 Good afternoon everybody, thank you for coming and welcome to the US Institute of Peace. My name is Joe Hewitt, the Vice President for Policy, Learning and Strategy. And we are delighted to have this event here today about something that I think we can all agree is important and that we haven't quite figured out yet. But with the help of two really important studies, we think we're gonna make some traction in understanding a little bit more about what works with regard to youth programming interventions. Let me just first say something about the center I lead with this amazing team that I lead at USIP, the Center for Policy, Learning and Strategy. What we are trying to do at USIP is help it reach its goal to be a convener for the purpose of shaping our field with regard to the best knowledge and the best learning that we have. And a big part of that mission is bringing together audiences like the audience here today so that we can have a useful exchange about what the latest knowledge looks like and a particular area of interest that we all agree is important, what we can learn from that and then how we can convert that into what's next. And so that's very much what we're up to today while we talk about youth programming and two projects, one in Afghanistan and one in Somalia. When we think about the problem of violent extremism or generally violence in a lot of countries that are considered fragile or conflict-affected countries, we often look at youth populations as a place where our interventions can be most useful. We know that youth populations can often be vulnerable to extremist recruiting. We know that or at least we have a sense that where there is poverty and the absence of opportunities for economic prosperity, those conditions can enable more support for violence, less support for the existing government. And so therefore where we can counter those conditions and do something to better the conditions for youth populations, we might be able to do something to turn the tide on those conditions. There's really compelling logic for why it is that youth programming interventions that are designed to improve economic prospects, designed to improve things like vocational training and secondary education should work. The logic is very compelling. Poverty and the absence of economic opportunity leave youth populations vulnerable. So employment and improved economic prospects should help shift away support from insurgent groups and towards the government. And more than that, we know that the opportunity costs to join insurgent groups go up for youths with improved prospects and improved lives. That's the logic. But what do we know about how this works? Does it work and what conditions does it work and so forth. So to help us with that today, we are very fortunate to have Besa Tesfaye who is a senior researcher at Mercy Corps to tell us about two important projects that Mercy Corps implemented to evaluate youth programming in Afghanistan and Somalia. USIP was very pleased to provide the funding support for these projects. The projects, as I understand it, were carried out in partnership, at least one with the Institute for Peace and Justice at the Joan B. Croc School for Peace Studies at the University of San Diego. And what we will hear about are the two studies. So they're slightly different, one in Somalia and one in Afghanistan. And then we'll be joined by a panel who I'll introduce to you once we get everybody assembled after Besa's presentation to open up a conversation about what all this means. We've got a great audience in the room and so at the same time, I wanna make sure that we leave plenty of time for your questions, your reactions and your comments. And so while you're listening to the conversation, compile your questions and think about how you'd like to contribute to the conversation. And then hopefully when we're finished, you'll walk away with something new that you can take back to your colleagues to help improve the kind of programming that we do in this area. So with that, I am really delighted to turn things over to Besa. If you could please join me in welcoming, Besa Tesfaye. Thank you, Joe. Good afternoon, everyone. It's really a pleasure to be here presenting on these two studies, which is actually the culmination of four years of work on the part of Mercy Corps and many other organizations. So it's very exciting to be at the stage where we can finally share results. I wanna start first by extending my gratitude to the US Institute for Peace for not only hosting and organizing this event, but as Joe mentioned, also being a big proponent of this research, actually funding these two studies and generally being an advocate for evidence-based programming. So I commend you for that and thank you. I also wanna thank the other panelists who will be joining us, who are the real experts on these two countries that we're gonna be talking about and that can help contextualize these research findings on these two countries, Afghanistan and Somalia. So I'm gonna take the next 15 minutes or so to set the stage for this discussion by sharing the results from the research that we did in Afghanistan and Somalia around youth violence and stability. And then I wanna open this up to a conversation about how this can inform programming and policy on these topics. So to begin with, I wanna say a bit about what motivated us to do this research. Over the past few decades, the aid community has witnessed two very important trends that have made us really question where we should be investing our resources and funding. The first trend is this one, which shows that generally global levels of poverty have decreased in recent decades, but at the same time, we see a gap in terms of development outcomes for countries. This graph shows poverty rates between 1981 and 2005. And as you can see, countries that have experienced significant levels of conflict have not been able to reduce poverty levels. Poverty has actually increased in those countries. Countries that have not experienced conflict have reduced levels of poverty in that time. And to give you a specific example, Somalia, one of the countries that we're discussing, Somalia's per capita income has dropped by more than 40% since the 1970s due to war. So this first trend just points out that in order to deal with the issues of poverty and underdevelopment, we really have to address conflict. The second trend that we have observed is this, which is a massive shift in the demographic profile of many countries that we're working in. This is specifically for Somalia in 2010, where roughly 70% of the population is under the age of 30. And this is by no means unique to Somalia. You see this kind of youth bulge all over the world in places like Sub-Saharan Africa, Southeast Asia and the Middle East. According to the UN, already more than, there are more young people in the world today than any other time in history. Roughly 1.8 billion people are between the ages of 10 and 24. And the vast majority of them live in low income countries that are affected by conflict. Now this is not necessarily a bad thing of itself, but this coupled with a number of other factors like poor governance, lack of economic opportunities, unemployment and so forth, can lead to a very volatile situation as we've seen. And this is a key risk in a lot of conflict affected countries as Joe mentioned, because we know that marginalized youth are often the ones to be recruited into violent movements. So this second trend has generally pointed out that not only do we have to focus on conflict, but we also have to put youth at the center of development programs in order to address this issue. So given these trends, a major focus of many policymakers, development actors in general, has been how to prevent young people from joining armed groups or armed movements. And the conventional wisdom to date on how to do this was that if you gave young people access to jobs, they'd be less likely to join in violent movements. And this has been a largely unquestioned assumption resulting in large investments in job programs aimed at building peace. In fact, a recent review by an organization in Berlin found that the UN, the World Bank and ILO had implemented and funded roughly 2,400 job programs aimed at promoting stability since 2005. So it's a very large investment that's founded on this assumption. There are however many problems with developing a response to conflict based on this assumption. And one of them is that in practice, there's very little evidence to support this claim that by giving young people access to jobs that you will actually be able to dissuade them from joining violent groups. And the reason why there's little evidence is because there have been very few studies attempting to study this at the micro level, looking at individual decisions to support or engage in violence. Secondly, some studies that have tried to tackle this question have been