 Well, I thought I would just start by reading something very briefly. This is a senior policymaker in a country who said in 2011, for too long tax reforms have been approached ad hoc without regard to their effects on the evolution of the tax structure as a whole. As a result, many parts of our tax system seem to lack a rational basis. Conflicting objectives are pursued at random, and even particular objectives are pursued in contradictory ways. OK, which country do you think that is? UK and the United States. The UK, actually that's a quote from a UK policymaker in the Institute of Fiscal Studies, but it could actually apply as much to the United States. So I read that quotation out, because I think we all agree that perhaps even the Nordic countries have the perfect tax system, and this is a major area of policy, particularly for our concern, which is the developing countries. But also the tax system you get is a product of technocratic issues, the optimal choice of taxes, but it's also, as our speakers, very openly put, a choice and outcome of political economy and political factors as well. But it's also actually a product of data. Sometimes we seem to reach decisions around taxation on the basis of a large amount of ignorance about what we have there as a tax system. And that came out very strongly for me across all of the presentations. I just sort of like to comment a little bit about the political side. I think it's very interesting that the work of the ICTD has really married the political science perspective together with the more technical economics. Because economists like me, we reached for this social contract idea. So I went around for a long time spouting away about social contracts. You pay your taxes, you get an effective state. And then Mick comes along with a view based on some evidence, but we don't have enough evidence. Well, maybe it doesn't actually work in that way. And one reason it might not work is that actually we don't have a single unified taxpayer that, as Mick said, there are many different types of taxpayers out there. And they each have different sets of behavior. And we need to understand that behavior so much more. So that's something that's come out very important from this level of research. Moving to the kind of the level of tax benefit modeling, I mean, one great strength of the tax benefit modeling is set up by UCA. And it's something you see in the developed countries that you really get a grasp on the political choices. And it's very important when that tax benefit modeling, the results of that are presented in the ways that the average person can understand and the average politician can understand. So that, for example, when George Osborne sits down in the House of Commons having delivered a tax decision, Institute of Fiscal Affairs, or some fiscal studies in the UK, at least give you some sense of what that means with a different strata across society. And that's important not just in a technocratic sense. It's also important as a democratic tool. Now, the two further points are going to make. One is on macroeconomics and one is on aid. And the macroeconomic point was that it's very struck by Liz's graph, showing the slump in corporate income tax revenue in South Africa. And indeed, we see a slump in tax GDP and the level of taxation across Africa as a whole in many developing countries around about the time of the 2009 financial crisis. We're now heading into another set of macroeconomic difficulties because commodity prices are slumping. They're taking down the revenues that governments hoped they would have in large amounts. Alternatives have to be found. Governments are going to be entered into periods of stabilization. When governments enter into periods of stabilization in a Russian crisis, they rush for taxes that are often very regressive in terms of the poor, the easiest taxes to grab, and you can end up in severe difficulties for poverty and social inclusion that way. And, you know, one good thing about research into policy is that we could at least help governments in making those tough revenue decisions at least be a bit more informed about the social consequences in a way that we were, say, 30 years ago when we had some of the big stabilization crisis. The second thing, and this is my last point, Kari, is the role of aid and taxation. It's very common in a policy debate for people to say, oh, governments get aid revenues, therefore they don't need to tax. But equally, donor agencies have been involved in a lot of tax reform over the years and still are. The amount of technical assistance that bilateral donors have given, but also the IMF. So in some ways, it's also not surprising that tax revenues can go up simultaneously along with foreign aid. But it can be the case that donors tend to sometimes over-push countries towards raising a domestic tax base, because donors also have to understand that countries have to do that in a way that's politically compatible with their own political settlement. And again, that's where we need this mix of research that we've seen demonstrated today, both the politics, the economics, the data, and whatever else we've got. Thank you. All right, thank you very much. I'm gonna start to do two mistakes. So this is data, probably based on the IMF or the World Bank tax GDP ratios. But it's not my data, I borrow it from Bestly and Persons. Second, I'm gonna just mention, I wanna talk about an average country. So at the lower bottom to the left, you have this blue dot. So that's where you have the low income countries. In the middle, we have the middle income countries. And then far to the right, we have the high income countries. And you see this is kind