“A man without money getting a sexual partner? It doesn’t exist in our community”: male partners’ perspectives on transactional sexual relationships in Uganda and Eswatini

Abstract Research on transactional sexual relationships has largely focused on women’s perspectives. Better understanding the men’s views—especially regarding relationships with adolescent girls and young women—can inform HIV prevention efforts. In 2017, 134 in-depth interviews were conducted with the male partners of girls and young women aged 19–47 years, 94 in Uganda and 40 in Eswatini. Respondents were recruited at venues such as bars where men and potential partners meet and through other young women. Most respondents believed that providing money/gifts was the way to establish relationships with women in their communities, a context that some found undesirable. Young women were mainly perceived as actively pursuing transactional sex for material goods, but respondents also described economically impoverished women who were manipulated into relationships. Men described conflict with longer term partners as a driver to seeking younger partners, who were more compliant. Transaction dominates the male partners of adolescent girls and young women’s understanding of sexual relationships, and inequitable power dynamics are reinforced by seeking younger partners. However, some respondents’ discontent with this dynamic suggests an opportunity for change. HIV prevention programmes should directly address the underlying drivers of transactional relationships (e.g. gender norms) and work with men who question the practice.


Introduction
Transactional sexual relationships-which refer to non-commercial, non-marital sexual relationships motivated by the assumption that sex will be exchanged for material support-are prevalent in many east and southern African countries (Stoebenau et al. 2016). While it is common for material support to be exchanged in sexual relationships, transactional relationships may be distinguished from others when the primary motivation for starting or staying in the relationship is material support (Moore, Biddlecom, and Zulu 2007;Stoebenau et al. 2016). Research in east and southern Africa highlights how transactional sexual relationships often take place within agedisparate partnerships, in which men provide money or gifts to women who are substantially younger than themselves (Luke and Kurtz 2002). A global review of transactional sex among youth found lifetime prevalence rates among young women in sub-Saharan Africa that ranged from 5% among 12-to 17-year-olds in Cameroon to 85% among 12-to 20-year-olds in Uganda (Krisch et al. 2019). A longitudinal study with young women aged 15-24 in Malawi found that 20% reported transactional sex over a 12-month period (Gichane et al. 2020).
East and southern Africa comprise around 6% of the world's population, but over half (54%) of the total number of people living with HIV in the world (UNAIDS 2019). Adolescent girls and young women (i.e. those between the ages of 15 and 24) in the region are disproportionately at risk of HIV, with about 6,000 of them acquiring HIV per week. Young women of this age accounted for 26% of new HIV infections in 2018, despite comprising just 10% of the population (UNAIDS 2019). Research has demonstrated a strong link between transactional sex and HIV-related risk, including inconsistent condom use (Luke 2003), gender-based violence (Dunkle et al. 2004), multiple partnerships (Dunkle et al. 2007), high alcohol consumption (Norris, Kitali, and Worby 2009), and socioeconomic disadvantage (Gichane et al. 2020). Transactional sex has also been associated with HIV seropositivity (Ranganathan et al. 2016) and HIV acquisition (Kilburn et al. 2018).
Transactional sexual relationships within the region take place in a context of distinct gender norms for men and women (Wamoyi et al. 2011;Fielding-Miller et al. 2016). These include dominant notions of masculinity associated with control over partners and the ability to provide financially (Maclin et al. 2015) and substantial power imbalances between men and younger women (Pulerwitz, Mathur, and Woznica 2018). At the same time, growing economic autonomy among women has affected the dynamics and transactional nature of partnerships (Kabeer 2005;Hunter 2007). Further exploration of the gender dynamics associated with transactional relationships will help inform our understanding of this phenomenon as well as the HIV response.
A number of studies have documented motivations and meanings of transactional sexual relationships from the perspective of young women (e.g. Luke and Kurtz 2002;Dunkle et al. 2004;Wamoyi et al. 2011;Bandali 2011;Fielding-Miller et al. 2016;Wamoyi et al. 2019). One line of research highlights such transactional sex/relationships as a manifestation of economic inequality, with women's economic needs as the main driver. Other research has highlighted women's desire to improve their social status, or the desire for love and intimacy, as important motivations (Stoebenau et al. 2016).
