CGD Podcast: Racing to Preserve Life-saving USAID Programs with the Project Resource Optimization Team

Erin Collinson: Welcome to the CGD Podcast. My name is Erin Collinson. I'm the Director of Policy Outreach at the Center for Global Development in Washington, DC. We're recording this in May, having witnessed an unprecedented upheaval of US foreign assistance over the last few months. There's still a lot we don't know about how everything will shake out in terms of the relative near-term programming of US foreign assistance and the future budget and appropriations process. We're seeing the early consequences of the US administration's actions, from terminating a large share of USAID awards that we're supporting all manner of development and humanitarian work, to dismantling USAID as an independent agency. 

Amid what has felt like a lot of grim news for development progress and the development sector, I've been heartened to hear about efforts to save critical aid programs, salvage development and humanitarian expertise and know-how, and rally support for continued US foreign aid leadership, often fighting misinformation in the process.

I'm excited to be joined today by two guests, Caitlin Tulloch and Rob Rosenbaum, who are members of a small team called Project Resource Optimization, also known as PRO or the PRO Initiative. That kicked into gear in response to the cancellation of USAID grants and contracts supporting humanitarian response and development programming around the globe. Caitlin and Rob previously worked at USAID and have some background in assessing cost-effectiveness.

That experience enabled them to hit the ground running to identify cost-effective programs that had been supported by foreign aid and were suddenly thrust into uncertainty. It gave them standing to point philanthropic donors eager to meet the urgent needs in the direction of opportunities to address immediate gaps and deliver impact. CGD is pleased to be hosting PRO's analytical work, though the team retains its independence and separately provides a free brokering service to donors.

I'm excited to talk with Caitlin and Rob about their motivation, methodology, what PRO is and isn't, their conversations at CGD so far, what lessons they're taking from this experience, and any bearing they think this work should have on longer-term conversations about the future of foreign aid. Caitlin, Rob, thank you for being here. Can I ask you to kick us off by telling us about your motivations for launching PRO and how you went about answering the question of what to prioritize in a fairly chaotic environment?

Rob Rosenbaum: Erin, thank you so much for having us here. We're really excited to join this conversation. We got started initially in February, shortly after the stop work orders and the executive orders were put in place that were halting all foreign assistance, including terminating or at least halting pretty much every award or grant that was coming through USAID. We originally started because there was a small group of donors who reached out to us.

They themselves had set aside some funding to try to figure out how could they most immediately solve some of the biggest challenges that were coming. They wanted some help deciding where to start and what to focus on. I think we pretty quickly realized there was likely a lot of other folks out there facing a similar sense of what I'd like to think of as analysis paralysis. With so many programs lost, how could they decide where to give a relatively small amount of funds and actually make a difference?

Fortunately, Caitlin and I, and a few other folks that we've been working with, have the tools and the skillset to help make those decisions based on our training and practical experience, reviewing and evaluating foreign assistance programs to help determine what programs will lead to the most lives saved per dollar. We started and decided pretty quickly to focus on life-saving programs, specifically in the health and humanitarian settings, and not on other important areas of work like education, for instance, because the need was so emergent and there were literal lives on the line.

Caitlin was really the mastermind behind our methodology and how we stood up the approach and the process to reviewing a huge, huge, massive dataset to come up with a small list of the most appropriate projects put forward. Caitlin, maybe I can kick it over to you to talk us through the methodology.

Caitlin Tulloch: Absolutely. As Rob says, I think one of the big challenges here was the number of different opportunities that you knew donors were going to be asked to compare. Comparing different opportunities in terms of how much progress they could make on a particular outcome is exactly what I've spent a lot of my career doing. What we're talking about is cost-effectiveness. How much impact per dollar you can have from among a defined set of alternatives.

The hard part is that all of a sudden, and this was in the days where it was really uncertain what would be canceled, what would get a waiver, what the long run fate of any of this was. There are thousands of possible projects that people could be picking up, and making a comparative judgment with that many options on the table is just really hard. We started by what we called a top-down process, where we actually downloaded the entirety of the fiscal year 2024 funding obligations from foreignassistance.gov. Lovely example of public data.

