 Thank you so much for inviting me to this session and thank you so much, Peter, for the introduction and also the Antonio's introduction to the special issues. I will not dwell with this session because that has been well covered by both Peter and Antonio. But just a few words about the salience of pre-colonial states which is the focus on the paper I will present. It is a matter of fact and you know that well that Africa's pre-colonial states they cease to exist as political units a long time ago. And the kings and chiefs were co-opted and subsumed by the colonial state. There seems to be some revitalization in some countries now. But these pre-colonial states they left deeply held identities that also today structure political decision making. Marta Fielbart has a really interesting book focusing on West Africa, Senegal as a case, documenting this. In recent research we also, some very prestigious research also shows that pre-colonial centralization is found to be strongly correlated with development at both the national and local level. You are probably most familiar with Michelopoulos and Papineau's work from 2013 in Latin America. But the underlying mechanisms through which these pre-colonial institutions affect long run development still remain weakly understood. This is just a presentation, a list of some of the recent work we did just to inform that this is, it is a very exciting work but there is still work which there are a lot of interesting things to do. And colonial, the legacy of colonial powers and the pre-colonial legacies. This is really exciting and thank you so much Antonio and Kunal for inviting me also to contribute to this special issue. This research is funded by the Research Council of Norway. The paper is published this year and I will not dwell into the technicalities here, just guide you through our motivation, arguments, results and possible policy implications. So why Uganda? Well Uganda has a long history of strong pre-colonial institutions and they continue to play an important role in the lives of ordinary people. And by focusing on one country that allows us also to study or exploit within country variations in pre-colonial institutions which are not affected by national institution or the identity of the colonial rule, the British in this case. And as also has been an important discussion point during this conference is that low tax compliance continues to undermine domestic revenue mobilization in Uganda as well as in a number of other Sub-Saharan African countries. The pre-colonial organization in Uganda was implied both you have centralized areas, kingdoms and more stateless societies. In the central and southwest there were centralized societies, kingdoms. The most famous one is of course Uganda but there were a number of others. While in the north and east there were more or less stateless societies. There was no organized structure in the societies. So the motivation for this paper is to understand the core roots of tax compliance norms and the literature has provided an array of theories and evidence attempting to explain these variations in citizens tax compliance. Partly you are all familiar with the economic deterrents theory. You have the fiscal exchange approach, you have issues related to fairness, trust in government and so on. This is well documented in literature and including the work Mary Mali and I and Ingrid Schussen did in 2014 but Bessley and others have done a lot of work on this. These studies largely focus on the quality of institutions and social norms as key determinants for tax compliance. But still they do not provide adequate explanations on the root courses of these variations in quality of institutions and tax compliance norms that lead to these differences both within segments of the populations and between areas within countries. Taxation in pre-colonial centralized kingdoms in Uganda was quite a complex and well developed issue. For instance in Uganda there were four main taxes. Just listen here, you can get more details in the paper. Interesting the number four war exemptions that were paid by men who did not want to participate in the war. Generally women and unmarried men were exempted and there were in-kind taxes and so on. The kings needed these tax revenues to pay for their armies to support the Royal Court and also frequent banquets and also to administrate the newly conquered areas. During colonization British indirect rule maintained the autonomy of the chiefs in administrating the locals including collecting taxes. And a quote from Harry Johnson, a special commissioner in Uganda he wrote the letter to the colonial secretary of state in 1990. He said that the power to rule for the Baganda is closely connected with the power to tax. We must give them those snail subsidies less if we did not this could cause a terrible uprising the quelling of which would be very difficult to stop. So in the kingdom taxation was a part a key element of people's identity, feeling of identity belonging to a state. So the objectives of the study is first to examine the legacy of pre-colonial centralization on tax compliance norms of citizens in contemporary Uganda. And here we compare pre-colonial centralized ethnic homelands and stateless ethnic homelands. And thereafter we aim to analyze what are the underlying mechanisms. First we examine whether the persistence in citizens believes about the need to obey authority matters. Because in the centralized areas or kingdoms the state had an organized force to uphold its authority. Second we examine the persistence in citizens' perception of trust in local institutions. Because the centralized states they shape local institutions. They have a bureaucracy, a capacity to administrate the areas. And third we examine the persistence of social cohesion or horizontal trust. The pre-colonial kingdoms were endowed with collective identities stretching across villages in their areas. Well the data we use are firstly we combine the geo-referenced anthropological data in pre-colonial ethnic homelands. It's the Murdoch 1959-1967 which is now also very much used in this type of studies. Historic using the Murdoch approach. And then we combine it with micro survey data from several rounds of afrobarometer survey. We use here rounds 3 to 6, 3 plus 4 plus 5 plus 6, yes. And we use regression discontinuity analyzes in neighboring ethnic homelands with different levels of pre-colonial centralization. Here this is based on Orange Bay by the work of Michelopoulos and Papinot. So the empirical analyzes here well the main dependent variable that measures tax compliance norm. We get that from the afrobarometer survey I think also 10 minutes. I think Marima also used this question in your work. And there the respondents are asked to state whether they think tax evasion is well not wrong at all, wrong but understandable, wrong and punishable. And then we use a binary indicator set equal to 1 if the respondents choose a statement