Yet comparatively little is known about the men's views concerning transactional sex, and how similar or different these are from young women's views and experiences. The limited number of studies focused on men have emphasised the high frequency of the phenomenon, predictors of transactional sex (e.g. alcohol, higher socioeconomic status), and the existence of gendered expectations that men should provide for women's material needs in sexual relationships (Dunkle et al. 2007;Jewkes et al. 2012;Wamoyi et al. 2019). This article draws on the voices of men in two sub-Saharan countries-Uganda and Eswatini-to understand men's perceptions, experiences and motivations for transactional sexual relationships with young women in their communities. Exploring the issue in Uganda and Eswatini is particularly important, as both are countries hard hit by the HIV epidemic. In Uganda, UNAIDS estimates that 5.8% of adults aged 15-49 are living with HIV, with 48,000 new infections among adults 15 and older in 2019 (UNAIDS 2021b). In Eswatini, HIV prevalence among adults aged 15-49 is estimated to be 27.0%, the highest in the world (UNAIDS 2021a).

Methods
Between April and August 2017, a total of 134 in-depth interviews were conducted with the male partners of girls and young women aged 19-47 years, 94 in Uganda (in Gulu, Mukono and Sembabule districts) and they were 40 interviews with male partners of young women in Eswatini (Matsapha and Siphofaneni). These areas were chosen because they represent a diversity of urban/peri-urban/rural and socio-cultural contexts in each country. The study took place in the context of the DREAMS partnership (Saul et al. 2018), which aimed to reduce HIV incidence among young women in 15 countries with substantial HIV epidemics, including Eswatini and Uganda. This article focuses on data from male partners of young women in the first year of programme implementation.
Eligible participants included men older than 18 years of age with at least one female partner aged 24 years or younger, in the past year. In both countries, respondents were recruited at venues where men and potential partners gather, such as bars, lodges, boda boda (motorcycle taxi) stands, construction sites and marketplaces. In Uganda, we also recruited the male partners of adolescent girls and young women enrolled in DREAMS, through the programme's implementing partner organisations. Venues were identified via focus groups discussions with local DREAMS programme partners, HIV service providers, and community opinion leaders. After being approached by an interviewer who shared a brief study description, and expressing interest in the research, potential participants were asked to complete a screening form to confirm eligibility (i.e. being at least 18 years old, and having at least one female partner aged 24 or younger in the past year). The screening form also asked whether the potential participant had a partner aged 25 or older, plus where he lived, and type of work he engaged in, with the intention of minimising the chance of social desirability bias affecting willingness to participate. Male partners of DREAMS programme participants were identified and contacted by study implementation partners (after permission to contact had been given by the young women herself) and were referred to the study team if interested in participating.
Interviews were conducted by six experienced research assistants in each country. All research assistants were male in Uganda; half were male and half female in Eswatini, in line with local investigator recommendations and research assistant availability. All received extensive training in qualitative interviewing about sensitive topics.
A semi-structured guide was used which explored respondents' perceptions of men's relationships with girls and young women in their community, including transactional sexual relationships, their own relationship dynamics, and their HIV prevention behaviours and service use. Interviews were conducted in local languages (Luganda and Luo in Uganda and siSwati in Eswatini), lasted between 60 and 90 minutes, and were audio recorded. All participants signed a written consent form and were interviewed individually in a private location. Participants were provided refreshments and were compensated for transport to attend the interview (at an average rate of US $5 per study participant).
Interviews were transcribed and translated into English by the research assistants. Transcripts were coded in Atlas.ti v7, following principles of thematic analysis (Creswell and Creswell 2018). A codebook was constructed based on topical codes (topics previously defined in the interview guide), as well as emergent codes (topics which emerged during fieldwork). The codebook, finalised by the principal investigators, resulted in 23 main codes. To maximise internal validity, three team members for Uganda data, and two for Eswatini data, coded the same three transcripts, with any differences in coding identified and discussed until consensus was reached. The remaining interview transcripts were each coded by one of these team members. Research team members then developed 3-to 5-page summaries for key code reports, which were then reviewed and discussed by the study team. Comparisons were also made between transcripts based on key characteristics (e.g. by country, and by younger versus older respondents). Preliminary themes were shared with in-country stakeholders (including policy makers, programme managers and investigators) during a half-day data interpretation workshop with approximately 50 stakeholders in Uganda and a Technical Working Group meeting in Eswatini. Feedback during these sessions helped further refine final themes and write-up of results.