conducted mostly in post-conflict environments or generally stable and peaceful environments rather than in contexts where violence is relevant. And so the results might be less relevant in those contexts. And lastly, even when studies have tried to study this relationship, they've only looked at the relationship between the programs and the outcomes without really trying to understand the mechanisms behind the outcomes. So over the past few years, Mercy Corps has really tried to tackle this question head-on and tried to address all of these limitations and the research that we've been doing. And we've focused specifically on Afghanistan and Somalia to very conflict-affected countries to really measure the impact of some of our youth programs in those two countries and be able to provide some evidence around this issue. To do this, we carried out impact evaluations of two programs. The first program that we looked at was a program in Somalia known as the Somali Youth Learners Initiative. This is a U.S. government-funded program, a six-year program that had the broad objective of supporting or promoting stability in Somalia. And the program had two key interventions that we were interested in understanding through our research. The first was to increase access to secondary education, which translated into working with the Ministry of Education in Somalia to build and rehabilitate schools across the country. We also focused on improving the quality of education, so we did a lot of teacher training and even provided salary support to teachers to incentivize them to teach. And we did general capacity building for the ministry so that they would be able to carry out these activities in the long term as well. And then in addition to that, there was another component to the program that focused on civic engagement for youth. And this included providing young people with training on things like leadership, communication, nonviolent action, and then actually giving those youth the resources to be able to mobilize in their communities and carry out civic engagement activities and advocacy campaigns. In Afghanistan, we looked at another program which was known as the Invest Program. This was also a longstanding US government funded program. We had implemented this first in Helmand and then in Kandahar. The objectives of this program was really focused on improving the economic well-being of vulnerable youth. And to do so, we focused on vocational training. So we targeted young people who were unemployed, who had been displaced, who had no source of income, and provided them with the ability to enroll in vocational training courses that span from three to six months. And then once they had gone through those vocational training courses, we also provided them with additional soft skill training to prepare them for the workforce. So that was the program. But in addition to that, we were also interested in understanding how the introduction of cash through an unconditional cash transfer might also affect their attitudes around support for violence. So although it was not part of the program, we also provided some young people with a one-time cash transfer of $75, which is approximately four months of wage for an unskilled worker. So it was a pretty significant cash transfer. In terms of methods, we essentially did two different types of evaluations, similar but different. So in Afghanistan, we carried out a randomized control trial, which is, in some ways, the gold standard for evaluation. And we had three different groups that we were comparing, young people who just were involved in vocational training, those who received cash, and then those who received both vocational training and cash. And in addition to this, we also had a control group of non-program participants in the research. And we looked at the outcomes of these young people at three points in time at baseline before the program began, at end line when the program ended, and then six to nine months after the end of the program. And our sample was roughly 2,600 youth from Kandahar where the program was implemented. For Somalia, we were not able to randomize, so we used a quasi-experimental evaluation design. We were comparing youth who were in the educational component with those who received both education and vocational education and civic engagement. We did a one-time survey in March 2017, a few months before the program ended, and we matched youth who were in these three groups to make sure that they were comparable or very similar in terms of their socioeconomic and demographic background. And our sample size for this study, there were actually two studies that came out of this. In total was about 2,000 youth in 11 locations across Somalia, including parts of South Central Somalia that had been under the control of Al-Shabaab a few years earlier. So what we were trying to measure was support for political violence or participation in political violence. And the question that we often get is how do you actually measure somebody's level of support or participation in political violence? And the answer is we use surveys to measure those attitudes. And there are obviously some flaws to this. The biggest one is this issue of social desirability bias, which is the tendency of people to answer questions in a manner that will be viewed favorably by others. And you can imagine if you're asking someone whether they support Al-Shabaab, for example, most likely they'll say no, even if they do, given that they're being asked this question by a data collector represented by Mercy Corps. So it's a very challenging problem that we've had with surveys. So one thing that we've been trying to do to get over this bias is introduce new techniques known as survey experiments. Which are basically indirect ways of eliciting truthful responses on sensitive questions. And one way that we've done this is a method called the random response approach, which is essentially we give the respondent a spinner similar to the one that you see on the screen. And the respondent will answer the question based on where the spinner lands. So if it says answer no, they answer no. If it says yes, they answer yes. But if it lands on answer truthfully, they will actually be told to answer the question truthfully. And because the respondent is the only one that sees where the spinner lands, they have less of a disincentive to answer truthfully when it actually lands on answer truthfully. And statistically there's a way that we can actually compute the truthful responses based on what we get. So this is like one of the new ways that we're trying to get more truthful responses on these questions. So once we carried out these studies and looked at the results across all of the different contexts that we were researching, we found some very interesting patterns that you see in this table here. First of all, we find that the interventions, the standalone interventions that just focused on vocational skills or education in general were not effective. So secondary education had mixed results. So in some cases it actually increased support for violence. In some cases it reduced support for violence. So across the board it was not consistent and therefore it's hard to conclude what the effects are there. For vocational training, the results were more conclusive. They actually, vocational training had no effect on people's attitudes around political violence in Afghanistan. And then when we look at cash by itself, that actually had a negative effect. It increased support for political violence six to nine months after the end of the program in Afghanistan. The second trend that we see is that the interventions that are more multidimensional that included two different components, so secondary education and civic engagement or vocational training and cash, those interventions tended to be more successful at reducing support for political violence. So this was encouraging in general because it showed us that there are certain types of interventions that can move attitudes around political violence, but what it didn't tell us was why these interventions actually worked in some cases. So to answer that question, we did some additional research, both qualitative and quantitative, and we found that the common factor in both cases was this issue of governance. So the interventions that were successful were the ones that actually shifted people's perceptions of government or their perceptions of agency and ability to influence government. So education and civic engagement in Somalia actually increased young people's confidence and nonviolent tools to affect change while vocational training and cash improved people's perceptions of government's responsiveness. The latter was a bit surprising because we didn't really understand why cash might have that effect on people's perceptions of government, but we believe it could be because people actually credited the government for making