 Good morning. My name is Vashiru Vasuta Ramudle. I would like to welcome you all to the fourth of six online policy dialogues as part of the Southern Africa Towards Inclusive Economic Development Essay Tide program research into policy series. Essay Tide is a collaborative research, policymaking and capacity building partnership between the National Treasury, UNewider, the South African Revenue Service, the Department of Planning, Monitoring and Evaluation, the Department of Trade, Industry and Competition, Trade and Industry Policy Strategies, the International Food Policy Research Institute and the European Union. The program also includes a number of local and international universities. The work of Essay Tide has been generously supported by funding from the EU. We are forever grateful for this support. So since its inception in 2017, the program's goal is improved economic research for informed evidence-based policy to promote inclusive growth in South Africa and the region. It is a result of a unique collaboration between local and international officials and experts under six work streams, namely enterprise development, public revenue, macro modeling, inequality, climate and energy and regional growth. You are all able to access all this valuable research done over the past three years on the Essay Tide website. Today's policy dialogue is hosted under the work stream, Turning the Tide on Inequality. We will seek with our panelists to unpack the difficult subject of funding social policy priorities amid economic inequality in South Africa. We will begin with a synthesis of research findings produced under the Inequality work stream by Murray Lamprock. He's the Professor of Economics at the University of Cape Town and a non-resident senior research fellow at UNU Wider. To you, our audience members, a very warm welcome to this policy dialogue this morning. We encourage you to be part of the conversation by writing your questions on the chat box or raising your hand and we will give you the speaking rights so that you can pose your questions to our panelists. They'll be more than happy to respond to any questions that you might have for them this morning. To kick things off, we've got that synthesis presentation from Murray. You can take it away, Murray. Thank you. Thank you very much. Gracie, you're going to put up the slides. Let me welcome you on behalf of our work stream for team. Next slide, please, Gracie. Turning the Tide on Inequality, it's been a very rich partnership between the National Treasury, SAAS, other government departments and the research community in South Africa and internationally, as is reflected by our panelists today, to focus on some key aspects to add value to what we know and what it implies for policymaking about South Africa's inequality situation. It turns out to be an incredibly important context for the broad work of SA Tide, but also context for very specific things. We've had two key prongs to our research work. One was about general income and wealth inequality at the household level. How is the society working? We have also focused quite hard on how the labor market is actually working, and what we've brought to the South African discussion really is the ability to use some of the tax data as well as the other rich survey data that we have in the country to try to understand these dynamics in very nuanced ways, which is absolutely crucial. There's international recognition of the importance of understanding inequality as a lens on understanding how the society and its labor markets are actually functioning and in order to empower citizens to be as productive as they can be and for a society to flourish. These issues are crucial. Obviously, we've had a key prong on government policies for inclusive growth. Inequality sits at the heart of mediating between growth and poverty reduction and good social outcomes. The inclusivity of growth hinges upon an understanding of inequality. As I said, we've pushed very hard on data and data projects and making them available to the research community. We've also run a number of projects that get stuck into the actual researchers within Treasury and across government in terms of demographic modeling that they've got a very frontier understanding of what the population is doing in our country and a frontier understanding of some of the modeling tools that they can use in their work to plan policies well. Next slide, please. Thanks. So to quickly take you on a tour through some of our big findings. There are many papers at the bottom of my presentation, which we'd be very happy to make available to you. I've got them all listed with the links to the UNU wider websites. We're not going to go there now. I'm doing a very high level review. But on the economy and the labor market, it sits center stage, actually, in mediating inequality in South Africa. All the health inequalities, all the education inequalities, all the locational inequalities where you live have a bearing and play through the labor market and we'll only crack our inequality problem through the economy and through the labor market at the end of the day in a sustainable way. We have company tax data available and we've matched that with employee data. And that really gives us a unique lens that this country hasn't really had before on the demand for labor. And also on how our industries and how employment actually happens, how growth happens, what the structures of our industries are, how they interact, do they price by markups, how employment happens. It's absolutely fundamental. And we've got a much richer texture out of the work in our extreme, but also the whole of SA time about that. And it's not a simple labor market. We see, for example, that a lot of the wage setting happen could be located within firms. In other words, firms have some discretion about how they set their wages and firm specific factors have a play in wage setting. But so too do sectors. And there's some really excellent work in the workstream showing that their local level labor market effects as well. If industries are very competitive, very non competitive in a certain area, but it's a strongly unionized area, that leads to a certain operation of the labor market in that area that's completely crucial to the development of a potential of that area and employment creation in that area. It makes the minimum wage discussion complicated. So what the workstream has brought to bear is the pressing necessity for us to have a realistic grounded on data picture of how the South African labor market actually works. Not some theoretical fiction, but that grounded picture. And so the minimum wage discussion, some of the papers on that have shown, for example, that that one needs to look at the impacts of minimum wages in a very dynamic way that minimum wages themselves do not inexorably lead to losses of jobs. It depends on those local level factors and things, but it also depends in longer run on on the second round on the spillovers on the further investment that might take place. Very good work done within the workstream on the employment tax incentive, which is the government policy to try and to try and lower wages for employers in order to stimulate employment. And the workers made an amazing contribution of actually grounding South Africa's evidence policy on on the ETI on on best practice. This country deserves best practice methodologies to work out whether the ETI is actually working or not. And it's a and it's another complicated story. It's it's partly working. These tax data come over 19. And we've got real time data coming in through the UIF payments and through through company submissions into the tax data. And it's the only picture we have of the country's labor market in real time. And I think that the project was very useful to to the country in trying to articulate some of the the emergency relief measures, but there's much greater potential to exploit that. And that's certainly one of the plans that that sauce and is a tight have for the future. Next slide, please. So the work program articulates extremely high levels of inequality in income and wealth. We knew that already, but there's a texture to the work that's done. Our group is on the panel today, as is Leo, and they can talk to some of the work that they've done fundamental work in the work program. But there's also work done on the top end of the income distribution by Ingrid Woolard and others. That's added a