We started using the filters. It's actually a really rich dataset. I know CGD has spent a lot of time with it recently, too, but using the different fields of information that are in there. When we downloaded the FY24 dataset, it's 22,700 rows representing $35 and a half billion in obligations. When I say a big set of opportunities, potentially, I really mean it. It has all of these great features in the data. It's tagged by country, by sector, not only by sector, but by an even more granular field called purpose, which is breaking apart health into HIV versus malaria versus other things.

As well as the names of the implementers in many cases, and the size of the obligation in that year, and a description of the program. Our objective was really about identifying programs that were likely to be lifesaving and among the most lifesaving bang for your buck. We filtered first by the purpose for which a particular obligation in this dataset had been put forward. Out of the 90 or so tags in there, there was about 20 that we felt were really closely aligned to these health and humanitarian objectives.

It's not saying everything in there was exactly or equally lifesaving, but that probably the stuff we were after was within those 20 tags. We also filtered by country and focusing really on low-income countries, eventually we put a ceiling in at about $8,000 income per capita per year in PPP. That's going to help us focus on the context where both mortality and extreme poverty are likely to be concentrated. That's our top-down search. When we applied this, what I thought was a pretty aggressive set of filters, you still end up with something like 2000 rows and $9 billion in spending.

From that point on, we really had to rely a lot on what we've called the bottom-up search, where other members of our team were out there talking to people, interviewing experts, doing a media review of something like 200 articles about canceled programs. That helped us both triage the most important projects out of that long list of 2000. Sometimes also helped us identify projects that we might have missed by simply filtering a big Excel sheet doesn't capture all the information you would want.

There were projects that we might have missed that we were able to find thanks to this not-so-data-driven, much more human process. Part of that was talking to the implementers themselves, learning about what projects they had and information about those projects. From that point on, we could drive from this filtering into assessment. That's where we started to look at the cost effectiveness for implementation projects, the ROI, how much we would learn, what you could do with that information from research projects, as well as their efficiency.

Erin: I can understand how, with the dataset that large, you felt compelled to be pretty ruthless in defining this universe. Thinking about the realities, the likelihood of securing philanthropic resources quickly and at the needed scale to really deliver, but focusing on impact. As you noted, many of the entries on your first cut list are life-saving work, things like global health services or providing emergency food aid. I know you also left room to consider high-impact research. Can you say anything about the kind of research you see as particularly important to preserving, and why?

Rob: Absolutely. I'm so glad you asked this question, and that is in part due to my minor bias into this space, having worked in this field for quite some time. Also most recently, my role at USAID was at a place called Development Innovation Ventures or DIV, which was basically like the R&D shop for the agency where we had funding to go out and explore and try to identify novel approaches and technologies that could improve the way that US foreign assistance and programming in low income countries in general could improve both the performance and the efficiency with how it's done and to improve outcomes in whatever sector or field that it was working in.

I think it is safe to say, regardless of whatever happens from here, whatever the future of foreign aid may look like, we are entering into an environment with considerably fewer resources. As much as current approaches are on a cost-effectiveness basis, we can't mortgage the future to solely invest in the present. We have to keep looking to the future and to figuring out how can we do things better, how can new technologies help bring more life-saving services to people on the ground, and how can we make every dollar stretch even further?

The only way to do that is through rigorous evaluation and is through a discovery process that enables new ideas and new innovations to be able to grow and then to be really tested and rigorously tested in new and different environments. When you ask the type of research that we're going after, right now for PRO, we are focusing primarily on studies that have rigorous methods. Primarily randomized control trials, so there's some exceptions to that.

Looking at interventions and looking at studies where the outcome of that study and of that evidence generation, we think could fundamentally change the way in which development is delivered in that particular context. There is an additional advantage to some of the research studies that we're looking at, which is that they have a lot of their fixed costs already covered because they were already started under USA, but the study was halted. In some cases, that means that the participants have been identified, the design of the study has been finalized, it's gone through IRB.

In many cases, they've already done a baseline assessment or even a midline assessment. It's a relatively marginal contribution of funding that can lead to the full impact and the full evidence generation from those studies when the alternative is that we lose everything and then go back to square one. In the most emergent sense and what we're looking at from a PRO perspective, I think we're really trying to target these places where there is a large leverage play for investment, and also having a fairly high bar for the value of the evidence that's generated on the other side.

Looking beyond PRO and beyond this emergency all-hands-on-deck approach, I think it is really important that we, as a collective community and development community, continue to invest in research and continue to invest in innovation. If we don't, we're just going to be stuck in the same cycle as we are now. I think it's really important that we have a pathway and a vision and, frankly, a light on a hill of funding opportunities for folks who have new ideas who are living in these communities that may be able to do things even better than we're doing now and to have a pathway to getting there.