tree and is set to 0 otherwise, yes. And then we have to control for individual differences we control for age, gender, whether people are residing in urban center, employment status, education, well, religion. And then we have a number of geographical controls at the level of ethnic homelands. For instance distance to the center of each ethnic homeland because institutional might be more influential at the central level than in a more remote area. Capital city whether people are residing in capital city or other areas, distance to national border and so on. Land area that has also to do with the economic diversity. We also control for pre-colonial and colonial ethnicity level variables including the intensity of exposure to slave trade which may have long-term effects on people's trust in each others, trust in the current authority. We control here for also pre-colonial economic development depends on agriculture and also colonial institutions like railway station, missionary stations and such which have impacts also on people's relation to each other and trust in institutions. And there are also a number of other controls which you can delve into when you read the paper and the benchmark results. Here we have three columns and the first column one there we control for a set of historical ethnic homeland variables which includes exposure to slave trade, indicator of pre-colonial, economic development, pre-colonial social organization, colonial investments and so on. And then we column two that controls for ethnic fractionalization and individual perception about the treatment of their ethnic group. And then the third column includes an index for an individual's perception of the quality of local service provision and also we have night density as part of the variable here we look into. But in all columns they include individual level controls like age, education, gender, religion and so on. And the result show that pre-colonial centralization is associated with a higher tax compliance norm than the stateless areas in Uganda. And here we also do an RDA analysis where we say is this an average for each kingdom or can we find some differences? Does it depend on where people are located, close to the ethnic, to the borders between the stateless and non-state, between the stateless and the centralized areas. So we control for that and we find the same answer. This is, we control also for response in the Afrobarometer living 50 kilometers from the borders between the centralized and non-central states, 100 also same answer. This is reporting the results from the response located 75 kilometers of the borders. And the same correlation between pre, there is a correlation between centralization and tax compliance norms. It remains positive and significant in all the columns. So what are the underlying mechanisms explaining this? Well first, obedience to authority. In centralized policies the state had an organized force to uphold authority and could uniformly apply policies throughout the given territory. So in such a system governance may then lead people to develop a culture of obedience to political authority. And here again we use questions from the Afrobarometer survey. Trust in tax authority, the courts, the police, the government. This is the question related to trust in tax authority from the Afrobarometer. The same similar question when it comes to courts, police and government. And we find similar results here that the individuals in the historically centralized parts of Uganda are more obedient to authority and the government in power than individuals in the historically non-centralized parts. Then we examine the role of trust in institutions and tax, because tax compliance cannot rely on coercion alone. Pre-colonial states could also encourage quasi-voluntary compliance through the accountability of leaders. And the persistence of accountable leaders in historically centralized areas, they may affect contemporary fiscal contract between citizens and the government by strengthening institutional trust. And we test this mechanism by looking into the relationship between pre-colonial centralization and individuals trust towards the central government and also various institutions. This probably cannot be read there from down there, but the results say people in the historically centralized parts of Uganda are less trusting of the state, the police and the tax authority than the share of respondents in the historically stateless part. So although individuals in the historically centralized parts are more obedient to authority, it is not necessarily based on a trusting relationship. Then the third mechanism we look into is the social cohesion or horizontal trust. The pre-colonial kingdoms use different ways to bring together individuals with different backgrounds, plans, and make them acknowledge authority and also comply with the rules. So social cohesion among the different clans were also strengthened by means of religion, political rituals and so on. And this helped very much to increase solidarity among the different clans and also created a cohesive polity. So we test if this social cohesion indicated by first ethnocentric national beliefs and second interpersonal trust whether that has persisted in the historically centralized parts of Uganda. I think I just moved the results here and so we do not find a significant result for the ethnic centric nationalism, but pre-colonial centralization is significantly correlated with a higher level of interpersonal trust. So conclusion, pre-colonial centralization is positively correlated with higher compliance norm and this is explained by the legacy of first pre-colonial states capacity in upholding authority and second a strong social cohesion through higher interpersonal trust, but not through trust in the central government and public institutions such as the tax authority. So people in the historically centralized states are more obedient to political authority which in terms shapes their compliance norm to general rules including paying taxes. So mistrust in tax administration can coexist with a relatively high level of generalized trust. Policy implications, well, even though people in historically centralized parts of Uganda have mistrust towards the central government and public institutions, they may still be willing to follow rules and pay taxes when they live in a setting with higher interpersonal trust. Social and economic policies to increase trust in public institutions can therefore help to improve tax compliance in Uganda and the success of such policies to improve trust in public institutions will of course depend on the leader's effort to acknowledge the past. And finally, Uganda was not the only country which had pre-colonial centralized states. You had it in Ghana, with Ashante, you have Great Zimbabwe, you have in Senegal, West Africa number. So these results may also be of relevance for studies in other countries of Africa. Thank you so much.