Intersections of financial agency, gender dynamics, and HIV risk: A qualitative study with adolescent girls and young women in Zambia

ABSTRACT Objectives: Recent research demonstrates that economic interventions may positively effect HIV risk among adolescent girls and young women (AGYW) in sub-Saharan Africa. Some evidence reveals potential associations between financial decision-making and bargaining power in sexual relationships. However, this evidence is mixed, nuanced, and limited. This paper explores how AGYW in Zambia understand financial agency and its effect on intimate relationships. Methods: In-depth qualitative interviews were conducted with 30 females aged 15–24 years residing in Kalingalinga, a low income, high-density residential area in Lusaka. Data were analysed using thematic content analysis. Results: Participants spoke of the ability to earn and spend money as reality for some and aspirational for many others, intrinsic to cultural and religious caveats influencing perceptions of agency for women. The transfer of financial independence to sexual agency within relationships was viewed as a mechanism for HIV risk reduction; however, male sexual privilege was an obstacle irrespective of financial decision-making. Conclusions: Programmes aiming to enhance financial agency for AGYW have the potential to reduce HIV sexual risk. Yet, to be most effective, integration with gender-transformative programmes is needed to address norms of male dominance that keep AGYW in positions of vulnerability.


Introduction
In 2016, the Joint United Nations Program on HIV/AIDS (UNAIDS) pledged to reduce new incidence cases of HIV among adolescent girls and young women (AGYW), ages 15-24, to fewer than 100,000, requiring a 75% reduction from 2010 (UNAIDS, 2016). The focus on AGYW is based on the disproportionate number of new HIV positive cases found among this demographic, particularly in sub-Saharan Africa (SSA) (Dellar et al., 2015;Karim et al., 2017;Kasedde et al., 2013). UNAIDS reported approximately 5000 new HIV infections per day in 2017, 66% of which are located in SSA, and nearly 20% of the 4400 new infections per day that are ages 15 and older being females between the ages of 15-24 (UNAIDS, 2018).
Gender inequality is a critical underlying factor on HIV risk among AGYW in SSA (Bhana & Pattman, 2009;Harrison et al., 2015;Higgins et al., 2010;Jewkes & Morrell, 2010;Krishnan et al., 2008). From age-disparate relationships Schaefer et al., 2017) to transactional sexual encounters (Jewkes & Morrell, 2012;Wamoyi et al., 2016), as well as the occurrence of sexual violence Stockman et al., 2013), HIV sexual risk for young women in SSA cannot be well understood without fully exploring gender inequality and how it influences sexual behaviour.
Societal norms that reinforce male dominance in sexual relationships can put females at greater risk of HIV. For instance, attitudes on male sexual privilege can perpetrate sexual violence against AGYW (Barker et al., 2010;Pulerwitz et al., 2010). Studies show that experiences of sexual violence increase the risk of HIV for females Mathur et al., 2018). In a populationbased study in Botswana and Eswatini, gender inequitable attitudes were independently associated with higher levels of male-controlled sexual decision-making, perpetration of rape, unprotected sex with a non-primary partner, intergenerational sex, and multiple/concurrent sexual partners (Shannon et al., 2012). Similarly, in South Africa, a lack of female decision-making power in sexual relationships has been found to be associated with inconsistent condom use (Pettifor et al., 2004) and incident HIV infection , whereas increased relationship power among females in Kenya has been associated with lower levels of sexual violence, greater condom use, and higher knowledge of partner HIV status .
Several recent studies have tested the correlation between economic strengthening interventions (such as cash transfers, savings accounts, or financial literacy) and HIV sexual risk among AGYW, identifying positive correlations between receipt of these economic strengthening interventions and reduced sexual risk behaviour (Baird et al., 2012;Cluver et al., 2013;de Walque et al., 2012;Pronyk et al., 2008;Ssewamala et al., 2010). However, some evidence reviews have questioned the efficacy of such interventions, wondering if they are broad enough to address the structural inequalities faced by young women (Gibbs et al., 2012) or if the self-reported behaviour data limits the accuracy of the findings (Swann, 2018). There is, therefore, a need to further understand the complexities of AGYW autonomy and how that autonomy impacts on relationships and sexual HIV risk through qualitative enquiry.