These activities were reviewed and approved by the Population Council's Institutional Review Board (IRB), the Makerere University School of Public Health IRB, the Uganda National Council for Science and Technology, and the National Health Research Review Board (Eswatini). Table 1 presents the characteristics of the 134 participants (94 from Uganda; 40 from Eswatini). Respondents' ages ranged from 19 to 45 years in Uganda and 20-47 years in Eswatini, with a mean age of 28 years in both countries. Nearly all participants were working, and most had completed at least a primary education.

Findings
Most men in Uganda were married or cohabiting (80%), and Ugandan respondents indicated that getting married was an important step towards successful adulthood. Most had married in their early 20s, usually to a woman three to five years younger than themselves. Multiple young women partners outside marriage were also frequently reported by the Ugandan sample, both short-term and longer-term transactional partners. Long-term transactional partners were often also referred to as wives by respondents, though they typically had not gone through a formal marriage ceremony nor did respondents mention being in a polygamous marriage per se (Gottert et al. 2018a). About half of respondents reported short-term transactional young women partners. These short-term partners were often seen as having the potential to become a longer term partner/wife. Regardless of the respondent's own age, partners were usually adolescent girls or young women when the relationship started, and thus as men added partners over time, the age difference between them grew.
In contrast, a minority (23%) of men in Eswatini were married or cohabiting, and marriage was not seen by respondents as a lifelong objective. Yet most men in Eswatini also reported having multiple sexual partners, including both shorter-term and longer-term transactional partnerships with younger women, with these partnerships often being concurrent.
Overall, despite the difference in marital status, the themes highlighted by participants were broadly similar across both country samples. For example, men expected sex within these transactional relationships-regardless of whether they were short or longer term-although the daily interactions in the relationships could range from those explicitly or exclusively focused on sexual activity to socialising in public and spending a substantial amount of time together.
Men saw money and gifts as the main way in which to establish and maintain sexual relationships with women Participants in both Uganda and Eswatini commonly shared the view that the main way in which relationships between men and women were established and sustained within their communities was through transactional exchange, in which men assumed the provider role. This applied both to primary partners and other longer-term relationships as well as to short-term partners. Respondents indicated that it was nearly impossible to initiate any kind of sexual contact with a woman without giving money or gifts. In this kind of situation, men perceived that, in the eyes of women, their value and worth was determined by how many gifts and how much money they could offer, or how well they could support a partner.
A man without money getting a wife or sexual partner? It doesn't exist in our community. I've spent quite a long time without seeing it happen and the reason for this can be explained through the ladies' desire for money … . It's not easy for a person who doesn't have money to establish a relationship with these young girls, because these girls desire beautiful things a poor man cannot afford. ( Men worried that relationships would dissolve if they could not provide financially, a concern that applied to both short-term and long-term partners. From this perspective, relationships possessed a certain fragility, as the threat of a relationship ending figured in men's minds. Because of this, men were emphatic that they could not risk losing access to economic resources and income. Even though women may have participated in paid work, men perceived themselves as the main providers and perceived that their female partners expected this as well. Respondents described how tension or conflict could arise with partners over this issue. Girls nowadays, my brother, are after money. If a woman is married, she expects the husband to have money and if it is the other way around, she will immediately leave the husband and settle with another man who has money. (Eswatini, aged 24, unmarried with two concurrent long-term young women partners) We were in a relationship for close to ten years and have five children together. We were happy but she became disrespectful, she always told me that I was poor. When my Aunties sold my portion of land, she told me that that was the only wealth that I had and now I was even poorer, so she decided to leave. (Uganda, aged 40, married to 18year-old, no other recent partners, speaking about his ex-wife) Participants believed that men with the most resources would be those with the best chances of establishing relationships with women. Older men were commonly described as having more resources-and thus being an advantage over others in securing relationships. This was sometimes described with frustration, particularly by younger respondents: So, the old men snatch the girls from the younger men, because the girls say no matter how much I may love you, if you cannot give me some things or cannot solve my pressing needs, they need to look for it from [older men]. (Uganda, aged 22, married to 18-year-old woman, with no other recent partners) In the same way that participants understood men's role as provider as a key aspect of their masculine identity, they often attributed a need to be supported financially/taken care of, and women's interest in money, to a female identity.
