1 00:00:00,260 --> 00:00:06,270 Let's see. For this particular seminar, if you have opinions on some of those before, 2 00:00:06,270 --> 00:00:13,080 just to explain the theme of this episode as actors and audiences and maybe even gradually towards the end of term conflict resolution, 3 00:00:14,640 --> 00:00:20,530 don't suppose is here to the University of Essex fully at Rice. 4 00:00:20,580 --> 00:00:24,060 We've had a bit of discussion about your life story about how you got to this particular subject. 5 00:00:25,180 --> 00:00:30,210 No, I won't tell you, because otherwise it would spoil the sort of basis of the of the discussion we're going to have. 6 00:00:31,110 --> 00:00:36,419 But I think what's interesting about this subject is it's a very important way of 7 00:00:36,420 --> 00:00:41,639 reminding us about connections between those research centres or disciplinary studies, 8 00:00:41,640 --> 00:00:46,700 which seem also separate. And I'm not saying it's also a good news story. 9 00:00:47,130 --> 00:00:52,290 So if you think you were right, so before I start that, 10 00:00:52,320 --> 00:00:57,900 I'd take the opportunity of thanking Robyn this and I think you culture war centre for inviting me over. 11 00:00:58,410 --> 00:01:03,780 I'm very excited to be here and to present this project which is joint work with children, 12 00:01:03,790 --> 00:01:12,540 record funding from the her school in Berlin and its kind of Trojan Horse copycat or scapegoat unpacking the refugees terrorism nexus. 13 00:01:12,990 --> 00:01:16,649 And as you can guess from the title in this project, 14 00:01:16,650 --> 00:01:27,240 we are interested in understanding the security implications of security consequences of refugee flows, particularly for host countries. 15 00:01:28,200 --> 00:01:34,290 Now, you may wonder why you're interested in this particular topic, and there are several motivations. 16 00:01:34,320 --> 00:01:36,000 So first of all, 17 00:01:36,000 --> 00:01:45,810 the refugee crisis from which Europe is emerging continues to be at the core of domestic and international political agendas for many reasons. 18 00:01:46,380 --> 00:01:55,320 One reason is because in recent years, we have seen an unprecedented increase in the number of forcibly forcibly displaced persons. 19 00:01:55,320 --> 00:01:58,860 So refugees, asylum seekers, internally displaced persons. 20 00:01:58,860 --> 00:02:04,770 And today these people make up 1% of the world population, which is quite staggering. 21 00:02:05,550 --> 00:02:13,080 But also looking closer to home. By the end of 2015, between one and 2 million people fled to Europe, 22 00:02:13,080 --> 00:02:20,550 coming from countries that experienced high levels of violence and terrorism such as Syria, Afghanistan and Iraq. 23 00:02:21,090 --> 00:02:30,149 And as a result, we have observed a pretty heated debate in politics and media about the security risks of hosting refugees, 24 00:02:30,150 --> 00:02:37,410 and particularly since a lot of refugees and asylum seekers flee from comfort zones with high levels of terrorism, 25 00:02:37,770 --> 00:02:47,040 the risk of possibly importing terrorist violence, the risk that, for example, terrorist organisations could use these refugee movements, 26 00:02:47,040 --> 00:02:56,580 this population movement could infiltrate them as a Trojan horse to conduct terrorist attacks in refugees from destination destination countries. 27 00:02:57,630 --> 00:03:06,520 Now, if you look at public opinion, a lot of these fears are actually reflected in in public opinion and how people view this this problem. 28 00:03:06,540 --> 00:03:10,710 These are the results of the survey that was conducted in 2017. 29 00:03:11,700 --> 00:03:20,370 In most of the Western developed countries. And what this graph is showing is the percentage of respondents that agree with that statement that there 30 00:03:20,370 --> 00:03:28,090 are terrorists pretending to be refugees who will enter the country to cause violence and destruction. 31 00:03:28,110 --> 00:03:35,010 As you can see, this proportions are pretty high, between 50 and 80% in most in most of the countries. 32 00:03:35,850 --> 00:03:45,240 So clearly, we may agree or disagree with this list, but these fears are real by whether they are grounded or not in evidence, but they are real. 33 00:03:46,110 --> 00:03:55,530 And then, of course, policymakers as well. This is actually this this picture was retweeted by President Trump himself and clearly 34 00:03:55,530 --> 00:04:01,530 illustrates the US administration attitude toward refugees in this case coming from Syria. 35 00:04:01,530 --> 00:04:03,269 And more broadly, yes, 36 00:04:03,270 --> 00:04:12,780 President Trump repeatedly stated that allegedly numerous foreign born individuals have been convicted or implicated in terrorism related crimes, 37 00:04:12,780 --> 00:04:20,669 and some of these allegedly came to us through the refugee resettlement program and also angry at Prime Minister Viktor Orban, 38 00:04:20,670 --> 00:04:25,410 defining refugees and migrants as the Trojan horse of terrorism. 39 00:04:27,060 --> 00:04:29,790 Even the intelligence agencies seem to agree. 40 00:04:30,240 --> 00:04:37,830 This is the Europol EU law enforcement agencies in the latest report wrote that jihadist terrorist organisations already 41 00:04:37,830 --> 00:04:46,830 exploited the flow of refugees and migrants to send individuals to Europe to commit acts of terrorism and may continue to do so. 42 00:04:48,330 --> 00:04:56,820 Now, I've shown you a number of claims related to public opinion policymakers and intelligence agencies about this issue. 