 Okay, so yeah, this is a paper in progress on information and mobilization and and demand for redistribution in South Africa and is joined with Patricio Piraino as well who is there and Eva Begna. So Eva and me we are the in Hamburg But also affiliated to Saldro at UCT. We were in Saldro previously and Patricio is still there So this this research Started when we were all there essentially. Okay, so I want to spend a second on the motivation because I think it's important to bring you on board So I first want to do some general motivation about why focus on Distribution and demand for the distribution and then a more specific motivation of our paper So for the general Importance of redistribution. I think we heard in the keynote speaker Talks like talk about redistribution and how it matters for inequality right for Brazil for example, and then Other people also talked about the distribution and I mean you might know better than me But I from what I recall of the top income papers at the beginning. They were making a big point about how taxation Was important for keeping the reductions in the quality for a long time in in France and in other Poissy the country's right so the distribution I would say it's fair to say it matters like in a clear way in the short run But it can also matter like for explaining like broad differences big differences in the quality So so okay, so what what then determines redistribution and and so that I think maybe Dominant view for a long time was that actually inequality would be one of the key determinants of redistribution and it was in such a way that so from this Meltzer and Richards type of Idea that when equality is very high the median voter is poor relative to the mean and then he will she will Demand loss of redistribution. Okay, so in some sense the story was reassuring type of story so that in the end like inequality from this political economy point of view is a bit self-regulating So when you get like lots of inequality Then the median voter becomes very poor and then demands more redistribution then inequality decreases But in already, I don't know 10 or more 15 years like this this Story about the effect of inequality on redistribution has been put into question like severely so I would say that now Well, I would say the consensus now is that we don't really know but certainly we don't find in the data a strong relation between inequality and redistribution in the way said in the in the theory And so and so this applies both in a cross-section So it's not the case that more unequal countries tend to redistribute more and also over time It's not that we observe that as inequality increases we've said consequent Increases in redistribution, right? And this is relevant internationally and also is relevant for South Africa, right? Because South Africa is a country with as now we all know some Huge levels of inequality, but the levels of redistribution are not so big average so then What then determines redistribution, right? So there's been this one specific paper of quite some relevance by Kutzymko and others this Emmanuel Scythe is also there that looks at the role of information For demand for a distribution, okay, so basically what people demand whether the policy preferences for redistribution for taxation And so this paper takes that sort of I think a relatively natural step in asking can it be that then One of the reasons maybe why redistribution doesn't React to changes in equality may be because people are not so aware of the degrees their levels of inequality or where they are in the income distribution or They have wrong information So they did a survey experiment, which is also what we did and I will explain shortly a bit what it is and And basically what they find is a bit puzzling in some respect, which is that Is that providing this type of information, so this is for the US, sorry Affects the concern for inequality so people that are treated Given all this information about inequality and what happened to inequality in the US and stuff They report to be more concerned about inequality but then when asked about policy preferences about the desired taxation They are not very much affected. So it seems that you give information and then people gets concerned But this doesn't follow so the policy preferences don't follow demand for distribution doesn't follow from this concern So it would appear that there's some missing element between being concerned and Wanting to change things, right? And this is where our paper enters, right? So our argument I mean in some sense It's very intuitive and in other sense is hard to pin down in a very precise way So we're still working on this sort of precise theoretical argument I'm going to give you here more than intuitive But I mean it's a very straightforward argument that it comes from social movement theory Like these sociologists were talking about this in the 80s I think something That is useful for the puzzle that I just mentioned, which is that Grievances are not enough for making people go to action, right? So people may perceive lots of injustice, but that's not enough for make them Do something about it, right? And the missing element in there That literature is mobilization. So you need some sort of social force that Tells you So yes things that I'm fair and this is not okay, but we're here it can be changed We're going to do something about it and it's possible to change Okay, and so you need someone that mobilizes people in order to to really then affect this This idea and I think that also this element is also relevant for South Africa because in South Africa we have seen In the last 10-15 years Very strong discourse at the leadership from the ANC and other parties on jobs jobs jobs jobs is really I think I would say I think it's fair to say it's kind of big part of the social discourse in South Africa and Then to the detriment of of inequality of restitution So it shows up inequality and redistribution does show up from time to time, but it's not so much dominant so in some sense what we are saying is that The arrival of a party like the Malema party, which is a very redistributive I mean it's a party with a very clear redistributive Policy should affect what people say about Redistribution that's in some