these interventions possible and therefore saw the government in a more favorable light. So to summarize overall, the studies indicate that development programs can move attitudes related to violence, and that's something that we are very, very encouraged about, but there are some important caveats. The first is that education or TVET vocational training by itself doesn't seem to address the underlying drivers of violence, and in some cases can actually be counterproductive. What is more promising as we saw in our research is this combination of interventions that address both lack of skills and lack of opportunities as we saw in Somalia, or that had both long-term and short-term focuses, such as the program in Afghanistan which gave young people cash to meet their short-term needs but also invested in their long-term capacities through vocational training. And then lastly, we're seeing that governance is really a key factor in a lot of programs and that we can actually shift perceptions around governance without doing traditional governance programs, traditional governance programs being democracy, human rights programs, even with these types of more economic development programs, we can actually influence people's perceptions around governance and that seems to be key for shifting attitudes related to violence. So I'll leave it there and invite Joe and others to come up, but I just wanted to say that this research really was an effort of many people, so I just wanted to acknowledge some of our partners. For the research in Somalia, we worked closely with the Joan B. Crock Institute for Peace and Justice. We also worked with a research firm based in Nairobi known as Samuel Hall. For Afghanistan, the principal investigators on this research were Jason Lyle from Yale University, Kosukei in May, and Yang Yang Zhu from Princeton University. And last but not least, we also have to credit our field teams that allowed us to look at these programs and evaluate them. So I just wanted to highlight those people, but thank you very much. Beza, thank you for that. So you have given us a lot to dig into. And I think the way we'll kind of try to organize this conversation is I'd like to ask maybe just a couple of follow-up questions from the presentation and then get a little bit into the country context as well. And so we'll move down the table in order. So before I do that, let me introduce everybody that you see here and then we'll turn to opening up a conversation. So Beza Tesfaye, you know. Immediately to the left of Beza is Abdu Lahai Halaki, who is a senior policy advisor at the International Rescue Committee on Africa. We're really delighted to have you. He is gonna be able to tell us a little bit about the Somalia context. And then to Abdu Lahai's left is Scott Warden, who is the director of Afghanistan Central Asia Programs here at the US Institute of Peace. And then to Scott's left is Aubrey Cox, who is a youth expert on youth programming here at the US Institute of Peace. And so what we'll be able to do is move our way down the table, talk a little bit about Somalia, a little bit about Afghanistan, and then Aubrey's gonna help us put it all together. No pressure on you to tell us a little bit about how all this fits into what we know generally about youth programs. And then we'll open it up to you and we'll get a conversation again, your reactions, your questions that you might wanna pose to the panelists. So, Beza, I'm really curious behind all of these important findings, you were very careful to point out that in the end, what we know is based on these surveys that get done where we're asking really sensitive questions to respondents. And so you told us a little bit about that experiment that might induce Anand's honesty. And because the person asking the questions doesn't know what the spinner landed on. So I'm gonna press a little bit on that point because I think this is a foundational kind of question about the believability of the data. And so I wanna try to act like a critic here, although I don't think I am really, but I'm gonna act like a critic. Why wouldn't somebody just say that they're never interested in supporting an extremist organization, regardless of what the spinner comes up to? Yeah, why wouldn't they, because as I understand it, the person with the clipboard doesn't know what the spinner lands on. So somebody could still, in a sense, disobey the rules inside that research design. Is that something that you guys thought about a little bit? Or tell me why the concern that I'm raising is something that can be addressed? Yeah, that's a good question. So we've been experimenting with that approach, the random response and other survey experiments. And we've actually found that the random response is really effective because we are able to compare responses from the same respondents when they use that, versus when we just ask the question directly. And we see that in general, people are more forthcoming. They're more likely to express actual support for violence when they have the confidentiality that comes with the spinner than when they don't. So we can actually, for the same person, show that they're more comfortable when they have these experiments. That said, it's not guaranteed 100% that they're not gonna maybe understand what's going on and try to essentially not comply with what we ask them to do. But the point is that by giving them the spinner and ensuring them, explain to them that the data collector's not gonna know what their actual response is, they don't have an incentive to not answer truthfully. So it's just based on that assumption. Great. A quick question just to lay down the facts a little bit. Did the beneficiaries in these programs believe the programs were being run by their local governments, either Somalia or Afghanistan, or did they think they were coming from Mercy Corps? It was a bit of both. So the programs were actually implemented by Mercy Corps, but we worked very, very closely with government counterparts. So the Ministry of Education in Somalia, the Ministry of Youth in Afghanistan. So for example, in Afghanistan, the vocational training courses took place in vocational training centers managed and owned by the government. It wasn't a Mercy Corps facility. So there was always that presence of the government in the programs. And then you can see what I'm doing. I just wanna sort of get some facts on the table that people might be curious about. One other thing, and this often gets asked about programs that are especially focused on transforming attitudes and behaviors with regard to insurgent groups or extremists. How do you know that your programs aren't filled with young people who would have never been transformed, who would have never been recruited by extremists in the first place? How do you know, in other words, that you're really just getting moderate type individuals and the extremists are avoiding joining your program at all? Wow, that's a very good question. So I think part of it has to do with our targeting approach. So for example, in Afghanistan, the vocational training program targeted specifically vulnerable youth. So youth who had been displaced by war, youth who were unemployed, who had very little sources of income. So by most definitions, these were the most marginalized and vulnerable youth. So if you believe that vulnerability, that type of vulnerability might make someone more likely to engage in violence than we try to target those youth in the research and the project. Great, great, thank you. Abdullah, you've worked on CVE policies and programs in all sorts of capacities in the Horn of Africa. You've seen these findings from Mercy Corps. Does the research differ or confirm the kinds of things you've learned when you talk with youth populations in places like Somalia or other places in the Horn like Kenya? Thank you, Joe. I think I'm gonna answer that question probably in a backward way. One of the things that working on CVE, particularly in the Horn and around Somalia, is, and I think this speaks to the larger problem or the gap within the CVE as an area of study, is lack of rigorous empirical evidence. I think one of the things that is coming out in this report is the level of integrity that has been maintained and the level of rigor that was employed in getting some of the data. So I think not very many of the studies that have come across really speaks to this issue with the level and the depth of rigor that we've seen with this. Most of issues that you hear about, whether it's about education or whether about why are youth in place are more vulnerable or susceptible to joining groups like al-Shabaab or even Daesh now increasingly in the Horn, a lot of it is largely extremely anecdotal. You'll hear the usual push-pull, the role of ideology, and then it doesn't move very fast. I think Kenyan government has, right now the whole process for CVE has been devolved. You have the national CVE policy, now