Elie Hassenfeld: [00:00:00] Hey everyone. I'm Elie Hassenfeld, GiveWell's co-founder and CEO, and I'm joined today by Dan Brown. Dan is a program officer, and he leads GiveWell's new areas team.

GiveWell's research team is organized around causes, so we have a team that focuses on malaria, a team that focuses on water, livelihoods, nutrition, vaccines, and more. And Dan leads the area that looks at programs that are outside of GiveWell's current scope, trying to look at things that we have not looked at in the past.

So, far in these conversations, we've toggled between the big picture, how is GiveWell responding in general and how are we thinking about the cuts at a high level, and then zooming in on some specific examples.

Today, we wanna zoom in on a specific grant that we made to CHAI and PATH, two international organizations, to support technical support units. We call these TSUs, they help ministries of [00:01:00] health in countries respond to the impacts of the US government funding cuts. And Dan led the investigation of these grants, and he is gonna explain more about what these are, how we investigated them, how this is similar and different from what we've done in the past. And these are grants that we co-investigated and co-funded with Open Philanthropy.

I think these grants are really interesting, because they might surprise some people that they're the kinds of programs GiveWell supports. I think a lot of people sometimes think about GiveWell as primarily Top Charities, and the majority of the funds that we raise and direct do go to our Top Charities.

And then we're also working on programs outside of the Top Charities that we think are high impact, very cost effective. And these grants to technical support units are a really good illustration of that other grant, the kind of grant that's supported by our All Grants Fund.

And as you'll hear in this conversation, we are less certain about the impact that these TSUs are going to have than we are about the impact of our Top Charities. But [00:02:00] because of the need and the demand that we're hearing from governments for this program, we believe that they have the potential to be incredibly impactful. And so I think it's a really good illustration of how GiveWell has evolved as an organization over the last few years that we're now in a position to make this kind of grant.

Because so much of the funding we direct is to fairly straightforward programming, delivering malaria nets or vitamin A supplementation, we have a good sense of the impact of those kinds of programs. And so when we support a program like these TSUs, we're asking this question, not only, are these TSUs good? But also, do we feel comfortable allocating several million dollars to the TSUs, and not to vitamin A supplementation or malaria prevention?

I think that kind of discipline that we have organizationally, but also in the models that we build and the analysis that we undertake, really helps ground us and puts, you know, in my mind, the right kind of pressure on us to think critically about these [00:03:00] more uncertain grants that we're making.

And then this grant investigation, in particular, is also a good example of how we've been able to move relatively quickly to respond to the urgent needs that have come up in response to the US government cuts. To date we've made a total of $23 million in grants in direct response to these funding cuts. The cuts happened about three months ago, and governments were requesting additional support from CHAI and PATH, and we were able to move fairly quickly to provide the support so they could take action and make better decisions going forward.

So really excited to talk to Dan today. Good to talk to you, thanks for joining, and would love to hear about what these TSUs are.

Dan Brown: Hi Elie. Yeah, really excited for these grants. The technical support units, they're small teams of three to five people typically. They're based in country and as you mentioned, they're gonna provide technical assistance to the ministry of health to help them respond to the US government funding cuts.

Those in-country teams will also be supported by another small global team, [00:04:00] both CHAI and PATH who are providing the support. We funded CHAI to work in five countries, Burkina Faso, Nigeria, Malawi, Uganda, and Zambia. And then we funded PATH to work in the DRC, so six countries in total.

The idea here is that the TSUs will provide additional capacity to basically generate evidence and other inputs that the government wants to support its planning and prioritization decisions in response to the cuts. All the decision making itself will be owned by the government, the government is leading this work. The work itself will be tailored to the government's priorities. But we think there are sort of broadly speaking, four main ways that the TSUs might be able to support the ministry of health in these countries.

So, firstly they'll help the ministry of health to identify the impact of the funding cuts across programs and health systems. There's sort of an information gathering component to this, or at least aggregating information. Secondly, they'll try to identify cost efficiencies in existing programs to make remaining funding go further.