43 00:04:57,240 --> 00:05:03,680 But we believe there are reasons to actually be. Cautious or sceptical of some of his claims. 44 00:05:04,370 --> 00:05:12,350 First of all, the first question is whether or not these perceptions may be actually hampered by cognitive cognitive biases. 45 00:05:12,380 --> 00:05:16,400 For example, the German chancellor, Angela merkel, 46 00:05:16,700 --> 00:05:23,540 acknowledged that it's possible that some refugees who have fled the security zones could pose a security threat for Germany. 47 00:05:23,570 --> 00:05:33,170 However, Germany was already a target for Islamic terrorism before the many refugees came to Germany, especially its 32,050. 48 00:05:34,220 --> 00:05:39,260 So the idea here is that there may be other reasons why countries are targeted by Islamist 49 00:05:39,260 --> 00:05:45,020 terrorists or other foreign foreign terrorists that have absolutely nothing to do with refugees. 50 00:05:45,880 --> 00:05:51,610 And so the question here is whether the conflict, they just characterised to some degree attribution bias. 51 00:05:51,620 --> 00:05:59,509 So basically the tendency of individuals to place excessive emphasis on the characteristics of 52 00:05:59,510 --> 00:06:06,200 certain people perceived as outgroups right and scapegoats them at the expense of more plausible, 53 00:06:06,200 --> 00:06:11,510 perhaps situational or environmental factors, such as, for example, some countries foreign policies. 54 00:06:13,520 --> 00:06:16,280 What would this existing research have to say? 55 00:06:16,310 --> 00:06:23,060 How about this question about the possibility that refugee flows are a security threat for the countries that they host? 56 00:06:23,780 --> 00:06:27,049 Well, when we looked at research, we found there was a gap, 57 00:06:27,050 --> 00:06:32,420 a significant gap actually in systematic research that could shed light on this particular question, 58 00:06:32,870 --> 00:06:40,610 because there are some studies that have looked at the association between the number of refugees present in the country and the risk of terrorism. 59 00:06:41,120 --> 00:06:48,169 But when it comes to explanation, these studies tend to theorise mechanisms that are purely at the domestic level, 60 00:06:48,170 --> 00:06:50,239 such as conditions in refugee camps, 61 00:06:50,240 --> 00:06:59,540 which, however, we don't have in developed Western countries or some degree of competition for resources between the refugees and the locals. 62 00:06:59,900 --> 00:07:07,549 But it's there is very little about the extent to which terrorism is diffused from one place to another, follow it. 63 00:07:07,550 --> 00:07:10,670 And basically through this is this refugee flow. 64 00:07:10,670 --> 00:07:14,930 So basically the extent to which terrorism is imported right from from abroad, 65 00:07:14,930 --> 00:07:18,560 which is sort of the framework that we are adopting here, this diffusion framework. 66 00:07:18,650 --> 00:07:23,760 Terrorism moves from one place to another through these population movements. 67 00:07:24,410 --> 00:07:30,709 But also, when we look at the outcomes in these studies, we typically see a rather aggregated outcome. 68 00:07:30,710 --> 00:07:37,850 How many terrorist attacks happen in that particular country, and we don't know much about who perpetrates these attacks. 69 00:07:37,850 --> 00:07:41,030 Are these really foreign actors and where do they come from? 70 00:07:41,030 --> 00:07:44,060 Do they come from where the refugees are coming from or from elsewhere? 71 00:07:44,360 --> 00:07:51,620 Or are these local actors that have nothing to do with the refugees but also against whom are these attacks targeted? 72 00:07:52,370 --> 00:08:01,009 And so as a result, it becomes really difficult to actually systematically evaluate some of the claims that persist in in political and media debates, 73 00:08:01,010 --> 00:08:09,200 as I've shown before. And finally, the third reason to be sceptical has to do with the fact that the debate, 74 00:08:09,200 --> 00:08:13,550 both the scholars debate, but also the political debate, tends to be incomplete. 75 00:08:14,210 --> 00:08:20,090 And what is missing is the fact that current debates tend to disregard the fact that refugees 76 00:08:20,090 --> 00:08:26,000 and asylum seekers are often themselves the targets of violence in the country where they go, 77 00:08:26,330 --> 00:08:30,680 only the countries that they're fleeing from, but also the countries where they go. 78 00:08:31,370 --> 00:08:38,860 And we seen in Germany 2016, more than 2500 refugees and asylum seekers, victims of attacks. 79 00:08:38,990 --> 00:08:48,500 More broadly, we have observed since a lot of the refugees come from Muslim majority countries, a wave of anti-immigrant, anti Muslim violence. 80 00:08:48,590 --> 00:08:54,140 These are just some examples from Canada and from from the U.K. but there are several more. 81 00:08:55,100 --> 00:09:03,740 Which brings us to the research question. Do refugee inflows actually lead to an increase in terrorist violence in the host state? 82 00:09:03,950 --> 00:09:08,810 And most important, if so, if there is some association, how? 83 00:09:09,080 --> 00:09:13,820 And under what conditions do refugee inflows increase the risk of terrorism? 