, social psychology, and evolutionary biology. He has published across reproductive medicine, cyber-psychology, and scientometrics. Recent work examines sex differences in nonbinary gender identification, decisions in assisted reproduction, and preferences versus choices in online dating. Stephen’s research informs public policy, health, and gender economics, using experimental and behavioural methods to understand non-market behaviour and complex individual decision-making in practice.

References

Alvarez

Campo

(2014) The influence of political conflicts on country image and intention to visit: a study of Israel’s image. Tourism Management 40: 70–78. https://doi.org/10.1016/j.tourman.2013.05.009

Arezki

Cherif

Piotrowski

(2009) Tourism Specialization and Economic Development: Evidence from the UNESCO World Heritage List. International Monetary Fund. 09/176; IMF Working Paper.

Bindel

(2003, July 12). Sex tourism as economic aid. The Guardian. https://www.smh.com.au/national/sex-tourism-as-economic-aid-20030712-gdh329.html

Brennan

(2001) Tourism in transnational places: dominican sex workers and German sex tourists imagine one another. Identities: Global Studies in Power and Culture 7(4): 621–663. https://doi.org/10.1080/1070289X.2001.9962680

Brents

Hausbeck

(2007) Marketing sex: US legal brothels and late capitalist consumption. Sexualities 10(4): 425–439. https://doi.org/10.1177/1363460707080976

Cabezas

(2008) Tropical blues. Latin American Perspectives 35(3): 21–36. https://doi.org/10.1177/0094582X08315765

Camacho

Ramos-Herrera

(2025) Does tourism reduce income inequality? Tourism Economics 31(3): 381–401. https://doi.org/10.1177/13548166241262349

CIA The World Factbook (2025) The World Factbook. The CIA World Factbook. https://www.cia.gov/the-world-factbook/field/exchange-rates/

Crick

(2012) Resplendent Sites, Discordant Voices. Routledge. https://doi.org/10.4324/9780203059425

10.

Forza

(2024) Prostitution Legislation and Sex Trafficking Joined Dataset. University of Oxford.

11.

Frankel

Stein

S-J

(1996) Regional trading arrangements: natural or supernatural? The American Economic Review 86(2): 52–56.

12.

Graburn

NHH

(1983) Tourism and prostitution. Annals of Tourism Research 10(3): 437–443. https://doi.org/10.1016/0160-7383(83)90068-3

13.

Graham

(2020) Dyadic regression. In: The Econometric Analysis of Network Data. Elsevier, pp. 23–40. https://doi.org/10.1016/B978-0-12-811771-2.00008-0

14.

Gurevich

Herman

Toubal

, et al. (2024) The domestic and international common language database (2024–2; school of economics working paper series). https://ideas.repec.org/p/ris/drxlwp/2024_002.html

15.

Hakim

(2015) Economies of desire: sexuality and the sex industry in the 21st century. Economic Affairs 35(3): 329–348. https://doi.org/10.1111/ecaf.12134

16.

Hall

Brown

(1996) Towards a welfare focus for tourism research. Progress in Tourism and Hospitality Research 2(1): 41–57. https://doi.org/10.1002/(SICI)1099-1603(199603)2:1<41::AID-PTH23>3.0.CO;2-6

17.

Helmke

Levitsky

(2004) Informal institutions and comparative politics: a research agenda. Perspectives on Politics 2(04): 725–740. https://doi.org/10.1017/S1537592704040472

18.

Hillis

Leavey

Kewley

, et al. (2022) Sex tourism in an era of globalisation, harm reduction and disease migration: a new conceptual model. Tourism Review 77(2): 592–606. https://doi.org/10.1108/TR-04-2021-0184

19.

Huber

(1967) The Behaviour of Maximum Likelihood Estimates under Nonstandard Conditions. Proceedings of the 5th Berkeley Symposium on Mathematical Statistics and Probability, University of California Press, Berkeley, In press.

20.

International Migration Institute (2025) DEMIG VISA data. Determinants of International Migration. https://www.migrationinstitute.org/data/demig-data/demig-visa-data

21.

International Organization for Migration (IOM) (2025) Global migration data analysis centre (GMDAC). Migration Data Portal. https://www.migrationdataportal.org

22.