counties are now devolved. I think the ones along the Kenyan coast areas that were really affected by al-Shabaab because of the nearness to the Somalia border, they all have some of this. But reading through most of them, it hits you that none, not none, but most of them are not anchoring really strong empirical evidence. So my answer to the question is yes and no, yes in the sense that not very many of them, most of them the gap is largely around lack of, you know, thorough empirical evidence. I mean, I can just give one example. The areas where al-Shabaab has recruited more in Kenya is largely along the northeastern part that is closer to Somalia and they are largely occupied by people of the same heritage and also the southern part that is largely Muslim, not largely but to a larger degree, but there are areas also, if you look at all the available demographic data, socio-economic data are areas that are fairly marginalized, right? But there are some part, the northeastern part, no, the northwest part of Kenya, that the data is almost the same for socio-economic data. But why are they not going? Such questions are hardly answered. So I think what I wanna say is thanks to Masiko, I think these opens up opportunities for us to begin treating CVE with the same level of thoroughness, the same level of empirical evidence if we want to shift the thinking. Because a lot of it, I mean everybody agrees, push and pull factors, they are baskets of issues. And I also just really enjoyed how the study coupled not just education, but it coupled education with civic engagement opportunities. And that is where you started seeing the shift in the data because they will start having different perception as far as the government is concerned. A lot of it is, okay, I'm not getting services from the government, I might as well go and join Alshabaab or Daesh, I'll leave it at that for now. That begins to get. That begins to get, that observation, to something that came up that I thought was interesting that Beza started to hint at, which is that they found some divergent findings in Somalia about the difference in contexts between places where maybe those basic services hadn't been established and then the results that they see versus places where those services have been established and then they get different results. So Beza, could you talk a little bit and help us make sense of how those results were very different and what you think that means? Sure, so I think you're referring to our results from Somalia, I mentioned they were mixed and specifically what we found was that educational services by themselves helped to decrease support for violence in south-central Somalia, but when we looked at Somaliland, the semi-autonomous northern part of Somalia, education actually increased support for violence, so we see these two divergent results and we tried to make sense of this, again, based on more qualitative and quantitative analysis and what is evident is that people's perceptions of government in Somaliland tends to be quite different because it's more of a functioning state. There have been services provided there for some time, so in some ways expectations are higher on the government than they are in a context like south-central Somalia where there have been very few existing functioning systems due to the war, so it's just a reminder that the context in terms of government capacity and responsiveness can also be an influencer of how successful these programs would be. Great. Scott, so on Afghanistan, so you have many years of lots of expertise on Afghanistan, there's some evidence put on the table here today that technical vocational education plus unconditional cash transfers has the potential, it seems, to transform attitudes about violent political opposition. Does that mean we've cracked the code on Afghanistan? What parts of the code have we cracked and what parts of the code sort of remain uncracked when we think about Afghanistan? So if you could sort of put this in context for us for Afghanistan and how this all comes together. Great, thanks. There's a lot of uncracked code, that's the answer. I wanna make a few remarks on this. I mean first, compliments to Mercy Corps and the study because in Afghanistan, other conflict affected environments, it's exceptionally rare to have an intervention that is measured with socially scientific rigor in such a thoughtful way and there's so much speculation as to what works and what doesn't based on anecdote, based on circumstance. So I think this is really commendable and important in its own right and I think a lesson and a model for kind of future development interventions for stabilization interventions to try to emulate this type of approach that we can get more information and more variables and this is a good start for that. So I wanna start by saying that but also then to segue into, but there are a lot more variables and I think that I see this in the larger body of literature and there is a growing body of literature just on Afghanistan on stabilization counterinsurgency approaches because fundamentally what you're measuring through this is not just its poverty reduction but what is the impact that it has and attitudes toward insurgency and attitudes toward governance and there are definitional questions but fundamentally that's the stabilization winning hearts and minds counterinsurgency effort which we invested several billion dollars on in Afghanistan alone. So I think that one thing I would say is while you have the kind of interesting counterintuitive finding that providing cash transfers actually over the, I would call it the medium term being only six to nine months or nine to 10 months increased support for the insurgency or reduced support for the government. So that's questionable and I think you gave some good answers for maybe why but overall, I mean this supports a certain common sense that the more interventions that you have the more aspects of the problem that you're trying to tackle by both vocational training and cash and I think you could add to that list the more likely you are to design something that's gonna increase support for the government if you do it in a thoughtful way but that said, the larger body of literature and I don't think that this goes against it is that those effects are one hard to find. What's the magic combination? Because of course you have one finding that says you have a negative result even though you're giving cash anomaly from the government. The other thing is that the effects are short term and so this doesn't have a two year, three year marker but overall studies on stabilization show that there is a short term benefit for well designed research and then it ends. And I think that raises the larger question of how do you get this to be sustainable and transformative over the medium to longer term. I think another thing the literature shows is that there is not a simplistic answer and you're not positing one here but there's not a simplistic answer of better economy, better wealth equals less support for the insurgency. There's a combination of social, political, security factors that go into this and they are shifting and changing and they shift and change a lot in a dynamic conflict environment. So when you think about that the task becomes a little bit more daunting and obviously we haven't figured it out in the macro sense because stability is I would say decreasing in Afghanistan. What to do I think this all points toward your last conclusion, one of your takeaways which is the governance issue. I see this intervention, other short term stabilization interventions as bridges from a highly unstable situation to something that's more sustainable and long term but if you don't have a buttress or a footing on the other side, it's a bridge to nowhere. And unfortunately that's often what you have because the underlying grievances, the underlying sources of instability which have to do yes with lack of security, lack of justice, lack of inclusion to local governments or lack of perception that a region is accounted for in national governance. Without addressing those things which are hard, which are longer term solutions, you're just gonna have a short term blip in support and then you might have a backlash in terms of well where was the follow through. And so that's I think where this is a useful kind of jumpstart, it's the beginning of a bridge but the larger picture is figuring out what are the governance institutional developments, democracy developments that need to be present so that this little intervention can have a longer term future. I'll pick up on this point about the importance of being able to do something quickly. What the Mercy Corps work suggests is there are some things that can be done quickly. Scott points out that that's not by itself sufficient to get to long term stabilization or long term successful society. But I wanna connect this to another piece