Subscribe
Share
Share
Embed
Welcome to GiveWell’s podcast sharing the latest updates on our work. Tune in for conversations with GiveWell staff members discussing current priorities of our Research team and recent developments in the global health landscape.
Elie Hassenfeld: [00:00:00] Hey everyone, this is Elie Hassenfeld, GiveWell's co-founder and CEO. Last year we directed more than $400 million to programs that we think are highly cost-effective, and most of those grants are the kinds of programs you'd expect us to support. We identify a program, we evaluate its impact and estimate its cost-effectiveness. And then if all that holds up under scrutiny, we fund it.
There's another category of grants that's also important, and those are value of information grants where we are trying to estimate the cost-effectiveness of improving the information that we have. The idea here is pretty simple. Sometimes we make grants that will support our making better future grants. These could be research grants where they enable us to learn more about the effect of a particular program. So imagine our funding a randomized control trial of a program that we're interested in. There could also be a pilot grant where we think a program or an organization is promising, and [00:01:00] providing some early funding will enable us to gain information about how that program could work in practice.
We have done these kinds of grants for a very long time, but now we're doing much more of it. So, for example, many years ago we gave early stage funding to a small organization called New Incentives that helped it build up its program. That early stage funding was not cost-effective in and of itself, but it helped create the option for us to give more later. And similarly, when New Incentives grew, we were able to fund a large randomized controlled trial of its program.
And that trial convinced us that its program was significantly increasing immunization rates in Nigeria where it worked. Today, as of today, we've now directed more than $120 million to that program, which we estimate will save about 45,000 lives. But back in the day, this kind of grant was more of a one-off, largely based on the qualitative case.
Now as we've grown our team, we have more capacity to consider these value of information driven grants, and we're making [00:02:00] them more and being more systematic about how we evaluate them.
Last year in 2025, we made 18 grants totaling $39 million, where we quantified the value of information and that estimate made a difference to our making that grant. That's about 10% of our total grantmaking by volume. We evaluate these grants conceptually similarly to how we evaluate any grant. When, for example, we grant money to a water chlorination program, we estimate how many people will be reached, how consistently they'll use chlorinated water, and what effect the program will ultimately have on health and mortality.
With the value of information grant, we try to get to the same outcome. We estimate the likelihood that the grant causes us to give more money cost-effectively in the future. Maybe it causes us to move money from a less cost-effective use of funds to a more cost-effective use of funds.
So today I'm going to talk with Julie Faller, one of our program directors, about how this works in practice. Hey Julie. Thanks for being [00:03:00] here. Thanks for chatting.
Julie Faller: Yeah, it's great to connect, Elie.
Elie Hassenfeld: So we're gonna look at a few examples that illustrate different kinds of questions these grants can answer.
So we're gonna look at support we've given to an organization called Give Directly, which just directs cash to very poor people in low-income countries. Here we're looking at piloting approaches to its program that could increase its cost-effectiveness via programmatic adjustments.
Number two, we're going to talk about support that we've given to deliver diarrhea treatment and to measure its effect. Diarrhea remains one of the leading killers of children worldwide, and it kills about a million people per year.
And number three, we'll discuss a study that we've supported to better understand consumption of wheat that is fortified with iron in India. This is to address anemia, which is a problem because it can cause fatigue, shortness of breath, and other symptoms. And we think anemia in young children is particularly problematic because it can inhibit their cognitive development.
So, Julie, where do you want to start? What do you [00:04:00] want to talk through first?
Julie Faller: I would like to start by talking through some research that we are supporting currently with GiveDirectly, looking at the effects of program variations. And so just to set the stage a little bit, GiveDirectly delivers unconditional cash transfers to poor people.
And we've been interested in the program for a long time, have supported it through GiveWell grantmaking. And then a few years ago we looked back at evidence about the effect of unconditional cash transfers that showed that not only did the poor households that received transfers benefit from those transfers, but also that there were positive spillovers to people living in nearby communities—that basically it seems like having those unconditional cash transfers allowed the economy overall to grow a bit. And even households that didn't get the money benefited from it. That work that we did reassessing the evidence made us think that GiveDirectly’s core program looks about three times as cost effective as we had previously [00:05:00] thought before looking at that evidence from spillovers.