Asset Theory provides a useful framework to understand financial agency, namely tangible accumulation of assets is theorised to affect behaviour by promoting future oriented thinking, increased self-efficacy, and enhanced human capital (Sherraden, 1991), promoting health-seeking behaviours. Sen's Capability Approach echoes similar concepts around access to resources, utilisation of those resources, self-efficacy, and well-being (Sen, 1993). It could be reasonably assumed that increased economic resources and/or increased financial decision-making power subsequently translates to greater decision-making power within sexual relationships, followed by a decreased exposure to HIV risk.
Therefore, there is a need for qualitative research to understand the nuances within the construct of financial agency and its potential limitations to reduce sexual HIV risk among AGYW. An enhanced understanding of these intricacies described by AGYW can inform HIV prevention interventions and offer insight on the value and limitations of economic strengthening interventions.
The purpose of this study was to conduct a qualitative assessment of financial agency among AGYW in Kalingalinga, Zambia, in the context of HIV. In this study financial agency refers to both individually earning income and having decision-making power over how earned income is managed, saved, and spent. It is used interchangeably with 'financial independence' and 'financial autonomy'. The following two research questions were explored: (1) How do AGYW understand financial agency within their own lives and communities?; and (2) How is financial agency perceived to influence, or not influence, sexual behaviours and HIV risk?

Study setting
This qualitative study was carried out in Kalingalinga ward, a low-income and high-density residential area approximately 8 km east of Lusaka's central business district and within Lusaka province, which hosts the second-highest HIV prevalence rate within Zambia at 15.7% (Government of Zambia, MOH, 2019). Despite period housing and social amenity upgrades since the 1970s, the physical infrastructure remains largely under-resourced, characterised by poor-quality housing, drainage and sanitation facilities and limited piped water supply. Economically, the majority of Kalingalinga residents work either in the informal sector or in low paid jobs within the public and private sectors. Many are unemployed. Healthcare in the area is mainly provided by the local public health facility -Kalingalinga health centre. It is one of several high-volume HIV clinics in the country providing free HIV treatment and care, including the provision of free antiretroviral treatment (Government of Zambia, MOH, 2012).

Sample
Participants were purposively sampled across Kalingalinga to reach 30 AGYW, with 10 girls from each age cohort (15-17, 18-20, and 21-24). Purposive sampling used a maximum variation approach (Palinkas et al., 2015) to achieve as much variability as possible with respect to the highest level of education obtained and marital status, as these were considered critical for understanding various viewpoints (Do & Meekers, 2009;Hargreaves et al., 2008;Singh et al., 2013). The study was a nested qualitative study with AGYW who had previously participated in a quantitative survey assessing HIV risk factors among AGYW on this site. The site is one of 21 locations within Zambia where the multi-sectoral DREAMS programme was being implemented to reduce HIV risk among AGYW. More on the DREAMS programme can be found elsewhere (Saul et al., 2018). Individuals were considered eligible for the study if they resided in Kalingalinga, were female, were between the ages of 15-24, and had previously participated in the quantitative survey. Only one resident per household was eligible to participate.

Data collection
Qualitative in-depth interviews were conducted in March and April 2019 with each of the 30 participants, in a location where the participant could speak privately. Discussions were approximately one hour in length and directed by a semi-structured interview guide to explore key themes relevant to the research questions, including community expectations for young women and income; how financial decision-making affects power dynamics within sexual relationships and how financial independence influences one's power to protect against HIV. The guide consisted of initial prompts followed by probing questions to understand how answers varied by age, to obtain concrete examples of stated beliefs and to better understand the values underpinning responses.
All interviews were recorded with the permission of respondents. Some interviews were carried out in English at the request of the respondent. Others were carried out in Nyanja or Bemba and subsequently transcribed to English. Back-translation checks were carried out by bi-lingual speakers.
