More on Cash Transfers to Reduce HIV among Adolescents

December 04, 2015

My recent blog on cash transfers as a tool for HIV prevention among adolescent girls and young women left out results from a number of recent evaluations that illustrate the importance of program design and, in particular, targeting the transfers to the poorest households in getting results in wellbeing.

Tia Palermo, a social policy specialist with the Transfer Project at UNICEF Office of Research-Innocenti and UNC-Chapel Hill, wrote with an update, which I’m pleased to share with her permission:

Widely demonstrated to reduce poverty and improve food security, school enrolment, selected health outcomes, resilience, and productive activities, the emerging consensus has been that cash transfers may also have the potential to facilitate safe transitions to adulthood, including delaying sexual debut, pregnancy and marriage, and reducing HIV risk among adolescents. Furthermore, concerns about unintended consequences, such as that cash transfers incentivize increases in fertility, people will spend the cash on alcohol and tobacco, and recipients will stop working have been largely debunked.

Four recent studies emerging from India and South Africa have challenged the prevailing narrative about the benefits of cash transfers for adolescents and raised concerns about unintended consequences. One study suggests that a conditional cash transfer programme in India targeted to households, conditional on certain behaviours among adolescent girls increased the likelihood of marriage as soon as the girls turned 18. Another study, also from India, found that a cash transfer programme conditional on childbirth delivery at a health facility was successful at its stated aim but inadvertently increased the probability of pregnancy or childbirth over a three-year period among women in target areas. A third programme in South Africa (HPTN 068), targeted to adolescent girls and their caregivers conditional on girls’ school attendance, aimed to reduce HIV infections, and a study evaluating this programme found no impacts on either school attendance or HIV rates. Finally, another study, CAPRISA 007, provided life skills training and cash incentives with the aim of reducing HIV and HSV-2 infection among rural high school students in South Africa was found to reduce HSV-2 incidence but had no impact on HIV incidence.

So what does all this mean for cash transfer programmes?

To understand the implications of these findings, we should consider three key points: 1) programme characteristics such as objectives, delivery and targeting matter, 2) context matters, and 3) cash transfers are not a silver bullet. Once these points are taken into consideration, the potential for cash transfers to facilitate safe transitions to adulthood remains clear.

Programme objectives, delivery, and targeting matter

Programme objectives, delivery, and targeting are important, particularly if we are to compare apples to apples. Programmes in India were specifically designed to motivate certain behaviours (e.g., institutional delivery and delayed marriage). They were targeted/narrow programmes consisting of one-time payments for complying with desired behaviours. Conversely, government-run cash transfer programmes in Africa are designed with the broader aims of reducing poverty and investing in human capital. To achieve these aims, African government programmes are targeted to the poorest and most vulnerable households, and consist of regular-interval payments which continue over time. This regularity of payments allows households to become more forward-thinking and invest in children’s schooling and productive activities. One-time payments which reward specific behaviours are unlikely to have the same impacts. Regarding concerns about fertility incentives, a programme which aims to increase institutional delivery (and thus women can only qualify when pregnant) is likely to have different impacts on fertility than programs which aim to improve food security and investments in human capital more broadly. In fact, rigorous research from Latin America and Africa demonstrates no, or very few, impacts on fertility of conditional and unconditional cash transfers to date.

Targeting, or decisions about who receives the programme benefits, also matters. The South African programmes (HPTN 068 and CAPRISA 007) did not explicitly target the poorest households (indeed, girls in HPTN 068 had to be in school and know how to use computers to qualify and youth in CAPRISA 007 were sampled from secondary schools), but large-scale government-run cash transfers almost always do (it should be noted that there is a large-scale, successful government-run programme in South Africa). Since girls in both treatment and control arms of the HPTN 068 study attended school at high rates throughout the duration of the study, and since CAPRISA 007 sampled from in-school secondary students, schooling as a protective pathway was unlikely in these contexts. Conversely, evidence from government programmes across Africa have shown that cash transfers have large, positive impacts on secondary school attendance. So if school turns out to be protective against sexual debut, risky behaviours, and marriage, then we can still expect other cash transfer programmes (either those with different targeting mechanisms, aims, or in other contexts) to impact these outcomes through the schooling pathway. HPTN 068 did however find that the programme reduced the frequency of sex and risk of unprotected sex and intimate partner violence, and CAPRISA 007 reduced HSV-2 incidence, suggesting that there were positive impacts of these programmes. Further, a study examining the government’s Child Support Grant in South Africa found that cash transfers reduced adolescent pregnancy and sexual activity .

Context matters

Across countries and regions, characteristics such as social norms (e.g., gender preference in schooling, norms around marriage age, etc.), levels of poverty, and quality of government institutions may differ, and all of these are likely to affect how cash can influence outcomes. For example, before we discount the ability of cash transfers to delay marriage, we should consider contexts of marriage markets. In India, it is customary for brides’ families to pay dowries to the grooms’ family. Conversely, in Africa, brideprice is more common, whereby the bride’s family receives a large payment or goods at the time of the marriage. So although a programme which financially rewarded families at a girls’ 18th birthday for not having married her off earlier induced immediate marriage (likely because families found themselves able to pay a dowry), this does not mean that smaller, predictable cash transfers over a sustained period cannot delay marriage. Indeed, the latter type of programme addresses structural poverty and may shift underlying behaviours. Thus, these programmes may delay marriage through pathways such as reduced school dropout and reduced economic insecurity, a driver of early marriage.

Not a silver bullet

Finally, it is important to remember that cash transfers are not a silver bullet. While they do allow households to meet basic food needs, become more forward-looking, and invest in human capital by demanding more services (education and health), programmes can only achieve these aims if adequate services exist in the first place. Thus, cash transfer programmes need to be implemented in conjunction with improvements to supply-side factors like availability and quality of schools and health facilities, as well as improvements in markets and economic opportunities.

The good news and looking forward

Cash transfers won’t solve all problems, but there is still good reason to believe that they can facilitate safe transitions to adulthood. New evidence from the Transfer Project demonstrates the ability of government cash transfer programmes in Africa to delaysexual debut,pregnancy, and improve adolescents’ mental health. Other evidence in the region also suggests that cash transfers may also reduce adolescent girls’ engagement intransactional sex, reduceHIV infection, and delay marriage. Researchers from the Transfer Project are working with governments and local partners to evaluate impacts on the safe transition to adulthood infive African countries. Data are still being analysed, and a lot ofnew evidence is still being generated on the ability of programmes to reduce risk of violence and risky behaviours, delay pregnancy and marriage, and examine impacts on schooling and work. So even though a lot of questions are still unanswered, there is good reason to believe that these programs may facilitate safe transitions to adulthood.

The Transfer Project works to provide rigorous evidence on programme impacts in an effort to inform future programme design and scale-up. Find more information on the Transfer Project’s research on cash transfers here or on Twitter (@TransferProjct).


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.