This made us think like, okay, that opens the door to there being potential program variations, where we might want to do significantly more grantmaking to GiveDirectly. And so we recently funded a grant that's going to look at three different variations in the program and help us think about whether it might be more cost-effective with those variations.
One of the variations is trying to give grants to businesses in the areas where cash transfers go. Another is trying to specifically target very poor young adults and see if those young adults use the transfers more productively than sort of less targeted transfers. And then the third is looking at whether having a physical infrastructure asset, which is a bridge, a footbridge, lets those cash transfers be more impactful than they otherwise would.
Elie Hassenfeld: Can you say a little bit more about how these work? You know, to what extent are they, we're supporting the program and we think there's like good evidence already that these adjustments [00:06:00] would make it more cost-effective versus gathering evidence is really critical to know how promising these programs would be for scale up.
Julie Faller: Yeah, I think gathering evidence is pretty critical to these. So all of these are pilots. So these variations haven't been tried before in the past and we're taking different evidence generation approaches to them where with the first one, the business transfers variation, we're looking at embedding that within an already existing randomized control trial.
And that variation itself is also gonna be randomized. And so that will generate really high-quality evidence about the impact of those program variations.
Elie Hassenfeld: So you said transfers to businesses, it's embedded in a randomized controlled trial.
So what that means to me on the evidence generation side is it's relatively cheaper for us to get information from this pilot because there's an existing trial infrastructure that we're able to patch into, is that right? And then what's the nature of these grants to businesses that we're trying to evaluate? Like, what's the thing that could be particularly promising about [00:07:00] this?
Julie Faller: Yeah, I think that's right. That being able to piggyback on the existing trial means that there's already a plan to collect evidence. There's already sort of the academic partners who are thinking about issues of study quality and we can embed this trial within that. So there is that sort of just letting us piggyback on the evidence generation that's already happening. And then on the transfers, so basically if you think back to your early Econ 101 kind of days, you can think about the transfers to households as basically being a big infusion of cash in the local economy.
And so when people get cash, one thing that they might do with it is want to spend it by buying things at local businesses. But in the part of Malawi that GiveGirectly serves where this trial is taking place—and in many parts of poor and developing countries—businesses are pretty small and they don't necessarily have the capacity on hand to be able to respond to a sudden increase in demand in the community.
And so this trial is going to say, [00:08:00] okay, let's try three different things. In one of them, we're going to tell businesses that the cash transfers are coming. So we're just going to let them know, hey, the money is gonna be entering your community, you might have more customers, and see what happens. And another one we're going to give them grants that are basically calibrated to the size of the business. So a capital infusion for the business itself and the business owners can do what they want with it to grow their business. And then in the third—this is cross-randomized design, so there are all these variations—and the third we're going to both tell the businesses that the demand is coming, that the cash transfers are coming, and also give them this capital infusion. And the idea is to see, are any of these sets of conditions setting the businesses up better to be able to profit and to be able to really channel that influx of cash productively.
Elie Hassenfeld: Got it. And the question is like if by enabling them to handle the infusion of cash more productively, it accelerates the [00:09:00] benefit that comes out of these cash transfers.
Julie Faller: That's right. Like, are you basically unlocking economic growth or allowing better economic activity in the community than you would if maybe you had an infusion of cash but people aren't able to buy the things they would want to or they have to go further out, and so it ends up that more money is spent sort of less productively.
Elie Hassenfeld: Then are the tools for measuring this effect strong enough that we'd pick it up? If businesses are like, oh, everyone in the community's gonna have more money, I'll raise my prices X percent because I think they'll be willing to pay higher prices than they were in the past. Would we know that? Is that something to be concerned about?