Thirdly, they'll conduct cost-effectiveness analysis, which the ministry of health can then potentially use to help reallocate funding to more cost effective [00:05:00] programs. And then fourthly, they'll help support the ministry of health to crowd in new funding for health programs, which could come either from external donors like us or through other parts of the government budget.

I think the relative focus across those four work streams will vary a lot across the six countries based on government's priorities, how much work they've already done to date. But that's broadly how we see the TSUs working.

Elie Hassenfeld: It makes sense why now is a particularly important time to have the TSUs. Assessing the impact of funding being lower, looking for cost efficiencies, aiming to direct funds to more cost effective programs, and crowding in more funds are particularly important, but the countries you mentioned, their ministries of health have been resource constrained for a long time. Why do you think this is something that wasn't done before?

Dan Brown: So I think to some degree this is stuff that CHAI and PATH have been working with ministries of health on already. They've got close partnerships with the ministry of health in all the countries that we are funding and have had that for many years, and in some instances have been working with the kind of cross-cutting health departments that are responsible for some of this type [00:06:00] of work. So I don't think it's really, really new.

I think what the TSUs hopefully will do is help intensify that support at a time when it's really needed. And I think certainly both organizations are seeing this as something that could be catalytic and then lead to more intense level of support over a longer time period than just the next year or so.

Elie Hassenfeld: Got it. That makes sense. So I'm imagining these teams, or the way you described them, it's five people who move into the ministry and are supporting its decision making and providing data and analysis so decision makers can use that to do better in this time.

I'm curious, like what do we know about the kinds of people who staff these teams? Where are they coming from? Who are they, what's their background? Are they CHAI employees, PATH employees, or something else? To what extent do you know that?

Dan Brown: I think it'll be a combination of existing CHAI or PATH employees and new hires. I think in terms of the sort of skill sets that we're talking about here, I think it'll be a combination, you'll probably need some people who are spending most of their time engaging in meetings and working with other stakeholders to understand government's priorities. You'll need also some people with maybe more of the [00:07:00] technical capacity to do the cost effectiveness modeling, the evidence generation.

So I think it's gonna be a combination of different skill sets, and that's why you know, we've got these sort of three to five person teams with a sort of variety of people within them.

Elie Hassenfeld: And then, do you know from past experiences do these kinds of people tend to be nationals of the countries they're working in, or are they like outsiders who are brought into the country to sit in the ministry and do this work?

Dan Brown: My understanding is they'll all be nationals of the countries they're working in. There's this kind of global team as well that will work across countries, which will be providing sort of cross country support, but the teams themselves in country will be nationals I would imagine most exclusively.

Elie Hassenfeld: Got it. How will the global team interact with the individual country TSUs?

Dan Brown: So I think that's still a bit to be determined, and the reason I say that is that what we have a good sense of is the overall work streams and also how this might play out in practice and how it might help support the ministries of health.

But I think the sort of concrete work plan for what the day-to-day is gonna look like will vary a lot and will depend a lot on government's priorities, the work they've already done on these work streams. This is actually a sort of slightly unusual grant for GiveWell in the sense that we've tried to move very quickly [00:08:00] given there's a real need for this support right now.

And so we've kind of accepted a slightly lower level of certainty on what the day-to-day will look like. We've actually requested a detailed work plan as an initial milestone for the grant that we're expecting to get in first few weeks and obviously CHAI and PATH will agree that with the ministry of health.

I suspect there could be a lot of support on maybe the cost effectiveness modeling side, I think that might be work that is typically done in cross-cutting teams within CHAI and PATH. I think both have teams that are sort of health systems teams, health financing teams who've done a lot of impact modeling in the past.

So I suspect that sort of expertise might live a bit more centrally, than in the individual country offices. And I think there also may be some disease area specific expertise that lives with CHAI and PATH's global teams that, perhaps, can complement some of the contextual knowledge that the country teams have themselves.

So I think there'll be sort of natural ways that the two teams complement one another, but in terms of exactly how they're going to support, I think that's something we'll learn in the first few weeks of the grant itself.

Elie Hassenfeld: Right. And that makes a lot of sense. So you're talking about these handful of people in each country, and you really want to have people who have or can easily form relationships with the government, who are nationals.