The research team included one female Zambian interviewer, one male Zambian field coordinator, and one female bi-lingual staff for transcription and translation. Another researcher from Columbia University conducted the data analysis, which was cross-checked by the U.S.-based principal investigator of the broader DREAMS study and the Zambian field coordinator.
Participants were first introduced to the study and provided consent for their participation. For individuals under the age of 18, assent was obtained as well as the consent of a parent or caregiver. Researchers reiterated that participation was voluntary and would have no effect on access to benefits of current or future programmes. Prospective participants and their parents/guardians were told they were free to withdraw from the study at any time and were assured that all efforts would be made to ensure confidentiality. Participants were compensated 50 Kwacha (approximately $5 USD). Study procedures were approved by the Population Council's Institutional Review Board.
Local ethics review and research authorisation were also obtained in Zambia (ERES CONVERGE IRB No.00005948 and FWA:00011697).

Analysis
Thematic analysis of qualitative data was applied to incorporate inductive and deductive analytical approaches (Creswell & Poth, 2017). Content analysis and inductive thematic analysis (Thomas, 2006) were used to identify new themes emerging from the transcripts, given that the concept of financial agency and its relationship to sexual HIV risk reduction is largely understudied. At the same time, deductive analyses were useful to explore themes that do exist in the current literature such as the theory of gender and power in the context of HIV risk (Wingood & DiClemente, 2000). Preliminary themes were identified and initial codes used to highlight relevant quotes. The lead analytical researcher then collated codes to generate four emergent themes. Data were also analysed by age groups/bands to understand variability by developmental stage.

Results
Sociodemographic characteristics of the sample (n = 30) were as follows: 33.3% of the sample were in each of the following age cohorts: 15-17, 18-20, and 21-25; 83.6% of the sample was single, 26.7% were attending university, 53.3% were attending either primary or secondary school and 20.0% were not in school, split evenly between those who no longer attended after completing secondary school and those who never attended secondary (Table 1).
This study sought to examine the constructs of financial agency, gender, and HIV risk reduction among AGYW 15-24 years old, in Kalingalinga, Zambia. Four distinct themes emerged, relatively unilaterally across age bands, and are described below.

Theme #1young women's perceptions on income generation
Young women spoke of the importance of earning their own income, particularly beginning around the ages of 18-20. While women this age may still receive some support from their parents, respondents wanted to earn their own money to ease financial pressures on their parents and purchase personal items without parental oversight. Other respondents considered economic activity necessary to reduce dependence on a male partner, either for their perceived lack of reliability or the position of vulnerability. A 23-year-old University of Zambia (UNZA) student offered, 'men are not really reliable'. A 16-year-old Grade 8 student noted, 'Yes, they [AGYW] need to earn their own money … You can't just wait upon your husband on an empty stomach from morning to evening for him to get off from work and come give you money.' Others were influenced by various social pressures to engage economically. A 23-year-old stated, 'This time in our country you can't just sit as a woman. You have to be doing something. We see it on TV.' Sources of income varied by age cohort. For AGYW under 18, reliance on parents/guardians was common and income generation viewed with suspicion. A 15-year-old Grade 8 student stated, 'If I start doing anything to earn my own money my mum will say it is bad for me because I am little.' Similarly, an 18-year-old Grade 9 student said, 'If she [mother] sees me with money, she will ask where I got the money from, so it is better I ask her for money.' Conversely, respondents from lower-income backgrounds began working from a young age.
Economic activity, particularly in the informal economy, was common for women ages 20-25. For women at university (26.7% of the sample), plaiting and blow-drying hair were typical, as was re-selling clothing or cosmetics. In the home villages of university students, young women bought goods at cost and re-sold for a mark-up.