lot of texture to what we understand about the country. And here's the state of the nation from a paper that's produced looking using the NITS dynamics. It paints a picture of a country that has a has made some progress on poverty alleviation. But we still have focusing on 2017. Now, a large chunk of 42% of South Africans are chronically poor. They poor every single time that this longitudinal survey visits them. They trapped in poverty in a sense and policy must be articulated to deal with that. But then we have a huge swathe of the country that is not necessarily poor in any given period. That's a 19.4% but extremely vulnerable and precarious and can't invest in the future. And so we have a vulnerability in the country and we have people who are just poor but the next time they're not. So there's this vulnerability aspect to life in South Africa that the research program falls in a lot of the texture with that, on test scores, on health equity, on all sorts of things. There's fine grain data and a thin middle class. And that's one of the key points to note for today's proceedings. If you define the middle class in a sense that they can serve the function that a middle class is supposed to serve, they settle, they're forward looking, they're investing in their children, they can buy houses. We've got a very fragile middle class and an elite. So Tomar Piketty, it's an inequality trap where many of our citizens can't fulfill their potential for this country and for themselves. So when Tomar Piketty gave the Nobel, the Nelson Mandela lecture in 2015, he spoke about South Africa as really top of the class on inequality and his way out of the experience that we can think of. In other words, what we're battling within the country, how come we haven't cracked the back of our inequality is one of the reasons for that is because of its level and the way it reproduces itself and the work stream does a great job of articulating that. So the whole is worse than the sum of the parts. I guess that's the point. These inequalities in education and health and in the labor market and in spatial where you live and in your housing and in your water supply all intersect to create the situation. So the whole is worse than the sum of the parts. That's me quoting myself. Yeah. Okay. I'm being chevied. Thank you for that. I appreciate it. This is not a voyeuristic exercise and it's not an exercise to just shout about the level of inequality. It's an exercise to understand our inequality at the level of detail that we can actually confront it. So here's a slide from the work that was done on the top end incomes. And what's useful here is it's only from the 95th percentile upwards and it basically shows what income sources actually are used, come in, are available to those people at the top end. And this is the bulk of our tax-paying individuals in the country and this is what the income looks like. And so it's very useful to understand this and it wasn't really understood. We didn't have the information. The project has made great gains in detailing these sources of information, in detailing wealth as well. Okay. Next slide please. So here's the, it's the final slide and it's the key anchor slide to transition into the session then. It's taken from some work that was actually funded on the fringes of the project by the French Development Agency, by Maya Goldman and Ingrid Willard and the CEQ Institute. But what it shows is that we do have a rich menu of social policies in the country and they are pretty well targeted to where they're supposed to be. And so the slide depicts, you can see for example that the bulk of educational benefits measured in money terms, spending if you like, the allocation of our budget, is appropriately going to the lower deans. And you can see that that's true of health as well and social security contributions, free basic services are small but they're not insignificant. They're part of our policy landscape. And as you can see where they sit, let's turn to the tax side. Focusing on the direct taxes, that's the pink line. You can see that the direct taxes are coming from the ninth and the tenth there. So the eighth, ninth and tenth, but predominantly the ninth and the tenth there are. This is our inequality coming, wrapping around. We've got a very narrow tax base because of our distribution of income, because of the thin middle class. This is the context for our country. And so on the one hand, we've got quite an extensive social expenditure program that's pretty well targeted. On the other hand, and it's doing what it can, but the intersection of these inequalities is not getting any momentum going on inclusive growth. On the other hand, our inequalities are coming around to severely limit the revenue generation from personal income tax. And that's a dilemma. We're now in a fiscal crisis. COVID has happened. Our social policies have been absolutely indispensable in coping with COVID. But we now really have some fiscal issues to confront. But we also have to confront our inequalities as well. And that's the context for today's session. So thank you very much. Thank you so much, Marie, for that insightful presentation. You've given us a lot to think about and to discuss this morning with our panelists. But before we carry on and introduce our panelists, I think one question for me, just to also look at the overall work stream, the inequality work stream, what would you say are some of the three or four principle lessons from this work stream that you have found over the past three years? You did mention that you've got a number of papers that you will be sharing with us a little later on. We can access those papers as well on the website. But maybe talk to us about the three and four or four principle lessons that you've learned over the past three years. Right. So one is what we're pushing on today. And so we really wanted to do exactly what we're doing this morning because we need to, government has a range of policies that can be tweaked and they can be done better. Right. And so what we have learned is that there is a high return to getting each of the individual policies working as well as they can. Very good research done on how to raise the quality of schooling, for example, improving access to the health system. Those things matter. But at the end of the day, the key lesson was the one that I flagged in my presentation is that the whole situation comes together. So there's no magic bullet. That's one of our lessons. Right. So when we're thinking about policy in the country, we do need to improve the quality of our education. We do need to look at the coverage of our social protection system. But it's their articulation, how inequality works is that all these inequalities intersect to keep us where we are. And so there's a positive lesson for that in the policy space that if the policies can be coordinated, you'll get a very high collective return from a set of social policies that are well coordinated and designed to articulate together. That's the one lesson. The other lesson from the labor market work is that one shouldn't be narrowly focusing labor market policy on wages. The whole industrial structure, employers and employees and the searching unemployed are very active participants. So if government provides a better environment for work and we've got very and we've got appropriate labor regulations in place to mediate the employer-employee relationship, the more we can improve the infrastructure and the environment for business, there will be a response. There will be a response to that. So in some senses, the employment creation discussion needs to think about a very simple process where they facilitate the employment relationship. But our inherited legacy is a problem in South Africa for bringing in new entrants, small firms, the informal sector, formalizing the informal sector, they face many, many tough challenges because of the institutionalized nature of the way industry happens. Let me just stop there. Thanks. Thank you very much, Marie. Now to bring in our panelists this morning, we've got Aroub Chattity. He's a research manager on wealth and quality at the Southern Center for Inequality Studies at the University of the VITS at VITS University and Mamiki Lyolo, a senior official at the South African Revenue Service. And we also have Leo Chaitka is a PhD candidate in economics at UC Lorraine is also a research fellow of the World Inequality Lab. So they've got a lot