of research that just came out that talks about this same issue in a different way and then I'll flip it over to Aubrey but just to reinforce this point. The Cameron Commission on Fragility and Fragile States that just came out talked about that one of the really important things that we have to start doing in Fragile States is figuring out how to find quick solutions first. Too often we're trying to devise development strategies that are highly complex, designed to try to transform societies over a five year plan and that's just simply too ambitious and you've gotta figure out how to get some quick wins. Well, one of the things that might be a lesson that's coming out of this Mercy Corps work is that it is possible to gain these kinds of wins, make short term transformations of attitudes in youth populations as a result of these types of interventions that poise a community or a society for what might be next. And of course, Scott rightfully points out, it's not clear what's next exactly but if you've in a sense created conditions in youth communities or youth populations where perceptions of governance have shifted slightly that maybe creates a political opening to introduce reform or perhaps make it easier to bring in populations that might have been spoilers. Maybe you've got some traction there to do a little bit more. So there's, I see this bridge between what Mercy Corps has documented here and this larger kind of strategic observation that the Cameron Commission just put out with fragile states. So that's just a footnote, I think, on this part of the conversation. Beza, I don't know if you want to respond to anything else from Scott before I kick it over to Aubrey or if you want to hold for a second, I'll leave it to you. Okay, so Aubrey, we promise that you would help make sense of all of this for everybody. You cover youth programming for USIP. You have probably some context to help us with here. And so for those in the room who aren't familiar with the general youth peace and security agenda and how UN Security Council Resolution 2250 defined that for us, can I ask you to tell us a little bit about that context and how this Mercy Corps work feeds that agenda? And then maybe I'll leave it like that for now. I don't want to give you a big multi-part question. Sure, great. And then we'll follow up. So as Beza mentioned, engaging youth in peace building, security efforts or development efforts is nothing new. In some countries, 70% of the population are young people and it's been a priority for a long time to engage youth in these efforts. But in 2015, the UN Security Council Resolution passed UNSCR 2250, which really formalized the international community's efforts and urged member states to formalize like individual government efforts to engage youth in peace and security efforts. And they've shifted the language that we're using when we think about youth peace and security. The resolution aimed to reframe the narrative around the youth bulge as something that can be hopeful as member states leverage this enormous demographic. It worked to provide methods where governments and international organizations or ways approaches where these organizations can engage youth at different levels and in different ways. And so that's kind of the framing of 2250 and how the international community is talking about youth peace and security. And I'll talk a little bit more about the specifics in a bit. But before it was passed during the debates, the Security Council members were saying basically, yes, inclusion is important and engaging youth is important, but we're not convinced. So in order for the bill to be passed, the Security Council members mandated that a progress study be undertaken and that the field of youth peace and security increase the body of literature around what works. So these two studies directly contribute to that effort, which is growing since 2015 with strong momentum. And so by contributing these studies with kind of rigorous methodology and evidence of what's working and what's not working, they're contributing to shaping that new, the new field of youth peace and security. The Security Council resolution identifies five key pillars for engaging youth in peace and security. So those pillars are participation of youth, protection of youth, prevention of youth to violent extremism or to join from joining violent groups, partnership with youth in peace and security efforts and the disengagement and reintegration of youth when they return from fighting or from violence. So the way I see these two studies linking to those five pillars is really through the prevention lens and the partnership lens, which the prevention lens talks about creating enabling environments, providing key services and opportunities for youth to engage in peace and security or development efforts. And the partnership is really about engaging young people as actors with agency in peace and security. And so the Somalia study, for example, was a demonstrated a combination of the two, which is kind of the top down reform of education and service provision, but also the bottom up of engaging and partnering with youth in the civic space. Leaving off with that last comment is helpful because again, for somebody who's new to understanding how youth programming works, what they've heard here today is something like the following. The way to transform youth populations is to provide some education, could be vocational or like in the case of Somalia, it sounds like broader secondary education and couple it with something else. Civic engagement opportunities, unconditional cash transfer, and then a transformation of some kind happens. But that's not all, that doesn't represent the full spectrum of interventions we can do with youth populations. There's some other things that we can do that can also be effective. And I wonder if you could talk a little bit about what are those other things? What do other, what else do we know about working with youth populations that go beyond these two linkages that you just made here? Can you tell us a little bit more about that? Sure. So after the resolution was passed as an effort to contribute to the body of study, some of our partners in the youth peace and security space kind of spearheaded by search for common ground led a survey, a global survey that ended in a report of the effectiveness and the presence of youth led peace building efforts in conflict affected communities. And that survey found that youth are actively engaged and that they're doing programming at minimal cost and high impact. And so engaging young people who are already leading peace building organizations is a very cost effective way to lead these peace and security efforts or engagements with youth. But that's also to say that engaging youth as leaders or as partners in peace and security isn't the only way that young people should be or can be engaged in peace and security efforts. There also has to be a shift by those who have power. So by governments, by international institutions in the way that they're engaging youth. So by providing a youth lens to a stabilization or counterinsurgency program, the way that these studies did. So it's shifting the mindset and the approaches of people in power, the governments or international organizations, but also partnering and elevating the role of people at the grassroots level, the young people who are already doing this work effectively. I wanna be sure to kick the conversation over to our audience very soon. And so while you're thinking of some questions to ask or some comments to add to the mix, I wanna give maybe Abdullah a high an opportunity because we talked about the governance issue and we didn't give you a chance to talk about maybe what that looks like from the Somalia perspective. And so I'm curious if you've got some things to say along those lines. I just want to pick it from where Scott has left. He said, these processes are complex, these processes are complicated, right? I mean, we are talking about Somalia country that has not had a stability since 1991. I was in primary school then in Kenya. So social services have broken down. I mean, you can imagine where you don't have anything going on, you've born, kids that were born, then they are now in their mid 20s. So Al-Shabaab developed in a space like that, in a space where you don't know who's coming to collect your trash, you don't know if your mail will be delivered, you don't know, none of those facilities are not in existence. And the spillover to the Kenyan side that doesn't get a lot of traction is the fact that those areas along the border, the northeastern and the southern part of Kenya that borders Ethiopia, Somalia, sorry, has the same problem. The government presence is largely limited in places where the government is. The relation between the community and the government is largely mediated through force. I mean, if you look at that part of the country, it's