Irrespective of education level, the concept of earning money was universally appealing and spoken about in relation to independence and power. When asked how a woman who makes her own money feels about herself, a 22-year-old UNZA student commented, 'She feels great because she doesn't need to depend on anybody.' Another student noted, 'I feel proud and it builds up one's confidence … You feel like okay, if I can do this thing, I can do even more.' A 16-year-old Grade 8 student commented, 'She feels happy because the money she is earning is genuine. It's not like she earned it through sleeping with a man. No, she is free because she works for that money.' Respondents of varying educational and socio-economic status reiterated that a woman of independent financial means would be 'proud' and 'confident', no longer facing an 'inferiority complex.' Many women, particularly those who were attending university, viewed financially independent females as role models and felt they could also be the same for younger women. A 23-year-old UNZA student commented, 'I think if girls are getting independent, you create a better future for the ones coming behind you.' Younger women also noted earned income as empowering. When speaking about a hypothetical woman who earns her own money, an 18-year-old Grade 9 student noted, 'She feels good because she works for her money and buys what she wants.' While all but two respondents indicated that earning their money was preferred, several noted that others in their community did not share the same aspirations. A 20-year-old Grade 12 student suggested that her friends would prefer to rely on boyfriends, saying that some of them were too 'shy' to find 'piecework'. It was widely acknowledged that boyfriends were another source of income for AGYW, in addition to or in lieu of parents. A 23-year-old reiterated that 'sugar daddies' were always readily available for girls that desired material items beyond their financial means, Similarly, respondents noted that not all parents encouraged their daughters to end transactional relationships, particularly when there were few alternatives to earn money or support the household. These findings articulated the notion that the aspirations of AGYW can hit stumbling blocks when faced with external dynamics of gender and power.
Theme #2female financial agency and social norms In and around Kalingalinga, the modern ideals of female economic activity intersected with more traditional norms of women being reliant on men, first in a parent/child relationship and then within marriage.
Respondents noted that while their generation considered this economic independence to be an admirable quality, women in their home communities who were not financially independent would have mixed feelings. Respondents shared that financial independence could be looked upon with jealousy and suspicion, questioning whether a woman with her own financial means was engaged in some form of transactional sex. A 19-year-old Grade 11 student commented, 'some people will think she is a good person while others will think she is a bad person … they will be looking at her as a selfish person'.
Respondents also indicated that some older people within the community held cultural and religious beliefs that women were to be subservient to their husbands that contrasted with their views on women's financial independence.
I have to always lower myself whether I am wrong or right. Even when I am right I have to apologize to him. If they see me doing that [earning my own money and deciding how to spend it], I will be called a 'bad girl'. 25year-old UNZA student (married) If a woman is independent and not married, they [the community] will think either she is of loose morals or she boasts because she has money-she thinks she can do (whatever she wants) and just live her life without a man. 23-year-old UNZA student In contrast with a university setting where financially independent women were viewed as admirable, lesser-educated women who earned their own money and were not reliant on men were insinuated to be sex workers. Over three-quarters of all respondents reiterated this common perception, reflecting the reality that sex work may be the most viable option for economic selfsufficiency as well as an absence of female leaders within these communities (making it difficult to imagine a self-reliant female as anything but a sex worker).
The opinions on how men view financially independent women were mixed. Nearly all respondents commented on how intimidated men would be by a woman who earned her own money and could care for herself. A 22-year-old UNZA student, speaking about her hometown of Kasempa, 'If you are earning your own money, people will actually approach you to say you should tone it down … and tell you that you will not get married [because] married men will be running away from you.' A 23-year-old UNZA student felt that men would find financially independent women 'rude'. 'They are unapproachable because they have everything … they will be like 'no, she won't be giving me the respect that I need if I was to approach her'. Indeed, participants noted that financially independent women were referred to as 'boss ladies' in university-educated contexts.
However, there were signs that norms were evolving, particularly in university contexts where several respondents mentioned that some men would find such industriousness attractive, suggesting that plenty of men love women who are independent and challenging.