84 00:09:13,850 --> 00:09:18,320 This is the question that we're trying to answer in this in this project. 85 00:09:18,560 --> 00:09:22,830 So to answer this question, remember, we adopt this diffusion framework. 86 00:09:22,830 --> 00:09:32,330 We're trying to understand to what extent terrorism travels through these population movements, specifically movement of refugees and asylum seekers. 87 00:09:32,360 --> 00:09:38,780 So we we developed three theoretical pathways, three possible diffusion pathways. 88 00:09:39,620 --> 00:09:42,770 The first one reflects pretty much the current climate. 89 00:09:42,830 --> 00:09:47,000 Why did refugees constitute a potent force for terrorism? 90 00:09:47,030 --> 00:09:50,480 We tried to sort of theorise this a little bit more. What is the logic here? 91 00:09:51,170 --> 00:09:54,120 Now we know that terrorist organisations have a number of goals. 92 00:09:54,150 --> 00:09:59,510 They have long term goals, but they also have more short term proximate goals which include. 93 00:09:59,820 --> 00:10:06,510 Attracting international attention for being able to advertise their cause more broadly internationally, 94 00:10:06,510 --> 00:10:13,430 provoking their foreign targets, increasing recruitment and transnational terrorist attacks. 95 00:10:13,430 --> 00:10:24,299 So attacks that these groups conduct in foreign countries have been demonstrated to help these organisations accomplish some of these proximate cause. 96 00:10:24,300 --> 00:10:25,470 They could be quite effective. 97 00:10:26,240 --> 00:10:33,000 The same type of attacks against foreign states, particularly when terrorists have to cross borders, they're actually very difficult. 98 00:10:33,000 --> 00:10:42,329 They're not easy at all to conduct. So the idea here is the refugee flows could create this direct physical linkages between the refugees, 99 00:10:42,330 --> 00:10:46,590 country of origin and also the terrorists, country of origin and the destination. 100 00:10:46,590 --> 00:10:54,270 Sorry, them, the host. The states basically allow terrorist organisations to extend their networks and facilitate 101 00:10:54,510 --> 00:11:01,170 their movement to other countries and that the perpetration of terrorist attacks there. 102 00:11:01,630 --> 00:11:07,230 Moreover, when we look at this is this applies most of the last point applies mostly to developing 103 00:11:07,230 --> 00:11:11,309 countries where we have refugee camps and there are studies that have shown that 104 00:11:11,310 --> 00:11:16,049 the radicalisation of refugees this more is more frequent and more likely when 105 00:11:16,050 --> 00:11:21,510 refugees enter in direct contact with terrorists that move across across the border. 106 00:11:21,510 --> 00:11:29,550 So here simply the idea is that terrorist organisations will use will exploit these refugee movements to conduct attacks. 107 00:11:30,280 --> 00:11:36,719 The host country will infiltrate literally. This was the second mechanism or the second. 108 00:11:36,720 --> 00:11:48,000 The second pathway is what we call copycat. And here the basic intuition is that terrorist activities by organisation abroad can inspire local 109 00:11:48,000 --> 00:11:54,130 domestic actors to resort to terrorism through processes of emulation or demonstration of fact. 110 00:11:54,170 --> 00:12:01,620 Various forms of inspiration, essentially. But inspiration is not sufficient to empower groups to conduct attacks. 111 00:12:01,620 --> 00:12:08,790 And the reason is access to resources, training, expertise, skills and other types of weapons. 112 00:12:09,150 --> 00:12:19,320 So basically the idea here is that a successful conduction of terrorist attacks requires requires a number of resources that are not readily, 113 00:12:19,560 --> 00:12:27,660 readily available. And as a result, they make it more difficult for domestic actors to resort to terrorism, in essence. 114 00:12:28,200 --> 00:12:34,079 So the idea is that refugee flows to the extent that these flows can be infiltrated by actual terrorists, 115 00:12:34,080 --> 00:12:45,240 they facilitate the movement of terrorism, specific capital to host countries, which in turn empowers local actors to act, to resort to terrorism. 116 00:12:46,470 --> 00:12:53,280 And this the rationale is because there are more material opportunities as a result of this movement of terrorists. 117 00:12:53,280 --> 00:13:00,780 So basically the idea is not only you have terrorists moving through these refugee flows, but also resources and, you know, 118 00:13:01,260 --> 00:13:09,600 material and human capital that travels through these flows and empowers local actors with grievances against the state to resort to terrorism. 119 00:13:11,070 --> 00:13:17,100 And finally, what I consider to be the most interesting, in fact, pathway is what we call scapegoat. 120 00:13:17,730 --> 00:13:26,850 If you're really looking at how local actors, local citizens from groups will respond to the arrival of refugees. 121 00:13:27,540 --> 00:13:35,880 And here the idea is that large scale arrivals of refugees and asylum seekers can generate perceptions of security threats in the host country. 