 Thank you very much for inviting me to this conference. It's a pleasure to be here. And I will present a paper which is based on the South Mod models. And the idea of this paper is to see whether it's possible to introduce reforms into the existing systems in African countries and to do it in a budget-neutral way and to achieve high equity in the distribution of social spending. So that's the topic of this paper. As probably I don't need to explain in this audience that a huge problem is that in low and middle-income countries, they typically have higher growth rates than say advanced countries in Europe. But this growth rate barely affects poverty levels. And the issue is lies in the lack of social protection. Although cash transfers have expanded in low and middle-income countries over the past decades, still the coverage and benefit levels are modest and insufficient. And according to the ILO data, in Africa, only 18 percent of the population is covered by at least one social protection benefit compared to 45 percent globally. So that's a huge difference. And in the welfare state literature, the African regimes are typically classified as insecurity or informal insecurity regimes as compared to the welfare states that are common in European countries. There is quite a large number of studies that assess the impact of social spending on the well-being outcomes, especially at the macro level, the country level, but they rarely account for the distribution of social spending across the population and whether this social spending benefit the poor or not. So this study contributes to the emerging literature that looks at how the distributional equity in low and middle-income countries could be improved through changes in tax-benefit policies. So this has been already said in this audience. So there is a tool which helps us to assess the impact of changes in social transfers and in taxes on well-being outcomes, and it's a tax-benefit micro-simulation model. So the principle is that they use micro-data from surveys, data on gross incomes, and then they build the, try to replicate the policy systems in the countries, and, you know, varying this, changing these components, we can look either at the impact of demographic factors, distribution, labor market distribution, or the effect of policies. So we can isolate the impact of policies using these models. I have looked into the literature, and there are two types of tools that are currently available to researchers in low and middle-income countries, and these are CICU models. The model is built by the Commitment Diabetes Institute, and there was quite a lot of research using these models. They looked into the distributional outcomes in African countries, and they also tried to simulate some policy reports by expanding cash transfers in these countries, and they looked into how this reforms could be funded from the national sources, for example, BAT increases, and who will be a removal of aggressive subsidies. And then there's another strand of research, and this strand uses, you know, models built on the uranium plate for South Models for low- and middle-income countries, and so this paper is listed here. So far only one study goes of child welfare outcomes, and in this paper I'm focused on child welfare outcomes, child policy, so it's one of the contributions of this analysis to the existing literature. So research questions, how distributional equity in low-income countries would be improved for changes in tax payment policies. I have two case studies, Mozambique and Zambia, Mozambique is a low-income country by the way, Zambia is a low-income country, and the questions that I'm asking in this study what is the distributional input of the existing tax payment policies in these countries, which types of policy reforms would bring about better distributional outcomes, and can these policies be funded from the internal revenue use so that they are sustainable in the future. I'm using South Models for Mozambique and Zambia, developed by UN New Wider, and my baseline is 2018. Baseline means that the policy that I'm submitting is 2019, although the data, the survey data comes from older periods, for most of all it's 2014, 2015, and for my present it's also 2015, but this is controlled in the model using growth factors, which I'm using for the income as my welfare indicator. I have adjusted this measure because there are some use and big power issues with zero incomes, so I used the expenditure to impute unreported incomes, and I also used the increase in my income measure. And a couple of disclaimers regarding the analysis, so the simulations assume full direct tax compliance in the formal sector, so the taxes are modeled for people in the formal sector, for employees in the formal sector, full indirect tax compliance across the distribution, and full take-up of benefits, which might be a bit problematic, but at this point I don't know how to basically account for non-take-up, given the lack of data. So the provision of non-cash programs is out of the scope of this model, which might also be considered an issue, and the analysis is point in time, and it doesn't incorporate behavioral effects in the sense that it's the morning after effect of these simulations. So this slide shows the income concepts that I used in the analysis, so we start with market income at social transfers, deduct direct taxes and social insurance contributions, and then we arrive at disposable income. The next step we also deduct value-added tax, and excises, I'm not wrong, the indirect taxes, and we have the measure of post-fiscal or consumable income, and that's the measure that I'm going to use for when I will present my results. So the other table shows the poverty lines in their express national currency, I'm mainly using the first one at $1.9 per day, and the reason is that the threshold is quite high, so even with this measure, if you look at this slide, at this table, the percent of population below poverty line is, at consumable income is 61.5 in Mozambique, and 66.6 in Zambia, so it's quite high poverty threshold, so I didn't consider others for that reason. So this picture shows the status quo, so the distributional impact of taxes on social transfers in 2019 in these two