Jaisuekun

Sunanta

(2021) German migrants in Pattaya, Thailand: gendered mobilities and the blurring boundaries between sex tourism, marriage migration, and lifestyle migration. In: The Palgrave Handbook of Gender and Migration. Springer International Publishing, pp. 137–149. https://doi.org/10.1007/978-3-030-63347-9_9

23.

Kahyalar

Seetaram

Fethi

(2024) Tourism and the shadow economy: long-run and short-run implications for resource allocation. Tourism Economics 30(3): 749–766. https://doi.org/10.1177/13548166231181261

24.

Kizilkaya

Kocak

Ulug

, et al. (2024) Pandemic‐induced uncertainty and tourism demand: evidence on the moderator effect of democratic institutions. International Journal of Tourism Research 26(4): e2725. https://doi.org/10.1002/jtr.2725

25.

Kocak

Cobanoglu

(2025) World uncertainty and international tourist arrivals: does good governance make a difference? Current Issues in Tourism 28(3): 341–345. https://doi.org/10.1080/13683500.2024.2309151

26.

Kock

(2021) The behavioral ecology of sex tourism: the consequences of skewed sex ratios. Journal of Travel Research 60(6): 1252–1264. https://doi.org/10.1177/0047287520946106

27.

Leheny

(1995) A political economy of Asian sex tourism. Annals of Tourism Research 22(2): 367–384. https://doi.org/10.1016/0160-7383(94)00082-4

28.

Holmes

Noone

, et al. (2020) Sun, sea and sex: a review of the sex tourism literature. Tropical Diseases, Travel Medicine and Vaccines 6(1): 24. https://doi.org/10.1186/s40794-020-00124-0

29.

Mayer

Zignago

(2011) Notes on CEPII’s distances measures: the GeoDist database. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.1994531

30.

Morley

Rosselló

Santana-Gallego

(2014) Gravity models for tourism demand: theory and use. Annals of Tourism Research 48: 1–10. https://doi.org/10.1016/j.annals.2014.05.008

31.

Neumayer

(2010) Visa restrictions and bilateral travel. The Professional Geographer 62(2): 171–181. https://doi.org/10.1080/00330121003600835

32.

Nørfelt

Kock

(2024) Leveraging evolutionary psychology for tourism research: identifying and addressing key challenges. Journal of Travel Research 64: 1262–1270. https://doi.org/10.1177/00472875241277530

33.

Oppermann

(1999) Sex tourism. Annals of Tourism Research 26(2): 251–266. https://doi.org/10.1016/S0160-7383(98)00081-4

34.

Porta

Lopez-de-Silanes

Shleifer

(2008) The economic consequences of legal origins. Journal of Economic Literature 46(2): 285–332. https://doi.org/10.1257/jel.46.2.285

35.

PRS Group (2025) The international country risk guide methodology. https://www.prsgroup.com/wp-content/uploads/2022/04/ICRG-Method.pdf

36.

RAND National Security and Research Division (2025) About the RAND database of worldwide terrorism incidents. Database Scope. https://www.rand.org/nsrd/projects/terrorism-incidents/about/definitions.html

37.

Richter

(1989) The Politics of Tourism in Asia. University of Hawaii Press. https://doi.org/10.2307/j.ctv9zcjr9

38.

Rivera

(2015) Sex Tourism in the Dominican Republic and Other Developing Countries: A Study of Political, Economic, and Social Determinants. Doctor of Philosophy: Graduate Program in Global Affairs. The State University of New Jersey.

39.

Rosselló Nadal

Santana Gallego

(2022) Gravity models for tourism demand modeling: empirical review and outlook. Journal of Economic Surveys 36(5): 1358–1409. https://doi.org/10.1111/joes.12502

40.

Salom

(2008) The worlds oldest profession: evolutionary insights into prostitution. In: Duntley

Shackelford

(eds) Evolutionary Forensic Psychology: Darwinian Foundations of Crime and Law. Oxford University Press, pp. 121–135.

41.

Silva

JMCS

Tenreyro

(2006) The log of gravity. The Review of Economics and Statistics 88(4): 641–658. https://doi.org/10.1162/rest.88.4.641

42.