Julie Faller: I think so. So the effect of these transfers on prices generally is something that we're concerned about and is like a core thing that we're trying to measure across this study, where of course a basic concern is that just like more money into the economy, whether or not businesses know what's coming, lead to price increases that are just eating away at the [00:10:00] benefits.
Elie Hassenfeld: Right. Got it. Okay. So that was one example was the businesses, and then another one you gave was the footbridges plus cash.
Can you say more about that pilot and how it works and how you think about that one?
Julie Faller: For the other two pilots, we're not doing randomized control trials. We're doing like much more pragmatic operational pilots where the idea is to see, first of all, can we actually do the intervention and then maybe generate a little bit of pre-post evidence of, is this suggestive of more impact of cash transfers?
The idea of the footbridge one is similar to the business transfers one in that it's like, is there something we could do to unlock and let people use their cash transfers more effectively. And the hypothesis is lots of rural communities are kind of cut off, and providing footbridges might let them access markets more easily. But there are also just sort of fundamental questions about trying to coordinate something like a pretty substantial physical infrastructure investment with [00:11:00] cash transfers. Is this doable? And so I think that that sort of operational question is one of the key learnings we expect to get from that pilot.
Elie Hassenfeld: And so does that mean that like in the first case, the businesses case, we say, okay, we believe that we can deliver this operationally, we just need to test the effect. And then in the case of the footbridges plus cash, we don't know if we can even deliver it operationally, so it wouldn't make sense to pay for the big randomized controlled trial, but if the operational part of it were to go well, and it still seemed promising, would the next step be some more rigorous assessment of the impact of the program?
Julie Faller: Yeah. And we've already started thinking about what a more rigorous evaluation might look like in that case.
Elie Hassenfeld: Yeah, it's really interesting because it's in some ways bringing GiveWell into this place that's very cool. Where it's like, historically, what we basically did was look for organizations that were delivering evidence-backed programs.
So, you know, prototypical example is there's a lot of evidence that malaria nets, when distributed, reduce malaria prevalence and then [00:12:00] save children's lives. Let's find an organization that's delivering the program. You know, hence we find Against Malaria Foundation and direct money there.
And then the next thing you could do is say maybe there's a program that has a lot of evidence behind it, but an organization that doesn't exist, can we help create and support the development of an organization delivering that program. And here we're going like all the way in a sense to the beginning, like in one case, supporting the trial of a particular program.
In another case, even testing the operational feasibility of this like infrastructure plus cash program. And then it's very cool that we're in the position, if it is successful, to carry it all the way through from pilot the idea, to support rigorous research, to then, you know, work with GiveDirectly to scale it up more widely if it seems cost-effective and promising.
Julie Faller: Yeah, agree. And I think in this case also the very idea of what are these pilots that could look more promising came out of work that we funded GiveDirectly to do, where we asked them to do scoping work to say, come to us with a list of things that you think could be good and then we will work together to figure [00:13:00] out what is the smaller set of things that we want to take forward.
Elie Hassenfeld: Can you just talk a little bit more about that process and where these ideas came from. Like how much were things that we thought about ahead of time versus things they brought to us?
How obvious in some sense are the things we're moving ahead with? How surprised were you by like, what came out of it and to whatever extent how did the, you know, the GiveWell and the GiveDirectly teams work together to arrive at the list that ultimately we decided to support?
Julie Faller: Yeah. GiveWell and GiveDirectly have a pretty long history, so we know each other well as organizations. What's interesting to me, I'm not the program officer who directly funded this work, but when I was working with Erin on this initially, we were thinking that some of the program variations that might be most effective would be leaning a little bit more into possible health effects of transfers. And so I had in my mind that maybe the variations that we end up funding might look like trying to target transfers to pregnant mothers or trying to make [00:14:00] sure that we're serving households with young children, things like that.
And I have less insight into the iteration process and why it turned out that those weren't the ones that we moved forward with. But I think through the process of GiveDirectly digging a little bit deeper, thinking about what operationally made sense, what made sense with their model, and then us thinking through the benefits of each of these a little bit more, we ended up thinking that these variations that are more focused on the kind of core economic story of cash transfers made sense.