[00:09:00] And then, I just think about our research team and the scope of activities that we want to be able to cover and the scope of skill that we need to have. It's modeling and evidence, assessment and communication, and then across like HIV, and malaria, nutrition, water, et cetera.

We're a 45 person research team, now we're talking about these very small teams. And so it makes sense that they would get support from CHAI's or PATH's much larger footprint and knowledge base.

I know you don't know exactly what the TSUs are going to do, but what are the kinds of things they could do? I mean, is there any way we could get concrete, either with potential examples or, you said that CHAI and PATH have done things like this in the past. Do we know anything from their track record about the kinds of impact they could have on government decision making by sitting in the ministries this way?

Dan Brown: Yeah, definitely. I think we have certainly some guesses as to how this could work in practice and how it could add value. So maybe I can talk through some of those for each of the four work streams that I mentioned earlier.

So, on the sort of information gathering piece, understanding what's the impact of the funding cuts across different programs and systems. First thing to say is, I think this work has actually been done in several of the countries that we are providing the funding for and might not be a [00:10:00] focus across all six. Something that governments obviously were trying to do immediately as soon as the funding cuts hit.

To the extent that it hasn't been done, I think it's not gonna be sort of firsthand data collection. I think this is gonna be about meeting with government employees, government health workers, staff from other implementing organizations, and trying to aggregate information they have on where the caps are as a result of the funding cuts, and kind of pull that together into dashboards, reports, things that are very user-friendly for the ministry of health for when they're making planning decisions. It's just a whole load of cuts across a whole load of different programs and disease areas, and having that in a way that's very presentable is just gonna make decision making so much easier.

On the cost efficiencies side, analysis to identify ways to cut costs and existing programs that doesn't reduce impact. I think there's a couple of main buckets here for where we think there could be some impact.

The first is, understanding ways that services can be integrated across different programs. What I mean here is, donors have often funded parallel systems for different disease areas. Sometimes that might be as extreme as having distinct groups of health worker for different disease areas, HIV, TB, [00:11:00] but there may be ways to reduce costs without sacrificing impact by integrating things like training and supervision of health workers, perhaps some aspects of the supply chain. So are there elements of warehousing and transportation of commodities that you can pull together across different programs that allows you to cut costs. Similarly, for data systems that are used to track stock outs or help with procurement decisions, again, in maybe ways to integrate those across different programs that allows you to reduce costs without much impact on the sort of ultimate benefits of the programs.

And the other piece on cost efficiencies that I think is looking for ways to transition activities from relatively expensive US government funded contractors to local organizations and local universities. So I think that's two of the big pieces around the cost efficiencies.

Just turning to the other two work streams that the cost effectiveness side, I think that will concretely look like CHAI and PATH teams conducting cost effectiveness modeling. And then the outputs of that I think will help inform the government's prioritization decisions. So the government will probably have a pretty clear sense of its own priorities of what type of commodity or area it [00:12:00] wants to prioritize, but there's a ton of decisions within that, right? Of what intervention do you choose? How do you target the program geographically, demographically, and having an analysis of the most cost effective ways to do that, I think would be really helpful for governments when they're making those decisions.

And then on the final piece, the sort of crowding in more funding for health programs. One thing we've heard from TSUs, we've heard that the ministry of health is not always able to spend its full budget. And one potential explanation here is that if there's a lack of sort of real time information on expenditure, it could be hard to reallocate funding from one program to another if one program looks like it's gonna go under budget. So something here, again, this is a hypothetical example, but it could be the case that the TSUs help the ministry to develop a tool to track real time expenditure across programs and enable them to better shift resources to reduce the underspend of existing budgets.

There's another piece around supporting the ministry of health to pull together a case for additional health program funding, which it can use to advocate for additional domestic funding, potentially with the ministry of finance or other groups.

And then I think [00:13:00] that the final piece on this fourth work stream is that the TSUs we expect will generate a list of the highest priority, funding gaps for the ministry of health and present those to external donors, including GiveWell but other private philanthropists too. And on our side having a list of specific programs for which there are funding gaps, where we know that the ministry of health is really brought in and it's high priority for them, that's extremely helpful for our decision making. You can see ways that would materially crowd new funding in for really important programs.