Erin: You have had success in pointing donors to programs that face disruptions from the initial pause and cancellations of US foreign aid. I want to give you a chance to talk about some of the connections you've helped facilitate and what's happening as a result.

Caitlin: This has been one of the really wonderful parts of doing this work. As an analyst, sometimes amidst a crisis, you're not sure if you're the most useful person around. Actually, if what we can do is help tee up information that's actionable for people who have the resources to step in, that's hugely important. We've done this as a brokerage model. Given how quickly we've needed to do this, we're not coming up with a stack ranking of exactly the expected cost-effectiveness of each of the projects in our list.

What we're doing is trying to identify things that pass a certain bar, and there's going to be variety in there between health and humanitarian, between Africa and Asia, between research and direct implementation. From the point of our list right now, there's more than $80 million across about 40 urgent and vetted projects there. We're then talking to potential donors about their specific interests. There are some who say, "Look, I'm here to just maximum lives saved per dollar, tell me where that is, and I'll be interested."

Others bring a particular lens and a strategy from their foundation that they still need to work within, even in this exceptional moment. A nice example of that is, and I think this was featured in an NPR story, CRI is a foundation who focuses traditionally on child health in Africa. They came to us and were asking, "What are some of the projects that we can invest in where this would have a really high return on those core objectives we focus on?" Also, what can we do in this moment to ensure the continuity of services in places with really high need?

I think they had a very good understanding of the fact that this was not only an investment that was generating outcomes as it went, but also an investment to preserve capacity for whatever the future holds. With those things in mind, there was a project on our list. It's a project doing large-scale child malnutrition treatment and health services in an area of Mali that has very, very low coverage of government health services. Where there's enough displacement and conflict, that actually the implementer had expressed to us.

They were really concerned that if they were forced to fully shut down, it might be difficult to go back. That's having worked in a humanitarian organization, this dynamic of access and community trust is important, and it gets built up over years. We were able to make the case to CRI that this particular project was good, both by the numbers, but also for this kind of sense of an investment in preserving capacity for the future in a place where the need was really high and seemed likely to remain high.

That was one of our first ones, I think, really gave us a lot of confidence that we were putting out something that was resonating with people. More recently, I think last week, we started working on a match for a large-scale health project in Burkina Faso, also doing a lot of health services in a context with a lot of displacement. In that case, it's not a foundation but a wealthy individual, actually, who talked about that in their lifetime they expected to do a lot of giving.

With the current crisis, maybe more of that's going to happen sooner rather than later. That they've been able to mobilize the resources to put in to help both preserve human life and preserve the capacity to deliver services through this chaotic period. There's more stories like that, hopefully even more to come, but it's been a really wonderful thing as a match between us as analysts and the people who can make that analytical work mean something.

Rob: There's one project that I think represents a nice example of a few of the different elements of the work that we're doing here, and it takes place in Mozambique. It's from an organization called John Snow International or JSI, and it's focused on immunizations. I think one piece of our methodological approach that we maybe haven't touched on as much yet is how we've been also working with implementing partners or the large NGOs that deliver this work.

To really pare down a lot of the programs that were originally built up into something that is much more tractable and really eliminates a lot of the pieces of the program that aren't directly connected to the outcomes that we're specifically solving for in this context. In this case, the outcome that we're primarily focused on was increasing immunization rates. We were able to work with them to really pare down what they were working on and create a targeted program that would help deliver over 700,000 immunizations to children under one.

I think what was exciting about working with this is that in addition to being able to identify a project that was so clearly necessary and had such clear ability to track its cost effectiveness, we were also able to pull a group of donors together, led by Founders Pledge. They helped lead a process where we were able to crowd in four separate groups of donors all together to be able to make this happen. I think the other thing that really stands out to this, is when we were originally talking to the JSI team about this project, and they were explaining why it was so important.

Being able to enable their government to inoculate this many children is obviously important. It would save a lot of lives. The other thing that they highlighted as a huge amount of their work is really technical assistance in supporting the Ministry of Health in delivering vaccines. Everything from managing the supply chain, to actual distribution of some of the antigens for some of their campaigns, to supervision of some of the health workers in the clinics. Basically, what they told us was this is all stuff that the Ministry of Health can do, and all staff were working towards helping them get there.