Elie Hassenfeld: Yeah. And it's really interesting to me because it's such a good example of a case where we literally would sort of get the wrong answer if we were just left to our own devices.
And it's the fact that we have, you know, a hypothesis about something we want to test and we have an organization like GiveDirectly that we can work closely with and take their feedback, but then ultimately leads us to a more promising set of options because we're going back and forth with them.
And another thing that this raises up for me is just how time consuming all of this [00:15:00] is in the scheme of things. It's like when I go back to what Holden and I were doing in 2008 when we found Against Malaria Foundation, it's like, well, with a couple people, you can find an existing organization delivering a very evidence-backed program and you can evaluate that evidence and evaluate that organization.
But this is an intensive, active process of successive iteration and planning and working together to create something new that's really only possible with much more capacity and, you know, I also think like more subject matter experience to be able to navigate which choices seem promising with, frankly, relatively limited evidence.
You know, ultimately we and GiveDirectly have to make our best guess about the fact that, you know, the footbridges plus cash is a test worth running rather than anything else. And we might be wrong about that, but, you know, we're certainly better positioned to do that today than we were in the past.
Julie Faller: Yeah, I totally agree with that. And I think another difference is that at this point in time, we have a program officer who's focused [00:16:00] exclusively on trying to identify highly cost-effective grants to be made in the livelihood space, in this realm of trying to increase people's income or, yeah, like sort of material consumption benefits.
And so that means that that program officer, Adam, is both excited to see the results of these pilots and think about what grantmaking we might do there, and is also actively thinking about, for example, the question you raised earlier: Would we then go on to evaluate the bridges plus transfers with a more rigorous design?
And he's like thinking about, yes, what would be the next step if this looks good, across this, along with the whole set of investments he's considering and stewarding.
Elie Hassenfeld: I do want to like circle back on, how's it going so far? What have we learned? When will we know more? But presumably the future we are working toward is one where we've tested a few of these and we've identified ones that we want to direct funding to or would like others to direct funding to, to scale up further. And presumably in this domain, these could be very large programs that reach and help a huge number of people very [00:17:00] cost-effectively.
Julie Faller: Yeah, I think that's right. And so, I think I can answer both of your questions at once there, which is what have we learned so far?
The answer is not a lot. These pilots we just approved last year and for instance the business transfers are scheduled to be wrapping up around now as we're speaking. And so we don't expect to have the final evidence about the impact of those transfers, especially how their impact changed over time, which is important to our decision making, until the end of 2028.
Doing this kind of grantmaking, where you're really trying to plant seeds, it can be quite a ways into the future where we'll learn the most sort of final outcomes. Now, of course, there are midline data collection that's happening. We'll be getting updates along the way. We might choose to make decisions based on that little bit earlier, but this is like a long-term endeavor.
Elie Hassenfeld: Well, that's great. This is super interesting. Let's talk about another example. So the next case study we were going to go through was the diarrhea treatment example.
So can you just [00:18:00] tell us a little bit about what that program is, what that issue is, and how we're addressing it?
Julie Faller: Okay, so at a very high level, one of the reasons that diarrhea is a problem is because it causes dehydration, and very young children are at high risk of dehydration.
So there's a product called oral rehydration solution, or ORS, that basically works to help reduce the risk of dehydration when somebody has diarrhea by replenishing electrolytes and also helping them retain fluid even when they have that illness. And so ORS has existed as a product that can be used in the community.
Basically parents, you know, give it to children for decades. And the mechanism of action is very well understood. But not all of the children who could benefit from using this when they're sick with diarrhea, use it, especially in I guess higher-burden or more remote areas.
And so our question was, okay, is there anything that we can do to increase the use of this product? And we specifically [00:19:00] were interested in a method of just door to door delivery, so that if you're somebody who has a child under five, and a community health worker or a distributor comes to your house and gives you the product and tells you how to use it, how much does that increase the use of this product?
Elie Hassenfeld: Got it. And so, how's this working? How's this going to date?
