Tax and top incomes

Well, good afternoon everybody. Welcome to this event and thank you all for joining us. I'm Paul Johnson, I'm Director of the IFS, and this is actually really very well timed in many ways, this event on tax and top incomes. We've got a fantastic lineup today as we launch this piece of work as part of the IFS Keaton Review of Inequality, funded, we are very grateful to say, by the Nuffield Foundation. It's one of, as hopefully many of you will know, around 20 big elements of work which are going into this big review of inequalities, which will between them help inform us as we as a commission come to our conclusions over the next 18 months or so, but we're still in that phase where we're gathering this phenomenal set of evidence. Today we're going to be started by Helen Miller, who is Deputy Director here at the IFS and runs our work on taxation, and she will be talking about the core chapter, core piece of evidence that she with co-authors has written for the review. She'll be followed by Owen Zidar, who many of you will know as a huge expert on the taxation of top income and of capital gains. Owen is Professor of Economics and Public Policy at Princeton and is indeed joining us from Princeton, so thank you, Owen, for that, where it is early-ish in the morning. Owen will be speaking for 10 minutes. or so. He'll be followed by Edward Troup, who is formerly the head of HM Revenue and Customs here in the UK. He's formerly a tax lawyer and has actually had a long association with the IFS. Edward is currently on holiday in Calgary and it's about 6:30 in the morning there, so thank you particularly to him for joining us. Then by the magic of Zoom. Richard is Richard Blundell, who will be summing up and is a member of the Deaton panel. He's speaking to us from Toulouse in the south of France, where it's 2:30 in the afternoon. So we have an international panel, at least in terms of where they are currently situated. If you have questions, then please do use the Slido app, for which you should have a link. We also have plenty of time for questions at the end, and we are due to finish at 3 o'clock, and we will bring proceedings to a close at that moment. I will stop at this point and hand over to Helen, who will hand over directly to Owen, then Edward, then Richard, and then we will go into questions. If you do see questions on Slido that you like, then do vote them up because I will see the questions near the top of the screen before I see the other ones. I will shush and let Helen—

Great, thank you, Paul, and hello to everyone watching. I'm going to kick straight off today with a chart that many of you will have seen a version of before. This shows you the share of total income before any taxes are taken into account that flows to either the top 1% of adults ranked on their income or the top 0.1% of adults. And what you can clearly see over the last century is this U-shaped pattern, and in particular from the early 1980s, the share of income flowing to top of distribution was increasing. I think that trend was really— has been behind many of the concerns about top income inequality in general, as well as top income inequality. But if you zoom into this most recent period, so now I'm showing you exactly the same data just zoomed in to the early 2000s, You still see the increase in the run-up to the financial crisis, but top income shares have been flat since around 2013. That's a pretty big change in a trend that was happening for 3 decades. I think it's fair to say there's still more for us to know both about what was driving the trend before the financial crisis, why it's flat now, and what's going to happen in future. But as things stand in the most recent year of data we can look at, The top 1% of people in the UK are a group of around half a million people, all of whom have income over £130,000. Worth saying that some people move in and out of the top 1%, so they're not the same people every year, but that's how much income they have when they're in the top 1%. And collectively, that group get 15% of pre-tax income. That's a pretty skewed distribution. That's more income flowing to the top 1% than flows to the bottom 55% combined. If you look at the top 0.1%, that's a group of 50,000 people, all of whom have income of over half a million. So worth highlighting there that if you're in the top 1%, you'll probably feel closer to the middle classes than to the top 0.1%, because that group really do have incomes that are significantly higher than even the, even the rich. And that group, that top 0.1%, gets 6% of pre-tax income, so 60 times greater than their share in the population. So mostly what I want to do with my time today is tell you a bit more about where that income comes from, how it's taxed, and start a discussion about policy. Before I do, a quick word on just what I mean here by income and what we're measuring. So most of what I talk about, we'll be talking about fiscal income. Which we're measuring from HMRC tax records data. And you can think of this as being a very broad measure that captures almost all forms of income, so income from employment, from self-employment, partnerships, dividends, properties, investments, and pensions. And it's, it's kind of the best data you can use, and it's better than survey data in that it captures people at the top much more consistently and accurately. But it is worth flagging that it's not perfect and doesn't include everything. So there's a lot about this in the chapter, but to give you a high-level sense, the main measure we're using doesn't include capital gains. I'll say a little bit about that later, but one thing worth knowing is that taxable capital gains are very heavily skewed to the top, but untaxed gains, which are much bigger overall and which notably include main homes, are much more evenly spread. We're not capturing here inheritances or foreign incomes, which are going to be important, including at the top. And of course, one concern that always raises when you use tax data is tax evasion. And in particular, for the top, people worry about the kind of offshore assets tucked in tax havens kind of evasion. So there's really not a lot of data out there on this, but if you use data that is available from things like Panama Paper leaks or Swiss bank accounts, then a very back-of-the-envelope calculation would suggest that the top 1% income share, which I just told you was 15%, could