Theme #3 -financial agency and decision-making power In the semi-structured interview, respondents were also asked about financial decision-making power within intimate partnerships. Among university students, the thought of women asking permission from a man to spend money was viewed as 'harsh' and 'degrading', not considered common practice and frowned upon in a relationship. Conversely, joint decision-making was considered healthy and promoted within the community. The concept of creating a jointly agreed household budget was widely accepted.
Seeking permission from a male partner to spend money was viewed as more normalised in lesser-educated settings, particularly for married women. While not the experience of most universitygoing women, several had stories of friends in their home communities. They equated such dependency on a man with a loss of personal agency.
I think that sometimes it [seeking permission from a husband on spending] is out of fear. For example, he has given me money [and said] 'Go and do this and this'. Then I have made my own budget and failed to meet his expectations. Then what does he do? He scolds me. Then he starts saying 'You are useless. You don't know how to use money.' I think that can make someone seek permission from a husband … out of fear. 25-year-old UNZA student (married) Even for women where permission-seeking was common, nearly all respondents commented that household finance should ideally be a mutual exercise between spouses. Joint budgeting exercises were routinely brought up as the proper course of action within marriage and a sign of mutual respect. Yet, long-held views on hierarchical power structures still persist. When speaking about a married woman who earned and spent her money freely, an 18-year-old secondary school student noted, 'People would say that this woman doesn't fear her husband. She does what she thinks is right in her head and her husband doesn't rebuke her.' University-educated students also spoke about these cultural norms.
My friends would say, 'No, as a lady, you're not supposed to be so self-confident about yourself.' 'You're supposed to submit when you have a partner.' 'It is wrong for a lady to be so self-confident that you feel like you are the boss.' But when I think about it, it is not right for someone to just abuse your rights. I will stand on my ground of being confident.' 23-year-old UNZA student Largely, unmarried women were exempt from seeking counsel on finances. They spoke of freedoms and agency not afforded to married women. Conversely, non-college-educated married women were perceived to be 'trapped', often with a lack of understanding on how to become economically active.
It [financial independence] is very important because if you are earning your own money, you will have a kind of dominion over your own life. If you are earning your own money, nobody will toss you around here and there when it comes to men … Many people stay in marriages because they have nowhere to go. [They think], 'If I have to leave today I will have to hustle out there and I don't even know where to start'. 23-year-old UNZA student Other respondents echoed this sentiment, stating that they knew women who allowed men to make all of the decisions in their relationship out of fear that the man would leave. Conversely, financial independence was considered to afford females the power to reject unwanted, violent, or abusive relationships.
It is easy for me to say 'no' even if my partner gets upset because I know even if he leaves, I will still be able to stand on my own. But if I know he is my main source of income, I know that if he gets upset, then things won't be too well with me, so what do I do? I will give in to whatever he wants. -23-year-old UNZA student Theme #4 -financial agency, sexual agency and HIV risk reduction In addition to increasing the decision-making power for AGYW in intimate partnerships (from managing one's own money to exiting undesirable relationships), financial agency was also perceived to offer AGYW greater power to avoid HIV risk behaviours from the outset.
Transactional sex was widely described as a known practice in local communities for AGYW to address material needs. Transactional sex fell along a spectrum from more formalised sex work to age-disparate relationships to informal transactions between men and women. Irrespective of the form taken, such relationships typically limited the power of AGYW to negotiate safer sex practices. In these relationships, men held the economic power and sex was the mechanism to access it. A 22year-old UNZA student noted, 'A person who pays for your alcohol decides what is to be done that night.' When speaking about how women earned money in Kalingalinga, a 23-year-old woman who left school in Grade 9 noted, 'some earn on their own through plaiting hair. Some have boyfriends'. Similarly, a 15-year-old Grade 9 student commented, 'Some people, if they don't work and they want to earn money, they will start patronising bars to look for men to have sex with them in exchange for money.' Indeed, when participants were asked their opinions on women's financial independence and how it influenced HIV risk, the most prominent theme was a reduction in transactional sexual relationships.