I think we have that kind of level of detail where we have a sense of where we think the value could be here, going deeper than that and exactly what that looks like day to day is still a little bit to be determined but those I think, are the main channels.

Elie Hassenfeld: Yeah that's really interesting, and something that is probably not intuitive to people outside of GiveWell but we experience every day is that it's very hard to determine where there are funding gaps that actually need to be filled.

Having this fairly like systematic, consistent process because it's by these two organizations assessing and sharing gaps in countries, and there's six countries, but especially in Nigeria and DRC, two of the countries with the highest burden of [00:14:00] disease, especially among the diseases that we're able to provide funding against, could be really helpful in just being more confident that we know what needs are out there so we can go to donors and say like, here are the needs that we could fill with additional funding.

Dan Brown: One concrete example of this within the new areas team is, an area that we're certainly gonna prioritize this year is health system strengthening. It's hard to know when you are approaching a new area, where do we start? What do we think is most promising? And knowing that in a few months time we may have a list of very concrete, prioritized health system strengthening opportunities coming out of the TSUs that we could fund later in the year. It's enormously helpful and I think really helps us just understand where to focus our time.

Elie Hassenfeld: Yeah. So I wanna just ask a follow up question about the cost effectiveness piece. So, you said there's these four areas. Info gathering, so decision makers just have a basic view of the situation and what the cuts have caused. Cost efficiencies, and you give some examples of how they might reduce overall cost and make things more efficient. Crowding in new funding. But then on cost effectiveness analysis, I'm curious if you could give a concrete case where a ministry would have to decide between program [00:15:00] A and program B, or targeting approach A and B. Give me an example of the kind of decision a ministry could make to increase the cost effectiveness of their health delivery, relative to what they might do without that information.

Dan Brown: Yeah, definitely. And, we have a couple of concrete examples from CHAI from some of their past work, which I can maybe speak to.

One thing that the TSUs might be able to support on and have supported in the past is helping ministries of health, their applications to the Global Fund, which is a major multilateral funder. As part of that, they did some essentially sort of GiveWell style cost effectiveness modeling to try to understand where specific malaria programs, I think one was a bed nets program, I think maybe another was indoor residual spraying, but they looked at which provinces within this specific country would this look most cost effective? And as a result of that work, the provinces that were proposed to the Global Fund in the country's application changed to provinces where the malaria burden was significantly higher.

And as a result, we think that counterfactually, the program is going to happen in a different set of provinces to the provinces that would've happened without CHAI's support, in a way that's going to lead to [00:16:00] ultimately more health impact. So that's I think the kind of thing. And you can imagine similar, you know, it's not just gonna be about geographic targeting, but it could be about demographic targeting. It could be about which implementation model you use in practice. Maybe if you are thinking about bed nets, it could be about which choice of bed net you use, which manufacturer you procure from.

So I think it's those sorts of decisions. Within a set of government priorities, there's always gonna be a lot of details of how you go about implementing some of these programs and how you target them. And I think that's really where having cost effectiveness analysis to hand in a really presentable way is gonna be incredibly helpful for governments while they're making those decisions.

And of course, cost effectiveness isn't gonna be the only consideration. There's gonna be a ton of other objectives and political constraints and so on the governments will face, but having cost effectiveness as an input that's really easy to digest, I think will be very helpful for them.

Elie Hassenfeld: We've been talking a lot about what the TSUs are and how they could have impact and how they'll help make things better. And I'm curious now to shift away a little bit from what they are a little bit more to our process.

You know, you mentioned that the process of deciding to make this grant was [00:17:00] quicker than usual to some extent due to the current moment. I also think people think about GiveWell as primarily being Top Charities. This sounds really different, this is a pretty different approach that we're following in supporting this kind of activity which I think is really great because of the kinds of impacts that you described. But just, yeah, can you say more about how did this grant come about and how did we investigate it and gain the confidence to support it?

Dan Brown: Yeah, absolutely. So in terms of how it came about, I think I mentioned this briefly before, but our understanding is that in some countries, the ministry of health is already reaching out to CHAI, particularly, asking them for support on some of the work streams that we're talking about here, particularly some of the information gathering. And we know that CHAI and PATH are already very closely embedded with the ministry of health and have strong relationships there. So I think basically, sort of, as far as we understand, the origin of the idea really kinda came from governments and then we received proposals from CHAI and PATH, which we started investigating.