122 00:13:36,330 --> 00:13:43,350 But this doesn't always happen. We believe that this is particularly related to where refugees are coming from, 123 00:13:43,350 --> 00:13:48,509 whether refugees are perceived as threatening because they come from countries that have 124 00:13:48,510 --> 00:13:52,160 those transnational terrorist organisations that we know are the ones that have it, 125 00:13:52,170 --> 00:13:53,370 like ISIS, for example, 126 00:13:53,610 --> 00:13:59,220 are the ones that have incentives to conduct terrorist attacks abroad because there are some organisations that are purely domestic, 127 00:13:59,310 --> 00:14:02,340 they don't care about conducting attacks abroad, but some do. 128 00:14:02,730 --> 00:14:07,940 And when refugees come from these countries, they will be perceived as particularly threatened. 129 00:14:08,490 --> 00:14:16,410 And also when the host countries or the countries where the refugees go have in the past experienced severe transnational attacks, 130 00:14:16,410 --> 00:14:21,480 severe attacks perpetrated by some foreign groups, does it have to be related to refugee? 131 00:14:21,570 --> 00:14:26,250 All is the idea of the past a vulnerability that becomes activated or triggered 132 00:14:26,250 --> 00:14:30,900 again by the arrival of groups of outgroups that are perceived as threatening. 133 00:14:31,560 --> 00:14:34,709 And how do people respond to this perceived threat? 134 00:14:34,710 --> 00:14:39,480 Well, in this case, fear trumps solidarity. Sometimes people act in solidarity. 135 00:14:39,690 --> 00:14:46,200 Under this particular condition, we believe that fear will trump sentiments and feelings of solidarity. 136 00:14:46,200 --> 00:14:51,840 And as a result, refugee inflows will trigger violent terrorist response by local, 137 00:14:52,410 --> 00:14:59,670 local groups that are meant to intimidate the refugees, the potential terrorists, but also to send a message to the local government. 138 00:14:59,720 --> 00:15:08,960 It is not dealing effectively with the threat. So here we posit that refugees will become targets of terrorist attacks by local groups. 139 00:15:10,160 --> 00:15:17,000 But because we are sort of placing this argument within a diffusion framework, we sort of take this one step further. 140 00:15:17,480 --> 00:15:25,910 And we are basically looking at conditions first in the countries from which from where refugees are coming from, which I sort of anticipated or. 141 00:15:25,910 --> 00:15:33,649 Right. Because the idea here, the rationale underlying the current claims that refugees could be a chosen force or that refugees 142 00:15:33,650 --> 00:15:39,410 could empower local actors with grievances against the state to conduct attacks against the state, 143 00:15:39,710 --> 00:15:45,230 is that basically refugees enable terrorism by transporting people, a.k.a. terrorists, 144 00:15:45,590 --> 00:15:53,150 but also ideas and skills from terrorist organisations in the refugees origin countries to the host countries. 145 00:15:53,790 --> 00:16:01,729 Similarly, the rationale underlying our story about the scapegoating of refugees is similar that refugees induce fear 146 00:16:01,730 --> 00:16:09,910 of terrorism or induce terrorism in the host countries by local actors by transporting fear of terrorism. 147 00:16:09,990 --> 00:16:14,930 They are perceived as potential vehicles and as a result they are punished because 148 00:16:14,930 --> 00:16:19,280 they are perceived to be the cause of this increased sense of vulnerability. 149 00:16:19,960 --> 00:16:21,890 And so basically, if this is true, 150 00:16:21,890 --> 00:16:30,379 that only refugees from countries with transnational terrorist organisation could logically export this type of terrorism to host countries, 151 00:16:30,380 --> 00:16:36,530 because in a diffusion framework, if there is nothing, if there is no terrorism at the source that can be transported, 152 00:16:36,860 --> 00:16:40,910 then we cannot observe the diffusion to the host for the host countries. 153 00:16:41,300 --> 00:16:44,840 And this allows us to basically distinguish between different types. 154 00:16:44,840 --> 00:16:48,800 This is what we call centre effects between different types of refugees, 155 00:16:49,100 --> 00:16:55,909 refugees that come from countries that have these transnational terrorist organisation that could branch out and also generate 156 00:16:55,910 --> 00:17:02,569 that fear and refugee flows and refugees that come from countries that don't have these organisations which we expect. 157 00:17:02,570 --> 00:17:10,040 We should not have any effect on the diffusion of terrorism to host countries along any of the pathways that I just described. 158 00:17:10,970 --> 00:17:15,680 But it is more like what about the countries where the refugees go because they are not all the same. 159 00:17:15,830 --> 00:17:25,340 Right. And so we we are taught that developed countries benefit from more advanced counter-terrorism capabilities by the ability to 160 00:17:25,340 --> 00:17:33,559 basically screen these population population growth and screen these refugee movements for potential terrorist infiltration. 161 00:17:33,560 --> 00:17:40,610 But also they are better able to police their territory, police or border for prevent these types of attacks. 162 00:17:41,370 --> 00:17:48,079 And so because remember the Trojan horse and the copycat past, we really even showed the possibility or the actual, 163 00:17:48,080 --> 00:17:53,240 in fact infiltration of refugee movements by terrorist organisations. 164 00:17:53,630 --> 00:18:02,510 So we expect that the countries that are better able at prevent is basically breaking this potential channels to be less likely to absorb diffusion. 