It is in-born; the women were created to receive their needs from men … . That is their nature, where a woman is not with a man, she cannot survive that is why you can find a man with six or seven wives. (Uganda, aged 40, married to 23-year-old woman with another concurrent long-term young partner) Men perceived most adolescent girls and young women as active in seeking transactional relationships Overall, respondents perceived young women as strategically using men to access money as well as the material luxuries neither they nor their families could afford. They described how women wanted and sometimes explicitly requested meals out, alcohol, visits to entertainment spaces, mobile phones, airtime and expensive clothing, as well as transport, including in men's cars. Most men read women's interest in money as a desire for status, entertainment or a way to compete with female peers. Interviewees were critical of such motivations and saw them as characteristic of young women "nowadays" compared to the past. Many men tended not to see their own actions as the drivers of transactional sex. Instead, they saw women as active agents in pursuit of money or gifts via sex, and as having a clear understanding of what each party was doing when establishing transactional sexual encounters. Respondents shared the following perspectives: You know these days girls have so much love for good things. It is unlike it was in the past where children would not mind so much about good things such as good dresses … so they look for men who can give them such. (Uganda, aged 34, married to 23-year-old, reported no other recent partners) [Girls] want money and material things, especially clothes and fast foods that is what they like to have. They want to be seen wearing expensive clothes … [men use money] as a way to attract them. (Eswatini, aged 33, unmarried with both young and older women recent partners) One thing I'm sure of is … the women do not reject money; they suggest the price themselves … . When you show her the money, she will straight away ask for a place where you can meet. (Uganda, aged 35, married to 23-year-old, reported no other recent partners) While transactional relationships occurred among people across a wide range of ages, participants highlighted that they were particularly frequent between young women and older men. Interviewees thought that young women were interested in older men because they saw in them an opportunity for greater financial stability and support.
… every woman's goal around here is having money. So, unlike the young men who are just after sex, these older men are willing and ready to spend on them, which is exactly what these women are after. (Eswatini, aged 22, unmarried, reported two longer-term young women partners) It is also common knowledge amongst young women and their sex partners in this community that all relationships go with some financial support. Apparently, young women have a clear understanding of the size and frequency of the financial benefits obtained from older men versus that obtained from young men. (Uganda, aged 31, married to 20-year-old, reported other recent partners)

Men saw some young women as being manipulated into transactional relationships
While in general men saw adolescent girls and young women as active agents in transactional exchange, there were some groups of young women they felt were vulnerable to predatory behaviours of older men-particularly very young, economically impoverished and migrant women. In these cases, respondents were highly critical of men offering material goods to establish a relationship, often describing it as an act of "luring" or "tricking" the girl or young woman, who was seen as an innocent or vulnerable player. "Go-betweens" were sometimes enlisted to facilitate these relationships and were even "bribed" to encourage the young women to engage in a transactional relationship.
[Men] provide them with gifts up until sexual activity takes place. Bayabadweba, which simply means that they attract them with nice things and these adolescent girls are unable to refuse because of all the things the older man might have bought for her. (Eswatini, aged 34, unmarried with two long-term young women partners) These girls are just deceived by the men … . When a mature man buys her a drink, she will not refuse because she doesn't even have money to buy herself one. So, when he buys her a drink, he will want to make her 'pay' [sexually] for the drink … . They use [money] to lure the innocent young girls in relationships, they know that these young girls have many financial needs … . Young girls [also] come from the rural area seeking a better lifestyle, so they are easy prey for these older men … . (Uganda, aged 24, unmarried, reported one young woman partner) Participants also highlighted how men without money may deceive women into thinking that they have resources to provide. For example, a man might incorrectly claim that he owns something to impress a potential partner.
