 What is micro simulation? I think many of you will know it's basically a technique where you use a household survey that has information on family structure on income and expenditure of the household and then you bring these microdata to the policy system which can be coded in different types of software and basically you take the social security code and the tax code to the data in order to know how much is the tax load on each family and then you can of course bring that up with representative data to the national level so that you know the tax load how will that come out as revenue for the government but you also will know how much social protection costs the government. So this is important of course for government but it's also very good information research. Now you can also then given that you have such detailed data look at more specific questions like what type of household is paying how much and so on and so forth and then of course you can have look at specific measures that we all know like poverty distributional measures progressivity of the taxes. The beauty of these models is that also that you can look at policy reform so imagine you have a change in or you want to have a change in your tax code you will change the tax function and then you can look again at the tax loads and all these are the indicators that I was just talking about in terms of government revenue but also progressivity of the tax system and that's also where you of course have a first round effect but we call a first round effect in terms of effects not only in individuals but on the government and then actually the last bit that just came up takes a step further you're talking then about behavioral responses and that's where research comes in so the static model then taken by researchers as those you find sitting here and you can start thinking about and answering more complex questions. So the South Mod project has developed together with the teams in the local all the countries that you see here models for these countries and also has rebanned already existing models for Namibia and South Africa so in the audience there's quite a few mothers and fathers of the local tax benefit model and it's great to see how well this is going so here you see all the partners involved and we actually had a meeting yesterday pulling everyone together sharing our experiences after the first round of modern building. Now these models as I said are nearly all finished. We will update them all together and keep them up. We have trainings and bunches happening or they already happens and they will happen this year and we are in the process of building an active musical because these models will only be as good as we have an exchange going on between the policymaker side, the research side and of course then the model developers. So this is very important for the models and research is important to push the models for new innovation as much as answering the more complex techniques that go like questions of for example tax elasticity which is one of the questions you can also partner with these models to they'll never just be answered by the model itself but they'll always need a researcher who frames the question using more methods. So what we will see today is four applications of such models. The first one uses the model for Ghana sort of the Ghana model and looks at the relationship between extending extending social protection formalities so that's Lucas presentation. The second paper uses Tasmat or am I now shifting yeah the second one is on Tasmat so it looks for how well beneficiaries are targeted and what could be reforms which would be deliver a better result. Then the third one uses the South African model Samot and analyzes the universal child benefit hypothetical reform and how one could finance it. And finally using the model for Ecuador the last presenter will show how you can use such models to estimate vulnerability to poverty.