Julie Faller: So to help answer that question, we made a grant to CHAI, the Clinton Health Access Initiative, in 2023 to run another randomized control trial in Bauchi, which is a state in northern Nigeria, to help us answer the question, both does delivering the product increase the coverage, and also how long does it increase coverage. This is exciting versus the program that we just talked about because we are starting to get back results from this trial and we think that this could influence the grantmaking that we do this year.
So, what actually happened was CHAI procured large amounts of ORS and zinc copacks—ORS and zinc are packaged together to treat diarrhea—and [00:20:00] they worked with health officials to map communities, to help estimate the populations, to go out and verify that where we thought there were households with under five children, there actually were households. And then they hired about 3,000 community-based distributors to go to half of the villages in this state of Bauchi and distribute the zinc and ORS.
And what's also nice to know about this trial is that the control groups, the villages that didn't get the ORS and zinc in the first year of the trial, will now receive the product as the sort of like final bit of the trial. So one thing that's exciting that we learned is just that this was doable.
I kind of was surprised about the scale that we're talking about here. So through the first set of distribution, CHAI served about 865,000 children under five, again with those 3,000 community-based [00:21:00] distributors. In the follow-up data to try to figure out what the coverage rates were, we estimate that about 80% of households with under-five children were visited and received ORS and zinc through this program. So just like the fact that that was doable was an exciting update.
Then in terms of our research questions of how much does this increase the use of the protective product and for how long, the data analysis is still ongoing for that, but the preliminary data are showing that there are substantial increases in use of the product, especially in the short term after the product is delivered.
We still need more data, but it's looking like, as you might expect, that effect kind of drops off over time, which could make sense. You know, the product gets used, maybe people forget about it. And so another thing that we're considering as we wait for those results is maybe it might make sense to support a model where the product is delivered twice a year instead of just once a year.
Elie Hassenfeld: Like a couple of really interesting things about this to me is [00:22:00] first you know, there's this problem, diarrhea, that kills a million people per year, and there's a treatment for it that is extremely effective, but it's just not used by roughly like half of the children who need it. And the question is why would that be the case? It's cheap, it's effective, there's no reason that, you know, it shouldn't be used.
And so what's wild to me is that this has been true for such a long time, before GiveWell was even a full-time project. Like for some reason, oral rehydration solution was the thing that like I wanted to give charitable dollars to early on, and it was really frustrating that I couldn't find a place where it seemed like you could give money and just get more children this ORS that they need.
And it's really been a long road to reach the point where we are in a position to say, there's this big problem, we're going to try to find a way to cost-effectively address it, which is, you know, the thing that we're testing now. And I think this is one of the most, to me, like exciting evolutions at GiveWell.
[00:23:00] We talked a little bit about this with Natalie a little while ago in the conversation we had about vaccinations, where, you know, the sort of original GiveWell model was look for programs with evidence and then fund them. And then here in this case of ORS for diarrhea treatment, we're in a position to say, there's this problem in the world. We do not have a good reason to explain why children are dying because they don't have access to this incredibly cheap and effective treatment. And then, you know, we're just going through this effort of finding a way to fund an organization, to test whether the solution that we came up with with CHAI will be effective at reaching more children with ORS who need it, then using it effectively, and then reducing deaths from diarrhea.
Julie Faller: Yeah, and I think we expect considering whether we can do grantmaking in this space to be an important part of our work this year. And we're thinking both about this particular model with CHAI, but also with other potential implementers.
And so I think it's also just a very exciting evolution in terms of our ability to take evidence and identify [00:24:00] and build funding opportunities so that we can, you know, get the programs in the places where they'll be most cost-effective with the implementers who can do it well in that area.
Elie Hassenfeld: Yeah, and then we're also like learning as we go, so you mentioned this finding that is sort of new but we're focused on, which is the initial effect seems very high, meaning like increased use of ORS immediately after receiving it, but then it pays out over time, and then trying to figure out how to adjust it.
