We recently spoke with someone who wanted to donate to a GiveWell top charity. They were interested in getting the funding “out the door” and to program participants as quickly as possible.
But our top choice for funding today is Malaria Consortium’s seasonal malaria chemoprevention program—for work it expects to complete in 2022. The potential donor was puzzled. Shouldn’t we prioritize an organization that needs the money sooner?
We often recommend donations today that support programs a few years from now. This probably diverges from many people’s intuitions about getting funding out the door as soon as possible.
Three key reasons why funding today leads to more impact in the future
1. Upfront coordination and planning increases charities’ impact.
Highly effective charities tend to spend a lot of time preparing before they implement their programs.
The Against Malaria Foundation (AMF) is a GiveWell top charity that supports the distribution of insecticide-treated nets, primarily in sub-Saharan Africa. These nets are hung over sleeping spaces and prevent mosquitoes from biting and transmitting potentially deadly malaria. AMF says the ideal lead time for its work is 23 months. During that time, it takes the following steps:
- Choosing a location: AMF decides where it should direct funding to have the greatest impact. It considers malaria prevalence, the number of people in need of nets, and whether partner organizations can conduct distributions according to AMF requirements.
- Negotiating an agreement: AMF and the country’s national malaria program negotiate a net distribution agreement and seek government approval.
- Ordering nets: AMF negotiates with net manufacturers to place an order for nets suited to local needs, which may vary in size, color, and insecticide.
- Producing, shipping, and transporting nets: Nets are manufactured and shipped to regional warehouses.
- Visiting households: AMF’s partners visit households to determine how many nets are needed.
Some of these steps are practical requirements for conducting a distribution, such as ordering and shipping nets to the relevant location. Other steps increase the impact of the distribution. Though 23 months may seem like a long lead time, it enables AMF to coordinate with in-country groups, identify the highest-need areas, order the quantities and types of nets that are most suitable for those areas, and select partners that can conduct high-quality distributions and monitoring. We believe that the overall impact of donations is much greater than they would be if they were allocated more quickly but less thoughtfully.
2. Charities with longer-term funding are more likely to hire and retain staff.
Charities have told us that it’s hard to retain and hire staff when they are uncertain about their long-term funding. Staff may leave if they don’t have confidence in an organization’s future funding. Their replacements must be recruited and trained, requiring significant time from existing staff.
A charity leader who is concerned their program might not be funded in the future may decide to limit investment in hiring. They may make quick hiring decisions that lead to poor fit and reduced program impact, or further staff turnover. Or, they may simply not hire at all. Instead, existing staff may be spread thin to cover the vacant roles.
A charity leader with long-term confidence in their organization’s funding may be willing to invest in hiring. We expect charities with steady, excellent staff to have more impact than those facing high turnover and suboptimal allocation of staff.
3. Two to three years from now, highly cost-effective charities will likely still have more impact than faster-spending but less cost-effective charities.
Charities’ impact per dollar donated drives our recommendations to donors. We expect differences in cost-effectiveness to overwhelm differences in charities’ spending speed over a few years.
We estimate that Malaria Consortium’s seasonal malaria chemoprevention (SMC) program is 16 times as cost-effective as GiveDirectly’s standard (non-COVID-19) cash transfer program. Malaria Consortium distributes cheap medicine to prevent children under five from getting malaria. GiveDirectly provides cash transfers to very poor households. Both work in sub-Saharan Africa. GiveDirectly estimates that participants in its standard cash transfer program receive funds between three and six months after those funds are donated. Donations to Malaria Consortium reach program participants in two to three years.
It’s likely that two to three years from now, the costs and benefits of Malaria Consortium’s SMC program and GiveDirectly’s cash transfer program will be about the same as they are today. In 2022, we expect that Malaria Consortium’s SMC program will be 16 times as cost-effective as GiveDirectly’s cash transfer program, as we do today. Even though donations reach Malaria Consortium’s program participants more slowly, they are likely to have more impact.
There is a way to change this calculus. If you believe that reaching program participants sooner is more valuable, you could apply a “discount rate” to the cost-effectiveness of SMC. Cost-effectiveness analyses often include discount rates to reduce the value of future costs and benefits because they occur in the future. We have used a discount rate of 4% for programs that lead to increases in future income for participants.
In order for GiveDirectly to be more cost-effective than Malaria Consortium’s SMC program, you would need to apply a discount rate of more than 300% per year to Malaria Consortium. We think that most donors would not apply such a steep discount. We believe they would choose to give to Malaria Consortium on the basis of cost-effectiveness, even if donations are deployed more slowly.
What we’re balancing
We balance the benefits of providing future funding with the potential downsides of doing so by limiting how many years in advance we fund charities’ work. This preserves our ability to donate funds to more effective opportunities that we may identify in the future. We direct funds to charities that will be spent in two to three years, rather than in ten years, for this reason.
We constantly update our charity recommendations with new information. We also look for charities with greater impact per dollar. This work will likely change how we prioritize funding our existing top charities and may lead to new top charities. We don’t want to commit funding so far in advance that we would have preferred to give that funding to an organization we’ve recently concluded has greater impact per dollar.
Funding a few years out is a reasonable balance of the benefits and pitfalls of providing future funding. The optimal timeline for funding will vary depending on the program and implementation model.
Responding to urgent needs
While we generally direct funding to our top charities for future work, we retain a pool of funding that may be used for urgent needs: “Grants to recommended charities at GiveWell’s discretion.” We typically grant these funds quarterly to the charity or charities we believe can use them most effectively.
We might have the most impact by supporting a pressing need. Last year, we made a grant to AMF to support a time-sensitive need in the Democratic Republic of the Congo. We estimated that the cost per life saved by donating to AMF was the same as the cost per life saved by donating to the other charity we were considering funding, Malaria Consortium’s SMC program. However, by providing additional funding to AMF, we were directly increasing the number of nets it could distribute in the Democratic Republic of the Congo the following year. By contrast, our near-term funding decisions would not have impacted Malaria Consortium’s plans.
We are open to granting or informing charities of our plans to grant discretionary funds sooner than quarterly, if that would make a difference for a charity’s ability to carry out high-impact work. But many times, we choose to grant discretionary funds to support charities’ work two or three years from now—because that is the most impactful option.
Sources and footnotes for this post may be found here.