 Thanks a lot for being here today, thanks a lot for sharing this session. So today I will present a paper which is a joint with Jean-Luc Aouso from UPF, who is here in the audience, on the long run effects of the Truth and Reconciliation Commission in South Africa. So the vast majority of civil conflicts are basically episodes of recurring violence following an initial conflict. And to break out this cycle of violence, one kind of institution that has become increasingly popular in the last decade around the world are Truth and Reconciliation Commissions. So this is an institution which aims at establishing the facts, letting the different groups of individuals telling their stories and reaching a common narrative of the past. And by doing so, Truth and Reconciliation Commissions, or just TRCs, are believed to foster peace and to promote interracial trust. However, despite the wide usage and support that these kind of institutions get around the world, the effectiveness remains heavily debated. And so in particular they are accused of having detrimental psychological effects on the victims of emphasizing everyone's misdeeds instead of just the misdeeds of the main perpetrators and of ignoring some socioeconomic dimensions of peace-building. So in this project, we use one of the most iconic TRCs that has ever been created. And this is the one that was established in South Africa right after the apartheid to study what is the causal effect of Truth and Reconciliation Commissions on political beliefs towards institutions and interethnic trust. So after the end of the apartheid, South Africa underwent a radical political transformation. And one of the keystones of this political transformation was the establishment of a Truth and Reconciliation Commission in 1995. In the following few years the commission received the testimonies of more than 22,000 victims and it received more than 7,000 amnesty applications. The TRC had a massive influence on the country and also like just in other countries. And an important reason for the large influence that the TRC had on South Africa is the extensive TV coverage that it received to the point that it quickly became one of the most mediated events ever taking place in Africa. One particular mission that played a special role is the Truth Commission special report which was the flagship program on the TRC from the SEBC which is the national channel in South Africa. The program was aired every week between April 1996 and March 1998 and it quickly reached a really huge audience allowing the TRC to become more known by the population at large. So in this project we will basically look at how exposure to the work of the TRC through this specific program can have long run effects on political attitudes etc. So to be more specific about this we basically use an empirical strategy which is very similar to what Pedro presented like two presentations ago. So we collect data on all of the SEBC antennas that were airing this particular emission in 1995 and we use this information to reconstruct two types of signals. So the first one being the actual signal strength that you will receive in a given world in South Africa. We do that for the whole country. That's what you can see here on the first map, so bottom left. Obviously since signal reception might be endogenous we want to capture variations in signal strength that comes out solely from topographic obstacles and so on. So for this reason we also simulate a hypothetical signal strength free of any topographic obstructions. If you take the difference between the two it basically gives you the variation that solely comes from topographic obstructions, this is what you can see on the second map. And so we are interested in the relationship between exposure to the TRC and different measures related to political attitudes or trust. The main outcomes will come from different waves of the afrobarometer and we are specifically interested in two layers of heterogeneity. The first one being along the racial line, so does the TRC have different effects on the white population and the black population. And the second layer of heterogeneity we are interested in is about residential segregation. So does the effect of the TRC on the population differ according to segregation? Before showing you the main outcomes here we just use predetermined outcomes from the 91 census to show you that exposure to the TRC is uncorrelated with district level characteristics such as population, share of white or black individuals in the population, share of african speakers and so on. Okay, so moving to the main outcomes now, in this table we look at the relationship between exposure to the TRC and political attitudes and in particular we use one question of the afrobarometer where respondents asked for which political party they would vote if elections were held tomorrow. And we just classify political parties according to the fact that they might be predominantly black or predominantly white. And what we find and that's what you can see in columns one and three is that getting exposure to the work of the TRC does not really affect the political beliefs of black respondents. However when we focus on the white respondents, so that's columns two and four, we do find a strong effect on them. So white respondents who get more exposure to the work of the TRC become much more likely to support a predominantly black political party. So let's now turn out to the interaction with the residential segregation. So here the measure of segregation is a genie index based on the 91 census and this is calculated at the district level. And what's really interesting here is that, so let's just focus on column six here, is that yes, so white respondents have become much more likely to support a predominantly black political party when they get more exposure to the work of the TRC. But this effect entirely vanishes when they live in a totally segregated district. So this really suggests a possible complementarity between the TRC and also the possibility of having interracial contacts as measured by a genie index in this case. Next we look at trust in different kinds of institutions. Here we just show you results for trust in the president, in the parliament, and in the police. And we find broadly consistent results. The TRC having strong effects on trust in these different institutions. But again, two layers of heterogeneity that appear to be very important. First the effects on the white respondents are much stronger and secondly these effects become much weaker in highly segregated areas. We finally have a look at interracial trust and here the sample size is much smaller so these results are noisier and should be taken with a grain of salt. But we again find consistent results with exposure to the work of the TRC having a larger effect on white respondents and a weaker effect in segregated areas. Okay, let me jump to the conclusion. So here we showed you preliminary results from a work in progress, so in the future we plan to obviously complement this with many other data sources and to also look at the downstream economic consequences of the TRC. But a few results are already very clear. The first one is that truth and reconciliation commissions do seem to have an important effect on political attitudes and trust. The second one is that you do have important heterogeneities along the racial lines so in this case effects on the white population tend to be much stronger than the effects on the black population. And finally these effects vanish entirely in highly segregated areas which as I said earlier might suggest a potential complementarity between the work of the TRC and intergroup contact. And the second thing which is interesting in this case is that we need to keep in mind that South Africa remains a highly segregated country so this obviously plays an important role in the interpretation of these results. Thank you very much for your attention and looking forward to the discussion.