If we pull the plug right now, if we shut this down overnight, then we lose all of this potential to do this handover, and we lose all of the capacity to be able to enable this program to work over the next year and to reach this many children. As Caitilin mentioned in the Molly example, what we're really buying here is not just the impact for those children, but also the runway to be able to do a strategic and thoughtful handover of that program rather than just pulling the rug out from under it.

Erin: Candidly, I think CGD was quite excited to provide a platform for the work that you're doing because we believe it's important. Also, because many of our researchers were eager to delve into the data and interrogate your methodology and even step back and reflect on the value and limitations of cost-effectiveness evidence writ large. Can you share anything about what you hope comes out of this partnership with CGD?

Rob: Sure. Our number one goal of this partnership is trying to get as many of these projects funded off of our list and to get as many of these programs preserved as possible. That remains our North Star, that is our reason for being, is to try to preserve as much of the most life-saving and cost-effective work that was happening under USAID as possible. We are so excited to be working with CGD to be able to have this platform, to have your incredible comms team able to help support this mission, and to hopefully be able to see more of this work happen.

We're also really excited to engage with the thought leaders within CGD. You had talked about interrogating our methodology. We actually love that. In fact, some of the most fun conversations we've had so far have been just geeking out on the methodology. They were able to find a couple gaps in it, and we've been able to adjust our methodology to make sure it's super sound. That we really are achieving our goals and doing it in a truly structured and transparent way.

We're also really excited to be able to use some of the lessons and share some of our learning from this experience from trying to respond in this very, very rapid and emergency type of a setting to understand what it might portend for the future of foreign aid. To engage with thought leaders within CGD on this vision for what foreign aid could look like in the future, and how some of the work that we're doing here might help inform them.

Erin: We've talked about how this PRO initiative is a short-fuse effort to fill gaps that can give programs funded by USAID a chance, whether that's buying some time to move forward, transitioning to partner governments or potentially with different sources of funding or maybe just winding down more responsibly. I wonder if you can also help us understand what the initiative isn't, because I know we all see the importance of continued USAID leadership in providing foreign aid.

Caitlin: One thing it really isn't is it is not a set of opportunities that is going to be here for very long. One of the things we're tracking is what we're calling the cliff dates for projects on our list. It's clear not all of them are going to fall over the cliff immediately, but by the end of summer, I think most of them will be forced to shut down because of the absence of funding. I don't think we see this becoming a clearinghouse in which new projects come into the pipeline and go back out. It's really about this one critical moment and providing a bridge for capacity and for really life-saving services that very clearly has a sunset date.

Another thing, it's not, and I just want to be as clear as possible about this. It's not a replacement for government spending. We had to look through all of the different things that USAID was funding, public goods that I personally, as an analyst, use things like FEWS NET. There are so many things in the USAID portfolio that fundamentally a ragtag group of economists, no matter how scrappy we are, we're not going to go out and find the level of funding you need to keep something like the Famine Early Warning System, FEWS NET, ongoing.

That is a really unique kind of thing. It represents collective action. It represents a true public good in the sense that it benefits people, but the cost to any one individual of putting together that level of data would be way beyond the benefit they individually get it. It's a coordination problem to all of these things. We are not raising $2 million, $3 million at a time, going to be able to cover things like that.

I worry that we're going to really see how much of the core infrastructure of not just service delivery, but data systems, intellectual property research, and development. There is a reason, as an economist, you have governments to do specific things that private markets on their own are less good at doing. We're not claiming to fill that gap, and there's going to be a huge, huge gap left that will need to be figured out.

The last thing I would say that we're not is a normative judgment on what kind of sectors or types of programs should be covered by foreign aid in the future. We have made really aggressive triage decisions. I spent 10 years of my career working on the cost-effectiveness of education programs. There are no education programs in our list, and that's not because I think education is not a good or a cost-effective thing to invest in. We're grappling with a world where tens of billions of dollars of funding have been wiped off the map. In that case, we've had to radically, radically triage.

This is about appealing to donors who we think may step up and fill these particular gaps. As I say, that is different from a statement of belief about what was good or valuable in USAID's portfolio. In general, when I work on cost-effectiveness analysis, I say, "Economists are going to be good at telling you how to achieve something, but that's not a proxy for what is a good thing to try to achieve." Again, to the role of a government deciding to invest in basic education around the world, things like that. There is so much of value there and our list is not going to capture all of it.

