 Good morning everybody. Thank you very much for three very engaging papers you know, I think all of these papers really show the strength of of of Where we've come in the SATI project You know, none of these papers would have been possible a few years ago without access to the to the micro data And I think it's a real step change in terms of of tax research in the country So on the one hand, I'm really It's used by the quality of the work on the other hand, I'm incredibly depressed by all of the results All right, so I mean if you think about the South African context where in something of a fiscal crisis we On the on the expenditure side they They're certainly there's a discussion around massive austerity cuts from from government from government We need more revenue right I mean, there's there's no doubt about the fact that we that we need to find revenue and you've you've I think all of the Papers address that point of saying well, what are some of the ways in which we can sort of address tax avoidance Perhaps think about the behavioral effects and in the case of when owners paper actually have a massive massive reform Which would would close or reduce at least reduce a tax expenditure. So I'm but unfortunately, I'm not I'm not seeing huge potential in any of in any of these avenues So perhaps if we start with Nadine's paper, I Think this is in a way the most topical issue and in in tax circles in the country right now. So on the In terms of government departments that are looking for more revenue that constantly saying well Let's push up that top marginal tax rate 22% of of revenues coming from that bracket surely there's some money there and then you have Nadine coming on saying well Actually, you might have a completely perverse outcome And so I think you know, it's such important work I've Struggled to to quite believe that it's as dismal as as your result shows it would be it would be fantastic I think to to sort of throw it at at this problem More and more methods and really try and tighten that up because I do think right now. We're a little bit. We're quite we're quite stuck all right between Certainly social activists saying it the elasticity can't be one The papers saying well, maybe maybe it is and and until we we make some Movement there. So I think it's yeah, I think it's very very good work. I hope the result is wrong Let's put it like that On the second paper on on tax refund. So this is not something I know much about and I'm going to talk very briefly about it the thing that it's perhaps I just have a question here and Well one comment and one question so the first comment I was going to say was You clearly need access to the audit data, but you got that today. So yay I do think that's really going to help tremendously and try to understand what what what is really going on here the thing I didn't quite understand And perhaps I'm just under caffeinated, but what I didn't didn't understand was why Why the length of the delay Would result in you being less likely to apply for a refund in the future. I can understand why Applying for a refund It's going to be a disincentive right so you apply for a refund And then you're in a whole world of pain because you get audited or you get at least validated at the very least But I don't really understand why the longer you wait for the refund why that's why that length of time Is is is is a is a reason why you would then not apply for a refund in the future So perhaps you can address that when we come back and then we're known as paper. So I Completely understand that the motivation for the paper Tax expenditures are obviously something that are on the cards if one wants to think about more revenue And if you want to think about making the system more redistributive and I should put my hand up that first and foremost I'm an inequality researcher So obviously it's it's it's music to my ears to think about how do you make the system more more progressive My anxiety here though is that we already are in a system where the savings rate in South Africa is incredibly low The data you showed us suggests that 52% of taxpayers are not contributing anything to pension funds now Even in a very generous system where they're getting big tax breaks Even the even the rich are not making use of the full tax break so When you're getting a 45% tax break and you're still choosing not to put money into your pension fund then you're in Then something else is going on right? I mean and some of that could be that People are just choosing not to not to use local pension funds because they simply have no faith in the country And they they would rather pay the 45% tax and take the money offshore Okay, maybe It could just be that they feel so squeezed that that even at a 45 with a 45% tax break They don't feel able to contribute the full amount to the pension fund So I you know I worry hugely about the behavioral effects if one was to to implement a system where that that tax break Gets reduced overnight to something I could 26% tax break across the board You know, I can certainly well imagine that that the rich Would simply would then say well at a 26% tax break. I'm simply not willing to invest in local pension funds I will I will rather take the money offshore while invest invest differently and an environment Which is already savings constrained that that that does worry me. So I don't quite know how you address that in the paper I mean behavioral effects are outside of the scope of