… when you go to a man and he gives you money, that is when he will use you … When you use his money, you would feel guilt and fail to refuse when he asks for sex. If he wants to sleep with me and he gives me money, how do I refuse? He may be buying you gifts like expensive phones so where do you refuse when he asks for sex? But when you have money, you have the power to say 'no'. 20-year-old secondary school student A 19-year-old UNZA student echoed this point, 'Women sleep with different men to obtain money but if you are independent, you wouldn't engage … you would have a better way of earning money.' Respondents were quick to note that independent financial resources featured heavily in sexual decision-making. If financial resources were sparse, AGYW were more willing to engage in transactional sexual relationships to meet their needs.
In addition to a reduction in transactional sex, financial agency was also tied to other HIV risk reduction behaviours. It was widely acknowledged that women were viewed as the most conscious of and principally responsible for safer sex practices. Whether they exercised that perceived responsibility was dependent upon what they stood to lose from requesting compliance. For some, HIV testing was a pre-requisite to intimacy. An 18-year-old secondary school student noted, 'If he refuses, then the relationship ends. She will look for someone who will accept to do VCT [voluntary counselling and testing]. That means the girl has power.' For others, requests for testing or condom use put their relationship in jeopardy, which was feared by women who did not want to lose their partner, for economic or social reasons. A 23-year-old who left school in Grade 9 also noted that few men agree to wear condoms or get tested for HIV and that women have to insist on testing prior to entering into an intimate relationship. Yet, that bargaining was heavily influenced by the female partner's sense of power within the relationship, something that was viewed to increase with the individual financial agency. Notably, testing was identified more frequently as an HIV risk reduction practice for which females could bargain whereas condom use was less often suggested.
Despite the near-universal recognition that financial agency would lead to greater autonomy and safer sex practices in intimate relationships, there was also the perception that other factors affected HIV risk for AGYW. One of these factors was the blinding effect of love, particularly for younger women. Some noted how AGYW, particularly those under 20, may be more easily influenced by their partners. A 23-year-old UNZA student noted, 'I mean, it's young love, so you just have to feel like you should do whatever he says. He should tell you what to do and all that.' Others echoed that poor choice could be made when decision-making is influenced by love, even by financially independent women.
Others noted the structural inequities that affected HIV risk for AGYW. For instance, male sexual privilege was a key concept that presented in relationships, and in many cases, irrespective of a female's financial independence. Several respondents noted that men felt they had the 'right to sex', a sense of entitlement that was exacerbated in marital relationships. Respondents across the socio-economic and education spectrum noted that men in the community 'needed multiple partners' and that male sexual privilege also perpetuated experiences of forced sex, putting women at unique HIV risk.
Sexual power dynamics were described as having greater balance and less hierarchy in non-marital relationships. Yet when asked about sexual dynamics in marital relationships, most respondents noted that men were 'head of household'. An 18-year-old secondary school student noted, 'They say when you get married, the man is the one with the power in the house.' A 25-year-old mother also noted that unmarried women who earned their own money had power in their sexual relationships. However, in a married relationship, that same woman would no longer have control. It was also noted that physical violence was generally not tolerated in non-marital relationships, but that within marriage, it was more common. Collectively, these comments underscore that a woman's rights within a marital relationship appear to diminish after marriage.
One married woman was verbal about her frustrations with this inequity and shared her own mechanisms for secretly setting aside portions of her earned income. A 23-year-old married woman noted, 'I have my own airtel account where I put my money … He can't be auditing me. That is my business. If you show him all the money, he will start telling you what to do with it. So, I put some aside.' Such practices show an internal resistance to power dynamics within marriage, particularly for women who begin to earn their own money. However, it was not clear how such resistance transferred to her sexual relationship and her ability to press for risk reducing behaviours.

Discussion
This paper explored the construct of financial agency and its influence on power dynamics within sexual relationships, including those dynamics that affect HIV sexual risk. The findings offer numerous qualitative insights on the value and limitations of financial agency in relationship power and sexual decision-making for AGYW in urban Zambia. These findings also help to contextualize the limited number of studies that have aimed to quantitatively examine these relationships (Jennings et al., 2017).