of wealth, wealth of knowledge around this topic of inequality today. So to kick things off, I'd like to get a few pointers from our panelists today. What are your thoughts of the presentation that was presented or the synthesis presentation that Marie gave a little earlier on? I'll start off with you about this presentation. Apologies, Mashudi, if you can maybe just repeat your question. So there was a bit of a disturbance in our transmission. Oh, sorry about that, Mamiki. I just wanted to get your thoughts around the presentation that Marie presented to us a little earlier on. Okay. So I think how SARS fits into this discussion around inequality would probably be the best starting point from our perspective. As Marie has already indicated that our SARS involvement is that we're providing text anonymized text administration data to this essay type program. This allows us to sort of build a panel data set that can be used by researchers, can be used by policymakers to start informing decisions. So we are feeding into the various work streams that are part of the essay type program. And first of all, I've just mentioned now that we give access to individual and firm level text payer data for research use. And what the goal there is to create a well documented locational data sets with key variables that can be augmented with research topic specific variables. Now, why I mentioned that is this because as Prof has already indicated that it's now useful for us to start linking employment data, company data, firm level data, and start assessing what is working in South Africa, where are we good at, and where are we still needing to improve? And I think for us, as Prof has indicated that the whole is worse than the sum of the parts that areas where we are intervening on may seem to be addressing some of the social inequalities that we were tracking over time. And then some of the programs that are being implemented may seem to be targeting specific areas of intervention. But I think overall, we'd want to see that these programs are starting to translate to the growth of the economy to the growth of the cake to the side, the growth of the size of the cake so that on the other side, on the income side, on the overall budget side, that on the size of side, we can also start collecting more taxes to enable and allow those who are still on the lower end of the scale to be uplifted to provide more social net and give upliftment to those that require assistance from the state. So I think for us, the studies such as this sort of give us an impetus to understand what is going on with regards to our base, our South African text base, or overall national base of human capital. And it also gives us a sense of what are then the income priorities that government needs to start looking into, particularly our role is to bring in the necessary revenue that can then assist these social programs that we are now understanding today and discussing today. Thank you so much Mamiki. Thanks very much. And it's a real pleasure to be here. Thank you for inviting me. So yeah, just to respond to Murray's synthesis, I mean, the first thing to say is congratulations to Murray and all the partners on really driving a crucial agenda that covers a lot of research. And I think it's the synthesis of it that makes it really powerful. One of the clear areas that came out was the importance of data and how Mamiki mentioned the use of SARS administrative data, but how this has been so key to understanding not only specific spheres in and of themselves, but the relationships between those spheres. And I think that's why inequality more broadly is a really crucial way of understanding some of the societal concerns that exist in South Africa, because you're able to explore those relationships, you're able to look at some of the causal mechanisms that are further up the chain rather than with the issue in and of itself. So I look forward to seeing how this project continues and seeing what more data may come out of the various administrative bodies to see how some of the analysis and research can progress. And just to say congratulations again to everyone involved with the SATI project. Thank you so much, Aru. But last person would be and get some insights from Leo on what he thought of the synthesis presentation. Thank you very much for inviting and and congratulations again for this type of I mean for all the works achieved. I think while it's all encompassing and so there are many, many, many issues that are tackled here, I think this idea that there's no silver bullet is which was mentioned by Murray, I think is is a good starting point to try to address the issue. I think also in general it's one has to reckon that it's very hard to curb such high inequality and so that all instruments, all policy instruments should be considered and in sort of a consistent policy package. I really I'm really aligned with such message and maybe I would stress maybe one thing in particular is that if you want to precisely target your policy and adjust your policy and measure also the impact, I wouldn't I cannot stress enough the necessity for researchers and policy makers to have in general better data. I mean I think South Africa is already pretty well placed here. There's I mean all these research effort would not have been possible without access to access to better data but I think one key message also I'd like to stress on is that data is precise data and match like employer employees things like that are key to solve such problems. Thank you very much Leo. Kickstart the conversation really today looking at inequality. Thank you to everyone who's given their thoughts around the synthesis presentation that was presented by Murray. So we know that the pandemic has led the current pandemic that we find ourselves in has led to a sharp decline in tax revenues and in this environment are they feasible previously underutilized areas or from where additional tax revenues can be sought. I think this is a question that we directed to Mamiki and anyone on the panel who would like to respond to this. I think around the world there is a general acknowledgement that the COVID-19 pandemic has done quite a bit of damage not only to the economy but to other societal factors health, mortality rates etc. In our space I would only speak with regards to SARS. In our space of course that has put a strain in terms of our revenue collections. It has also put a strain on many other things that we have to administer as the agency or administering on behalf of the ministry. I think for us there are lots of lessons that we've taken from last year with regards to COVID-19 and there are certain strategic decisions that we've taken that allow us to move on forward to see how we best deal with the challenges that have been posed by the pandemic. So if I just maybe give a quick overview of our annual performance plan for 2021, we still continue on the path of encouraging voluntary compliance amongst the taxpayers with the idea that we exist to serve a higher purpose, that is to enable government to build a capable state that forces sustainable economic growth and social development that serves the well-being of all South Africans. So this is a cornerstone of the work that we do and in terms of our vision 2024 to be a smart modern SARS with unquestionable integrity and to be trusted and admired, there are certain building blocks that we are explicitly working on. There are about nine strategic objectives that we're working on. I'll just maybe pause and just highlight the fifth one that talks about SARS' employing smart leverage in terms of ability to, SARS' ability to extract value from our data. And in terms of that, that's objective, strategic objective number five. And I'll just quickly highlight to you what it entails is the increased use of data to improve the integrity to derive insights and improve outcomes. And here we're expanding and increasing the use of data, data analytics, artificial intelligence to create capability to understand the compliance behavior of taxpayers and to provide clarity, certainty and certainty we needed to make it easy and seamless to provide our services to first up, voluntary compliance, and then to also be able to detect the risks and trends and instances of non-compliance so that we can enforce responsibly. And why I'm mentioning this is because as I think Arupa has mentioned that the use of the data sort of has multiple uses across the many