the part that has got the largest concentration of the military barracks, you know. If you look at all the demographic numbers, transition from primary school to secondary school, length of road, that is changing now post, you know, 2010 constitutional processes where, you know, services and powers are being devolved to the local authorities. But still, it is within the milieu that you see people being enticed to go and join this group. So yeah, I mean, like, governance is a big monster. I mean, like, but if we don't start engaging with it, even if it's at a micro level, let's say, for instance, as this study has demonstrated, it's education, then we will just be paralyzed and we'll not be able to do much. So it behooves, you know, states, non-state actors that I cannot emphasize more. To begin thinking about this, a lot of the time, intervention tends to take largely a very formal kind of sense, but in some of these areas, and in communities like the Somali community, what has kept the community going is the level of resiliency that has been enhanced through the informal, non-state kind of network that is already existing. And so when we think about intervention, we'll be useful to begin thinking about non-state, informal network that will be able to do some of the state, some of the functions that state would have done, but still is not doing its fair share. It's, as I listened to you say that, I'm reminded of something Aubrey said that one of the big parts of what the youth, youth and security agenda is trying to do is create agency in youth populations. And what that sounds like it can do is build the capacity, if you will, for youth populations to operate in those kind of more informal networks that you just described to do things that are important in fragile states, to hold government accountable for its actions, to advocate non-violently for government change. I mean, all of those are capacities that youth populations would presumably need to build. And so any investments that create that capacity will help, in a sense, have a knock-on effect of hopefully pushing governments to reform in the ways that we think are desirable. All right, I promise we turn this over. So the way we're set up, there are microphones somewhere, although I've lost track. There they are. So if you could raise your hand, introduce yourself, tell us what institution you might be affiliated, if you like, and let's take one at a time at first, and then maybe we'll start bunching them up as we go along. Hi, my name is Mara Cronenfeld. I'm from International Youth Foundation. And I was really, thank you for the study. It was elegant and powerful and I really appreciated hearing about it. And my question gets a little bit about what Scott said about, is the effectiveness of the two interventions that had dual prongs just about multi-dimensionality? And the reason I asked that, like you had brought up, is that I know the World Bank has been engaged in a meta-analyses of all the impact and quasi-experimental evaluations about workforce development programs among youth, or youth livelihoods programs, and they had the same finding that one single intervention was not effective, but when there were multiple interventions at once, there was actually a statistical increase in the number of young people who graduated and had job-ready skills and got jobs and stayed on those jobs. So I wonder if it is that, and I wonder, what I wanna get at is whether the governance, if you're pinpointing governance, is it governance per se, or is it the feeling that we are taking care of, and are those separate or not, or are they the same thing? So that's basically my question. Yeah, so I think the fact that we see in both contexts that two interventions, a multi-dimensional approach, as you put it, is effective has to do with the fact that oftentimes the reasons why young people engage in violence is complex and multi-dimensional. So there's usually not just one driver or one factor behind it, like I don't have a job, so I'm gonna join Alshabab. So I think the idea that programs need to be multifaceted and dynamic and multi-dimensional stems from the fact that the reasons why young people engage in violence tends to be complex as well. So I think our research as well as what the World Bank is putting out is pointing towards that and also pointing to the fact that first, we need to really understand the root causes of why young people engage in violence and then design programs to address all of those underlying factors. And the second question about what governance means or what we're really capturing, I think it varies. So in Somalia, we were really looking at young people's sense of agency, so their ability to affect change within their communities and amongst their local leaders. So it didn't have to do so much about their perceptions of government, but more their efficacy and their agency. In Afghanistan, it had to do more with viewing the government as being more responsive to their needs. So it was actually a perception that the government was doing more, again, because they probably gave credit to the government for enabling those programs to take place. So there are multiple facets of governance as well that we're seeing. Okay, anybody else wanna jump in on this particular question? If not, okay, hold for a second. Did I see a, so all the way up front. And again, if you could just introduce yourself. Alexander Kravitz from INSITE, thank you for a report. Shifting geographically, I'm curious whether, based on your general experience, there's anything that you, from the report that you learned that could be applicable to the problem of gang violence in Central America in particular? And maybe to you, to Joe, I'm curious whether USIP is doing any work from the learning perspective to apply to Central America, to the problem of gang violence in Central America? On the issue of gang violence, I think it's a bit different, because when we talk about violent extremism, we see it as a form of political violence, broadly speaking. And maybe, I think that the underlying drivers might be a bit different than the type of violence that you see in places like El Salvador and Guatemala. We actually work in Guatemala on violence reduction programs. But it tends to be a bit different. And the youth that engage in that type of violence might also be different. For example, a lot of the work in the Northern Triangle around gang violence tends to indicate young people who engage in that type of activity come mostly from very vulnerable families. So their families have either dispersed and they're left by themselves. They're vulnerable economically. They don't have access to many opportunities. So we're talking about youth who are socioeconomically vulnerable. Whereas what the research has pointed to in terms of violent extremism are youth who are more driven by a desire for justice, who face grievances, who are, as the research pointed out, concerned with governance issues primarily. I don't want to generalize about the issue in the Northern Triangle because I do think there might be some connections with their young people's perceptions of governance and injustice as well. But largely, it tends to be more focused on very marginalized, vulnerable youth who have faced socioeconomic disadvantages rather than political disadvantages, let's say. So for the question about USIP's work on youths and gangs and criminality, Aubrey, I might put you on the spot in a second because you might have visibility on this. I don't know of anything off the top of my head. However, I will say, just as a general comment, it appears to be almost an apples-to-apples comparison. Whatever motivations lead somebody to be part of a criminal gang might be similar to motivations that lead somebody to be part of an extremist organization. But Beza's right to point out these organizations have different purposes. And so sorting out what the organization does, address political grievances in a country, carry forth an ideology, it might be an extremist ideology, and so forth versus what a criminal gang does, set aside those differences, the organizations take care of their people in ways that can be similar. And the motivations that individuals have for joining might be really focused on the way they will be taken care of inside those organizations. So I do believe that these two communities have a lot to learn from each other and don't speak to each other as much as they could. One of the last things I did before I left USAID and I'm looking at a former USAID colleague now who can call me off this or not, but I'll say the following. One of the very last things before I walked out the door and came over here was got the team at USAID to commission a study to see what we could learn about it. But I haven't heard where that stands and where that study, what the status of that might be, so Dara, feel free to chime in if there's something to contribute. But that is