165 00:18:02,660 --> 00:18:07,880 So Trojan horse and copycat mechanisms would be or phenomena would be less likely 166 00:18:08,090 --> 00:18:12,350 in developed countries precisely because these countries can break those ties. 167 00:18:12,890 --> 00:18:21,710 But what about the scapegoat? And importantly, remember, the scapegoat mechanism is based on fear, but it is not based on actual terrorists. 168 00:18:21,980 --> 00:18:28,430 It's enough to be afraid that there is a possibility that these refugee flows could be infiltrated. 169 00:18:28,880 --> 00:18:34,160 And these fears can be easily amplified by media, also by populist politicians. 170 00:18:34,550 --> 00:18:42,440 And on top of that, people that usually more scared, the more exceptional a phenomenon which, of course, we know is the case for developed countries. 171 00:18:42,940 --> 00:18:51,680 And so the idea is that the Schengen mechanism, the fear of refugees because of the risks and because of this potential infiltration by terrorists, 172 00:18:51,920 --> 00:18:55,160 is going to be more pronounced in developed countries. 173 00:18:56,360 --> 00:19:02,479 So how do we test these arguments? Our starting point is the global terrorism database. 174 00:19:02,480 --> 00:19:08,660 For those of you who are not familiar with it, this is the largest database of terrorist attacks worldwide, 175 00:19:09,170 --> 00:19:13,310 and it covers a time period from 1970 to 2060. 176 00:19:13,340 --> 00:19:18,440 So this is our starting point. And we have information based on this database on the perpetrators. 177 00:19:18,950 --> 00:19:24,200 And what we don't know based on the global terrorism database is where the terrorists are coming from. 178 00:19:24,200 --> 00:19:30,830 And this is why we now create this new dataset, which we called GTP Global Terrorism Database Homes, 179 00:19:31,310 --> 00:19:37,049 where we basically looked at all these perpetrators of terrorist attacks and called their home countries. 180 00:19:37,050 --> 00:19:42,200 So where they are primarily that established and with significant and but it's very important 181 00:19:42,200 --> 00:19:47,599 for us to be able to know where these terrorist organisations that conduct attacks brought, 182 00:19:47,600 --> 00:19:50,149 where they are established, where they're based. Because remember, 183 00:19:50,150 --> 00:19:55,340 we are differentiating refugees that come from countries that have organisation that 184 00:19:55,340 --> 00:19:59,560 have a tendency to conduct attacks abroad and refugees that come from countries. 185 00:19:59,710 --> 00:20:06,370 They don't have these types of organisations. Our outcome, the variables. 186 00:20:06,580 --> 00:20:10,930 We have three pathways the Trojan Horse, the copycat and the scapegoat. 187 00:20:11,530 --> 00:20:15,159 So our outcome variables are three different forms of terrorism. 188 00:20:15,160 --> 00:20:22,960 And specifically, these outcome variables reflect the observable and mutually exclusive implications of the pathways. 189 00:20:23,320 --> 00:20:27,820 And what this means specifically that for the current course, 190 00:20:28,210 --> 00:20:34,970 we look at the number of transnational attacks that take place in a potential refugee hosting country. 191 00:20:34,990 --> 00:20:40,389 What does it mean in practice? It's simply the number of attacks that are conducted by foreign groups. 192 00:20:40,390 --> 00:20:44,950 And then we are able to map these foreign groups with the countries of origin 193 00:20:44,950 --> 00:20:49,210 of refugees that are hosted in that particular in that particular country. 194 00:20:49,990 --> 00:20:52,420 For the copycat, remember, this is about emulation. 195 00:20:52,420 --> 00:21:00,280 There are local actors with grievances against the state that may be in power by the movement of terrorist groups through refugee flows. 196 00:21:00,280 --> 00:21:06,760 And so here we're looking at the number of terrorist attacks by domestic groups in a country against domestic targets. 197 00:21:07,600 --> 00:21:14,410 And finally, for the scale, because when we posit that refugees are going to be victims of terrorist attacks by local groups, 198 00:21:14,710 --> 00:21:19,750 we look at exactly that attacks in most countries, the target individuals who are actually refugees, 199 00:21:19,750 --> 00:21:22,780 but also individual like the asylum seeker or others, 200 00:21:23,380 --> 00:21:28,480 might want to share the same nationality as the refugees because we have information 201 00:21:28,480 --> 00:21:32,410 about the nationality of the victims of terrorist attacks in each country. 202 00:21:33,670 --> 00:21:40,690 Now our refugee data comes from the UNHCR, the United Nations High Commissioner for Refugees. 203 00:21:41,230 --> 00:21:48,790 And this is the nice dyadic data, meaning that we have information about where the refugees come from and the country of asylum, 204 00:21:48,790 --> 00:21:51,190 where they where they are hosted, where they go. 205 00:21:51,490 --> 00:21:58,030 I should specify that we are not looking here at refugees that are resettled in third countries where we can get the countries of asylum. 206 00:21:59,680 --> 00:22:06,880 And again, here we are able, because of our good home status, that we are able to distinguish between two categories of refugees, 207 00:22:07,240 --> 00:22:14,920 refugees that flee from countries that have these transnational terrorist organisations and the refugee they flee from other countries. 