You know, we're collecting this information as we go so we and CHAI can figure out how to adjust and make this program more effective. I'm also curious like how do you interpret that data? You know, overall, the data that's come back so far, does it seem relatively promising because the program's working and people are using it more early on, or, you know, are you kind of pessimistic because we see this effect of fade out and you're not sure how we'll address it?
What's your overall take based on the results we have back to date?
Julie Faller: Yeah, I think overall my take is that I am optimistic that there's funding that we can do in this [00:25:00] space. So, one of the things that we learned in the initial kind of baseline data collection is that diarrhea was more prevalent than we had thought when we made the grant. And also that ORS and zinc usage was lower than we had thought when we made the grant, in Bauchi in Nigeria. Both of those factors mean that increasing usage is potentially really impactful, even if there is, as seems likely, some fade out over time.
I should say it's also kind of interesting, we didn't initially intend to track usage over a full year, and we ended up making what we call a top-up grant, basically putting more money to be able to measure out to 12 months, when the research team came to us and raised, like, this is probably going to be an issue. And I'm now, as the preliminary results are coming in and we have very uncertain like tiny amount of data about potential fade out, I'm feeling really grateful that we did decide to look at effects over time so that we can make a more informed funding decision.
Elie Hassenfeld: It's always amazing to [00:26:00] me that, you know, it really seems like the preponderance of the evidence is decisions to put a little bit more money into collecting a little bit extra data usually looks pretty good because, you know, it's very hard in the moment to decide to give more or to spend money on data collection versus delivery. But then we do, you don't have to be surprised too often for it to be worth it. And I don’t know, I'm really glad that we're able to do that, that we have donors who support our spending on research because I think it's more costly in the short term, but allows us to direct funds more cost-effectively in the long term. And that finding seems very consistent over time.
Julie Faller: Yes. And I would just say going back to what you started with at the top about, like the fact that there's so many places that we could consider funding a program means that, yeah, it absolutely makes sense to pay a bit more to be able to make those funding decisions be well informed.
Elie Hassenfeld: One final thing that I think is worth touching on with respect to this program, and just correct me if this is wrong, but this program came out of work we've done with CHAI and what we call the incubator that we've supported there, where [00:27:00] the direct purpose is we work with CHAI and their team, and we try to look at a large number of potential programs that we could support and select the few for pilot implementation, early research, and then scale up with more rigorous evidence that ultimately we get to this stage. So you know, this program, to reach even this stage where it's being tested, is downstream of a lot of other work to compare it against other programs.
Do you have any notion of the scale of the funnel at each stage before getting to this program. You know, are we talking about like we looked at 10 programs and landed on ORS, is it a hundred programs, or how would you describe that to the extent you can?
Julie Faller: Yeah. It's definitely not like one in 10 programs that end up coming through this. There are many, many more programs that are reviewed and considered before we end up making a funding decision like this. This ORS and zinc trial was actually the first grant that we funded out of the incubator in 2023. And there [00:28:00] were many programs that were reviewed and considered at varying levels of depth before we hit upon this one as one that made sense.
Elie Hassenfeld: Great. Yeah, that's my sense too. Okay, so let's just move and talk about our third example, which is a little bit different and it's focused on getting some additional data about wheat fortified with iron in India to address anemia.
Julie Faller: Yeah. So this is data that we gathered to help inform a funding decision to Fortify Health. And so, again, just to set up the problem, Fortify Health is an organization that operates in India and their program model is that they partner with millers who package and sell atta wheat flour. And they partner with these millers to be able to fortify that wheat flour with iron. The wheat flour, you know, is fortified and then sold on the open market just to consumers, there's no sort of targeting or delivery support beyond just your normal go and buy your wheat flour.
And so we had a lot of questions about this that were along the [00:29:00] lines of, you know, how much fortified flour are households buying, and then within the household, who eats the flour? And that matters because different demographics are at more risk of anemia. And additional sort of questions that were just around that, basically consumer data. Like who's buying the product, who's eating it, and how are they using it?