different value chains of policy decision-making. So for us being able to plug in into the area where we know there is a dire need and that is data, data that can be tracked over time, data that can be used meaningfully to make informed decision and evidence-based decisions. So for us we've put that as part of our explicit objective number five to try and make sure that when we do try and make it easy for you to comply with your text affairs, when we do make it easy for us to service you and when we're dealing with those that are non-compliant that we are based on some level of data and information and intelligence so that there is a level of fairness to the process and there's a level of let's call it a scholarly rigor to the work that we do. So I think for us when we are now preparing ourselves to I'm not going to call it post-COVID because we're all going through different waves of the pandemic that having gone through last year we are now sort of trying to see what do we need to reinforce as part of our working package and what do we need to bring on board that's probably is new that perhaps last year or a year ago or two years ago was not something in our thinking framework. Thank you. Thanks Mamiki. Before you respond Arupa to some of the additional tax revenue measures that could be sought by government I'd like to invite our audience members to also be part of the conversation this morning to put your questions in the chat box or raise your hand so that the panelists can respond to any questions that you might have for them this afternoon so all of you are very welcome to do so this morning. Arupa maybe we go back to the question that I posted to Mamiki and she responded to quite lengthly maybe you've got something to add to it. Yeah thanks very much and I think it's a crucial question at this point in time so as Murray mentioned in his presentation especially bringing out some of the CEQ analysis a lot of the tax revenues are based on on taxing flows and of course over the course of the pandemic this has drastically been affected so one way of searching for for underutilized areas is perhaps switching from from flows to stocks in an attempt to perhaps access balance sheets to find these resources now as part of the SA Tide project Leo, Emery and myself we estimated the distribution of household wealth in in South Africa which is one of these sort of balance sheets a household sector balance sheets and we were able to do this because of because of the administrative data that came into the project that allowed us to to provide a much more accurate estimate even if there's room for for more data and more more accuracy and so this paper I think is available on on the SA Tide website but just to give you a broad sense of what we found we found that the top 10 percent of people in terms of wealth they owned 86 percent of all the wealth in South Africa and and that gets more concentrated as we go up so as an example the top 0.01 percent had a 15 percent share and on this is about three five hundred people three thousand five hundred people and on average this is each person had ownership of around 486 million rand but that's that's average and then on the opposite side the bottom 50 percent on average were in in net debt or had no wealth so to speak of so so this estimates of course this is an estimate it's not audited numbers and this can improve of course with the more and more data that comes out and hopefully more data that comes out from SAAS as well but but at that if we take into account some of the other sources that are available our estimates are roughly in line with with their estimates so as an example Afraja we underestimate that concentration in relation to Afraja so so we think this is a good base on which to estimate that that wealth distribution and in relation to question I think the the useful thing about this estimate is that it provides some info about a potential tax base so perhaps as you know we also on the back of this proposed that in the circumstance attacks on this asset base on this on this what we call a wealth tax it could be a useful policy proposal to help fund some of the urgent and continuing needs during during COVID during the pandemic and the urgent social needs that that Murray and other excellent researchers have have demonstrated now of course this is a this is a hugely polarizing debate and I think there's a lot of strong a priori beliefs but one of the one of the key things that hopefully we've been able to provide is some data to at least move some of the debates along a little bit and so based on this tax base that we estimated you know we think that we could that a potential tax like this could raise somewhere between 70 and 160 billion rand that's 1.5 percent to 3.5 percent of GDP and that's with taking into account between 15 percent to 50 percent of of evasion and avoidance so we have very conservative estimates and and our position is that of course there are challenges to this implementation challenges but but on two points we think that firstly because of the deep financial markets and the proactive systems that have been put into place both by Saab and the Reserve Bank South Africa is actually relatively well placed to to develop the systems with of course with political support and with with investment but also the potential revenue that could be raised in such an urgent situation is something that should lead at least to this policy being put on the table and and discussed but let me not talk any more about the wealth tax I think there's you know we've got a paper that's also out there and and a tool that can that can help you play around with rates and estimate what sort of revenue you can get one of the I think one of the crucial things that we're also trying to say is that if you change the perspective of the tax system to to look at wealth rather than labor market income then this opens up different avenues that could be that could be utilized and explored at the very least and and one of these examples is dividends of tax perhaps other things are looking at exemptions that are in place for pensions looking at estate duty and and broadly thinking about the distribution when focusing at the top end so you know some other excellent research that was done in this inequality stream by Bassier and Willard when looking at the income distribution they know that the growth in the top incomes versus a stagnant middle is due to capital incomes so that's incomes from wealth so taking the perspective of wealth could lead to looking at potential resources for taxation that up until now have been focused on on the income inequality from primarily from from labor market incomes so yeah let me leave it there thank you so much Arup and Leo as well you've been quite integral in in in in this paper that Arup is talking about is there anything more that you'd like to add when we was talked to the question on additional tax revenues that could be sought from the South African taxpayer yes well Arup mentioned all the essential points maybe and and it's true that my presence for this dialogue is essentially justified by the fact that we've studied the distribution of wealth and we've tried to estimate how much could be collected by implementing a progressive wealth tax so yeah I mean the thing I can really defend is that indeed I strongly believe that this is an avenue that should be considered and the figures we provide are here precisely to give an idea a Roth idea that despite all the uncertainties we have about how people will react and even the underlying distributions we have good reason to think that this would generate no I mean some significant tax revenues and so I think this proposal should really be on the table maybe maybe it's complicated for now maybe it cannot really be implemented the way we wish it would be that sort of things but I think this is really an avenue for policymaking there and I just would like to mention the fact that I mean there has been some historical cases in Europe there's still a wealth tax in Colombia now and for instance Argentina I mean of course like these are different settings and maybe sometimes the contexts are not really you cannot really compare them and so it's hard to derive like any lesson from complete different contexts but just to keep in mind that Argentina as a response to the covid crisis has implemented a wealth tax under 0.8% so it's like very similar to what we propose and I mean the tax