countries, and as we can see on the slide, and maybe it's not very surprising, but the existing systems reduce income inequality to some extent, so in case of Mozambique it's 12%, in case of Zambia it's 4%, but in general they're not effective as regards to poverty reduction, and this is due to the lack of social transfers, so if you look at poverty headcounts, changes in poverty headcounts, so for Mozambique it's basically, there are no changes, and for Zambia we can see even an increase in poverty headcounts due to the impact of indirect taxes, so this is a more detailed picture that shows the impact of taxes and transfers by populations of groups, so this is from Mozambique, deciles, age groups, household types, also by province the data allows to do this original analysis, we can see that the tax-benefit system is highly progressive, if we look at this picture distribution by deciles, but if you look at the total, this column on average the population does not benefit from the tax-benefit system, and this is mainly due to the impact of indirect taxes, and as far as the subgroups, demographic groups are concerned, all household types except for parents, loan parents with three, and families with three and more children are net losses in the system, now moving on to Zambia, the tax system is progressive, but it's less progressive than the system in Mozambique, social transfers are well targeted, but indirect taxes appear to be highly regressive, so you can see the share of indirect taxes are high in the bottom deciles, and again the conclusion is that on average the population doesn't benefit from the tax-benefit system due to the impact of direct taxes on the top deciles, it's quite high on the top deciles, and the impact of indirect taxes on the bottom deciles, so in this case in this country all population, all household types and age groups are net losses, and you know I'm aware of the time, so I'll move on to the second part, and well basically the main question of this study, the policy simulations, so here I'm looking at the types of policy reforms that could bring about better distributional outcomes, I'm introducing new transfers into the system on top of the existing arrangements, and there are two types of transfers that I'm looking at, reform one is means-tested transfer equal to the household poverty gap, so it's a perfectly targeted transfer for the poor, and the second one is a non-minst-tested transfer for each child up to the age of 18, so it's a universal benefit, but for children, and I'm also looking into two sources of funding for each reform, so the first source is increase in standard VAT rate, and I opted for a three percentage point increase in the standard VAT rate, which seems reasonable, or this would be equal to a 1.1 proportional increase in all PAT and turnover tax rates in these countries, so these reforms are budget-neutral, so basically the amounts of benefits are reduced to match this revenue, this available extra revenues that are coming from this increase in either in VAT or in personal income tax and turnover tax rates, and now I move on to the results of this analysis, so this is the impact of these two allowances on various types of measures, poverty head count, poverty gap with child poverty head count, child poverty gap and Gini index, and I found out that both types of allowances are effective in reducing national and child poverty gaps, but in case of the first allowance, that means asset transfer, poverty head counts go up, so in case of the second reform, universal child benefit, there is no problem in the sense, so the poverty head counts do not increase, so that's the first result, then looking at the distribution of this new allowances and gainers and losses in terms of the income deciles, reform one is of course more progressive, but here we assume perfect targeting that the benefits reach the poorest households, so of course here in this exercise it will be more progressive than the second one, however when we implement the second reform, a larger part of the income distribution benefits from this type of reform, the universal child benefit, and because they benefit from this reform it helps to offset losses due to increased VAT and or income tax rates, so when we do this kind of simulation it's really important to account for the fiscal side, so the revenues that are available for this analysis, so in this case it's beneficial because the universal transfer is more beneficial because more households along income distribution get this transfer, and in terms of net winners of both reforms, so these are children under 18 years and the most vulnerable household types, the household types that have highest poverty rates, these are couples with three and more children and lone parents with three and more children, while other age groups and household types see a reduction in their consumable incomes due to higher taxes, so I have two minutes and I will just make some concluding remarks, so this analysis shows that the existing tax benefit systems in these countries that were my case studies, they are really not effective in terms of income inequality and poverty, they reduce income inequality to a small extent, but in terms of poverty there is of course a large gap in terms of social protection, and models like South Mod they allow us to and they provide a platform which allows us to explore more effective means of redistribution in these countries, and in this paper I did a hypothetical simulation, a policy exercise to estimate the examples of the effects of this new cash transfers, and this both types of allowance is given the lack of social protection in the system, once they are introduced they are very progressive and they are effective in reducing poverty gaps, but they will require a much larger budget to be more effective in terms of reducing both child poverty headcount and poverty gap, so in terms of the design of these programs the the main conclusion that the strictly targeted allowance may be poverty increasing and a larger part of income distribution benefits from a universal child benefit which helps to offset losses due to the increased tax rates, so these are my conclusions, thank you very much and I will I will be grateful for your feedback.