208 00:22:15,370 --> 00:22:17,979 And this is important because, for example, 209 00:22:17,980 --> 00:22:26,440 what if refugees are selecting right countries where they go based on some factors that also influence terrorism? 210 00:22:26,830 --> 00:22:36,100 Now, basically, if refugees make a selection, a strategic selection about where they go, the idea that this should apply to both types of refugees, 211 00:22:36,100 --> 00:22:40,570 regardless of where they come from, because refugees are civilians, so they care about security. 212 00:22:40,930 --> 00:22:44,079 All types of refugees should care about about solitude. 213 00:22:44,080 --> 00:22:44,709 That basically, 214 00:22:44,710 --> 00:22:54,550 by comparing the effect that these two types of refugees have on terrorism in the host country are better able to test our arguments systematically. 215 00:22:55,600 --> 00:23:01,719 And finally, because there may be a number of other factors that influence whether or not a country experiences terrorism. 216 00:23:01,720 --> 00:23:08,830 In our analysis, which is quantitative, we take into account the GDP of a country, the population, 217 00:23:09,160 --> 00:23:16,180 the type of regime, democracy or not, the presence or absence of a civil war or even inter-state war. 218 00:23:16,340 --> 00:23:23,410 Whether or not the potential host country has conducted terrorist sorry is conducted an intervention in a civil war, 219 00:23:24,440 --> 00:23:30,090 in the presence of a civil civil war in the neighbourhood, the presence of terrorism in the neighbourhood. 220 00:23:30,100 --> 00:23:36,940 So whether a country is located in what we call a terrorism hotspot and sort of our unit, 221 00:23:36,940 --> 00:23:41,149 so many eyes will be looking at each country in each in each year. 222 00:23:41,150 --> 00:23:46,900 So we have a country in the side, but we start from a more disaggregated design. 223 00:23:47,410 --> 00:23:53,230 Now, you may wonder, how do we differentiate between receiver effects or between recipient country here? 224 00:23:53,590 --> 00:24:02,319 For simplicity, we have used the whether or not a country or city member as a shorthand to separate developed and developed country, 225 00:24:02,320 --> 00:24:06,550 because it's a shorthand for a number of other sort of characteristics. 226 00:24:09,670 --> 00:24:13,990 This is just the script purely descriptive what our aggregate data looks like. 227 00:24:14,000 --> 00:24:25,549 So here we have data from 1972, 2015, and the the solid lines, the red one is the shows the trend over time for domestic attacks. 228 00:24:25,550 --> 00:24:29,020 That's one of our three outcome variables. The No. 229 00:24:29,020 --> 00:24:34,000 One, transnational attacks, meaning attacks conducted in a country by foreign groups. 230 00:24:34,450 --> 00:24:39,220 And the last one is attacks against refugees and asylum seekers. 231 00:24:39,580 --> 00:24:44,889 And the vertical bars are basically the total number of refugees in the world, 232 00:24:44,890 --> 00:24:48,790 disaggregated by whether or not they come from countries with transnational 233 00:24:48,790 --> 00:24:54,519 terrorism and countries so that the dot for those of you who are interested, 234 00:24:54,520 --> 00:24:59,260 we have regression analysis with our last with the country in your effects but. 235 00:25:00,640 --> 00:25:08,980 I want to skip to the results first. So what do we find when we run this regression analysis for the Trojan Horse? 236 00:25:09,550 --> 00:25:16,720 We find that refugees from countries that have these transnational terrorist organisations do in fact increase 237 00:25:17,020 --> 00:25:22,600 their risk of transnational terrorist attacks by groups that are based in those refugee origin countries. 238 00:25:23,200 --> 00:25:28,089 However, there is no evidence of this Trojan horse effect in developed countries. 239 00:25:28,090 --> 00:25:30,580 So in other words, we only find an effect, 240 00:25:30,760 --> 00:25:39,730 the only kind of support for the Trojan Horse hypothesis in developing countries and very similar results for the coffee cup. 241 00:25:39,740 --> 00:25:47,799 We do find that there is an increase in terrorist attacks by local groups that benefit from potential terrorist infiltrations, 242 00:25:47,800 --> 00:25:55,300 but not in developed countries again. What we do find in developed countries that yes, there is an increase in terrorism, 243 00:25:55,750 --> 00:26:05,050 but it is terrorism that's perpetrated by local actors against refugees, and particularly against refugees who are perceived as threatening. 244 00:26:05,890 --> 00:26:14,020 However, when we look at the other refugees, the other category of refugees, refugees, they come from countries that are not perceived as threatening. 245 00:26:14,260 --> 00:26:21,440 We don't find any effect. We don't find that these refugees spread anything basic, that they are not associated with attacks by foreign groups. 246 00:26:21,730 --> 00:26:25,960 They are not associated with attacks by local groups against local targets. 247 00:26:26,290 --> 00:26:30,460 And they are also not associated with attacks against refugees. 248 00:26:32,140 --> 00:26:34,660 I remember that when I talked about the scapegoat. 