base is not exactly the same and there was not yet any report giving all the details so that we can really compare what we propose with what they did but just in general they did this and they managed to collect some some sizable amount of revenues even a little bit more than they expected so it's an experience that shows that despite what people sometimes think but without much data about this when you have a tax like that not everybody goes away not everybody flies and the investment do not stop suddenly that sort of thing so I wouldn't like to warn people having these ideas about the reaction to a wealth tax that of course some people may react and of course it provides some disincentive in some regards but sometimes the benefits overtake the the cost and so I think it should be considered precisely not just in light of some general ideas that we have about how people will react I think Mamikie will welcome the extra revenue that would come from the wealth tax maybe I'll bring it back to Mamikie and ask what kind of talks are they having within SARS on this wealth tax I know that national treasury would need to pronounce on the policy around it but maybe at SARS what kind of conversations are being had around the wealth tax? I think should you here I would definitely defer this back to national treasury for them to give a perspective on I think from SARS's perspective any revenue opportunities that present themselves we will pursue them to the full extent to the full extent that the law permits us to do so to to the full extent that the fiscal policy allows us to do so but I think it would be premature of me to sort of make a pronouncement on behalf of SARS on tax policy matters I am aware that we may have one or two colleagues from treasury that are part of this discussion perhaps they may want to step in and just give a sense of how far conversations are going so that I don't make pronouncements that I'm not I don't have the authority to do so. Indeed thank you so much for that Mamikie I know I've just seen my colleague Chris Axelson join the platform maybe he'd like to respond to the question on the wealth tax and how far those conversations are going we just ask Grace maybe to also just give Chris the hosting rights so he's able to respond to the question around wealth tax I'm not sure if Chris he does have co-hosting rights and you'll be able to respond to that Chris can we throw it to you and just to get an idea of how far the wealth tax conversations are at this point within national treasury. Hi Mashidu, hi everyone sorry to end late on another meeting and put me on the spot here but you know internally we are looking at this issue and the OECD has done quite a bit of work there's this nice work that's been done by the Rupin colleagues in terms of trying to get an overall estimate of net wealth and if there've been a few papers which we look at quite closely to see well you know how do these correlate with what we see and what we think about the total stock of wealth that is that is out there but it's very much trying to you know we're investigating this to see if we can come up with an internal position you know the minister hasn't stated anything publicly on this other than the fact that we are looking into these types of issues you know there are a lot of wealth taxes already in South Africa and so we also look at those trying to quantify see how we compare with other countries it's quite a complicated issue and we also need to look at from Saaza's perspective in terms of how possible this is to do you know Saaza had some difficult times in the recent past and you can see it and its impacts on revenue collections and you know Saaza focusing on their primary core mandate with the taxes that are there available at the moment so we do also have to take into account the administrative consequences of following a completely new tax base so we're we're researching we're looking at all these issues I think these types of debates are very useful for us so that we can get a different perspective and then we can try and provide you know an analysis and a view that we can put forward to our political superiors and that's basically how far we are at the moment thanks Mishidi. So if you're putting me on the spot but well you've responded to it quite well thank you so much Chris and to our audience members please please feel free to put your questions in the chat box or raise your hand for our panelists this morning they're all available to respond to your question and another we've got an hand up but thank you so much from Andrew don't just check your name properly I just want to get your name Andrew Donaldson thank you so much Andrew you can post your question to the panelists thanks. Hello yes I want to ask something about sort of the underlying public finance theory here so one of the first things you learn as an undergraduate sort of studying public economics or tax tax economics is that you know that a comprehensive income tax incorporates a wealth tax if you make I mean obviously make very simplified assumptions in that so but if you have a simplified assumption about the the states of the economy and the generalization of returns on assets then then then when you're taxing income fully you're effectively also taxing the the asset base that underlies that that income obviously the world is much more complicated than that and so in practice that's not what we have in practice we have income tax that is that is more narrowly defined and a range as Chris has indicated of supplementary taxes we have a capital gains tax we have a tax on on residential property that goes to municipalities some of our transfer taxes kind of like wealth taxes although they're really imperfect so what most countries have is an income tax that is incomplete and then some supplementary taxes that that are targeted at various forms of wealth what's being proposed is you know rather than continue with that reform on which we've made quite a lot of progress over the last 30 years we introduced capital gains tax we stepped up those rates etc etc that what's being proposed is that we should rather put our eggs in the comprehensive definition of wealth and and tax that at a flat rate it's extraordinary difficult to do and in South Africa it faces the immense problem that a lot of South African wealth is holdings of foreign assets and of course about half of the South African economy if you like the wealth that underpins South African GDP is owned by foreigners so so so you've got a whole lot of very complicated cross-border things that you're trying to do here and it's my question to a group and Leo is you know why would you really want to opt for something so hard rather than to continue with the progressive strengthening of our income tax system and it's and and it's complementary taxes which is if you like the the path that we've been on the underway for the last last few years and and and that has the support of of of of international tax advice thank you Sandra you can maybe yeah maybe i'll take this one thank you very much for your question and so there's a couple of points to this so first i'd like to make a disclaimer here i'm not exactly sure i understand what you mean by a complete income tax would encompass tax on wealth because you can think of well typically wealth tax that would achieve some collection that cannot be done through income tax so maybe it's it's you mean that a complete tax on income allows to tax future capital accumulation but not past capital accumulation so that's that's one point you can still in a way do more with wealth tax than you can do with a fully with a full income tax now i think there's a in a way that's maybe the first element of your question is in terms of responding to a crisis that affects the flows and that's what Arup was mentioning our idea here is that you might have some interest in looking for resources that are in the stock rather than wait for the flows to to to be renewed given that you're facing a crisis that affects them so that's to me that's a good reason to maybe dig this dig that policy issue and in fact historically it's been implemented in many cases as a means to mobilize sort of resources accumulated in the past but and evenly distributed across the population mobilized them through taxation to allow investment and so in to allow investment as a response to a crisis such as for instance the world war two will germany implement a wealth taxes response