So some of this work was already being done on nights and weekends, but there's a limit to the extent to which that can continue, and so that's where both organizations presented [00:18:00] proposals to us.

Elie Hassenfeld: Dan, like the starting point there in some ways is CHAI and PATH had these existing, fairly close relationship with governments, and then governments at the beginning of this calendar year, saw these cuts come through and said, CHAI and PATH, we see the kind of support you've given us and can give us. We could use more of it now given the situation we're in.

Dan Brown: That's right. I mean, it's not true in all cases. There are some countries where the government was very explicit they didn't want involvement from international organizations in that first round of gathering information, identifying what programs and systems have been impacted. So not true for all countries, but certainly in several countries that was the case. And in countries where the initial approach was to do things without any involvement from international organizations, I think those countries are now more open those that we've spoken to.

In terms of the investigation itself and sort of how it was different to what GiveWell would ordinarily do, I think the decision was made really much more on qualitative factors than on our cost effectiveness analysis, and that's probably a bit unusual for GiveWell. And I can explain a little bit more why in a moment.

The investigation itself, I think the single most important part of our investigation was speaking with [00:19:00] very senior officials at the ministry of health in each of the six countries to really understand, is this something that you want? Is this something you're asking for? How do you think this is gonna add value? And are CHAI and PATH the right organizations to do this, or are there other organizations who are a better place to support this work? And the work is government led, without really strong government buy in, it's just not gonna be effective, so that was I think the most important part of the investigation.

And then as part of that, we also try to understand practically what do CHAI and PATH's relationships look like with government? Who are they speaking to? What's their track record on similar kind of work?

The other sort of qualitative factors that came into this. I think we had a bit of back and forth with the organizations about which countries we think this work would be most effective in. And I think this really did come down to which countries have been most exposed to the cuts, and which of the countries with the sort of highest burden of preventable disease, where you can imagine that improving the planning and prioritization process is going to translate to a larger impact on health outcomes.

We did do some cost effectiveness work, I think it was more to sanity check does this seem like a sensible use of GiveWell funding, rather than [00:20:00] feeling confident that we have a best guess for the ultimate effect of the TSUs. The kind of channels in our model for how you think it's gonna be impacted are very closely related to those work streams that I mentioned. So, you know, if you find a 30% cost saving on a given amount of government expenditure, you get 30% more program. If you shift government expenditure from one program to another, that's 20% more cost effective, again, you get more value from that overall amount of government spend.

And the cost effectiveness analysis that we did to make ourselves feel confident that this is a sensible use of funding relative to standard Top Charity interventions and so on, is really just to understand how much government expenditure would you have to influence with the sort of analysis that CHAI and PATH are doing for this to look like a good use of funding? And for the cost of the TSUs in each country, it is in the region of a few million dollars of government expenditure would have to be shifted to be 20, 30% more cost effective, 20, 30% cheaper for this to look like a good use of funding. And so it became apparent quite quickly that there's a lot of upside here and that this does seem like a reasonably sensible use of GiveWell funding.

But again, the quality factors, I think knowing there's real government buy-in [00:21:00] for this and that CHAI and PATH have track record and have the right relationships within government, I think was the more central aspect of the decision making.

Elie Hassenfeld: There's like so many interesting aspects of that to pull out. First, just something that I think I've heard a lot over the years you know as like a potential criticism of GiveWell is, you know, hey GiveWell, you should be doing more to influence how governments operate, rather than just funding programming and commodities directly.

I think part of the issue is that it's very hard, you can't really like push the government around, like the government's gonna make decisions that it wants to make based on its priorities. And so this is a case where supporting work that is able to improve government decision making is feasible and promising because the government is saying we need this now, we want your help. And then we're able to step in and provide the funding for these units just as a very effective way to do that.