It's common for guys without money to deceive these girls … . You can tell her that 'maybe that building is mine', you can even borrow your friend's room … and you tell her it is yours. If you don't, you can't get her because you don't have money. (Uganda, aged 40, married to 24-year-old woman, reported other recent partners) Such men also deceive a lot, they can tell the girls that they have these things yet, the possessions could be owned by a relative so by the time they girl realises, she is pregnant by him. (Uganda, aged 33, married to 23-year-old, reported no other recent partners)

Men often look for younger women in particular
Interviewees described two main reasons why younger women were of special interest to men: because of their beauty, and because they were less likely to be argumentative than older women. Also mentioned a few times was that adolescent girls and younger women were likely too young to have acquired HIV, and thus HIV risk reduction behaviour was not needed with them. An example of the first main reason is as follows: They look fresh and beautiful as compared to the housewife that one could be having back at home. (Uganda, aged 33, unmarried, reported multiple recent longer-term young woman partners) Second, men stated that arguments with a partner could trigger the desire to seek relationships with other women; conflict and arguments with long-term partners were reportedly quite frequent. These other relationships were usually with younger women, who were considered to be more agreeable and compliant. Interviewees highlighted young women's availability in general, their willingness to listen, and their tendency not to contradict what an older man was saying as positive characteristics, in contrast to wives, who were depicted as controlling and/or independent-minded.
The young girls are easy to handle psychologically as compared to the older ones. They are easy to play around with and they are not controlling, and they give freedom to the man to operate. (Uganda, aged 35, married to 21-year-old woman, reported another recent partner) The young women listen and cooperate all the time, yet older women argue. When [men] call the young women, they are always available to spend time with them, unlike older [women] who will question every time and need explanation. (Eswatini, aged 29, unmarried, reported multiple long-term young woman partners) Only a few men described being in loving, trusting relationships without the expectation of moneywhich they highly valued Finally, a subset of participants described their relationships as characterised by commitment, love and trust without the expectation of money or material goods in return. They valued these relationships highly.
I feel so honoured to have her and she is not someone who loves money too much, she's someone I can spend the rest of my life with, which is why I am happy about her … . Actually, if a person loves your money, she doesn't really love you. (Eswatini, aged 27, unmarried, in long-term relationship with a 21-year-old woman, reported no other recent partners) My previous girlfriends only wanted me because I was still strong and could give them money. [My current partner] can be trusted. She is also caring; she provides all that I need especially when I was not moving without the aid of walking sticks … . She liked me because of the way I was talking to her and the way I approached her. She likes being around a person who makes her happy and laughing all the time. But she is also humorous, she makes me happy. (Uganda, aged 33, married to 23-year-old, reported no other recent partners)

Discussion
In this study we sought to understand men's perspectives and experiences of transactional sexual relationships between men and young women-a prevalent practice in eastern and southern Africa that influences the trajectory of the HIV epidemic (UNAIDS 2016). The most common theme from this large qualitative data set with men across Uganda and Eswatini was that men perceived that providing money/gifts was the main way to establish and maintain relationships with young women in their communities. They worried that their own relationships would dissolve without the ability to provide-be the partnership short-term or long-term. Only a small group of men thought alternatives to exchanging gifts and money for sex (e.g. a love-based relationship) were possible. These results reveal how men themselves-even from two different social contexts-feel constrained by these gender dynamics.
Most respondents perceived adolescent girls and young women as actively seeking transactional sex, motivated mainly by the desire for material goods, and saw this as a hallmark of current times. This is consistent with what young women themselves have reported in several contexts in sub-Saharan African countries (Wamoyi et al. 2011;Shefer, Clowes, and Vergnani 2012). These studies found that women are often active participants in decision-making about transactional sex, and they are intentional in using sex to financially support themselves and their children.
At the same time, factors influencing transactional relationships-as reflected in men's perspectives on limited relationship options and young women's role in seeking these relationships-are rooted in gender-based power relations and gender inequitable norms. Masculine identity has traditionally been associated with the ability to provide financially for others and is enacted by men economically supporting their partners and families (Hunter 2007;Maclin et al. 2015;Siu, Wight, and Seeley 2012). The inability to provide economically, therefore, both limits men's ability to find sexual partners (particularly for younger men), and distances men from enacting masculinity successfully. Young women's sometimes active role in pursuing men with money challenges dominant and often stereotyped narratives on expectations around relationship formation, yet is also often driven by economic need and gender-based differences in access to alternative sources of income.
