In rural South Africa, the strongest predictor of whether a young man has exchanged money or other gifts for sex is whether he has committed intimate partner violence or rape, according to an analysis of data from an HIV prevention trial.¹ Transactional sex is also associated with high socioeconomic status, adverse childhood experiences, having a large number of sexual partners and alcohol use. These associations hold regardless of whether the young man is the provider or the recipient of the gifts, and whether the sex is with his main partner or a casual partner.

The study was a secondary analysis of data from a study of an HIV prevention program that recruited 1,396 young men aged 15–26 from secondary schools in 70 villages in South Africa's Eastern Cape province. At baseline, participants completed structured, face-to-face interviews that included items about demographic variables, media exposure, childhood trauma, alcohol abuse (determined by the World Health Organization's screening questionnaire), resistance to peer pressure, coerced sex with men and attitudes toward gender relations and relationship control. In addition, participants were asked whether they had ever engaged in trans- actional sex, defined as heterosexual intercourse motivated by the provision (by either partner) of food, clothes, transportation, cosmetics, gifts for family members, school fees, a place to sleep, alcohol, a "fun night out" or money; they were also asked if they had ever been part of a transactional relationship—one in which either partner's involvement had been primarily motivated by expectations of receiving material goods. Finally, participants were asked whether they had ever engaged in emotional, physical or sexual violence against a main girlfriend and whether they had ever coerced a woman who was not their girlfriend into having sex. The researchers calculated descriptive statistics, and identified predictors of transactional sex using logistic regression.

After the exclusion of respondents who had never had sex or who had not provided an adequate sexual history, the final sample consisted of 1,288 young men. Most were 20 or younger (84%), and nearly all were students (97%) and had a main girlfriend (89%). On average, they had had seven lifetime sex partners; 73% had had at least one casual partner.

Although not the norm, transactional sex was not uncommon: About one in five respondents had had such sex with a casual partner, either as the provider of resources (13%), the recipient (2%) or both (5%). A similar proportion had been part of a transactional relationship, as the provider (7%), recipient (6%) or both (8%).

Gender-based violence was the strongest predictor of transactional sex with a casual partner. Men who had perpetrated both physical and sexual intimate partner violence with a main partner were more likely than those who had done neither to have had transactional sex, either as the provider of material resources (odds ratio, 5.6) or the recipient (2.8). Men who had committed a sexual assault outside of a relationship also had elevated odds of having had transactional sex as the provider (1.6) or recipient (2.2). Emotional abuse against a main partner was associated with receiving resources from a casual partner (2.3), but not with giving them.

In addition, the odds of having provided gifts in exchange for sex were elevated among respondents with an alcohol problem (odds ratio, 1.6) and increased with each additional year of age (odds ratio, 1.1), each additional unit increase in respondents' media exposure scores (1.3) and each five additional lifetime partners (1.3); they declined with each unit increase in respondents' peer pressure resistance score (0.8). The odds of receiving resources from a casual partner in exchange for sex increased with each five additional lifetime partners (1.3), but declined with each unit increase in peer pressure resistance score (0.7).

As it was for transactional sex with a casual partner, intimate partner violence was the strongest predictor of having been in a transactional relationship with a main partner. Young men were more likely than their peers to have given money or other resources in exchange for a relationship if they had a history of physical violence against a main partner (odds ratio, 1.6), sexual violence against a main partner (2.5) or both (5.0). Similarly, the odds of having received resources in exchange for a relationship were elevated among young men who had committed physical violence (1.5), sexual violence (2.3) or both (4.1).

Other factors associated with having been involved in a transactional relationship as the giver included having an alcohol problem (odds ratio, 1.9) and having ever been coerced into sex by another man (2.6); furthermore, the odds increased with each five additional lifetime sex partners (1.3) and with each unit increase in socioeconomic status score (1.2). The odds were reduced, however, among men who had 10 or more years of education (odds ratio, 0.6) or who were in the quartile of respondents with the most equitable gender attitude and relationship control scores (0.6). Factors associated with having been involved in a transactional relationship as the recipient included having ever earned money (1.7) and having an alcohol problem (1.8); again, the odds were reduced among men with the most equitable gender attitude and relationship control scores (0.4).

The fact that perpetration of both physical and sexual intimate partner violence was consistently the strongest predictor of trans- actional sex—regardless of whether the man provided the gifts or received them—suggest that "transactional sex should be viewed as part of a cluster of closely related violent and controlling practices," the researchers note. Although it may seem counterintuitive that receiving resources would represent a form of controlling behavior, these transactions may represent "financial abuse or exploitation," such that men who expect to receive resources in exchange for sex or companionship "become violent if thwarted," according to the researchers. Because the study sample consisted of young rural men who had volunteered for an HIV prevention program, the findings may not be applicable to the general population. Nonetheless, the results suggest that "interventions which seek to explicitly transform ideas of masculinity that privilege heterosexual success with and control over women will be more effective than those that address only individual risk behavior in reducing incidence of transactional sex, HIV risk and gender-based violence," the authors conclude.—P. Doskoch

REFERENCE

1. Dunkle KL et al., Transactional sex with casual and main partners among young South African men in the rural Eastern Cape: prevalence, predictors, and associations with gender-based violence, Social Science & Medicine, 2007, 65(6):1235–1248.