to this for instance so there's still a i think a key difference between taxing income and taxing wealth when you try to answer such such a crisis and i think it should be kept in mind when advocating for one for the other now our estimates take into account the fact that wealth is being held by the by foreigners so i also want to stress here that we do not include this wealth into our estimates so normally maybe maybe we're underestimating this share but we are using like sub data and so normally we're not sort of putting forward a policy proposal that would imply being able to tax wealth held by foreign citizens and now regarding the fact that it's complicated this i mean i can we completely acknowledge that this is indeed complicated but and and maybe in a way in a beginning one would not be able at all to implement a wealth tax the ideal wealth tax you know the one that would apply to the entire tax base and and take into account depth and things like that so we recognize that it is indeed complicated however i think it would be interesting to still work in that direction if we have a upcoming paper that is unfortunately not yet available for for reading but and we see that the if you look at the distribution of of tax paid by people as a function of their wealth we see that it's largely regressive meaning that people with more wealth tend to pay a smaller share of taxes as a proportion to their wealth than so richer people tend to pay a smaller share of taxes as a proportion to their wealth than poor people and but wealth is an ability to pay tax as well right so in a way there's also a matter of making the system more overall progressive i think there's this case this argument also that need to be thought of we don't have unfortunately figures yet to really show you show you this but it's been pointed out in other contexts such as in the us for instance i mean i'm taking a lot of time to answer this it was a like i could i could add a few few things more but maybe i i can hand in the yeah the mic all right thank you so much leader we've got more questions coming through from our audience we'll start off with marlis marlis pieta i hope i'm pronouncing your name correctly and then we'll also take c vireche so we'll do those two questions and then we'll get the panelists to respond and then we'll do the questions in the chat box afterwards thank you thank you very very much and briefly one comment to what you just said so in the tax data and we have a variable which indicates whether the submission may be of a tax practitioner uh number one and number two um we have an identifier anonymized identifier of this tax practitioner so um i think one way um going forward what stars could potentially do and i'm not sure whether they're already doing something like that but you know once um they realize um that a particular tax practitioner using one way of doing uh decreasing tax liability um that it could potentially then trigger audits for tax submissions by the same tax practitioner and that is maybe one thing that we could potentially look at um so essentially look at a tax practitioner effect um number one do you use make use of a tax practitioner yes no and then also just see whether it's linked and whether the tax practitioner uses the same methodology across his times um that's just something that jumped to mind and then um two questions that i have one for mamiki and one for chris um chris um i know we've spoken a little bit about this um so obviously there was a a new tax bracket that was introduced um a couple of years ago um and i know you have um started looking at the analysis what the effect was that um yeah if you could make shares just some of the results that you have um and maybe based on that um where do you think are we on the la on the laffer curve that was my one question and then the other question is for mamiki um so you know during this conversation we've been speaking about the availability of tax data and it's been an enormous effort um and it's it's been it's been so wonderful and and i have two questions um how do we ensure that we uh can get more accurate data going forward um and obviously only that is what is submitted to saris is the data that saris receives and that we that we get and can use for the analysis so my question is um how do we ensure and implement new systems at saris uh that can verify more information uh that are submitted to saris um so one way for instance is doing um integrated e invoicing tax systems where you know if a company says we paid uh 20 000 rand in rental um that you can actually verify that information automatically because the company uh who received the rental income for instance uh they would also indicate it as uh as income and obviously then just checking how reliable is is the information um so that way we get uh can improve tax compliance number one and number two we get uh better data um those are just some of the the initial thoughts thank you so so much it's been so wonderful thank you thank thanks margius see you beret you can pose your questions so the panelists can respond everyone i think the analysis of wealth tax is here is the normative approach but the empirical analysis too needed for a decentralization of wealth to meet to meet the volatility in economic spheres because the covid pandemic has revealed the weakness existing the straight actions uh because the many of the states in developing countries and and many other third world countries are faced uh the many difficulties to meet the the social aspiration of the underprivileged section i think even though we have the uh the holistic approach to the uh the wealth tax the progressive tax is i had no basically i had no optimistic about the the progress wealth tax because in many countries it is not uh working good for the decentralization or redistribution of wealth to the uh the real uh the poor peoples uh to upgrading their life conditions and other employment situation are not working properly because we need uh there is a alternative approaches or the implication of the wealth tax in adopting or embracing in many of the countries is too needed uh because uh the excessive influence of the multinational companies the state governments are the any form of governments are facing uh the difficulties to exercise their functions in a holistic manner thank you thank you sivir ish i hope the panelists were able to hear um sivir ish because there was a bit of an interruption um in his line thank you so much i think mamiki you've got a question that was posed to you as well as uh chris you can take it away thanks yes unfortunately for sivir ish i i struggled to hear what he was asking so i'll probably first deal with uh marliss's questions uh um regarding the availability of data how to ensure that you've got more accurate information so um maybe if i can just take you through some of the things that uh sarz has put forward in its um annual performance plan uh in terms of the first three objectives um i may have quickly spoken through it but maybe just to give you a context of what it means when we're saying providing clarity and certainty for taxpayers and traders of their tax obligations this means that we make sure that when people engage in the tax products that we administer we give them the necessary education necessary information so that when they start filling in the forms the way they're supposed to feel them they know what they're supposed to submit to us and that information is in fact correct so there is a whole subset of activities that we are undertaking just to make sure that we deal with that particular objective of um providing clarity and certainty to taxpayers and traders then the next one is making it easy for people to comply with the tax obligations obviously this means that whether they are interfacing with us um online they're interfacing with us e-filing um doing and trying to sort of minimize our face-to-face visits but at any point of interaction that we have with the taxpayers that it should be easy for them to interact with us and one of the things as an example that we've introduced to make sure that we do have um a lot more accurate information is that we introduced auto assessments so those of you who've just gone through with us in the past file season know that you would have received a message from SARS saying that with auto assess duties is what we think your tax um affairs are or your tax obligations are indicate that you agree