As you may have heard on a recent episode of our podcast, we’re launching an experiment: focusing more attention on programs that increase the economic well-being of people in extreme poverty. We are in the process of completing a search for a new program officer to lead an expanded livelihoods research team, and we plan to allocate up to $10 million for granting to cost-effective programs we find in the first year. Depending on the outcome of these efforts, we may hire additional researchers to focus on livelihoods.

GiveWell has historically directed most of its funding toward health interventions that avert death and disease, but those are not the only positive outcomes our grants target. We have long grappled with questions about how to value different positive impacts relative to each other. In particular, how much more valuable it is to save a life than to substantially increase someone’s economic well-being? Our expanded research into programs that improve lives will help us better reflect the diversity of relevant perspectives on that question in our grantmaking.

Why livelihoods and why now?

Our standard moral weights—that is, the values we assign to different outcomes—assume that saving a life is about 100 times more valuable (depending on age) than doubling a person’s income for a year (see our recent blog post on moral weights to learn more). This assumption has meant income-focused programs have been less likely than health-focused programs to meet our cost-effectiveness threshold.

But our moral weights are a necessary tool, not an absolute truth. Some GiveWell donors and staff, as well as some of the people affected by the programs we fund, place a higher value on income-increasing programs. To account for this, we’ll be funding livelihoods programs that would appear as cost-effective as our standard recommendations to a donor who values income gains twice as much as our standard moral weights.¹In other words, our cost-effectiveness threshold for livelihoods programs would assume that saving a life is about 50 times more valuable than doubling a person’s income for a year.

We have investigated and funded livelihoods programs throughout our history, and this new funding threshold and program officer role will allow us to build on that work. As we have more than doubled our research team in the past three years, we are now positioned to explore more of the impactful opportunities that we—and our donors—have long cared about.

Generating evidence

Over the past several years, we have funded a number of studies to generate evidence about the cost-effectiveness of livelihoods programs. Our new program officer will now take advantage of what we learn and make data-driven decisions about how, and if, to scale our support of similar programs. Below are a few examples of research we’ve supported.

Unconditional cash transfers

Unconditional cash transfers have been on our radar for many years; we first looked into GiveDirectly in 2011, when it was a new organization. Its flagship Cash for Poverty Relief program sends one-time unconditional cash transfers of approximately $1,000 via mobile money platforms to households living in poor regions of Kenya, Malawi, Mozambique, Rwanda, and Uganda.²There are no strings attached to the cash transfers, and only very basic eligibility requirements.

For several years, we’ve been funding research exploring the impact of this program design (see, for example, grants we recommended in 2019, 2020, and 2021). We also funded researchers at the University of California, Berkeley in 2022 and 2024 to support follow-up data collection seven or eight years after cash transfers were distributed as part of a randomized controlled trial (RCT) of GiveDirectly’s unconditional cash transfer program in Kenya.

Late last year, drawing on recent research and preliminary results from the RCT we funded, we re-evaluated GiveDirectly’s flagship program and now estimate it to be 3x to 4x more cost-effective than we previously estimated. We have remaining uncertainties about some of the parameters in our analysis, particularly spillover effects—that is, how the cash transfers affect nearby people who did not receive them—and we’re working to get answers. In December 2024, we recommended a $491,700 grant to the University of Oxford to fund data collection for a large-scale RCT of GiveDirectly’s cash transfer program in rural Malawi. This RCT will assess the impacts of the program on the local economy as a whole, which will allow us to learn more about spillovers from cash transfer programs.

Layered and targeted programs

We think that combining programs or targeting them to specific groups can sometimes increase their cost-effectiveness. So, as we look for cost-effective opportunities to fund livelihoods programs, we’re looking at ways that unconditional cash transfers could be distributed alongside other programs or targeted in a way that increases their impact.

We recently made a grant of $89,347 to GiveDirectly to scope whether variations of its program could be more cost-effective than its standard Cash for Poverty Relief program model. With this grant, GiveDirectly will explore program variations that could increase cost-effectiveness through better targeting (such as leveraging remote poverty mapping tools to identify the extremely poor areas where people still have access to local markets where they could buy or sell goods), layering additional interventions (like financial coaching), and optimizing transfer design (for instance, aligning timing of transfers with agricultural cycles).

The goal of the grant is to scope potential iterations of GiveDirectly’s program, with a view to piloting promising options later in 2025. This is representative of our experimental approach to expanding livelihoods work: providing a low cost, structured way to learn whether more substantial investment in funding and staff time is warranted.

Resources to support economic well-being

Unconditional cash transfers are just one way of improving economic well-being. We’ve also been investigating whether other sorts of livelihoods programs might meet our cost-effectiveness threshold. We’ve been looking specifically at programs that provide various resources or commodities or that increase people’s employment opportunities.

Reading glasses. We’ve recently funded an RCT to study the economic benefits of distributing reading glasses among working-age adults with close vision loss, and we’re considering other grant opportunities to support their broader distribution. We think that being able to see better is likely to improve people’s ability to work. However, the current evidence doesn’t answer some of our key questions about the long-term effects on household economic well-being,³In particular, we are interested in household consumption, which is measured by, for example, asking people what they purchased over the past week or two (setting aside large, infrequent purchases like a home or a vehicle). the lifespan of eyeglasses, and the effects of mass distribution programs. The RCT, which we co-funded with Livelihood Impact Fund, will be carried out by researchers at the London School of Hygiene & Tropical Medicine in India and Kenya. In addition to providing more evidence, the study will provide reading glasses to 20,000 adults with age-related vision loss.

Agricultural programs. In 2022 and 2023, we funded One Acre Fund to plan and then implement an RCT of a program that distributes tree seedlings to farmers in Rwanda and provides agricultural training. There is evidence from an RCT in Kenya that One Acre Fund’s program increases the number of trees planted after one year, and that these trees, which can be sold for timber, may generate additional income for poor households. We think the additional research we funded may address some of our uncertainties, such as the causal effect of One Acre Fund’s program on the number of seedlings planted and surviving past the first dry season and the long-term economic benefits for households. We received baseline results in late 2024 and expect results from the endline survey in the coming months.

Footbridges. People in remote communities that are isolated by an impassible river can have a harder time accessing employment opportunities and places to buy or sell goods and services, so in 2022 we funded an RCT of footbridges constructed by Bridges to Prosperity in Rwanda to see if the increased access they provide increases people’s income. We think it’s likely that the bridges will generate increased opportunities for employment and higher wages. In addition, we think the bridges may provide more opportunities for farmers to sell their crops, thus increasing their profits. However, we are uncertain about the magnitude of the income increases and how long they last. We received preliminary results earlier this year and are in the process of incorporating them into our cost-effectiveness analysis for the program.

Regardless of the results, each of these studies will provide valuable data to better inform future funding decisions for livelihoods programs.

Learning from our experiment

This increased focus on livelihoods opportunities with a dedicated team is an experiment, and we’ll adapt based on what we learn. In this way, our work on livelihoods programs parallels our approach more generally: to engage in rigorous research and follow the evidence, always focusing on finding cost-effective programs to save and improve lives the most that we can.

In 2026, we anticipate having clearer evidence about the effectiveness of our initial investments and a better sense of the potential impact of maintaining this increased focus on livelihoods programs. If the evidence is strong and we find programs that are similarly cost-effective to the rest of our grantmaking under this alternative moral weights framework, we may eventually expand the subteam. If not, we’ll reassess. We look forward to sharing updates on what we learn—and the new questions we’re asking—as this work moves forward.