of average. We call it, let's call it an average country, the line. As you can see, some of the poor countries are doing better than the average, some are doing worse. Now, I'm coming back to the reason why we want to raise the tax GDP ratios, as Jukka mentioned. It might be because we want to deliver more public goods. So we need to increase the tax GDP ratio. If you take the average country among the blue ones, so let's say they have 15% of GDP in revenue and then they get 5% of aid in support on GDP, maybe they borrow a little. So that ends up with the public sector around 25% of GDP. Now, if you want to start to achieve the sustainable development goals, or even the one we had before, the MDGs, we or UN were doing a lot of calculations on how much taxes or how much revenue do you need to reach those goals. So there is this pressure now on the poor countries to increase the tax GDP ratio to, in order to achieve these goals, forthcoming SDGs. So how do we do that? And I think the, some lessons, I think we got here now during the presentations. I worked a bit on taxation, not very much, but I think that is an excellent initiative because if we are supposed to do some kind of advice or do analytical work, we need high quality data. And as mentioned, we don't have that, we didn't have that before. So I think that is very good. And moving on to South Africa, I think also when you have that type of registered data, that also enables us to do this kind of analysis, which is very important. Because one way to increase your revenue, or I mean, there are two things you need to do, you need to expand your base, that is make sure that you have GDP growth, and then you have the rates. But we know the rates, we are, I think the kind of consensus among development economists is not to raise the rates too much, but more of a broadening of including more of those that doesn't pay today. So I'll keep the system simple, but bring in those that are supposed to pay. So that would mean that we could probably move up a bit here, but in the end, it's very much also the base that will determine how far you can go. And I think the project Yucca mentioned and all the presentations here gives a very, a lot of interesting thoughts and how to improve the capacity and the analysis in order to push those countries. You could perhaps push them a bit up, but then you have to move along the train. But we also, we don't have to, we should not forget the growth side. And I just want to conclude, this is another data-scary graph. So we have three countries. So I can say the blue one, sorry, the green one is Tanzania, the red one is Kenya, and the blue one is, not very clear, it's Sweden. My point here, I think this is the data from the medicine database. The average tax rate of Sweden, that was, okay, Kenya's around 25%, but we were quite rich before we reached that 25%. This series, at the left, we are in 1880. And then you have 100 points, so you end up in 1980. So I just want to final caveat here about this trade-off between taxation and growth. I think that is something I need also to think about. Thank you. And before giving, before going to discussions, I would, questions, I would ask you from the floor to keep your discussions very succinct and short. So please direct questions and not long statements, so please, over there. Very quick question both to Alex and to Mick. You both made us a strong point of the fact that the new data will be used to interrogate the situation in extractive industries and their taxation. Neither of you mentioned the Extractive Industry Transparency Initiative. I found in the countries where I've worked in mining that that has been a very valuable source to check on the quality and the content of administrative tax data. Do you not think that that is a good source, or do you have some suspicions and doubts about the usefulness of the EITI data in those countries that have signed on for that initiative? Let's take a second question and then we come back to answers. Oh, Hanna. I'm Hanna Rengina from the Finnish Foreign Ministry. I would like to have your view on the global UN tax body that Professor Stiglitz so forcefully defended this morning. Thank you. Okay. On Sir Studio, I want to comment. There was a question, I think, directed to Mick and Alex directly. And then the second, anybody? Sorry. Oh, sorry. The EITI, I don't want to be rude, but I'm going to be. I think I share what is probably a consensus view, at least on the civil society side, that the main value of the EITI to date has been in the mobilization of civil society rather than the data that's been generated. But in terms of the actual data, there's two things. One is that I think in general, countries that are signed up to the EITI are also generating better data. So there's less likely to be, in a sense, because of the valuable role it plays, there's less likely to be value in checking it or having it fill in gaps. But the other thing is the EITI, to its great credit, is now in a process of understanding the limitations of what it has asked for previously and moving, I think, possibly in the next couple of years towards a mandatory format of data. And what the data has always missed is a denominator. So you've had revenues without necessarily having the right denominators to put that against, which really undermines the corporate accountability side, the value of it. So I think it had a role, but it's more indirect, perhaps. On the UN tax body, Hannah, I think we've had that conversation. I worry about how much energy has been put into that when I think there may be other wins more easily available. And I worry about getting a body which is insufficiently resourced, actually, to do a great deal, but takes a lot of political energy. On the other hand, the fact that what the OECD is in the process of delivering in BEPS is increasingly, as we get into the political negotiation in the last stages, shutting out developing countries from a lot of the benefits really makes the case that the OECD just isn't fit for that particular purpose. And I think the fact that the lobby against it being more inclusive has had the success as it's had in the last few months is making the case for UN tax body increasingly hard to resist. Well, this is primarily to Yuka. What you said sounded extremely interesting. And I can even imagine other research in the tax area which could be fruitful for the future. For instance, the role of presumptive taxes. There are lots of situations where it could be too difficult to simply compute income in a way usable for tax authorities. And that would lead you to attack something else, which would be sort of a rough measure of, or at least would be correlated with income in a useful way. So that's one. And then following up on another remark, it would be very useful to figure out what the effective taxation of extractive industries is at the moment and how you can improve on the system by perhaps using another configuration of taxes. Thank you. Elena Perez from University of Western Cape. Yeah, my question is as well for Yuka. You say that around this tax benefit modeling, that was a bit puzzled. You mentioned that you're using representative household data. But in the case of South Africa, for example, we also hear that around less than 10% of the households or the population is part of the tax paying group. And that's the top 10,000 accounted at the peak for 70% of the revenue of income tax. So if there's that willingness to recognize there's not such a thing as a representative tax payer, how is it that we're now moving on to thinking about a representative household in the model? Thank you. Before giving the floor to our speakers, I would like to put one question directly to Yuka. You said Yuka, that tax and social protection arrangements should be considered as a system. But what about subsidies? In some countries, government is also subsidizing, for instance, petrol prices, which is basically a subsidy to the rich. Shouldn't that be also part of the system that you mentioned? Yeah, thank you for the questions and comments. Yes, the idea of resumption taxis is do you mean normative work or descriptive work on the impacts? Because to some extent, the paper by Henry Claven and his co-authors in forthcoming in JPE is about some sort of resumption taxis because it's on a, in Pakistan, the small firms can't escape the turnover tax. I understood that in the South African case, it's a choice. The firms can choose whether they are taxed according to the turnover tax schedule or not. But in the Pakistani case, the tax system treats the small companies by just assuming that their rate of return is something because it's based on the turnover directly. So in a sense, there's already some descriptive work on that. But how to start normative thinking of that? I'm very pleased to collaborate with somebody as clever as you are on that topic, because I don't have at the moment great ideas on how to do that. Sure, I mean, taxis on the natural sector and on the mining sector, that's a great topic for research. The thing is that we are a rather small institute and then we can't handle all the possible topics. But that's certainly something that we simply can't ignore when we talk about taxis, for example, in the African context. Then I was probably, there was a question about the tax benefit micro simulation model and the use of representative agent. I think I was unclear. What I meant was that we are planning to use survey data, which is representative of the population as a whole. So it has like thousands of households. So it's certainly not a representative agent model. It's a model which is exactly geared towards analyzing redistribution and distribution. So obviously there are issues regarding the data quality and especially to what extent the variables in the data sets, which are available, actually match with the tax and benefit legislation the countries have. So that is why we have already been engaged in an analysis of contacting feasibility studies on whether the idea of building a tax benefit micro simulation model for country X would make sense given the data and given what we know about the tax benefit legislation. So it's certainly not representative agent. And Kari, you are quite right. Yes, subsidies should be seen as a part of the system. And actually it is a big puzzle why, for example, those subsidies and things like petrol are used to that extent that they are. I mean, they are often motivated by distributional concerns that the poor households spend in proportional terms more on some of these items which are subsidized. But they could easily be, I mean, easily and easily, but at least you could come up with a policy where these subsidies would be replaced by transfers, basically, to those households which are below the poverty line and removal of the subsidies while financing these transfers would still leave money to the government, which it could give to some other useful purposes or then reducing other taxes. So there was an excellent talk by Michael Keane from the IMF last year in a wider conference on precisely this topic and some of the political economy considerations that can actually affect that choice. So why is it so hard to get rid of these subsidies? Right, I mean, this is just a point of information. But in terms of there's an interesting thing in terms of South African data that, for instance, in our second last income and expenditure survey that was the one conducted in 2005-6 where there were, in fact, detailed questions about income tax. Unfortunately, the data wasn't presented like that in the last survey. But in that survey, the aggregate income tax that households thought they paid was actually half what was collected by SARS in that year. So that's quite an interesting thing. It might mean that PAYE is actually a very good tax in some senses. And just the other point, I mean, just in terms of discussions on the kind of the extractive industry and so on, I think it also came up in conversations that in South Africa there's something running at the moment called the Davis Tax Committee. We actually have a member here sitting here in good will honor if anybody wants to speak. And there's been a whole series of reports on various aspects of, as they've been generated, being published, various aspects of the tax system. So there has been work done on mining and so on. And that report, it has been published a really, yeah. It was published about three weeks ago, so anyway. Just for interest. Yeah, just a couple of comments on this. The first one is what is the fiscal system comprised of? Yes, it's expenditures. Yes, it's subsidies. But also, but don't forget, tax exemptions, a very big part of the fiscal system of many countries, especially low income countries. I'll give you an order of magnitude. I'm not sure if it's true, but it's probably right that on average in sub-Saharan Africa, tax exemptions to investors might account for about one third of total tax raised. It's a massive figure. The other comment I wanted to make on was the point about research on mining taxation. Yes, all in favor, but the reality is it is very difficult to get the data. Not so much the data on tax on the mining sector as the data on the mining economy itself and what is going on. We know there is massive transfer pricing taking place and the figures that are actually available on the economies of mining projects, well, the publicly available information is little and often not very good. There is much more private information available in private databases that you can buy. So there is some scope, which is why I said we wanted to focus on this. There is actually some scope for hopefully extracting some of that data from private commercial databases and trying to bring it into the public domain. But I think without better data, there are plenty of economists around who'd love to do research on extractives because it's such a very, very big issue. But they run up into data problems very quickly. Thank you. I think we have time for two more questions. Jaco, first one, and then you. I'm Jaco Kangasimi from FinFAN, Finnish Development Finance Institution. And my question and comment really relates to that. Whether you are assuming that the contribution of companies to public financing and financing public goods actually comes through taxes. Those registered recorded taxes, since in many cases where we are involved, we actually see that not to be the case. I take two examples. First one, energy sector. In the energy investments we have recently made, independent power plants producing basically selling to the government-owned utilities and making very, very large contributions to public finances, but not primarily through the taxes that they pay, but through the fact that the power that they produce is cheap, very much cheaper than the current marginal cost of power that these government-owned utilities are these days paying to the sort of emergency power suppliers that are the alternative in some case. If we dare to go to really sort of post-conflict places where nobody has dared to invest in efficient energy investments in a generation, the ratio could be even that basically, you could produce power at a quarter of the cost or maybe even less than that than what is actually currently being paid for these emergency arrangements that they have. That was one example. The other example comes really from the poorest countries from rural projects where yes, the companies do pay taxes, actually dozens of different types of taxes and fees for all sorts of inspection and all sorts of fees to various government bodies, but the real action, the real money is what they are asked as part of their license to operate to do directly. So they are investing directly in schools, clinics, roads, bridges, wells, et cetera. That's what they are asked to do for the local community. That's from the point of view of the local people. That's bigger thing than the taxes that they pay. And actually from the finances of the company point of view, that may be the bigger money in those directly provided goods and services, rather than the various fees and taxes that they pay. Thank you. And the last question. My name's Imran Valodia. I wonder if Mick could talk a bit more about the work he's been doing on the informal economy and taxes. I think it's a really important issue because what one usually thinks about the informal economy is that it's people who are trying to hide from the tax system. And for many of us who work in the informal economy, we sort of don't see anyone trying to hide. So it would be quite nice to hear what he's doing and whether you have any sort of measures of effective rates of tax and things of that sort. Thank you. So who would like to comment on these two questions? Okay, let me, the first question first. Look, I have every sympathy with someone who's involved in the private sector sitting in these discussions and they hear tax and companies avoiding tax. And it sounds as if somehow we're blaming companies for this. I don't blame