or you don't agree with this and for us to be able to do that we obviously have to get a lot of third party data so there is a lot of um um information that we gather on behalf of taxpayers and we sort of link up a lot of data points so that by the time you interact with us we're able to already tell you what we know about your entire um tax situation or your tax profile so how do we ensure that we get more accredited for introducing such measures would obviously allow us to do so um the third objective that we've got in place is to make sure that we detect those people who are not complying and make it so hard for them and make it so costly for them so that by the time we sort of catch them either early on in the commission of the crime or later on that well or tax non-tax compliance that we already are putting together lots of information that allows us to then say you should have complied and this is what you're being punished for um the fourth fifth objective that I've already touched on is the make increased use of data to improve our integrity and to derive insights but I think more importantly number six it's modernizing our systems to provide digitalized streamlined services and I think for us that's a very important thing to mention that we are in starting to invest quite heavily in making sure that our systems are modernized and that we can sort of have an almost like a live connection to what our taxpayers are doing and be able to collect that information and be able to build a much more comprehensive picture about the taxpayers this allows us then to do many many things such as behavioral economic social data analysis so that we can affect profile things like behavior according to text practitioners behavior according to a group of taxpayers behavior according to localities behaviors according to many many different dimensions so all of that is as we are sort of investing in both our capacity capability human capability and also our infrastructure our data our systems that we will be able to incrementally work towards making sure that we encourage people to almost um voluntarily comply because they know that South has an eye in and tentacles all over the data sets that tell us a story about the text picture I trust that that maybe that helps in sort of giving perspective on that and with regards to integrated systems like e-invoicing yes I think there's a whole lot of solutions that are in place that we would need to obviously make sure that we plug and play into our system what is suitable for South African environment what SARS can handle in terms of its systems but it's definitely integrated systems is definitely what is in our in our performance plans thanks thank you so much Miki we'll get Chris to respond and then we've also got questions in the chat box so just also just mindful of the time we'll go through the question in the chat box a little later on as soon as Chris responds to this question thank you thanks machete so um the first question was just on the new tax bracket 45 percent and what happened there I'm sorry I can't come out and state results just yet we're still working on it and I think it should actually also be approved so it could be a bit of a sensitive one but to say that there definitely was a behavioral response I mean that was quite sure that we did see that and we have mentioned in some of the budget documentation that we published that we didn't get as much revenue as we thought we would from these post-income tax changes so that can sort of show you give you a small sort of insight into it if you look at some of the other research that's been done came to the paper on the taxable income elasticity and on that one he thought that the top rate of just below 45 was sort of as high as you can go um which flows into your next question on the laffa curve I mean I think in public discourse this isn't spoken about too well you can have laffa curves amongst particular tax instruments but not overall so if you look at for example the top tax rate you know if there is a behavioral response and people do try and shelter incomes at the top end you might well and get close to or experience effects of a laffa curve but then if you look at raising revenue from personal income tax say through not adjusting the brackets for inflation and the behavioral response is a lot lower and then you don't have as much of a laffa curve and the same goes for that and the fuel levy and other things so there's certainly it's not a case where if you raise taxes you won't get any additional revenue amongst all the instruments it's you really do have to look at it specifically um just to add on as well there I mean even if you do this analysis on the taxable income elasticity and you see that well perhaps we are at the 45 percent maybe it's at the peak of what it should be in terms of the laffa curve it's so dependent on what your tax base is so you know if you have and we did have quite a lot of deductions available at the time and then individuals use those deductions to reduce their taxable income suddenly you're not getting as much tax then but if you were to restrict those which is sort of part of what Andrew was saying um and then you do another increase later on the elasticity would be different and you might not be at that level of the of the laffa curve so it's very dependent on what we actually do with the tax base and we've been pushing forward um proposals to try and broaden the tax base and reduce a lot of these deductions and incentives um the other thing I just wanted to point out from Andrew's question was and this equivalence between a net wealth tax and comprehensive income tax and you can look at it as being exactly the same thing if you can tax all the gains to wealth it will be equivalent to a particular level of a wealth tax and that's why the OECD come out and they say if you've got a very comprehensive income tax and you tax almost all those gains from wealth then there isn't as much need for a net wealth tax however if you don't have a comprehensive income tax base and you don't tax the comprehensive income wealth then there is more of a justification for a wealth tax and I think it's useful to have those um you know that sort of insight and applicable when we think about this policy um the other thing is also as Mamikie was saying which we mentioned in the budget was Saza really trying to actually just boost the overall information about wealth using information they get from other countries using third party information that we get from banks and others because if you've got a better insight into overall wealth which is what we need anyway um then you can also have a better insight into what those income should be and you can audit the high wealth individuals better and try and get a greater tax on all that income so there is definitely benefits for us trying to at least get a picture of a better picture of wealth in the country thanks. Thank you very much Chris so we'll just respond to Isan Basir's questions he's got a number of questions on the chat box I'm not sure if Leo and Rupa have already read the questions in the um the chat box or whether you'd like me to repeat them to you and we also just need to also just um keep our responses short so we can wrap up the conversation it has been going on for um a little bit over our time I'm not sure Leo and Rupa have you had a look at the questions in the chat box? I have uh but Rupa I think you were raising your hand so just if you want to come up go ahead uh yeah it was just um a sort of general response to the comments before um so so of course there is this debate about the equivalence uh between capital income tax and and wealth and and I I completely agree with Chris that it's a discussion that needs to be had and um and I think that one of the key points I just want to raise is I think that the data systems that are required to improve capital income tax are um you know are almost identical to the ones that are needed to understand the asset base which uh which would also um be needed for for a wealth tax so I think just to reiterate I think that one of the key points that's coming from us is about is that we fully support this um the development and the capacity investment in in data systems for this regard I think you know one of the challenges around capital income tax as as Leo mentioned is that it's not an event um that the administration can rely on um and as with any tax including a wealth tax