The other thing that I think is really interesting to pull out is that this grant is much less certain, our analysis is more speculative. But one thing that I think is really powerful about the way that GiveWell does this research is we're always [00:22:00] comparing the more speculative, uncertain opportunities to this alternative use of funds, which is straightforward, high confidence interventions from our Top Charities. So it's like every time we're asking should we direct money, a few million dollars to the TSUs, we're asking, do we prefer that to our alternative use, which could be more incentives for immunization, more seasonal malaria chemo prevention, more vitamin A supplementation.

One of the things that gives me more confidence about our support of these kinds of opportunities beyond the analysis is just knowing that we're always asking that question. Like, how does this compare to the alternative?

Dan Brown: Yeah, I totally agree with that. And it's also a really a good sort of accountability mechanism, right? There's lots of things that sound like a really intuitive, great idea at the moment, and this jumped out immediately as a intuitively very good idea. But having that kind of standard to compare to, I think really just sort of force your hand to make sure that you' re really like interrogating the case for things that otherwise intuitively do sound quite good.

Elie Hassenfeld: We co-funded this grant with Open Philanthropy, and I wonder if you could just talk a little bit about the role [00:23:00] we played, the role they played, how that worked as we did this together.

Dan Brown: Yeah, absolutely. I think one of the real benefits of the co-investigation was that there's just a lot of stakeholders that we wanted to speak to, a lot of government stakeholders, other people in the global health world who've got experience with technical support units. And obviously wanted to speak to various PATH and CHAI country teams and really get a better sense of what this is gonna look like as far as we could. And that just meant if we want to turn around a very quick investigation, which we need to in real world terms, but also have confidence in the grant by speaking to lots of people, I think being able to co-investigate this with them, there were two people from Open Phil who worked on this with me, Nic and Rafa, and I think having three of us be able to take a lot of different conversations at the same time, allowed us to sort of move this on very quickly and also get input from lots of different countries at the same point in time.

So yeah, I think that was one of the biggest strengths of the co-investigation was just having that coverage across lots of stakeholder conversations, which again, was one of the central things that gave us confidence in the decision itself.

Elie Hassenfeld: Just to give a sense, how many of those conversations did we [00:24:00] have collectively, roughly?

Dan Brown: I think it was around 20 to 30 conversations, something like that, across different countries and again, other sort of global health stakeholders. So it was senior officials in the ministry of health in each of the six countries. We then had conversations with other funders who funded similar kind of work in the past, and who could potentially be interested in funding similar work in the future. We spoke to other people who have worked closely with CHAI and PATH in the past and have also got familiarity with technical support units and the kinds of activities that are involved to, again, get their sense of whether this is a good thing to be doing right now and how it would work with government and what the risks are.

Yeah and then conversations with CHAI and PATH's country teams, as well as the global teams. So yeah, I think it was probably nearer 30 conversations by the end.

Elie Hassenfeld: Okay. I'm really curious if you could tell me like, what does a good outcome for this grant, something like a 70th percentile outcome, what does that look like? And then like a disappointing outcome, like a 30th percentile outcome?

Dan Brown: Great question. So I think a good outcome, I have to think a little bit harder to put numbers on this, but I think having a clear signal that the governments are really bought into the work of the [00:25:00] TSUs, that the work that they did really informed government's decisions.

I think in practice, you know, that's not something we can necessarily evaluate with an RCT or something. This is going to be, I think, at the end of the grant, can CHAI and PATH point to specific examples where their analysis has identified cost savings in programs, or where the recommendations that they made has been written into government budgets, or other government reallocation decisions between programs have been done on the basis of cost effectiveness analysis that they put together.

I think another potentially good outcome here is that on the piece we discussed around crowding in additional funding for health programs, if we can see evidence that donors have actually stepped in and filled some of those funding gaps that were a high priority to the ministry of health, and that could also look like a scenario where GiveWell directly makes grants for work that has been identified by the TSUs later in this calendar year.

I think those are probably the main outcomes the sort of real sense and case studies and examples of government finding the advice and the analysis that they're producing to be helpful and concretely being [00:26:00] useful for decision making and actually specific things that can point to at the end of that, I think would be really great to see.

Elie Hassenfeld: Roughly when would we have some sense of you know, this has gone positively or has not quite gone the way we hoped?