Transactional sex can create heightened risk for acquiring HIV among young women due to power asymmetries in relationships (UNAIDS 2016). Not many men discussed a direct connection between HIV risk and transactional relationships, but some men felt younger women were less likely to have HIV due to their young age, which was thus a reason to seek them as partners. Further, masculinity in these contexts is at least partly defined by men's control over their sexual partners. A notable number of men in this study thought women who did not question their authority made better sexual partners, and that they were more likely to find acquiescence in younger women. This likely compounds the risk for younger women and those with greater resource limitations, as these women were likely to be constrained in their ability to take decisions. Previous research has found that girls and young women have a hard time negotiating the terms of sexual relationships with men who offer them money or other material goods, which contributes to HIV risk (Bandali 2011). Yet, other research demonstrates that both inequitable gender norms and low levels of relationship power correlate with higher HIV risk and partner violence (Pulerwitz et al. 2010;Gottert et al. 2018b;Pulerwitz, Mathur, and Woznica 2018). Differences in age exacerbate these gender-based power inequities.
While most men perceived young womens actively seeking transactional sex, a minority of respondents also described other men (often friends or acquaintances of theirs) as "luring" some young women in sexual relationships. This was especially true for particularly vulnerable subsets of young women, including very young women, poor, and recent rural to urban migrants. These findings reinforce how men can take advantage of existing gender-based power dynamics and norms to manipulate some young women into transactional relationships. It also suggests that some men (who usually reported on this with a negative tone) are uncomfortable with this dynamic, which provides a potential opening to work to shift norms and reframe this type of manipulation as unacceptable.
These findings offer promising avenues for future programmes and interventions, especially considering some men's dissatisfaction with relationships involving gifts and transaction, and their critical views towards transactional relationships where they perceived them to be exploitative (e.g. with very young girls). First, it is important to provide opportunities for men to openly share these views, and for men and women to collectively engage in critical reflection around this. It is likely not evident within communities that some men express concern about, and question, the status quo. A number of evidence-based, gender transformational programmes are available that encourage discussion about gender norms and power relations as they affect risk of HIV, sexually transmitted infections, and gender-based violence. Stepping Stones and SASA!, for example, work with men and women in communities and have been tested in many sub-Saharan countries, including in Uganda and Eswatini (Jewkes et al. 2008;Abramsky et al. 2014), and Program H and One Man Can are programmes focused on reaching men directly and engaging them in dialogue in small groups (Barker et al. 2010;Pettifor et al. 2018). That said, and despite the growth of work on gender-transformational programming in recent years, these programmes have rarely been implemented at scale. More consistent scale-up of evidence-based programmes is the next step-and challenge-in the field (Levy et al. 2020).
Another theme that emerged in the interviews was the connection between discord within marital and other longer term relationships and men reportedly seeking new partners (particularly younger partners). Several studies (e.g. Higgins et al. 2014;Jewkes, Flood, and Lang 2015;Gottert et al. 2018b) have suggested that many men and women in contexts such as those focused on here lack skills in couple communication and conflict resolution. Programmatic responses could include applying evidence-based models that strengthen men's, women's, and couple's skills to communicate more equitably and resolve conflicts in relationships (e.g. Doyle et al. 2018;Siu et al. 2017).
Study findings also highlight the need for intervention to support economic autonomy for adolescent girls and young women. Emerging evaluation results from the DREAMS partnership-which was initiated to address the contextual factors contributing to high HIV incidence among girls and young women in the region and included a package of life skills, health education, equitable gender norms promotion, and economic and educational support activities-suggest that this package may indeed be successful in reducing transactional sex (Mathur et al. 2019).

Limitations
Several limitations of the study should be pointed out. Despite the extensive training of interviewers to minimise response bias, participants' responses may reflect social desirability bias, given that transactional sex is a sensitive topic. Moreover, given the eligibility criteria for the study were that participants should report having an adolescent girl or young woman partner in the last 12 months, the views expressed may reflect those of the group of men that engage in such relationships and may not be representative of other men. To shed more light on both women's and men's perspectives, it would be useful to include couples in future research, to understand more about relational dynamics from both men's and women's perspectives simultaneously.

Conclusion
Directly addressing transactional sexual relationships, and their underlying drivers, could play a major role in curbing the HIV epidemic. Given the men's greater access to economic resources and decision-making power, and gender norms supportive of transactional sex, unless we confront these dynamics directly, high levels of HIV vulnerability are likely to persist.