And so to help answer those questions, we funded a household survey that was implemented by IDinsight in six cities in India. Basically went to households and asked them those kind of questions to help us reduce our uncertainty about how the flour was ingested.
Elie Hassenfeld: Can you just give a sense of where the data was coming from before that we were relying on, how credible that data was, and then maybe like how much of a change could this newer, better data source have on our bottom-line decisions?
Julie Faller: Because the survey addressed different parameters within our cost-effectiveness [00:30:00] model, there are different answers depending on the parameter. Some of the questions around, like, how much flour are people ingesting were based on, I believe, Fortify Health's best guess, as well as some additional desk research that we did. There was one assumption that was around how much is there plate wastage, which I think as a parent of young children, you might be familiar with this phenomenon, where you buy and cook the food and then somebody fails to eat it even though it's already on the plate.
Elie Hassenfeld: Not my children. Not, not one. That's I've, yes, a lot of, and lunch wastage too.
Julie Faller: Lunch wastage, yeah. Yeah. But so of course that matters because maybe it gets all the way into like people are actually cooking it, but then nobody eats it and then the iron can't be absorbed. And for that, we were relying on a study from, if I remember correctly, Malawi. So a study from like a totally different context that wasn't at all related to it, but that was the data that we had.
And so each of these parameters was kind of like [00:31:00] low to medium importance on its own, but we thought that together they represented pretty substantial uncertainty, like maybe people are eating less flour than we think, and no children or girls or women of reproductive age are eating the flour, and they're always drinking tea, which we think inhibits iron absorption when they do it, and like there's a lot of wastage. And if all of those things turned out to be true, it could represent a substantial decrease in the effectiveness of the program.
Elie Hassenfeld: I think this is something that like people don't fully grok about GiveWell in our work that I think is very salient to us, which is, we know that so often we are trying to make judgments about what to support based on the best data available, and that best data available is often really bad. And in the absence of good data, we're going to do our best to take the data we have plus, you know, our best judgment about the world to make a reasonable estimate.
But there's like so much opportunity to get better information that can make a really big difference to the bottom line and this kind of [00:32:00] work that could really have a huge effect on the kind of funding that we do for this program in India because it's just a much higher quality source of some of the most important inputs into the decision about how cost-effective this program is.
Julie Faller: That's right. Yeah.
Elie Hassenfeld: All right. Thanks, Julie.
--
Elie Hassenfeld: Hey everyone, it's Elie again, and I hope that conversation was interesting and informative. I think it, for me, really shows how much GiveWell has changed over time. Going from an organization that in our earliest days was really just able to look at the set of organizations that already existed and evaluate them, to the organization we are today where we can support new organizations, support research, support pilots, all sorts of things that we weren't in a position to do way back when. Today, we can do that with greater knowledge, better experience, and also greater systematicity about how we evaluate all the different [00:33:00] opportunities.
And I think it's still pretty unique, as far as I know, in the global health and development funding ecosystem for there to be an organization that is operating across as wide a spectrum as we are from early to late stage. Meaning, relatively few large funders are just trying to use money to deliver cost-effective programs to people who need them. So programs like malaria nets, and vitamin A supplementation, and water chlorination, and malnutrition treatment. So, you know, large dollar amounts to just deliver programs to people in need. And then are also able to operate all the way at the early end of the spectrum, in funding new organizations, new pilots, and basic research.
And I think that puts us in a really powerful position to try to make change because we are in some sense a funder who's happy to pay for cost-effective programs that can be delivered at scale. And therefore we're very good at thinking about the kind of research that would inform those later decisions. And so we're very grateful to the donors who support us that we're able to fill this, in [00:34:00] my opinion, you know, needed gap in the funding ecosystem.
I think another thing that to me really jumps out from this conversation is just how much need and how much opportunity there is for this kind of value of information driven work. And as we've grown, we're in a position to collect that data and make better decisions in the future. So, as always, thanks for listening, for your interest and engagement in our work, for your support of our work.
We really appreciate it. And if you have any thoughts or questions, please do reach out directly at info@givewell.org. Thank you so much.