Erin: We've all alluded to the fact that this has just been a period of incredible disruption for US foreign aid. I think most of us hope that things will start to move in a more constructive direction at some point in the near future. With that, we're seeing clear interest in reforming US international assistance. As we start to consider the future, what lessons from this exercise would you hope to use to inform that reform effort, that re-envisioning of foreign aid?

Rob: I think that one of the things that we're seeing right off the bat is the importance of incentives in how we design foreign aid programs and how much they matter. I think CGD has done a really good job, Rachel and others have written extensively about this, about the Christmas tree or the kitchen sink or whatever metaphor you want to use, approach to foreign aid that has become fairly prolific. Particularly in a lot of bilateral foreign funding or foreign aid development funding.

Which is to say that there is a project that may have started with one particular goal, let's say delivering immunizations we're treating children with malnutrition. Pretty quickly spiraled as more things got added on to it based on other interests or other goals that that program had. While those things aren't individually bad, when taken together, it can lead to very, very large awards. They are complex in managing and also very difficult to scale.

You're trying to scale so many different things together that it becomes much harder to reach a larger number of people with the core services that you know are going to drive the largest number of impact. What we've been able to see in our engagement with the implementing partners who have been running these programs for many years is how quickly they've been able to adapt and willing to adapt, recognizing the changing circumstances, and being able to say we're not going to be able to necessarily deliver all of these services anymore.

Here are the things that we know are most directly tied to impact on the outcomes that you're asking us to focus on. Here's how we can take a project that may have had programs A, B, C, and D and to say, okay, we're just going to focus on A and B, and we're going to really pare this down to those core functions. Also, we're going to be able to bring the price tag down, which for PRO is really important because we're looking at private philanthropy, which isn't able to cover as large of awards as the government was able to cover.

That isn't necessarily just to talk about a particular program here or there, but just to say that it's very clear that the incentives of the funders who design these programs and commission the implementing partners to go and do it really matter. That there is potentially a world in the future where we could get more simplified, we could get more focused and hopefully, as a result, drive the implementers who are delivering this work to being able to reach a much larger scale with those particular interventions.

Caitlin: I think it's worth emphasizing that when we're talking about simplified programming in the way Rachel Glennerster has written some pieces about this recently. That doesn't mean we're talking about vertical delivery of single interventions exclusively in the health space. I think we want to be clear, we're not saying moving all the way to the other direction. A corner solution on the other side is the optimal thing. In practice, the list that we've put together includes things that really don't fit that mold.

The JSI project, Rob, that you were talking about, has what I would think of as system-strengthening components. It's not like actually the JSI people are going out there and putting shots in arms. What they're doing is providing the very targeted technical assistance, the improvements in supply chains, the monitoring, the micro targeting to make that happen, and in a way, I think you can say confidently, is enabling more vaccinations than would have otherwise happened. That is an example of a system strengthening piece, but it's so tightly focused on this very specific delivery issue of a really cost-effective service.

It doesn't always have to be single services, too. Immunization is lovely. It's easy to write a spreadsheet about the benefits of that, but that Burkina Faso project that I mentioned is bundled health programming in really high-need areas. If I wanted to do a model of that, it would be relatively easy to look up the cost effectiveness of Kangaroo Mother Care, but there's no health organization in the world that delivers only Kangaroo Mother Care. They don't parachute in right after your baby is delivered, hand you the baby wrap, and then walk out. You do that as part of a bundle of antenatal services.

Again, I just want to emphasize, there's a middle ground between these Christmas tree, kitchen sink programs that have everything in them and this incredibly slimmed down version that I think is a caricature on the other side. There's really productive tension in that. Asking questions about what has the best marginal return on funds, that's not always this incredibly pared-down single intervention service. It can be more than that. I think that's the area that I'd be really excited to explore and to try and incentivize in the future.

Erin: I know you've been warned that we try to end on a lighter note here. My final two questions will reflect that. First, what's one weird or funny or just memorable thing that's happened to you on the job?

Rob: I'm going to dig way back in the memory trenches for this. Very early in my career, I was living in Cape Town, and I was working in the townships outside of Cape Town for an organization called Children of South Africa. Through that role, we were working with a lot of different children's homes and community-based organizations caring for orphans and vulnerable children in the townships outside of Cape Town. My job was basically to bounce around between all these different organizations and do grassroots consulting.