be 1 percentage point higher if you accounted for the income flowing to offshore assets. Now there's a huge amount of uncertainty around that, so I wouldn't emphasize that number in particular. I'm putting it on there just to flag that it's probably important, but it's not going to radically change the pictures I'm showing you. I'll also mention in passing, given that it's in, you know, in the news today, Although we're not capturing this foreign income here, and that's going to be important, a lot of the foreign income will be flowing to non-doms. So it would be important to think about the income in terms of getting a picture of income inequality for people who live in the UK, but at least under current tax rules, a lot of the foreign income won't be taxed unless it's brought back into the UK. So we'll be missing less on the tax side than we are just on the pure inequality side. That's just a word on data. I'd say most of— you should think of this really as we're capturing most income that people have, and I'm going to tell you about that income. So where does the income come from? This chart shows you for the top of the income distribution where income's flowing from, and the red line is showing you employment income. So the takeaway here is that for everybody, employment income— for each of these groups, employment income is the biggest source of income. Even for the top 0.1%, it's over half of, just over half of income. What's happening at the top that's different is that there is more what we're calling active business income, so that is income that's coming either from partnerships or from what we're calling closely held companies, so companies with fewer than 5 shareholders. So the top is different because there are more business owners at the top basically. There's a bit more dividend income and not much change in other investment income. So the top really is a story about wage earners are still important and business owners are more important. You can see that if I show you the same data but now just grouped by types of people, about half of the top 0.1% are wage earners predominantly, and about 30% get most of their income from running or working for their own businesses. We can look at the kind of industries these people work in. If you just take these categories look at wage earners, they are disproportionately working in finance and insurance, and that's particularly true for the top 0.1%. So for the top 0.1%, half of them are wage earners, and half of that group work in finance and insurance, almost exclusively in London. So there's a very concentrated group of people there. The self-employed who are at the top are largely people working in large professional services partnerships, so think big legal firms, big accountancy firms. Owner-managers of companies are more equally spread across different industries and actually across the country, so they're less concentrated in London than the other groups. So that very broadly is who's at the top. I'm now going to tell you 3 facts about tax at the top. The first is that income taxes, so including income tax and National Insurance contributions, are progressive. I've put some numbers on the slide here. Well, that basically means that average tax rates increase as income increases, so people on average, people who have more income pay a higher share of that income in tax. The second fact is that top taxes have increased since 2010. I'm not going to go through the whole list, but just to flag 2 big things that have changed. There's been a new additional rate of income tax for incomes above £150,000, so the top rate In the UK is now 45%, and dividend taxes rates have been increased too. So before 2010, the top rate on dividends was 25%. Now it's 38.1%. And as a result of those tax changes, top income inequality has come down. So this red line just shows you exactly the same data I showed you before. I've indexed it to 2009, but it's showing you the trend in the top 1% pre-tax share of income. If I now show you what happens after tax, you can see these 2 lines are moving in the same up until the crisis, and now they've departed. So basically tax has been doing more since 2010 to reduce post-tax income shares. So in 2018, the top 1% got 15% of income before tax, and after tax that was reduced to 11% of income. So taxes are doing more to reduce income inequality. at the top. But despite this overall change in taking more from the top, what really stands out still when you look at the top is that how much tax an individual pays isn't only determined by how much income they have, but by how that income is earned. There are big differences in taxes across income sources. So this chart I'm going to just use to walk through some of the examples of that. So this chart at the moment shows you for different types of people, the mean tax rate of that group in the top 1%. And one thing that immediately stands out is that wage earners, for example, have an average tax rate of 42%, whereas for a rentier, which we're defining as somebody who gets the vast majority of their income from passive investments, faces a tax rate of only 30%. Actually, if you include employer National Insurance contributions, you can see that a much bigger tax penalty on employment arises. So a wage earner is taxed at an average rate of 49% compared to, say, 42% for a self-employed person. Now, employer NICs can be a little bit tricky, and we could debate who exactly is, is paying that ultimately, but whether it's the employer who fundamentally bears the burden or the employee, it's clearly a tax that is putting a penalty on employment and encouraging people to use a self-employed or a company legal form rather than to employ people. And of course, to the extent that it is felt by wage earners through lower wages, then it's going to directly matter for income inequality of individuals. But actually, even this chart isn't showing you the full extent of all the tax differences. Another big one relates to capital gains. So we know from lots of other work that owner-managers retain a lot of income in their companies, and they later pay that income out in the form of capital gains. When they do, they can access a preferential rate of capital gains tax, which even after including corporation tax comes to just 27%. So for a lot of the company managers in this graph, the yellow and blue bars are showing what happens when they pay themselves