sense it was if if Malema is critical, which is something I will come back to later Okay, so this is the basic idea and then what we do exactly is we conduct a survey experiment in to Cape Town Townships So what's a survey experiment is just a normal survey where then you Provide some treatments to some people and not to others in a random way Right in order to assess the effect of these treatments on some stuff. You are going to ask them about So so in all we have now like we have around 1600 observations so the big bunch in Kailitsha, so it's a black African black township and Apart some 400 500 seen delft in the colored section of depth So we have two different thousands. I will talk about that later and the treatments that we give our Information on local inequality trying to replicate the findings of the paper I just mentioned this question go and the size and others and Then we give two other video to other treatment. Sorry that are supposed to generate this idea of mobilization Okay, we'll talk about this now And then of outcomes we have like we asked them about this concern for inequality and About policy preferences and then we have like one Outcome that it could be very interesting, but we haven't looked at it yet. So I'm not going to present the The research for this but it's like we ask people then if they They are they feel strongly about this redistributive policy whether they could send an SMS to the local counsellor or sign a petition to try to Oppressionalize better this idea of an action, but I'm not going to present yet We don't have yet this in figure that and one thing that is very relevant that we still have We're going to make a second do a second wave of this. I mean we have scope for adding things So the feedback would be really really great Okay, so let me talk very quickly about the the two places So so the two townships are Kailitsha and Delft as I said and Well the here I have some demographic statistics and here some outcomes So Delft would be the colored township and Kailitsha is the African black town. So essentially Kailitsha is poor There are less people that work and Madrid levels are relatively similar But then informal housing this like half of the houses in in Kailitsha are Shacks essentially informal second whereas in Delft they're all Former houses and then yeah, so the the Kailitsha votes A and C and Delft votes the A Overwhelmingly so the the local counsellors are A and C for Kailitsha and the A for Delft. So they differ significantly But it's interesting to give already a feel of the outcomes relevant outcomes in the two Townships so Most of the people think that inequality is a problem in South Africa, right? 90 percent Then we try to Like capture this idea of whether we have managed to shock this idea of mobilization by asking people about whether they think that High inequality is inevitable right and you see that around half Think that high inequality is inevitable and then we ask about Preferences for taxes. So the would you want to increase the top tax? So we ask it in a sort of very slow way So that people understand what we mean But this is essentially what what this is and then very few people would like to increase the top tax rate, right? and This is this bit is this basic income grant where there has been some debate in South Africa about it and there are More people maybe because they've heard of it more. There are more people supporting it And then we also asked whether people would like to have some Extra tax for whites, which was something that Desmond Tutu at some point raised and and Yeah, we see that it varies a lot by township So I just want to put one thing that I find interesting and already like reviews our argument, which is that You see that the poorer Township They tend to think more often that inequality is inevitable and they tend to Demand less often and increasing tax rates, right? and the and the same for the The this beggining on ground and the only big difference is really this tax for the whites Which is something that you would expect because the collards are generally less Supportive of this type of thing So I mean this obviously doesn't prove anything, but I mean it sort of illustrates our argument Okay, so let me show you then what here what are our treatments? So we have Essentially three three treatments one is to give information about local inequality so from the sensors we computed income levels and also Levels of ownership of certain assets like fridges and computers and so on and then we Showed them here in different In different neighborhoods of keta that are also very recognizable for everyone living there like Newlands comes Bay And here we have Google it to Philippi and then we had some colored also Asilon and we should play I think Okay, so we give this type of information This is the local information in the international information. We basically From the World Bank, so yeah, so from the World Bank we took some measures of the ratio of the share of the 20 percent to the 8 the top 20 to the bottom And then for some countries some important countries like the US and some countries that are around them And so we did it In a way that we wanted to make them pay attention to it We were first showing them this graph without South Africa and asking them to guess What the data says about South Africa and then afterwards we will show them the So I mean this is supposed to capture the idea that the high inequality in South Africa is a bit of an anomaly That was what we wanted to generate right with data that is real Obviously we selected these countries so we put Namibia, which also has very high inequality Okay, but meaning the idea which is I think a fair thing to say is that South Africa has unusually high Inequity and that inequality does not have to be so high and it's not so high Generally, okay, that was the thing we wanted to communicate And then we have video messages with we were giving on top of these two information treatments and we had three videos One from summa one from Malema and one from tutu and the three videos were Contempting inequality and saying that something should be done about it, right and And it was very easy to find the Malema video. It was We found the tutu video and it was hard to find the summa