The findings illustrated how AGYW associates financial agency with influence and confidence for women yet also highlighted how that power is viewed locally, namely in a way that is prejudiced by social norms that privilege men within society. Women earning and controlling their own money were viewed as 'boss ladies' or as trading sex for material gain. These viewpoints are long-held gender norms derived from Christian ideologies and an economic structure that enables men as the primary earner, keeping women reliant on their spouses for economic support (Evans, 2015). Deviating from those norms would call into question a female's 'respectability' (Evans, 2015), a concept that was echoed in this study. Yet, interviews also captured an evolution of social norms, particularly among younger and more educated residents. Recent qualitative research in Zambia has indicated some flexibility with respect to gendered divisions of labour, potentially instigated by a lagging economy, but nonetheless, one that has enabled women to participate more fully in the labour market and be seen as equally capable of leaderships roles (Evans, 2016;Foster et al., 2012). Thus, as more women enter the labour market, it normalizes and perpetuates a cycle of empowerment.
Whether and how financial agency influences HIV risk reduction was another key question of this analysis. Respondents overwhelmingly considered sexual decision-making power (including safer sex practices) an outcome of increased financial agency, however, male sexual privilege was still considered a defining feature within society that placed limitations on that power. Indeed, when considering theoretical frameworks such as Asset Theory (Sherraden, 1991) or the Capability Approach (Sen, 1993) in the context of gendered power dynamics, the relationship between money and self-efficacy may be less linear and fraught with structural roadblocks for females. Indeed, some evidence has shown programs for AGYW that situate HIV prevention within larger constructs of gender and power may be more effective at reducing HIV risk behaviour (Haberland, 2015).
These structural barriers illuminated the complexities women face regarding sexual agency. In many cases, the threat of forced sex was considered a reality and highly reflective of perceived male dominance in sexual relationships. At the same time, women were perceived to have a fundamental power in consensual relationships, specifically when it came to health-seeking behaviours, but those decisions were often heavily influenced by their economic needs and other considerations. Whereas women with fewer financial means were perceived to have less freedom of choice in partner selection and lower bargaining power over HIV risk reduction behaviours, women with financial independence, particularly single women, had the power to negotiate safer sex practices. Such women felt free to dismiss potential sexual partners if they denied HIV testing, supporting evidence that testing is consistently higher among women than men on the continent (Stephenson et al., 2013). However, it should be noted that this type of agency was inferred to be most powerful at the initiation of a relationship. Previous literature has noted that female agency in a relationship can diminish after a woman has consented to sex (Jewkes & Morrell, 2012;Pulerwitz et al., 2018).
Power to negotiate condom use was not as explicit, even at partner selection, as some respondents preferred sex without condoms themselves. Therefore, condom use remained more challenging and its connection to female agency more tenuous as respondents were more likely to describe female sexual agency as reflected in mandatory testing rather than the use of condoms.
Findings indicated that AGYW considered a decrease in transactional sex to be the primary mechanism of change that tied financial agency to sexual HIV risk reduction, aligning with previous studies connecting broader concepts of economic strengthening to reductions in transactional sex (Gibbs et al., 2017;Gichane et al., 2020;Jewkes & Morrell, 2012), a known risk for HIV transmission (Dunkle et al., 2004;Wamoyi et al., 2016). The explicit linkage tying financial agency to a reduction in transactional sex is the direct connection between AGYW and financial resources, enabling them to bypass relationships based on 'exchange', which tend to weaken the sexual bargaining power of females (Ranganathan et al., 2017). This suggests the power of economic interventions in HIV reduction programmes in the region, particularly interventions designed to connect AGYW with sustainable sources of income, entry into formal financial institutions, and financial management training.