dividends or wages in a given year. But a lot of them can actually access a much lower rate if they keep money in their company and later take capital gains. And in fact, if they delay the capital gains until death, they'll be taxed at zero. There are big differences here in tax rates within the top. So let me just finish my section with a few words on policy, and I'm sure we'll come back to this in the debate. Obviously, a question that crops up in this area constantly is how high should the top rate of tax be, and in particular, should it be higher? The punchline here is that I don't think there's one clean right answer to that question. It depends on a whole variety of factors including different preferences, so people will differ in their views on how high redistribution should be, for example. It also depends on what's driving top shares, so there's lots we could talk about here, but to give you just a sense of what I mean, imagine 2 extremes. If we think that the top actually is comprised of people who are putting in lots of effort, they're highly productive, maybe they're innovative, they're entrepreneurial, creating lots of value will think differently about the top rate than if actually the top is characterized by a lot of market power, excess profits, where just some individuals are lucky enough to be able to capture that. So it matters what's going on at the top. It also matters how people respond to taxes, and of course there's a whole variety of ways in which people respond, including real responses like how much to work and where to locate themselves, but also whole bunch of avoidance responses like whether to take income in dividends or capital gains and how to time income. And all of those things go together to affect what we think the kind of the best top tax rate should be. But I think it's interesting to note that if you take official HMRC forecasts for the UK, then it's predicted that increasing the additional rate of income tax, the top rate in the UK, would actually raise very little, if any, additional revenue. Now there's a lot of uncertainty around that statistic, but it's arising because high-income people are estimated to be very responsive to tax, so you tax them higher and they change their behaviour quite a lot. But for at least business owners, we have pretty good evidence that quite a lot of that is avoidance, so people can switch into capital gains rather than dividends, for example. So that's not a, it's not a fixed parameter of the system, how responsive people are to tax rates also depends on how you design the tax base and the rest of the tax system, so we need to think about taxes broadly here. So I haven't got a clean answer for you on exactly what I think the top tax rate should be, I think that's debatable, but I do think there's a very clear direction of reform that we should be pursuing, and in particular I think you can quite easily make the case that tax rates on business and investment income should be more similar, if not aligned to, tax rates on employment income. There are a whole range of reasons for that, but to give you a couple of, a couple of big ones, I think it's horizontally more equitable. So imagine you have an accountant, and one accountant is working in a big partnership and the other is working in-house at a company as an employee. They could be doing very, very similar work but face very, very different tax bills. That's inequitable. If we had more similar rates, we would quite quickly reduce a lot of the tax-motivated income shifting that we see. So people would be less likely to switch into capital gains if there was no tax advantage to doing so, for example, and therefore all the inefficiencies that come with that switching could, could be largely prevented. And I'll note in passing that although these lower rates on business income are often defended as being ways to target more entrepreneurship, they're not very well targeted. There's very little evidence that they actually do much good. Give you one example of that. We did some work looking at owner-managers of companies And although you see they hold lots of money in their companies, it does nothing for investment. That money isn't invested in productive capital, it's just held there and taken out as capital gains at a later date. But there is a but here. If you simply put up rates on dividends and capital gains, you would discourage some savings and investment, and that's because underlying those gains or those dividends, often there has been some, you know, financial investment in a new asset, for example, And if you put up higher taxes in some of those cases, you're going to discourage some investment. But without getting into the details right now, we can come back to this. The way to get around that trade-off between do you want high rates or low rates is to reform the tax base. And in particular, I'm thinking of things like capital allowances, the treatment of losses, the treatment of financing expenses. You can change all of those things to ensure that you minimize the extent to which taxes discourage investment and risk-taking. Once you've done that, if you fix the tax base properly, then with the reform base, I think there's a very strong case to align overall rates, including capital gains and including removing the uplift of capital gains at death. And I'll note in passing that because I think aligning these rates would stop some of the income shifting I've talked about, it would allow you to raise more from the top as well if you wanted to. It would make higher rates at the top more possible. Let me quickly summarize that whistle-stop tour I've just given you of the chapter. Those at the top get a lot more of their income from working either as employees in the financial services sector or for their own businesses. The share of pre-tax income flowing to the top was rising pretty consistently since the 1980s, but it's been flat in more recent years. We've seen a pretty big trend in the underlying— sorry, a big change in the underlying inequality trend. Income taxes are progressive, at least on average, and higher taxes since 2010 have reduced post-tax income inequality, but it's still really noticeable that governments are choosing to maintain these big differences in tax rates across different forms of income, and in particular maintaining low rates on capital gains, which is creating various inefficiencies and inequities.