Helping them build their books, understand how to manage funds, help them access resources, crowd in funding, bring volunteers, those kinds of things. I was constantly on the road going between them. The car that I was driving was this Toyota Tazz. For those of you not familiar with the glorious Toyota Tazz, is basically a lawnmower engine with a box around it. There was one day when I was visiting one of the organizations and was sprinting off to go to the next meeting and reached into my pockets and couldn't find my keys.

I looked everywhere. They weren't there until I got to my car and realized I had left them in the ignition. Naturally, I had to figure out how to get them. I started asking around if anyone happened to know how to break into a car. Nobody nearby us did. Somebody knew somebody who knew how to break into a car. We tramped through with a growing group of very interested bystanders.

We had a fairly large contingency of people walking through the streets of this township to find A, a coat hanger, and B, anyone who knew how to use it. We got back to the car. First off, nobody actually knew what they were doing, but everyone took turns breaking into my car until we finally got it. Finally got into the car, was able to start the car and go off to the next visit. I guess it taught me that it takes a village can really apply to almost any context in your life.

Caitlin: Oh, no. Now, I feel like I should have gone first because I don't know that I have a story that involves collective car burglary. I was in Somalia a couple of years ago doing one of the coolest things anybody ever let me do, which is running a workshop with state and federal Ministry of Health officials trying to optimize their package of maternal and antenatal care services using high-dimensional constrained optimization. We did literally have a grad student on the phone from the US who was rerunning his model as we like fed him parameter values.

It was an incredible experience. Again, they let me just go to Somalia to do constrained optimization, which is my favorite kind of optimization. We were in this conference room speaking to supply chain experts and antenatal care experts, health officials from multiple organizations in the government. You started hearing this banging on the window. Eventually, we realized that there was this troop of monkeys that was actually trying to break into the conference room to take our food.

One of my colleagues I was there with, who's kind of a neonatal care expert, was the most vigilant I've ever seen. She'd been very relaxed until that point, but clearly she was not having it with the monkeys. She spent much of the rest of that session standing at the door, making sure-- the monkeys did actually manage to break into the hotel lobby at some point and ran around. We in the conference room were safe. I have a bunch of pictures out from that conference room of the monkeys trying to break in for whatever reason.

Erin: Those are great stories. Finally, if you could instantly change any one policy in the world, what change do you think would do the most good?

Rob: When I think about what has happened at USAID and in foreign policy more generally, I feel like it can't be divorced from the context and frankly, the political economy that has brought us here. Our political system right now incentivizes and solves for extreme positions and extreme candidates in our quality. If I could wave a magic wand or change any policy, it would be to eliminate all political gerrymandering in the US, at least this is for US focus.

Or put differently, I would require that all political jurisdictions are drawn by a nonpartisan process. Particularly when presented with the facts of how much it costs and what we accomplish, I fundamentally don't believe that the median voter and the median American wants what we've seen happen here with USAID.

Caitlin: One of the things that has always struck me looking at USG funding foreign assistance broadly, is it's among the most generous in the entire world. As a fraction of the full United States budget is just tiny. It's less than 1% of the budget. This has been brought home to me by the fact that the cost of R&D, for example, the F-35 Lightning fighter jet is about $400 billion. That's just the R&D, not the production costs. Per plane, it is something like $80 to $100 million per plane that we now procure, which is roughly the size of some of the largest flagship USAID programs at a country level.

Again, 400 billion over many years indeed, but that's multiple times the entire budget of this federal agency. I believe in trade-offs and the value of numeral goods and making us think about things. My proposal has always been that we should have to denominate all of the appropriations for foreign assistance in equivalent F-35 units, and talk about how many F-35s, 10% of an F-35 that we're giving up to fund this country-level project.

Rob: At some point, I think we may need to talk about the door or the wing or something.

Caitlin: Exactly.

Rob: Some subdivision.

Caitlin: “This is one piece of cockpit glass worth of foreign assistance.” That is my proposal for making people think more seriously about the trade-offs.

Erin: Thank you again to Caitlin and Rob for joining us today. You can learn more about the work of the PRO team by visiting CGD's website, CGDev.org, or the PRO initiative's own site at Proimpact.tools. Thank you so much for listening.