video But this is really points at the thing that is not so easy to find like like ANC discourses that are specifically right now about inequality Okay So we wanted a bit of a cross-section of the type of messages that people the type of elites people get exposed to Okay So the results Let me go directly to the results. So this partly is nice partly is a bit maybe strange or Maybe not strange, but unexpected Maybe we should have known better, but then this is what happened. So I here I have essentially different outcomes in different columns and Here we have our treatment the local information international information and this the video on top of the local and the video on top of the international So first the effect on the This concern for inequality whether inequality is a problem, right? So to start with as I told you before the levels on average are relatively high already So it's hard to show it, right? But still it does so you give the local information and People so there's some effect that then people tend to think more often that inequalities are a concern, right? And you give the international information and also most people then it doesn't make them think a little bit that inequality is a problem Now the the sort of nice thing is that we replicate their results from this paper I refer to about About the effect of local information, right? So it affects the concern for inequality, but we asked them about Top taxes or this basic income grant. It has no effect was so this is as they find And the nice thing is that when we give them now this information about international equality It does shock strongly the perceptions of inevitability of inequality and a consequent change in in Preferences for for taxation as well as for this basic income grant Okay, so this is nice and this is the thing that it fits with what we thought now in reality our Mobilization the videos were supposed to shock people to think like Like yeah, we can change inequality, right? And make them think that it's not inevitable, right? And it has the opposite effect. So it's something like we were shocked We thought that maybe in some townships, you know, we thought that maybe in In delft it wouldn't be clear but we thought in Kylie I would work but no, it's really Like you give them the videos on top of the international treatment and it totally well not totally but a big chunk of it gets Undone so they see the international evidence. I see like yeah, it's possible to decrease inequality and then they see their Elite guys talking about that and say like no it's not possible to so But then what is sort of interesting and this we have to think what it implies is that then following this they also then Reduce their demands for tax and this is something like that. I think that precisely is what I was saying that we need to think Precisely, what is the channel that would generate this thing? It looks kind of nice because it looks a consistent pattern where changes in perceptions on even in the ability lead to changes in In in tax preferences both for the top tax and for this big all this year is not significant But it's essentially similar story and then the last thing is that so all this doesn't seem to have to do with this racial stuff So it's not that it affects this the any perceptions on these taxes on whites So I just want to say One more thing and then they conclude so one thing that is relevant and I really want to point out is that these research are really robust This these two things so here we break down by township and you observe the same patterns So here's inequality inevitable and this top tax preferences for Kylaija and for delft and okay for that we have like few observations and Results are sometimes not significant, but but the pattern is there very clear and for Kylaija as well and Also, another thing that is partly reassuring and partly concerning is that if you also break down by by leader video So summa Malema and tutu the results are there even for tutu, right? So it's it's really I think this is sort of Interesting there is something there. I think that Okay, so let me just conclude and So We sort of give further evidence of this idea that if you give information on on Inequality it makes people concerned, but it doesn't need people to to action and We find and I put it here in a sort of mild way So we do find that our mobilization treatments affect inevitability perception and in turn attacks preferences But we have a strong caveat now about this video treatment that it does the opposite of what we thought It should do and so our first idea was that the leaders in South Africa just have like no Credibility right so in the end they see these people and instead of getting mobilized they get demobilized There's a bit of an issue with tutu Because we now did some to try to understand a bit better what's going on with some focus groups showing them the video and asking Them to tell us how they felt about the video and it's clear that people like to do In both townships they like them they put that and they saw the video and they were mostly happy about it But there was a substantial chunk of people that felt sort of sad About it. So so it just felt like it's concerned He talks about inequality and like how there are these people that don't have anything and these other people that have a lot and That some people reported feelings are depressed and sad about it, right? So there's something whether and this it was the same with summa summa Many people didn't like him and my name also many people didn't like him, but But for summa also some people were reporting feeling this having more this Like the depressed type of feeling and few people felt angry which in some sense is what I was expecting more But people would just be angry by seeing these people talk about inequality that no doesn't seem to be the case more like more more like depressed So yes, as I say then we have a second survey ground I think what would be really nice is to really find some really credible leader that really would then Get us what we want right really generate people to think that that inequality can be changed So we have to then think a little bit like local leaders or even sports stars or this is sort of I mean the direction we have Yeah