With that, let me stop and hand over to Owen to tell you about All right, thank you very much, Helen, and thanks for the invitation to talk about top incomes and capital tax policy. I'm going to largely take a US view but make some comparisons to some of the things that Helen just mentioned. So I have 3 main points. The first is that when you look at the data, there's really a striking world of business owners who prevail at the top of the income distribution, especially if you go even further in the top 1%. top than the top 0.1%. The second point is that the groups that are able to achieve kind of this status are very selected. They typically tend to be white males who had quite wealthy parents. And then in terms of tax reforms that are feasible and of consideration, there is actually a lot more revenue, I think, than typical forecasters believe for the types of reforms that Helen just talked about, like raising capital gains. And so I'll go through a few reasons why I think that is. All right, so to start, I think that one key message that you should take away for thinking about top incomes is that often the discussion in the media focuses on some high-profile anecdotes. But I think it's important to see the forest through the trees. There are just many, many more so-called Main Street millionaires than those who show up on the front page of newspapers. So this graph shows some data from some work with Matt Smith, Danny Yagan, and Eric Zwick in our Capitalists in the United States paper. And what we do is we say, let's go to a dataset called Compustat— or sorry, to ExecuComp. which has the C-suite wages of all executives and CEOs, and just add it all up and ask, how much money is that collectively? And how many people does that represent? And it's on the order of $33 billion in 2016. And what you see when you do that comparison is that's just an order of magnitude smaller than the pass-through income, which is the business income of closely held firms in the United States, both in terms of overall tax base as well as the number of people. And that's also true for the top 0.1%, which is about half the size of this, this bar. So the first point is that there are just many, many more Main Street millionaires. And if you're thinking about the tax base, that's really quite key. You can see that if you look At kind of where income comes from in the United States in the top 10%. So this graph on the left is showing the share of fiscal income for, say, P90 to 91. So that's the 90th percentile. But 90% of them are earning their income from wages. But if you go to the top of the distribution, you can see similar patterns that Helen just pointed out. But in the US, there's a sharper rise in business income where it's about half of of income in the top 0.1%. And if you kind of zoom in on business income and decompose it into 2 components, one is from these pass-through firms that are not traditional corporations that you can associate with public companies, but instead, for tax purposes, have the profits and losses of their business flow through directly to the owners. You can see that they make up about 40% of the top 0.1%'s income, whereas dividend income is much smaller. Now, partly that's because some, some income is retained in businesses, but I think it's useful to see how important this group is, especially as you go to the very top. And as Helen showed, you know, you see similar patterns in the data in the UK. Now, what are these businesses? This table goes through and aggregates the total amount of business income for 4-digit industries of companies that have an owner in the top 0.1% and in the top 0.1% on the right. And you can see, these are things like doctors, professional and technical services, dentists, lawyers, specialty trade contractors, insurance, computer systems design. You know, generally, you know, pretty skilled service-oriented businesses that rely a lot on human capital and are not quite as physically capital intensive. And then at the top, you see some similar groups, but you also see like financial investment activity, oil and gas, auto dealers, physicians, doctors, etc. So this is a pretty wide range of companies. And if you zoom out a little bit, The main thing that I want you to think about is that there's a lot of mid-market companies in the $50 to $500 million with pretty high-income and wealthy business owners. And again, it's in the US, it's not just finance and real estate, that's about 20% of these entities. They're from a broad class of industries. And they're kind of a reflection of the rise of services and the fall of manufacturing. a lot of activity is in this form. And so when you're thinking about the top tax base, at least in the United States, it's quite helpful to think about these millionaires who are not just in New York and California. They're in a wide range of places owning kind of auto dealerships, beverage distributors, you know, just a huge range of Main Street companies. So that was the first point. The second point is that there are quite striking differences in terms of where the owners of these companies come from. And so in some ongoing work with Raj Chetty and John Van Reenen and Eric Swick, we're trying to track where high-growth business owners come from. And one graph that's in my discussion is plotted here, which shows as a function of how high income your parents were, the likelihood of you founding a company just overall, or a star company, meaning a company that's in the top 10% of employment and revenue growth, or a superstar company that's in the top 