Dan Brown: Yeah. So the grants themselves are for a one year or 18 month time period, depending on the country. The touch points through that grant period where we're gonna be speaking to CHAI and PATH and asking for some indicators of how the work is going, will be at three months, six months, 12 months, and then 18 months. So I think we'll have a reasonably frequent sense of how the work is going, what government's priorities are, whether the government is bought into the sort of advice and recommendation analysis that they're doing.

I suspect that some of the decisions themselves on the government side will need to be made very quickly. So I don't think it's the case that it's gonna be years until we see the actual results of some of the recommendations. But in terms of the ultimate impact on the health outcomes we care about, I think that would be a lot harder to understand and maybe that would be something that would manifest over a longer period.

Elie Hassenfeld: Right. So it's a 12 to 18 month grant, it's May right now. Maybe by the end of this calendar year we'd start to have some sense of how it's going, and then perhaps [00:27:00] by the end of the next calendar year, we won't know whether the health outcomes that we were aiming to improve have been improved, that's a longer term, but at least the effect on government decision making, government action, or other donors would be more salient probably by the end of next year.

Dan Brown: Yeah, absolutely. Yeah.

Elie Hassenfeld: Based on your response on what a good outcome would look like, what I'm imagining a bad outcome is the absence of any of these in a material way. So there's no clear signal that the government's really engaging. There's no evidence of cost savings. There isn't evidence of an impact on government regulation or what they're writing into law, and either we don't believe the gaps that we've seen and donors are not stepping in, or they don't seem compelling enough to fill and so in some sense we've done this work, but nothing comes of it.

Dan Brown: Exactly. Yeah, that's exactly right. And that's always I guess the like risk with some of these high leverage, non programmatic grants is, it is all dependent on whether the programmatic outcomes followed from the initial investment.

Elie Hassenfeld: Anything else before we wrap up, anything you didn't say that you wanna make sure to add?

Dan Brown: I think the only other thing I'd add is that we funded CHAI and PATH, [00:28:00] these TSUs in six existing countries, and these are countries where they have a lot of buy-in and where they're ready to work right now. We're open to funding additional TSUs in the future. And that could include from CHAI and PATH directly, and we know that they're scoping potential work in other countries too.

But we know there are lots of other organizations, as I mentioned before, who are providing technical support like this to governments. We didn't do a really comprehensive investigation into other organizations as part of this work because of the need to move quickly. But if there are people from other organizations listening to this podcast who are well positioned to provide this kind of work, we are open to considering other investigations into TSU in other countries in future. And if that is the case, then you can reach out to us at research@givewell.org.

Elie Hassenfeld: Cool, yeah. Thanks, Dan. it was great to talk this through with you.

Hey everyone, it's Elie again. I think this grant that we discussed in this conversation is a really good example of how GiveWell has evolved in the last few years, and also how we're focusing on responding to current events in US government [00:29:00] cuts.

You know, over the last couple of years we've really broadened the scope of focus for our research team. We've grown the capacity of our research team. We've built relationships with more organizations and with more people in country. And it's given us the opportunity to be able to look at a grant like this and make the bet that it's going to be a good use of funding and a better use of funding than something else that's really excellent, something like funding our Top Charities.

And then at the same time, we are trying to leverage that additional capacity and knowledge and networks to be able to move quickly in a moment when there are really high needs. And I think this grant in particular is so focused on helping governments respond better, like empowering and supporting them to do a better job in what is going to be a really challenging situation.

And so I'm just really glad that we're doing this, and I think it has the potential to be really impactful. If anyone listening to this has questions or comments, you can always share them with us via an email info@givewell.org. We'd love to hear from you.

And if you're wondering how you can help in the present moment, the best way to help is to [00:30:00] give via our Top Charities Fund, which enables us to direct funds to the four organizations that are our Top Charities, or to the All Grants Fund, which we will direct to the highest impact per dollar opportunities that we can find. And it includes grants like this one to the TSUs, which carry more uncertainty and risk with them, because we know less about them, but we do think they're very, hopefully very effective and impactful.

Thanks again for listening and for supporting GiveWell, and for participating in this. We really appreciate it. So thank you.