What continues to warrant discussion is how formalised sex work fits this model. For sex workers that obtain an income and have control over that income, they can be considered financially autonomous. Yet their power to negotiate safer sex practices may vary and may be dependent upon client demographics such as age and wealth quintile (Ahinkorah et al., 2020). Further, it is not uncommon for sex workers to be subject to physical violence in Zambia (Malama et al., 2019), dismissing any ability to negotiate the terms of engagement. Additional research is needed to understand how financial autonomy intersects with HIV risk reduction for women engaged in more formalised sex work, particularly in informal and unregulated environments.
Another prominent finding was how power and agency for unmarried women were considered to diminish once married. Religious and cultural norms on male headship within marriage heavily influenced financial agency and sexual decision-making. It was striking how much freedom and power were felt by single, economically active women, yet how they recognised that freedom would lessen once married. For many, the concept of male headship was not debated or critiqued, simply accepted as fact, an interesting juxtaposition of beliefs that embrace both a modern lifestyle with the remnants of a gender-based hierarchy of sexual privilege that has yet to be fully challenged. This finding echoes research in South Africa that shows young women can both challenge patriarchal norms in some spheres of influence while enabling them in others (Jewkes & Morrell, 2012).
The findings on joint financial decision-making within marriage were an interesting contrast to the stated lack of sexual autonomy for women. Unlike sexual decision-making, seen largely as a male role within marriage, budgeting and household finance were viewed as a shared effort. The limited literature on financial decision-making and HIV risk reduction for married women is somewhat mixed (Singh et al., 2013;Vamos et al., 2013). This literature, in conjunction with our study, suggests norms and roles within marriage may be compartmentalised such that women who are afforded power in one aspect of the marriage are not afforded that power in another. Nevertheless, given that joint budgeting is viewed as appropriate within a marriage in urban Zambia, practitioners may have the leverage to promote joint decision-making in sexual and reproductive health choices in a synonymous fashion.

Limitations
This study includes several limitations. First, while purposive sampling is often a key feature in qualitative in-depth interviews, it can only capture the opinions of a relatively small number of respondents. While sampling was purposive and informed by a maximum variation approach, sampling bias does exist to favour respondents that are willing and available to participate in the study. Further, given that the study was limited to former DREAMS participants, the findings are biased towards a sub-group of AGYW who have received some level of HIV prevention information. Thus, findings should not necessarily be generalised to a larger population. Further, the most substantive quotes came from university women, while younger adolescents provided fewer lengthy quotes to feature within the paper. Nonetheless, the comments and viewpoints of younger adolescents are reflected in the summary text. It should also be noted AGYW were intentionally asked about their perceptions of financial agency and HIV risk within their community (as opposed to within their own personal experience). This decision was made in an attempt to solicit more candid feedback on taboo topics but also requires the resulting data to be viewed as social representations of financial autonomy and HIV risk reduction. Even with this method of indirect enquiry, the data is still subject to social desirability bias. Respondents may have felt an internalised pressure to answer questions in a way that was perceived to be 'correct' among DREAMS participants. This study tried to alleviate potential bias by utilising a female facilitator with several years of experience leading group and individual interviews with AGYW in the area. The facilitator's guide and prompts reiterated the invitation to speak freely and assured confidentiality.

Conclusion
Financial independence is a widely held aspiration of AGYW in Kalingalinga. Such agency is, however, tied up with negative community-based perceptions about what it means to be a woman with earned income and control over how it is spent. Further, the transfer of financial independence to sexual agency within relationships has enormous promise as a mechanism for sexual HIV risk reduction; however, the realities of male sexual privilege remain an obstacle irrespective of financial decision-making power. Women's sexual agency was viewed as far greater in non-marital relationships as opposed to within marriage. Programmes that seek to enhance financial agency among AGYW have the potential to reduce HIV sexual risk. Yet, to be most effective and to address norms of male dominance, which can continue to keep AGYW in positions of vulnerability regardless of their financial power, such programmes should consider integrating broader, gender-transformative programming that addresses drivers of gender inequity in sexual relationships, including within marriage.

Disclosure statement
No potential conflict of interest was reported by the author(s).