In a world hemmed by patriarchy and poverty, cash transfers could be the missing link in SA's HIV prevention programmes.
I recently visited an informal settlement in the heart of Nairobi and met a group of adolescent girls. They spoke to me candidly about their lives in the context of high rates of HIV, teenage pregnancy and gender-based violence.
As I was talking about violence, they stopped me.
Yes, there was violence, but it wasn't as straightforward as this, they explained.
Some of their sexual relationships were consensual, they said, or at least as consensual as they could be, based on the limited choices they had in their lives.
One young woman, Felistas*, explained: "My mother needs to pay rent at the end of this month. Who is going to help us pay this rent?"
In Eastern and Southern Africa, we have reduced new HIV infections by 30% since 2010 - the highest decrease charted by any region in the world, 2017 UNAids figures show. But 800 000 people contracted the virus in the same year - a third of whom call South Africa home.
Now imagine the unimaginable: nine women between the ages of 15 and 24 are infected every hour in South Africa, our figures reveal. This translates to about 1 500 new HIV infections a week in this group.
It is easy to see why we at UNAids are concerned.
Take a look at the latest HIV data from the region
Adolescent girls and young women are still being left behind by the Aids response. We urgently require scaled-up and targeted programmes that o er women a combination of options to help them to stay HIV negative. These projects have to recognise that young women are not all the same and that they have different HIV prevention needs depending on factors such as their age, location and economic status, as well as the levels of gender inequality they experience.
Clearly, putting condoms into the hands of women is not the only answer.
At least, not the sum total of it.
Empowering adolescent girls and young women starts with education and continues with job opportunities and exploring innovative methods to get money into their hands in ways that do not involve risky transactional relationships like the ones Felistas described to me that day in Nairobi.
For years, researchers have investigated how cash transfers could form part of a combined approach to preventing new HIV infections in young women.
A cash transfer works much the same way as a social grant; a modest sum of money paid to a girl or her parents - from as little as $1 to $10 a month over a specified period. Getting the money can be predicated on meeting certain conditions, such as attending school for 80% of the time.
The rationale for conditional cash transfers is based on numerous studies that have shown that women who complete more years of school are less likely to contract HIV.
In Botswana, every additional year of schooling was associated with an almost 12% reduction in the risk of HIV infection among young women, a 2012 study reported in The Lancet Global Health journal. Researchers think that this, in part, is because succeeding in education builds confidence that women can use to help them to negotiate safer sex, 2005 research published in the journal of Tropical Medicine & International Health suggested.
In Malawi, cash transfers helped young women to stay in school, reducing early marriage by 40% and teenage pregnancy by 30%, a 2012 study published in The Lancet found. There was also a 64% reduction in the risk of HIV infection among study participants.
A 2016 study in rural Mpumalanga also found that school attendance was associated with a lower risk of HIV infection, research published in The Lancet shows. But with 95% of girls already attending school before they began to receive cash transfers as part of the study, the monthly cash sums had no effect on school attendance.
Like what you're reading? Sign up for our free weekly newsletter or donate to support great journalism.
The R300 a month did, however, make women less likely to have experienced violence at the hands of a partner - itself a risk factor for HIV transmission, 2010 research also published in The Lancet reveals.
Women in the Mpumalanga study who received the money were also less likely to have had sex in the past 12 months or unprotected sex in the past three months.
As part of the study, women received R100 directly while the remaining money went to their families.
So what did they spend this money on? Toiletries, clothing and school supplies.
But young women also bought things such as mobile phones, ringtones, hair appointments, sweets, make-up and even birth control; things that typical teenagers across the world would buy, a follow-up 2018 study published in the BMC Public Health journal shows.
Research published in the same year in the journal Aids and Behaviour found that the young women were also able to save, stay free of debt and improve their food security - and this was truest of those from the poorest families.
The evidence on cash transfers" direct effect on reducing new HIV infections is still mixed. But we know that young women's ability to take ownership of financial resources could empower them to stay in school, thereby avoiding HIV infection.
Cash transfers could also be a missing link in current HIV prevention initiatives.
But we cannot talk about empowering young women without working with young men and addressing harmful masculinities and risky behaviour. We need to have conversations with our boys and men that challenge harmful gender norms and teach boys to become young men who respect, value and love women in all their diversity.
Let's provide men with safe spaces to have conversations with each other and with their partners, families and communities. This, combined with innovations such as cash transfers, would make a real difference.
*Not her real name
Catherine Sozi is the director for the UNAids regional support team for Eastern and Southern Africa. Follow her on Twitter @CNSozi.