there are also avoidance and evasion strategies for that so um so as an example declaring dividends may be delayed and you go in that that money goes into retained earnings um and other similar strategies like that um so so I think you know there is a debate that needs to be had in that regard but the reason why we propose this this tax at this point in time as Leo mentioned is because the flows have have dried up and I think a final point just to say taking dividends tax as an example um that's at a rate of 20 percent that's flat that's um only looks at the dividend that's been declared so um there's a lot of room there to develop that because as it is now it contributes to the regressivity at the top end of the distribution um having information about the total wealth of the uh of the person who is declaring that dividend will then allow you to understand at what rates you can uh you can make a dividend tax to make it progressive but it's precisely things like a flat rate on on the dividend tax that makes uh that makes the overall tax system um slightly regress at the top I think there was also um another question um um uh from uh Sea Veresh um or perhaps it was a comment um but yeah I from what I heard I just like to respond to it but I think every country has its has its uh specificities and I think one of the things that we're trying to uh demonstrate is is that South Africa relative to to other I mean Sea Veresh said developing countries but relative to certain other countries is less data rich but compared to many others is is more data rich and I think this can really help um when it comes to understanding the an implementation of uh of the wealth tax but also just a general tax um set of taxes SARS has got uh and it has done an incredible job over the last you know a decade and more in developing those systems so I think the situation in South Africa is slightly different uh in that regard and and how successful redistribution is is also um needs analysis on the expenditure side and and whether the policies uh even if there is expenditure if they're the right policies um so that that question I think is is more complex and needs a much more in-depth analysis which which hopefully is what SA Tai is doing I think Leo wanted to respond to Issan so let me hand over Before Leo um responds to Issan's um questions in the chat box I'd like to say a big thank you to Marie he has to step off the the platform uh thank you so much for your presentation a little earlier on Marie and thank you for your time this morning so we'll just keep uh a group Leo and Rewiki and as well as Chris for another couple of minutes to respond to the questions in the chat box and then we'll definitely close the conversation thank you so much Marie for attending today thank you all very very much cheers cheers you can go ahead Leo okay um so maybe just to save time so I'll try to respond like as quickly as possible on the three points and maybe make a last comment about this comparison between capital uh between wealth tax and uh comprehensive income tax so first thing about monitoring compliance so maybe I can read the questions so that uh so it's about our wealth tax proposal about our estimates uh and the first question is about um what can we say about monitoring compliance um so that's a bit general so just a point of detail we use macro data for the estimated total wealth but do we use micro data to sort of distribute it across the the population so we really combine these these two uh now the macro data comes from SARB uh and uh with uh I mean pros and cons but um now regarding this very question about the compliance uh we are not really in a position to say uh how uh how well the administration of the wealth tax would be undertaken uh and of course it's uh it depends on the context and depends on administration and so I mean we hope like SARS would be efficient at doing doing such a thing but uh of course there are like many elements that we we many informations that we don't have as policy researchers uh just but we sort of take into account uh the possibility for uh part of the compliance not to be uh 100 percent that's of people might evade the tax people might over over um estimate their depth for instance because it's a tax on net wealth and that's an issue that was uh that's uh that's quite important in terms of implement implementing a wealth tax in in Colombia for instance so we sort of discuss these uh I think I'm already too long and that's only the first point so uh to address this in the estimation we factor in uh some uh approximation of by how much the tax base would be undermined by such behaviors uh under reporting and evasion that's all the things so we take that into account and we have different estimates depending on the assumptions we make about that and I really invite you to test all whatever scenarios we have you have in mind about these uh thanks to the simulator which is online can google like South Africa wealth tax you will be directed to the simulator um now question be whether this includes non-liquid assets yes it does like housing and so it is based on market value uh notes that uh it is more fair than property tax because property tax is uh based on market value but not taking debt into account such that with the same value people in debt when people not in debt pay the same property tax it's not the case with the wealth tax proposal and now see how this compares to other tax uh proxying wealth this is very large question we have a section uh fully dedicated to this into the into the paper so I encourage you maybe to to go through this um this part of the paper it would be hard for me to really sum sum that up what we can say in general is that it would sort of perform a little bit better in in aggregate uh because we're targeting targeting people really having wealth um but it's yeah it would require more thorough discussion so I think I cannot address this point here and last but not least I would just would like to mention that you do have indeed this equivalence between perfectly encompassing and well-designed uh income tax and capital tax so you can sort of impose a tax 100 percent rate sort of on a on a capital return by adjusting the rates on a capital and a capital income but you also have to take into account that maybe you want to have very very rich people to have actually less wealth after the tax so which which which what we call a confiscatory rate so you might want to tax them sort of more than what they earn from their wealth at least from a short period of time and this is a like sort of policy relevant to crisis I think so it should be taken into account as as a possibility I feel that the debate here has been sort of assuming systematically that a rate that would imply more than 100 percent of capital return is not a policy option but I think it is especially in times of crisis it's especially sorry especially in terms of crisis but also especially in a in a in a country where you have about 4 000 people that owns according to us estimates more than the bottom 50 percent so more than millions of people I think this sort of rebalancing through maybe sometimes confiscatory rates does make sense thank you very much Leo I see our audience members have follow-up questions but we do we have unfortunately come to the end of our session we have gone 30 minutes beyond time and thank you so much to the panelists for staying on to respond to all the questions from our from our audience members this morning thank you so much to yourself and Leo Arupa Miki as well as Chris who's on the line this morning a very big thank you to our audience and I'd like to make a big thank you to our official partners without whom this would not have been possible that's you and you wider the national treasury the south african revenue service the department of planning monitoring and evaluation the department of trade industry and competition the trade and industry policy strategies the international food policy research institute and most of all the european union for the continued commitment and invaluable financial support for this very important program so to all our audience members you can get all the research that was spoken about today and much more on the essay tied website and do look out for our next essay tied dialogue which is scheduled for the 29th of june and a big thank you to everyone who joined us this morning and do have a wonderful day further thank you very much