totally putting you on the spot, and I know. But in any case, it's a great question. I really thank you for it. So I put Aubrey on the spot and then we'll get to you. Yeah, so the research that I've done around the correlation between CV and gang violence shows that as Beza said, that the drivers are often different and the purpose of the groups are different. The recruitment processes have similarities and the leaving the violent groups has, they have similarities. I've seen that specifically organizations in Denmark are learning lessons from the kind of disengagement and reintegration piece of gang members and violent extremists. There's an organization that I can't pull off the top of my head, but they've actually basically copy and pasted a program that was created years ago to reintegrate gang members into society. And now they're using a very similar approach that includes mentorship, social, kind of psychosocial services and accountability to reintegrate youth who have been in violent extremism groups. In fairness to the people I've put on the spot. Dara, do you want to chime in a little bit about what you know about this and then Beza over to you. Hi, I'm Dara Katz from USAID. And we did indeed finish the study that Joe was talking about and also did another study with FHI 360 and Mercy Corps, which was more field-based looking at Guatemala, Morocco and Jordan. The latter two having CVE or nascent CVE programs and Guatemala, of course, having the citizen security gang prevention programs and what, trying to access the files in my head, but generally what the conclusions were was that each side could actually learn from each other. The initial considerations or thoughts was that CVE to learn from the gang prevention programming, but actually both could learn from each other. One of the things for the gang prevention programming was to look more systemically at what was going on in the environment, which is what CVE programs tend to start with this broader analysis of the drivers of recruitment. And then on the other side, the place-based and tertiary prevention targeting of the gang prevention program was very useful to the CVE programs and John is nodding his head so I know I'm kind of on track, but if others including Beza may want to add, but we did find that very useful and the conversations are continuing. I think they're not as intense as they were when that study was going on, but we do communicate and try to learn from each other. Thank you, Dara. Beza, do you want to add to that? I was just gonna say, yeah, we worked a little bit on this issue with FHI 360 and USAID and there are, like Dara said, there are lots of lessons that could be applied between both different groups, mostly in terms of targeting, measuring outcomes, monitoring and evaluation, and you're right, I think there is some overlap as well in terms of how these groups create a sense of belonging for young people that are marginalized, which could also lead to lessons. So yes, that's it. Thank you for that question. Other questions? There's a guy in the white shirt and the gray tie. Thank you and we'll bunch up for you now. Thank you very much. Mike Jayjeets teach in the George Mason University peace officer program, but also formally here at USIP, work with Scott. I walked in a little bit late, so I apologize if you addressed this, but in 2015 Mercy Corps did a study of the causes, root causes of political violence and found that poverty was not the root cause. And in fact, it was really injustice of use by security forces, corruption, that sort of thing. So I'm not surprised that you found that giving cash to people, which addresses poverty, wasn't really the way ahead. And it seems in terms of this discussion of whether it's a multivariate, multidimensional approach is the best one, or whether the one that actually increases, you mentioned governance, but the capacity of the individual to be able to deal with injustice. So I would just simply suggest that to me makes a lot more sense. But so the question going forward is, how would you test these, if you will, different theories of change? And one variable that I would be interested in testing is the level of development of civil society with whom the individual, the youth that you're giving these skills, governance skills and these ability to hold others accountable, but they have a place to test those skills and use them. Great, thank you for that, Mike. As we move down to the, let's have the panel address this one, and then after this one, we'll try to bunch a few questions together. So Beza and others, do you wanna talk about? Thanks, Mike, I agree with you. And that's a good opportunity for me to maybe expand a little bit on my focus on governance. And that's to say that in surveys, certainly anecdotally from my work in Afghanistan over many years, if you ask people what do they expect from government? This idea of service delivery has kind of permeated the development as well as the security kind of theory of change, where you say, well, what services do you expect? Number one is security, and number two is justice. And then after that, it might be education where there actually has been some progress. And then you get to the more kind of nitty gritty services, particularly in urban areas, but you get to roads, infrastructure, and so forth. But there is a hierarchy, and if you don't do the security and the justice, then people are gonna have a negative opinion or a positive opinion based on that far before you get to these other drivers, and I would say also economics and poverty being lowered down in the list, particularly in areas that didn't have services or that had high poverty to begin with. I think then what you get into is this, certainly in the Afghan environment, but I'm sure it must be true of Somalia and other similar areas where, and this is being a little bit cynical of me, but we kind of do what we can measure and what is easier, and then hope for the success along that because it is a necessary but insufficient cause. So we say, okay, let's start on that and make some progress. But fundamentally, I mean, if you looked at the larger picture in Afghanistan, you say, well, look, grand corruption, patronage that interfeins to allow the president's brother to be the governor of the province where this was done and to run it as if it's a mafia state, but he also fights the Taliban. That is a larger, harder to crack egg, but it's fundamental to the problem. And so these programs also have to be put into that context of there are larger reasons for a war going on and no, you need to start somewhere, but you can't move the needle by aiming for small interventions that address one cause when really people's attitudes toward government are, what did they read in the newspaper about grand corruption or what did they unfortunately witness in the mosque when somebody blew it up? And so that needs to be borne in mind. Quickly, so I think you're right, the results from these two studies are not that surprising given our previous research around the importance of injustice and governance as a driver. So it just sort of reaffirms that and builds on that research. In the future, how do we test this theory of change that governance is important? I think what we can do is continue to invest in these types of research studies, specifically around programs that are trying to address governance. So it could be programs like this that do it in an indirect way or programs that are directly trying to address corruption, security sector reform, perhaps other programs focused on governance could also have metrics to measure how successful they are. Let's see if we can bundle a couple together. So there are two people right in a row there. So that'll make it a little bit easy. Is there a third question maybe? And then way in the back, you'll be number three. Hi, Alex Peterson with Blumont. Beza had a question for you kind of on the, maybe an urban rural shift. So you talked about in Afghanistan it was based in around Kandahar city, which has very different security dynamics than the rural situation may have. I was wondering if there's anything that maybe had been learned from Somalia where you had a more, it seemed you had a larger geographic spread about kind of what the needs and the opportunities around youth engagement are just rural to urban. Okay, totally different cluster of questions I think. Just going back to some, which you both touched on Joe and Scott before sort of this, oh sorry, Adriana Davis with USAID. That'll give you some lens of why I'm asking this question. You know, you were talking about before Joe about the sort of, there's a good example of quick impacts and how maybe sometimes when you have, be able to have that flexibility certainly in these crisis and ongoing crisis situations that maybe it's not, we might call them post-conflict but it's still ongoing sporadic conflict sometimes. That rapid, you know, ability to be