Skowron

(2014) OpenLoveMap. OpenLoveMap.

43.

Snijders

TAB

Bosker

(2012) Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling. 2nd edition. Sage.

44.

Sou

JPU

Vinnicombe

(2023) Does governance quality matter for FDI-led tourism development? A supply-side perspective. Tourism Economics 29(2): 392–408. https://doi.org/10.1177/13548166211052814

45.

Sunanta

Angeles

(2013) From rural life to transnational wife: agrarian transition, gender mobility, and intimate globalization in transnational marriages in northeast Thailand. Gender, Place & Culture 20(6): 699–717. https://doi.org/10.1080/0966369X.2012.709827

46.

Svensson

Sundbeck

Persson

, et al. (2018) A meta-analysis and systematic literature review of factors associated with sexual risk-taking during international travel. Travel Medicine and Infectious Disease 24: 65–88. https://doi.org/10.1016/j.tmaid.2018.03.002

47.

Taylor

(2006) Female sex tourism: a contradiction in terms? Feminist Review 83(1): 42–59. https://doi.org/10.1057/palgrave.fr.9400280

48.

Thanh‐Dam

(1983) The dynamics of sex tourism: the case of Southeast Asia. Development and Change 14(4): 533–553. https://doi.org/10.1111/j.1467-7660.1983.tb00165.x

49.

Tinbergen

(1963) Shaping the world economy. International Executive 5(1): 27–30. https://doi.org/10.1002/tie.5060050113

50.

UNESCO (2025) About World Heritage. UNESCO World Heritage Centre. https://whc.unesco.org/en/about/

51.

Wang

Tziamalis

(2023) International tourism and income inequality: the role of economic and financial development. Tourism Economics 29(7): 1836–1864. https://doi.org/10.1177/13548166231177106

52.

Weichselbaumer

(2012) Sex, romance and the carnivalesque between female tourists and Caribbean men. Tourism Management 33(5): 1220–1229. https://doi.org/10.1016/j.tourman.2011.11.009

53.

White

(1980) A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity. Econometrica 48(5). Available at: https://doi.org/10.2307/1912934

54.

Williams

Hall

(2000) Tourism and migration: new relationships between production and consumption. Tourism Geographies 2(1): 5–27. https://doi.org/10.1080/146166800363420

55.

World Bank (2023) Gini index methodology. Gini Index | Data. https://data.worldbank.org/indicator/SI.POV.GINI

56.

World Bank (2024) Official Exchange Rate (LCU per US$, Period Average). World Bank Methodology. https://databank.worldbank.org/metadataglossary/world-development-indicators/series/PA.NUS.FCRF

57.

World Bank (2025) GDP per Capita (Current US$) | Data. World Bank Group Data. https://data.worldbank.org/indicator/NY.GDP.PCAP.CD

58.

World Tourism OrganisationWorld Travel & Tourism Council (2012) The Impact of Visa Facilitation on Job Creation in the G20 Economies. World Tourism OrganisationWorld Travel & Tourism Council.

59.

World Tourism Organization (2024) World tourism barometer volume 22 issue 3. https://doi.org/10.18111/wtobarometereng

60.

World Trade Organization (2025) Regional Trade Agreements Database. The World Trade Organization. https://rtais.wto.org/UI/PublicMaintainRTAHome.aspx

61.

(2023) The effect of tourism on the shadow economy: the level of economic development is key. Tourism Economics 29(8): 2129–2142. https://doi.org/10.1177/13548166221126577

62.

Ying

Wen

(2019) Exploring the male Chinese tourists’ motivation for commercial sex when travelling overseas: scale construction and validation. Tourism Management 70: 479–490. https://doi.org/10.1016/j.tourman.2018.09.014

63.

Ying

Wen

Goh

, et al. (2022) Perceived constraints to sex tourism overseas: scale development and validation. Asia Pacific Journal of Marketing and Logistics 34(7): 1503–1523. https://doi.org/10.1108/APJML-09-2021-0640

64.

Yotov

Piermartini

Monteiro

, et al. (2016) An Advanced Guide to Trade Policy Analysis: The Structural Gravity Model. World Trade Organization.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB