 Now, I'll also introduce the discussant for this session, Anna-Lena Oppo, who's a Leave Her Hume fellow at the LSE. And her work focuses on inequality, perceptions, redistributive preferences, and social protection in the Global South. And she's also a former focal point of the Government Revenue Dataset Project at UNU Wider, which I think is why we're here. So Anna-Lena, you'll have about 10 minutes. And then we'll get some responses from the panelist. And then we'll open the floor to questions after that. Thank you for the introduction, Kyle. And thank you, everyone, for these very interesting presentations and for this session. I think I agree with Christian on what he said in the beginning, the value or the importance of the perspective on data, which underpins a lot of the research that we heard about it, and policy analysis and advice. OK. So we had these two presentations that primarily focused on the data project. And then one that was a bit more applied, telling us a story that is very important and interesting. And what I also really liked about the presentations that we see in that, we understand sort of that these data collection processes are also facilitators of a lot of collaborations and partnerships. So on the first presentation, made by Christian, on the tax expenditure, I was one of the people in the room who had to get a clear definition of what tax expenditure is. So thank you for that. And what I also found very interesting is what you mentioned in the beginning, that it can work as an important incentive for certain policy goals, right? You mentioned, for example, the link to climate change and so forth. So that was very interesting for me to see as well. What I wanted to ask, if you can give us some further reflections on the distributional effects of tax expenditure in terms of direct and indirect benefits. Who's benefiting? Who's missing out? And also, you mentioned that you want to have a rational use or promote a rational use of tax expenditures. So I was wondering if you can give us more thoughts on what rational means in that context. And then I took some notes where I have to think about what I meant. In terms of, oh, yeah, I found it very interesting when you spoke about the universal benchmark as well, sort of as almost a non-positivist approach of seeing all the catalogs, what can be taxed, as in other contexts that's not even on the radar. And hence, it's not accounted for. So I was wondering if there is also a rationale to have this kind of like, it almost reminded me of the tax effort debate on all the possible types of taxes that you could have and why they are not relevant in other contexts. So perhaps there is some reflections or merits to talk about this. And on the data, I wanted to ask when there's only aggregate data, is that a matter of reporting or is it a matter of politics as well or both? So just interested in whether there's also sort of like an incentive to keep some information covered, let's say. Thank you, Evgenia, also for the presentation as a non-tax expert. I cannot give you a lot of technical input or questions, but I really liked the global minimum tax so that was also something that I wasn't familiar with. And what I also found very interesting and when you showed that the size of the cake as in tax revenue is still increasing, it's just the composition that changes, right? Which of course raises important questions on the tax burden, who benefits, right? In terms of profit shifting, I also wanted, as a sort of like outsider to this field is like where do they typically shift to? I mean obviously in terms of places where they're not taxed as highly or sort of who stands to benefit from this as well. And I was thinking if there is a link as well in terms of like say the global minimum tax would sort of be a remedy for this disincentivizing this behavior if there's a dialogue to be held between you and Christian on the tax expenditure sort of as a tool of setting incentives, right? And on the last presentation what I really liked as well is like seeing sort of that link between data research and policy and practice, the partnership it builds, the workshops that are held like linked to these data collection efforts. Having worked a little bit on data harmonization, I know how tedious it can be. And sometimes it almost seems a little bit like an artificial process that is mainly geared to like ticking the same boxes. So I was wondering if you have also in these consultations if you have like some element of this pick and choose or trying to kind of sanitize also some certain maybe political elements of like making it fit and what information may get lost. And similar to what Christian introduced with the transparency index, I was wondering if it makes also sense to have that sort of in terms of incentivize better reporting perhaps, but maybe I'm too idealistic. And I think yeah that the last question I think is on the last slide, which is a little bit maybe more on the content if you have any insights as to why tax revenues have stagnated in the time period that you've been looking at sort of. Yeah, thank you to the presenters and then I'll leave the rest of the time for the floor. There any questions which you will moderate, right? Okay, so yeah, if the panelists want to just sort of go one by one and respond to those remarks and then we will open the questions up to the floor and if there's any online as well. Thanks for your comments. Regarding the distributional effects of tax expenditures, as I have said evaluations on tax expenditures are rare actually. What we see is that in most cases where evaluations take place that tax incentives tend to be more regressive in their distributional impact than progressive. This said there are some tax expenditures as for instance the income tax credit in the United States and similar measure in the UK that have a progressive impact with regard to income and also with regard to employment. So it's difficult to give a universal picture but quite often actually tax expenditures that do not aim at having a distributional impact but aim at promoting growth or incentivizing investment or promoting certain types of behavior do this by having also a regressive impact on it. And this leads me to the question that you asked about the rational use of tax expenditures. From our perspective, a rational use of tax expenditures would be able to answer two key questions. One question is is the tax expenditure under review effective in terms of reaching their stated goals and of also not having undesired side effects? And is it superior compared to other types of public policy tools in particular direct spending? And you have governments that include these types of questions in their pre-assessments of tax expenditure. So when you want to set up a tax expenditure you have a certain record of guiding questions in the Netherlands and Ireland for instance where these questions are being asked and we have to provide some type of evidence with regard to that. Universal benchmark, it's a very nice question also sitting next to OECD. It's important institution when it comes to setting up and seeing through international standards. It's not likely that in the near future we will have a universal benchmark. There are certain initiatives of having regional benchmarks. ECOVAS for instance in West Africa is promoting, I'm sorry, East Africa is promoting or is it West Africa? West Africa, sorry. It's promoting this type of regional harmonization with regard to benchmarks. From our perspective you could have some intermediate things in the sense that you could have a joint understanding of what constitutes a benchmark measure or what constitutes a tax expenditure. To give an example, almost all countries would agree that a threshold income below which you are not taxed with a personal income tax is part of the benchmark system. So here we would say, okay, there is an international consensus on that. But with regard to reduced rates for basic goods for instance in services, we have diversity. Germany does not consider reduced VAT rates as a tax expenditure. Most other countries would do that. And it would be nice to have a joint understanding with regard to those measures as well. But there will always be some kind of variation. And the last question was if the issue of aggregate data versus detailed data is reporting or politics, reporting is politics. Setting up those reports is a political decision. Governments are transparent because they choose to be so either out of their own understanding as being a good government or because there is some kind of pressure from civil society for instance. And in this sense, if they do not provide provision level data but only aggregate data, usually it's not a data issue. It's an issue of a decision to present the data. Anna-Lena, thank you very much for your questions. Regarding the first questions, who benefits from tax burden? I mean, if corporations shift profits to tax havens or to low tax countries, what happens actually like the main beneficiaries are the owners, the shareholders of corporations and the owners of capitals. And if we speak about losses of such movement of burden, like for example to individuals, like from the research of Gabriel Zuckman and also Thomas Piketty, we see like the examples of even France, like really rich country. In 2018, it was the situation that top 10 percentage of rich people, so it's like about rich people, they paid in terms of the taxes less than the rest 90% of people. So like we speak about like poor and middle class. And the same extra situation we see in US, that's why like if it could amplify inequalities, it actually lies on the shoulders of like poor and middle class. The second question, why do they shift profits about corporations and who benefits? So as I like mentioned, they mainly shift their profits to tax havens. And as Annette, for example, mentioned today during her presentation, there is also the gravity effect that they shift to jurisdictions that are near them. So mainly, but they have like this jurisdictions with the low tax rates. About global minimum tax, would it desensitize such behavior? It's the same as about BEPS project. It could actually desensitize them, so it could actually move some of the profits from tax havens. But there is no evidence I mean now because there is no data from the notation of global minimum tax. So it's going on. But there is a set of the literature that actually simulates the global minimum tax implementation. And what it shows that in comparison with like developed countries and developing countries, developed countries could benefit more. And the reason of it is that for example, for developing economies, they rely a lot on investments and like attracting investments, especially with the special economic zones. And such a policy, global minimum tax, could actually desensitize investors like to invest in the government, like in their countries. And there was the discussion of the rate that should be implemented. So now it's like 15%. But for example, in the book by Gabriel Tsukman, he suggested like 23% if I'm not mistaken. And African countries also like suggest the level of 20% and higher. So yeah, there is a huge discussion and yeah, about global minimum tax. Thanks. Thank you very much for your comments. I think I took down two questions and forgive me if there was. I think there might have been a third, but I was already trying to think what the answer to the first one was. Please remind me if I missed one. So the first question was what information gets lost in the harmonization process? That's an excellent question. That's why I was already starting to practically think what the answer might be. And then the no less complex, but perhaps more answerable question of why tax revenues stagnated in 2010. So on the first question, absolutely, information gets lost in I suppose the first issue that we have in the harmonization is what constitutes a tax? So the OECD has a definition of tax revenues and it has also thankfully expanded and more into non-tax revenues as well. But for now the big discussion is or a common discussion with country focal point is whether or not what they might consider a tax instrument actually is a tax instrument. And because of the OECD's definition it means that some countries revenues, for example, from natural resources don't go into their tax GDP ratio which is very contentious because then they see it looks as if they're performing much worse than they would consider themselves when they look at the tax GDP ratio. So first of all, the OECD definition of taxes as compulsory, unrequited payments to general government sounds sensible but that leaves various things out. And so I'm pleased, as I say that we're now but yet non-tax revenues as well for Africa and the Asia Pacific because that allows us to get a more complete revenue picture. And so we'll continue to do so obviously. Then within when it comes to different tax types I think there are fewer discussions the discussions are perhaps a bit more straightforward because the OECD classification is for the most part very clear and it's pretty well disaggregated, it's pretty granular. So where there is perhaps disagreements with the national focal points often the national focal point may carry the day because it is up to them ultimately whether or not we publish their data. But for the most part, I think this is very, very marginal because for the most part the classification is very clear. Now, as to why tax revenues have stagnated in developing countries, well I'm going to extend the OECD as well since the global financial crisis. First of all, I think a couple of caveats to that. First is the way the tax GDP ratio is formulated. Obviously it also depends on what's happening with GDP. So if an economy grows at 10% a year and tax revenues grow at 10.1% a year it might not lead to a great jump in your tax GDP ratio but it would be unfair to say that tax revenues is stagnating, they are there, they're sort of holding fine. So it shouldn't be necessarily the case that there's been no growth in fiscal space in developing countries during the 2010s on the flip side. Of course there's interest payments have become more onerous. Actually you can easily say that fiscal space is narrow. So one of the, we look surprised at the pandemic even at the proportion of increase in tax revenues in African countries that has been eroded by the growth of debt servicing costs and it's about two thirds. So that, and that was a three time dynamic so you think it'd be even higher. And so there are these different dynamics to take into account. At the same time these averages are, you are looking at an average for 33 different countries in Africa now. There's a lot of heterogeneity, not only in the level but also in the trajectory. So a country like Togo in 2010 was a couple of percentage points beneath the African average now is a couple of percentage points above. It's been a hugely successful period for them involving a lot of reforms to tax policy, to tax administration, huge political commitment is a great story to see. On the other hand you have countries where there's been declines as a percentage, and revenues as a percentage of GDP. Often we found that this relates to mineral producers or oil producers where they had huge falls in line with falls in oil prices. Now maybe in 2021, 2022, these will pick back up but these are incredibly volatile revenues and so there's a lot going on beneath that story. I would say all that being said, persistent informality is a huge constraint on the expansion of revenues from tax sites that are currently, as we saw in that final slide on the tax structure. If you don't address informality, your PIT revenues won't grow very much in neither of those social security contributions and that hopefully has interest users and social protection researchers. And then there's also just the slow pace of economic diversification in many of the countries that are covered and so that's, I think, one of the things that you'll need while you're searching, it does so well at, it provides that big picture. It looks at revenue mobilization from the perspective of the overall diversification and development of the economy and I think that's something we should know about. And so that's it. Thanks so much. So we have somewhere between five and 10 minutes left. So at this point it would be good to take some questions. Firstly, are there any questions online? Sadly not. Do we have any questions in the room? We have two, three, four, okay. If they're brief, we could take, let's take the first two that went up, so Arhelge and yourself and then we come back. Christian's confident he can handle four questions. Thank you so much. I have first a question to again, yeah. Very interesting presentation, both interesting but still is good. I have much to learn how to do a proper presentation. My question is related to what do you have information on the sectors of these multinational companies? What sectors are most aggressive when it comes to profit shifting? And second, where are the headquarters of these multinational companies? Then I have one question to Alexander. There are now quite a number, several government revenue data sets available. You have the IMF, which has existed for decades, I think. You have the Univider, government revenue data set. Now you have the OECD. And when you look into and try to compare, you find also that inconsistencies here. And you find also that the user friendliness of these data sets are different substantially. I must say that I'm most impressed of the Univider government revenue data set. If I want to make a presentation, I go there. Very intuitive, easy to use. And my question is, do we need all these data sets? What is the added value of producing more amounts? Wouldn't it make sense, given that there are also inconsistencies in the data between these different databases, that one aimed to join forces and develop one kind of harmonized type of data set, which actually had more reliable, credible data and which then would be more cost-effective to operate? Thank you. Yeah. So there was one just in front of me. I can just say who I am. I am what gets that from the CMI Bergen Norway. Hi, my name is Christina and I work for the Extractive Industry Transparency Initiative, where the secretary is located here in Oslo. My first question piggybacks on your comment. Are 57 implementing countries report tax and non-tax revenues, disaggregated by company level and type of revenue? And we map those to the GFS classification. So what's the value add of another classification because this one has been established and all our countries are already doing this work and doing another mapping is extra work and we should be analyzing the data. So I want to be a little bit provocative. What's the value add? Because we were already doing this and we have been for almost 20 years. And for you, Christian, I had a question on one of the things that we see happening in our research rich countries where we have state-owned enterprises that play an important role. There's this issue of quasi-physical expenditures. I