249 00:26:34,660 --> 00:26:42,430 I said it matters whether refugees are perceived as threat, and it matters also how vulnerable the local populations feel. 250 00:26:42,940 --> 00:26:51,840 And one way for us to get at this perception of vulnerability is to look at what is the most severe prior transnational attack, 251 00:26:51,850 --> 00:26:54,820 the attack perpetrated by some foreign group in the country. 252 00:26:55,330 --> 00:27:02,770 And the idea is that when you are vulnerable, before you are targeted before by for some foreign actor, the presence of our troops, 253 00:27:02,770 --> 00:27:07,930 of the arrival of our groups that are perceived as threatening because it could make you vulnerable again, 254 00:27:08,620 --> 00:27:13,900 they could bring back that experience, increases hostility and fear. 255 00:27:14,170 --> 00:27:20,800 And as a result, what we find here is that the more, the greater the perception of oops, sorry, 256 00:27:21,580 --> 00:27:34,300 the greater the perception of past vulnerability here, the more severe the prior attack from the most severe attack in the previous five years, 257 00:27:34,990 --> 00:27:42,450 the more likely it is that as the country hosts more refugees coming from countries perceived as, 258 00:27:42,790 --> 00:27:46,510 the more, the larger the number of refugees perceived as vulnerable. 259 00:27:46,520 --> 00:27:51,610 Given that you are, the larger the number of refugees perceive the threat to be. 260 00:27:52,030 --> 00:27:59,440 Given that you also have high vulnerability, the more likely it is that refugees themselves are going to become a. 261 00:28:00,250 --> 00:28:10,180 So basically, when local populations feel more vulnerable and the inflow of refugees from countries that are perceived as threatening and as a result, 262 00:28:10,420 --> 00:28:19,059 the refugees themselves are perceived as a threat, translates into a larger number of attacks against the refugees by the local groups. 263 00:28:19,060 --> 00:28:27,910 So basically refugees become, in a way, scapegoated for this increased perception of vulnerability and insecurity and insecurity. 264 00:28:28,480 --> 00:28:33,700 However, when we look at the other types of refugees where the refugees in right here, 265 00:28:33,910 --> 00:28:39,820 they come from the countries that don't have these terrorist organisations, so are perceived as less threatening. 266 00:28:40,150 --> 00:28:44,469 We don't find that there is much like that anything and it becomes slightly negative. 267 00:28:44,470 --> 00:28:51,640 So we don't find that those types of refugees are associated with more attacks against refugees, but the other ones are. 268 00:28:52,210 --> 00:28:58,750 So the bottom line here is that when of course, a number of reasons why refugees and asylum seekers might, 269 00:28:59,200 --> 00:29:07,360 you know, might not be welcome in those countries, why a local population might feel hostile to them. 270 00:29:07,960 --> 00:29:13,090 But what we are doing our attention to here is the security driver side of this hostility. 271 00:29:13,450 --> 00:29:17,020 And maybe there are many reasons and we do explore some of them, 272 00:29:17,290 --> 00:29:26,710 but we do find the security matter when people feel physically threatening threats and sorry, they become more aggressive toward toward refugees. 273 00:29:28,000 --> 00:29:33,430 Now, you may wonder, okay, you've shown us some results, but what if there are some other factors, 274 00:29:33,430 --> 00:29:37,540 what we call an observed confounders, things that we can't observe, but they are there. 275 00:29:38,050 --> 00:29:44,530 They make, for example, a country more likely to host refugees, but also more likely to be a target. 276 00:29:44,530 --> 00:29:49,030 And one thing the obvious thing that I can think about is the geographic proximity. 277 00:29:49,210 --> 00:29:56,650 Maybe these host countries just close. It's too close to a terrorism hotspot and that also brings refugees. 278 00:29:56,650 --> 00:30:05,549 So it's basically close to the problem, but it's not. Refugees are causing terrorism in this particular and this particular country. 279 00:30:05,550 --> 00:30:09,030 Oh, what if refugee flows are infiltrated along the way? 280 00:30:09,090 --> 00:30:14,079 So to address this possibility, we develop what we call placebo outcomes. 281 00:30:14,080 --> 00:30:19,700 And what simply well, this simply means we have looked at first transnational attacks. 282 00:30:19,700 --> 00:30:25,019 So attacks by foreign groups that are already there originate from countries, for example, 283 00:30:25,020 --> 00:30:31,380 neighbouring countries whose refugees however are not currently present in the host country. 284 00:30:31,680 --> 00:30:36,569 So if, for example, geographic proximity is simply driving these attacks, 285 00:30:36,570 --> 00:30:42,780 we should see that this particular outcome is significantly associated with with refugees, 286 00:30:43,230 --> 00:30:47,940 with them, with the presence of refugees, even though the two are not causally related. 287 00:30:47,940 --> 00:30:51,570 But we don't find basically the thesis that this is the case. 288 00:30:52,020 --> 00:30:55,860 And also, what about xenophobia? What if the problem is just any foreigner? 289 00:30:55,860 --> 00:31:04,620 Right. We look at actually attacks against foreigners from countries that actually are not sending refugees in that in the host country, 290 00:31:04,620 --> 00:31:08,190 any other sort of attacks against foreigners unrelated to refugees. 