companies for this. The problem of the evasion of taxes by companies big and small is the result of complex interactions between governments and companies and taxpayers. It's not the fault of any one person, it's a system. Even if sometimes we campaign against companies, I think anyone understands it's a complex system. And I have every sympathy also with companies that invest in political risky situations and they are asked by government to make direct contributions of various kinds, infrastructure, education, et cetera, and they do that and they say they're making their social contribution and I think very often that's true. But what we're left with in the extractive sector is a situation where in very many cases, companies are taxed on the basis of contracts that they write directly with politicians, not on the basis of tax law. Sometimes they are in fact exempted even from filing taxes. And therefore, and suggestions that somehow they're evading taxes. Of course you can reply, well we're making various other kinds of contributions but how do I as a tax payer really know what your other contributions are and whether this is equivalent to the taxes you should be paying. So in my view, and this is not blaming companies, I mean we have a very complex system but we need to shift towards a situation where extractive projects are taxed in exactly the same way as other economic activities not treated as very special. Because when they're treated as very special there are massive amounts of what we politely call side payments involved and you know that really needs to stop at some point. Sorry, very briefly on Emory, let me just, we can talk later. I mean two points about informality. One is I think, especially for sub-Saharan Africa the very interesting question for research is not so much the informal sector but is informal taxes. And a survey we've done in Sierra Leone suggests that the average rural citizen pays more in informal taxes than they actually pay informal taxes to the state. The other point, I just want to make a point about the concept of the informal sector because I think this has misled us for a long time. When I say informal sector and possibly you say people think of someone who is selling cigarettes in the street or banging bits of metal in a backyard or something. From the tax point of view that is not the problem. These are not taxable people. The informal sector from my point of view is any sector of the economy where there are regular large transactions made in cash. And if you redefine the informal sector that way the informal sector is not located in small backyards it's located in the offices of hospitals and doctors and other professionals and many other businesses. I mean that's what the informal sector is, large cash transactions. So I'll stop there. Thank you. I have to disagree a little bit with Mick. Just, you know, to keep in practice. I do want to put some blame on companies. You know, I think Mick has made very well the point that it is a complex situation always. But I've spent too long being told we just play by the rules. By companies who are actively and very expensively lobbying to make sure the rules do not involve them either paying any more tax or being any more transparent. So I think there are cases where we actually should put some blame. But it's quite specific to that. Now in general, you know, I think the cases you raise are very reasonable ones and I think it's certainly true that we lack transparency about valuable payments that are made by companies. But then I say, you know, let's step back and on balance. Do we think that what we don't see involves in general companies paying, you know, their fair share but not wanting to support transparency or companies paying a bit less than that? And it's hard not to think that the second of these is more likely in general. So I'm certain there are companies that are, you know, paying a great deal in those social contributions in that direct way. But equally, I suspect there are a few countries, if anywhere, on balance that outweighs the tax that isn't being paid that should be. So I would take a slightly less optimistic view of that. Just a brief comment on the question of formality. I think while we have an emerging body of evidence from the Latin American context, much of it was cited by Santiago Levy in his talk yesterday on the impact of expanding social protection systems on the extent of formal work. You know, much less on the impact of taxes on the formal sector, on the share of the formal versus informal work in a set of other developing countries. So again, this is a key margin of response in the developing countries, but the body of evidence would be, what we have at the moment is really not great. So more work should be done on that. Thank you. I think our time is up, so I would like to thank our presenters, the persons to the Tony and Jörgen who made also the responses. I think it's obvious on the basis of what we have heard that we will need more data on taxes. It's good to hear that that UNU wider is getting also involved in research. We have heard the commitment of wider what to deliver. We are looking forward to that. And I think also I would like to mention that not only is the tax data important for the countries when they plan their tax policies, but also for us donors who work in some of these countries, such as for instance Mozambique, Tanzania, and especially in relation to the extractive products where we're very, I think it's important to know what income there will be from these extractive products and how that income is going to be used for the government in the context of financing for development. Thank you very much everybody.