what you can do in pit mod and and and say mod But perhaps a little bit more Perhaps a bit more Contextualization in the paper around what is savings behavior look like how many people are saving for retirement? how many people cash in their retirement funds at the and When when they change employment, so perhaps just a little bit More on the side of thinking about pension fund reform or pension fund reform so much but pension pension savings behavior and less just about the sort of mechanics of of of tax expenditures So sorry, I had one more comment from Nadine So I think it's sort of it's a it's a general point of which is interesting across the papers so Nadine put up a slide right at the start where she had per capita income of of countries and The top marginal tax rate and then the takeaway there was that South Africa is not a particularly rich country But has this very very high top marginal tax rate Sure, that's true But I think one wants to contextualize that within saying and by saying that this is a highly highly unequal society Right, so when you think about about who are who in this economy are the people paying the 45% tax rate? Those are not people living in a poor country, right? Those are people living in in a Part of the economy which is really much closer to to an upper income country than a Than a you know than an upper middle income country So perhaps just a little bit of of caution there, but yes, so overall a picture in which people seem to be starting to To not and not pay their vat and full which is not what we were seeing ten years ago, maybe Where we don't think we can push up the top marginal tax rate anymore and where I'm somewhat negative about the about the future of tax Expenditures, so yeah, thank you Thank you Ingrid maybe before we go to the Questions I would like to invite the the presenters to just comment and give some their perception of the Thanks, Ingrid for these comments. So I share your depression So of course when we started this project we kind of had hoped ourselves for a different result Yeah, so so we are very aware that the country desperately needs revenue and it would have been great if this Was like also in terms of the data would have been like a like a helpful reform So I fully understand that this is a very unequal society and I see like also a symbolic value in having a Progressive tax scheme and such an unequal society I think nevertheless, it's also valuable to look at the data and check with the data whether the the aims that Were set out with that reform were achieved I think the advantage of our identification strategy is that it's kind of transparent so we can think about I think caveats and we will try to carefully address them in the in the paper, but So Chris access and as part of the author team and he also was of course kind of Alarmed with the results and so it's just so we we think about it. It's not a paper yet It's a project and we're working on that But at least our from everything that we've seen so far we would not think that the country collected more revenue with that reform Hi, yeah, thank you. Thank you for the comments just to quickly reply to the question of like why would length of delay right Effect things so I mean mainly there are two reasons right the the classic one is So again, and this is similar to the way people think about carry forward in corporate income tax systems, right that If you don't adjust for time value of money, that means you're just changing the tax rate effectively and that of course can have Changes on incentives on the margin, but then You know, we think that is probably relatively tiny relative to the effect of Delays on credit constraint firms Right, and if you're credit constrained having the money today versus having it two months from now can Can can make a big difference in terms of solvency in terms of the things you can do Etc. Etc. So Yeah, something we I had didn't show today is You know, we don't see big investment adjustments if you look at sort of six month or one-year chunks But we do see them in especially in the month-to-month reporting. So which is sort of suggestive that what might be happening is People are sort of spreading out the investment rather than and whether that is a real response or it's a reporting response It's it's very hard to say from the data but Yeah, clearly, you know, and I was talking to Julia Mascani yesterday who was like clearly There is plenty of evidence that people are for going the so, you know, more than I don't know And I get you know worrying about revenue is is a real thing But I guess that's not how we were thinking about this, right? That's That's just sort of it's interesting. I think it applies to South Africa and applies to a lot of other places and It's sort of something we know relatively little about Yeah, thank you Ingrid for your comments. I think she makes very valid valid points Perhaps just with my insight into the data And you know also what Nadine Is find and in her paper and what we have seen With the sources of income and and the deductions of the high-income Taxpayers is that they they definitely do do tax planning and tax do play a role in how they declare income and and and and how they then utilize Exemptions and deductions and I think what's very important to realize Also, there's a lot of opportunities for tax arbitrage. If you like in a high-income earner Many of them are also like directors of companies and they've got the company tax rate are