flexible like you see in like OTI or other kinds of programs. But then with, you know, the need for more transformational, long-term, complex projects that are really gonna get these root structural issues. I guess my question is more about, I see a sort of making it a black or white question and it's one or the other and I see right now it's sort of going more towards the quick impact short term but as you noted, that could lead us to backsliding or maybe not really getting, maintaining those gains or being able to scale them up or have really transformational impact. So how can we avoid, I think that's sort of a false choice. How can we avoid it being one or the other especially because sometimes then if we get, if it starts being very short term, that never ends. We just stay in constant short term mode but not necessarily, you're doing short term for suddenly eight years, you know. So how do we like even from within avoid sort of this false thing to make sure we have the rapid flexibility and the things that maybe work tangibly short term but make sure, you know, it's that sort of emergency like relief to development continuum again but how do we make sure we're not sacrificing one for the other or the dialogue and the narrative isn't only well it's only this one or it's only that one but to make sure we get the both without losing either because, you know, there are limited funds for sure but I don't wanna lose the transformational at the expense of this is a quick sexy project that the ambassador loves because it looks great but you do need to have, they do have value, you know, as well, these tangible impact flexible programs. Okay, good. One more after that, please. Hi, Marina Peuge from the Washington Institute for New Year's policy. My question was within both studies, did you notice within the youth demographics a particular sweet spot where there was the highest impact or effectiveness for the multi-dimensional programs such as a combination of maybe age or level of education and on the other side was that particular ceiling where it wasn't as effective based on like if there was a particular demographic trend in that? So there are three questions and just to kind of sum up Alex is interested in maybe differences between urban and rural contexts. Adriana is wondering, she put a lot on the table but just to distill, how do we avoid just keep doing short-term impact, trying to go for short-term impact, how do we eventually get to the long-term recognizing there are a lot of things with youth populations that do take a while to transform how do we get there and then Marina is asking us really to look further down into the data and have an understanding so within the sub parts of the sample, where are we making the biggest impact and what do we know about that and where are we a little bit limited in making impact in the little parts of the sample? So I leave it open to the panel to grab whichever one is in your wheelhouse. I'll start with the first and the third. So in terms of the rural urban divide, you're right in Kandahar in Afghanistan, we looked mostly within the urban population because that's where the program was being implemented. I don't know if we can compare that to Somalia but in general you're asking about differences in youth needs based on the urban and rural divide. I mean in general it's much more difficult to program in rural areas in a place like Somalia because of not just logistical challenges but also the security situation and Somalia tends to be more challenging outside of the major urban areas, especially in the south. So in some ways I guess those are the areas that might benefit more from these types of programming. With the SYLI program we were intentional in working in all parts of the country including in the south where as I mentioned some of these areas had been previously controlled by Al-Shabaab, very rural and difficult areas that the government did not have access to for a long time. So I think that's the most I can say about that question. For the other question about the disaggregation, we did try to do that not based on the profile of youth but kind of based on the location. So in Somalia we disaggregated by how conflict affected the communities where we implemented the program and we found that the program was actually more successful in more stable environments and in the more conflict affected areas it was less successful. We saw less impact in the violent areas. And we also disaggregated by the length of the program so how long a young person had been exposed to it and not surprisingly we found that young people who had been in the program for a longer time, so about three years or more tended to be impacted positively by the program versus those who had been in it for a shorter period of time we saw less impact there. I might underscore what Beza said or what the Afghanistan study showed that often when international organizations or governments go in for kind of quick wins and kind of photo op engagements that they actually can further marginalize youth and do more harm than good or kind of counter the good that they were trying to do. So sustained engagement is something that is becoming more prioritized in the youth peace and security kind of field and then also in terms of the rural and urban divide. I have one thing that when you think about true inclusion or inclusive youth programming you think urban and rural but you also disproportionately young women are excluded from youth programs and so the importance of the international community reframing the definition of youth to be young men and young women versus kind of the natural young men that people often consider. I can tackle the second question which was about short term, long term and how to blend that. I think there are a couple examples to think about. I mean one, this has been one of the more highly touted programs for development in Afghanistan which is the national solidarity program and now this is citizens charter. That is a small grant program to local communities for small scale development projects often physical infrastructure. But in the absence of local governments or governance the way the program got around that obstacle was to introduce a governance component along with the assistance so that there is a community council which is first selected which has quasi-democratic engagement with the community. They choose the projects, there is oversight and monitoring by the community and the money is given and it's implemented locally. So that's a way of saying well in the absence of some state structures or institutions let's create the different parts of the picture in the program itself and then deliver it over the long term and that seems to make good basic development sense. I think that in the larger and even longer term scale the key is to work with local governments or national governments. One of the problems of Afghanistan is you have multiple donors all of whom want to control their own money which represent different pieces of what is already been described as a complex puzzle and the Afghan government hasn't had sufficient, they have low capacity to be sure but they haven't had sufficient visibility or control of projects actually kind of integrate that in a way where you can take an economic development project and a justice project and a governance and an infrastructure project and because you're there locally you can figure out how those pieces fit together to actually engender support. Those levers are controlled by different actors at different times and so to the extent that you can work or build the capacity of a local government to work through them I think that proves to be a better long-term solution. We are right at time. I want to thank the audience. Thank you for spending time with us today. This bringing together various people from various communities, various communities of practice this succeeds when you come away with something that you've learned that's new, you take it back to your organization, maybe share it with others. That's what success looks like when we bring folks like you together. So I hope you do have something new to bring back to your respective offices and share it with your colleagues. It's really been a pleasure to spend time with you. I want to recognize Ruben Gringard who helped us with the early part of the thinking to help figure out a way to craft a conversation around these reports that would be of interest to a bigger audience. So he was really essential in bringing all this together. I want to thank the people who helped us with the microphones and the audio in the room. Everything was seamless as usual. And I especially want to thank our panelists who have spent time with us and especially Beza for sharing this great report. I've learned a lot. It's been really good for me. So if you would please join me in thanking everybody and have a great afternoon. Great, nice job. Oh, it's great.