was wondering if this is also part of the data set or the issues that you're working on because we see a lot of countries struggling with understanding what that is and what the importance of that could be. Thank you. Do you want to take maybe one more question before responding or is there one more? Daniel? Okay, thanks. Sorry, we're eating into eating time. So. Some of my questions might have been raised already. So I would go back to them and let me add this question might apply to Alexander and probably to some extent Christian. First of all, thank you for those presentations. One of the frustrating things a researcher might encounter is when you're working with tax data and then you find out that there are some countries of interest but there are no data points for those countries. Or you might find some countries of interest but if you look at the period, you notice that they are gaps. I know over time there's been some improvements but probably I would like you to reflect a bit more on what the challenges are and how this has been addressed over time. Okay, thanks. Let's say between one and a half and two minutes response each and then we'll wrap up from there. Thanks. Thank you very much for your questions. I would be brief actually because the first question was what sectors are more aggressive multinationals? So yeah, we have this data but I need to look at the data back in order to answer you properly. So we have this data, that's right but we don't really show this in the paper that is aggregation among sectors because the most important thing was about aggressiveness like multinationals. And the second is like headquarters of this multinationals. So what do we have? So as you remember probably from the slide that we like in the end have 20,000 of subsidiaries among multinationals and the exact number was 842 or 43 I think. Like they belong to domestic multinationals. So this is desegregation but again like the headquarters, I mean the country so we need to look back in the data. Like we have it but we need to look back. That's it. But thank you very much for all these questions. I think that we need to add this to the paper. It's worth it. Thank you. Okay, yes, thanks for the question. Yeah, you from EITI, you know of course that the extractives sector poses all kinds of challenges with regard to transparency and data collection and tax expenditures are of course not alien to the situation quite on the contrary. We know that many tax expenditures in the extractives sector are granted through agreements and they do not appear in the reports. So this is one issue and the second issue with regard to state-owned enterprise and quasi-fiscal expenditure. At this moment we do not have it on our screen actually and I think it's also again an issue of data but let's follow up directly afterwards. I'm very much interested in hearing your thoughts about this and with regard to data gaps from our perspective with regard to tax expenditure the key issue is the quality of reports. This decides over the data that we have in our database and what we do is we see this as a major policy challenge at the national at the international level and try our best to advocate for better reporting. That's what I can say about this. Thank you very much. Unfortunately, that's all we have time for so thank you very much. Thank you very much for your questions. I'm not sure this is not the appropriate moment for me to make a comparison or explain the strengths and weaknesses of the respective databases. I would say two things. Because these have been sort of discussed as well I don't want to quote people without that portion but I think it has been said frequently that the the data space today is unrecognizable from where it was 10 to 15 years ago that there has been a great development in the quality of data available. Now, as ever, development is uneven. It's a process that will move forward, step back, there'll be rationalization. It's a work in progress. I can say that from the OECD in 2018 I think we covered 80 countries which is barely enough to call it global. Now, by the end of the year, we'll be at 130 so we're getting there. And I'd say that we're also improving the quality. We have the opportunity, thanks to the partnerships, thanks to donor support, we have the opportunity constantly to be improving the quality of data. And I think Carl would accept that, would agree that there's filling the data points and there's actually filling it well with good quality data. And I think we benefit from this when we pool our experiences and our expertise and we have these different people who are perhaps accessing different information. And over time, the data that everyone is having access to is getting better and it's not necessarily, it's not an easy conversation to have on a panel but we are, the OECD global revenue statistics initiative is a response to a country demand. We have no persuading power on countries to join. We invite them and if they see the value of the initiative, then they join. Equally, the regional partners could walk away at any time. Equally, the donors on whose support we rely might not see the value of it. So I think we will carry on doing what we're doing and we will do it to the best of our abilities and in the broadest consultation possible knowing that exactly you are having these discussions and it's confusing and you might wonder why. But I think it's beholden on us to make, to have a clearer answer to that picture, to that question perhaps over time than we do today but in any case, I do implore you to just contact us to reach out because the differences you might see, we are working on it all the time and I would say the inconsistencies are much, much smaller and as a final note, we have a bridge table in our publications where you can combine, compare the OECD classification with the IMF for example and just see exactly where the commonalities lie and also all our documentation is at least transparent and it's publicly available and so you will be able to see if you're curious as to where the differences lie. It's not the best explanation I've ever given but I hope it suffices in a certain level. Thank you. Okay, thank you very much to all of the panelists, to the discussant and to the audience. I think it's been a really interesting session and I think that text is nicely up to the coffee break so thank you everyone.