currently at 27% so I'm positive that tax planning With taking forms in terms of capital income or other ways because I mean if you like a business person you can declare dividends and There are different ways to actually earn your income and declare your income and dividends has got like a final withholding tax on it so I think just in the context of Behavior by by those that are actually high-income earners They do have more flexibility to actually structure the Salary income and how they're going to be taxed in the personal income tax system on the other hand This is that if you look like Ingrid has shown Highlighted that if you think they can they can actually deduct up to 350,000 and currently the average is about the hundred and seventy five thousand I can keep I think I've got a number on how many are actually at the 350,000 and and it's it's really you can count it on your one hand so it's clearly not You know high-income earners are not going to deduct the full 350,000 So they're not doing tax planning to get the maximum out of the tax incentive That also tells you a story is there are other ways to actually minimise Tax liability, but if you think we are so generous And I mean it is a real concern that only 48% are providing for old age You know when they retire one day and you look at the averages that are claimed something is out of sync and I think yes behavioral changes are very important But I think is there's also need to have a really look if we can't or should not broaden the tax system further and and rather than I've got the notion that we shouldn't have put the 45% that the one and a half million mark But rather that the one million mark So why would you bring in such a high marginal rate and then just cover a small high-income earners percentage? We if you above a million in South Africa, you are actually High-income high-income earner, but that's another point for discussion. Thank you Thank you very much for the interventions now we Can open the questions for the for the members here in the room and Then you can also see whether there are some other questions online Thanks for all the presentations. I have a question on the VAT paper and it's I'm just thinking out loud here so The registration threshold in South Africa is relatively Low it's even lower than the UK in Italy and if I were to make a guess I think maybe 200 or so of the companies will pay 60 70 if not 80 percent of the entire VAT and These will be managed by the large taxpayer unit and On top of that if you are large exporter most likely you will be in a free economic zones where maybe you will not be into the VAT at all Or administer it differently so the question is then you know this skewed distribution how would that affect your study and then going to the point of the Discussant which I agree on I like your motivations on distinguishing between credit constraint firms or not And maybe you could do that explicitly the analysis would be actually very added. Thank you Thank you. Let's take an adorn. I think John Stones It was the second. Yeah, then you come to No, John Stones was it? Okay. Thank you so much for the great Presentations I have some questions and a few comments. The first one is on the marginal tax rate when you increase It's really amazing the findings. So it's like the people on the top percentile decide to plan further and they feel probably you just end up impacting their behavior and So probably instead of seeing more revenue as you increase their rate They just decide. Oh, this is unfair. They find ways of planning better. That's very insightful. Thank you so much Then my second comment is about the refunds in our country. The refunds is about a budgetary issue We always have Budget ceilings of how much we can refund So whether the refund delays or doesn't delay is about really availability of budget And it's interesting to imagine that when you delay it, then The people are likely not to to claim more refunds but also we have sometimes a challenge of invoice tradeings many of the refunds are not actual Refunds so sometimes that delay you need to over scrutinize it and make sure that these are valid refunds So that you don't lose revenue. I wonder in your in your study. Have you looked at other? Aspects that may influence like invoice trading. I don't know whether it's a case that you considered Then the issue about The pension funds for us. It's by the way compounds I don't have a say whether I should contribute to NSSF for pension It's automatic when I'm employed in a given government I have to pay NSSF and it has to be it's not it's mandatory like But the way we calculate it is I you first get my gross pay and then remove the taxes And then that's when you remove the amounts for NSSF and pension. So in our case, it's as if it's not exactly a tax expenditure because the tax is already removed Is that is it a different kind of scenario in South Africa? Thank you so much Yeah Yeah, just next to him Thank you. I think it's good papers My question especially because we're grappling with the issues of revenue as the discussant says I think on the refunds pre 2018 post 2018 And first-time file is I think it could be easily explained by the fact that in 2018 post you do away with field audits for first-time file is where in the first instance You wanted to verify their physical presence of that particular taxpayer and 2018 post you use certain technologies to be able to say I can I can For sure tell who you are and