291 00:31:08,670 --> 00:31:16,469 If the problem is foreigners, we should see that having some refugees increases attacks against against all other foreigners. 292 00:31:16,470 --> 00:31:20,400 Because the problem is not the refugees such as somebody is. The foreigners are not good. 293 00:31:20,760 --> 00:31:28,079 But we don't find that this is the case again. And what if the relationship goes the other way around? 294 00:31:28,080 --> 00:31:32,630 What if, for example, that the host the countries that host refugees, 295 00:31:32,640 --> 00:31:40,800 they they basically calibrate their proclivity to host as a function of how how threatening the refugees are. 296 00:31:40,860 --> 00:31:45,809 Right. Or what if destination countries are chosen by the refugees based on security? 297 00:31:45,810 --> 00:31:54,540 Can we actually develop a fairly sophisticated instrumental variable approach that allow us to basically overcome this this problem? 298 00:31:55,110 --> 00:32:02,310 I'm happy to talk more about how we do this in the Q&A, but it's basically to take into account the possibility of a reverse causality. 299 00:32:02,760 --> 00:32:10,470 Again, finding this is that this is not the case, that our results are robust even when we we stop this alternative estimation. 300 00:32:13,530 --> 00:32:17,249 Finally, we have conducted some additional tests because you may wonder, 301 00:32:17,250 --> 00:32:22,290 for example, what if these results are driven by Muslim versus non-Muslim refugees? 302 00:32:22,290 --> 00:32:29,460 What if this is not about terrorism in the refugee sending countries, but is really about are these Muslims or not? 303 00:32:30,240 --> 00:32:39,209 We looked at that and interestingly what we found that whether or not refugees are Muslim or not has no effect on the Trojan Horse. 304 00:32:39,210 --> 00:32:43,800 The infiltration of terrorists is no effect on the copycat. 305 00:32:43,830 --> 00:32:50,100 However, and this is very sad, refugees from Muslim majority countries are more likely to be targeted. 306 00:32:50,700 --> 00:32:57,840 They are more likely to be scapegoated. They are more likely to be victims of terrorist attacks, especially in developed countries. 307 00:32:58,770 --> 00:33:03,390 So there is a scapegoating problem when it comes to the Muslim refugees. 308 00:33:03,900 --> 00:33:09,270 We have replicated our results looking at the lethality of terrorism as opposed to the number of attacks. 309 00:33:09,270 --> 00:33:16,200 And we find essentially to say we have looked at differences before and after 911, we don't find any. 310 00:33:16,710 --> 00:33:20,820 We have look at the history of terrorism in the countries where the refugees go, 311 00:33:21,180 --> 00:33:30,660 a result of what if the problem is not terrorism in descending the refugee setting or origin country, but is about civil war? 312 00:33:30,900 --> 00:33:36,690 It's just the story of US Civil War. Interestingly, we don't interestingly, we don't find that this is the case, 313 00:33:36,690 --> 00:33:43,530 which really matter if there is a civil war is how much terrorism by foreign groups there so basically civil war with 314 00:33:43,530 --> 00:33:49,980 transnational terrorism as opposed to any civil war which is in line with what we expect with this diffusion story. 315 00:33:50,850 --> 00:33:55,709 What if refugees are targeted because they have actually been engaged in some form of violence? 316 00:33:55,710 --> 00:34:00,360 Why they are punished? They're not scapegoated. They have literally punished for some bad behaviour. 317 00:34:00,990 --> 00:34:05,850 We don't find that this is the case. And again, not because there is no evidence of violence by refugees, 318 00:34:06,090 --> 00:34:12,060 but because we find the refugees are targeted regardless of whether they engage in violence or not. 319 00:34:12,810 --> 00:34:19,530 And finally, we replicate analysis with a dyadic design where we basically have the origin country and descending country and 320 00:34:19,530 --> 00:34:27,660 we look at at the relationship between the two and find that our results can be replicated with that design. 321 00:34:28,260 --> 00:34:34,829 So I will conclude, what did we find here? We found very limited evidence of a Trojan horse in a copycat. 322 00:34:34,830 --> 00:34:38,070 The fact there is some, but only in developing countries. 323 00:34:38,550 --> 00:34:46,920 Developed countries seem to be immune, but not because there is no problem by not because refugees or asylum seeker are never a problem, 324 00:34:47,400 --> 00:34:52,440 but because the developed countries have already done what is necessary to prevent this risk. 325 00:34:52,650 --> 00:34:59,130 Right. So there is no systematic pattern. This should lead us to be worried about developing countries in. 326 00:34:59,600 --> 00:35:08,600 Terrorism through art, through refugees. However, the current debate in politics and media are missing a key issue, 327 00:35:08,900 --> 00:35:14,810 which is the security of the refugees themselves and this phenomenon of scapegoating of refugees. 328 00:35:15,320 --> 00:35:23,210 So as a result, policies and discourse of the future, hostility and even hatred toward refugees are not making countries safer, 329 00:35:23,540 --> 00:35:26,990 but have the result of causing more terrorism, not less. 330 00:35:27,590 --> 00:35:31,220 And with this work concludes and I look forward to your questions, sir. 331 00:35:31,250 --> 00:35:31,580 Thank you.