where you are Then that sort of leads to two delays But what I wanted to just make sure of I understand at what point do you call it a delay? Because there's legislative provisions Around what time do you have to actual process it? Thank you Yeah, let's have those three then after we'll take in the next one. Can you start Jack? Yeah, thank you. Thank you very much for all of these questions So let me start from the first one and let's talk please more because you know we It's a little troubling We you know we we started talking way Nona back in back in April when we presented this in in Johannesburg in Johannesburg and We we unfortunately still have trouble understanding exactly what it is that we're counting in terms of waiting time and the latest information we have seems like These are really waiting times that have to do with interest payments Etc. And and we we really really really need to understand precisely exactly what it is that these dates We're looking at exactly represent and what it is. We're measuring The audit data. I mean right we got access to it today So we we literally have not looked at it at all But we are very much looking forward to and of course that would include on my understanding is payment data We'll see That would also of course be be very interesting to look at If you have further insights into like I said, let's absolutely talk because if you have further insights as to why exactly The changing waiting times happened That would be that would be extremely interesting to us and then you know we're thinking of doing this sort of like a donut right leaving a hodl in the middle and looking sort of further left and further right from October 2018 precisely because We imagine there was some backlog in the first month or two and and we'd rather we'd rather do away with that Even though, you know, it means giving up sample size Um Regarding budgetary issues and invoice trading, you know as usual when it comes to evasion if we knew it was happening it wouldn't be happening and So it makes it a little harder than just saying well, is there invoice trading, you know, they're like I said There have been Apparently there have been sort of there's anecdotal evidence that it does happen Although it's it's very hard to say how widespread it is and sort of that is sort of Why we think it's it's perfectly legitimate that this is sort of a trade-off that VAT that the tax administrations have to make right You can't just be handing out the refunds willy-nilly But on the other hand What we're looking at is if you wish is the other side of the coin of that is like, okay but are you gonna just hold on to the refunds forever and You also don't want to do that, right? So so that's sort of where we're coming at this from regarding this cute distribution of VAT payers in in South Africa that is a very interesting comment We had not really Engaged with that other than thinking about sort of the different sides of exporters and everybody else, right? That is sort of something that we we've We've engaged with But you know, I I mean of course in terms of tax payments The the biggest taxpayers are the ones that are most relevant for obvious reasons Given the effects that we're seeing on investment or that were you know, there's the motivational evidence We're seeing on regarding effects on investment One fear and you know can credit constraint on Regarding credit constraint we can link it to corporate tax data where we can see Interest payments and so we can see whether you have loans or not. So that is definitely something we're planning to do but You know, it's nonetheless not necessarily Obvious that you don't care about the investment of of the tiny firms, right and in the sense that In terms of what this could mean for firm growth in terms of what this could mean for sort of the dynamics of firms going forward limited time we need to have Just one minute Okay, I hope I understood the question, right? Perhaps just saying is is that if you look at how pension deductions are claimed Usually if you like low or low middle income your ability to actually provide for your time and deduction and make use of the tax tax incentive are more limited and What happens also in practice is is that? In if you are with an employer that there's got compulsory contributions to a pension pension contributions deductions It's usually limited to a certain percentage of your of your gross income So they are like Limitations on how much you can contribute but we also got retirement and new it is so if you want to supplement your Contribution you can do it with like retirement and new it is and many people have done that in the past That they will before the final tax year ends will actually put more money to make use of the of the of the tax break Given what we see now in the high income earners and and and the averages that are claimed there It seems that it that's not so much driven to go to the maximum the 350,000 to make use of say a 41 or 45 percent marginal tax rate So I think this is that it's not only the tax system that matters when when you save or when you Plan your your your taxable income and your tax liability Especially for those that are high income earners. It's like Ingrid is alluded to you know, they might decide to rather take their money off. Sure. Thank you Yeah, thank you very much and we appreciate all the presentations and we hope that They also the audience has learned a lot about what we are doing. Thank you so much