1% of employment revenue growth 5 years after founding. And what you can see is that for the vast majority of People whose parents were kind of in the median incomes, you know, they're just much, much less likely to start a very fast-growing company. And so we think that's an important thing to note, that when we're thinking about business owners, they're much more likely to come from certain backgrounds, they're much more likely to be male. We also have some, some results on race, and, and that's something that we really need to work on in terms of broadening opportunity. both in terms of having a chance to get to the top, but also because of the social spillovers of who you might hire and give opportunity through the labor market. The last point I want to make is on what are policymakers considering in terms of what's feasible and what we can do. And so Helen mentioned one of the recommendations of raising rates on capital gains. And that's also frequently debated and currently under consideration in the United States. And one thing that happens in this discussion is that when you try to do that, often the official scorekeepers say it's just not going to raise that much revenue. But I think that is a bit misguided. And I wrote a paper with Larry Summers, Natasha Sarin, and Eric Zwick enumerating some of these reasons. The summary of it was that many of the prior studies that try to estimate how responsive people are to capital gains changes really look at a very narrow window and so pick up a lot of avoidance. But a lot of that money shows up in the medium term. And, and so trying to have a longer horizon to estimate the revenue effects is quite important because you might overstate how responsive and how How little revenue you'll get. There's big changes in the composition of capital gains, and similar to what Helen was saying, you know, bringing rates closer together provides some backstop to the rest of the tax system. So if there isn't a lower rate option, this is— this can be attractive. And base broadening reforms like having step-up basis at death that Helen mentioned, you know, will likely decrease the elasticity of the tax base. And so just to, in the last 2 minutes, give you a sense of how big these magnitudes are in the current thinking. What this graph shows is a comparison of 2 things. The first is just the mechanical effects of raising capital gains by 5 percentage points. And in the United States, there are projections from the Congressional Budget Office that capital gains realizations and qualified dividends over the next 10 years will amount to about $15 trillion. And so if you just say mechanically, if there's no behavioral response to raising capital gains by 5 percentage points, how much money would that raise? You'd get about $750 billion. In contrast, the scorekeepers from the Joint Committee on Taxation, which is like the HRMC in the US, they say it'll only raise a little over $100 billion, which is really quite striking in terms of how much behavioral response they're kind of implying is implied by shrinking $750 down by almost a factor of 5 to 7. And so based on some recent estimates that I have from work with a graduate student at Princeton, which just came out in AER Insights, we think that you could actually raise hundreds of billions more if you account for estimates that incorporate this medium-term retiming That even though money kind of is avoided in the year or 2 window, if you take a 10-year view, which is how budget forecasts in the US work, we think that the current scorekeepers are much too pessimistic in terms of the revenue potential of capital gains. Thanks. I will now turn it over to Edward.

Thanks, Owen, and thanks, Helen, for a great paper and a great presentation. I think Well, I'll say something from the policy and tax administration background because that's what my career has been in. The conclusions of Helen's paper were very much that there is progressivity, but there's— and that there's therefore a reasonable amount of vertical equity, but that the horizontal equity between different taxpayer groups and different types of income is really quite limited in certain respects. What I'll try and do is to say something about not what that progressivity should be or what the equity should be, but what the constraints are for policymakers and tax administrators in trying to achieve particular outcomes. I think that, you know, first of all, looking at the policy side of things, looking at the wealthy, saying something, you know, in a sense obvious about the wealthy is the wealthy have choices. The wealthy are different from us. They have more money. Not all of the money, or a much lower proportion of their cash and money, has to be spent on immediate needs, and therefore both what they can do and the extent to when they do it is much greater than for the lower range of taxpayers. There are therefore greater activity— greater range of activities, and with them come a greater range of choice and hence for arbitrage on the tax system. So there are quite a lot of constraints on policymakers, both in chasing down what those options may be, chasing down what taxpayers are actually doing, and the ability for taxpayers to manage their outcomes, which can vary over location and can vary over time, for instance, through the use of trusts, for use of roll-up investments, as Owen and Helen have said, using capital gains, which makes policymaking that much more difficult. So just looking at the limitations on policymaking, first of all, that the space, because of the nature of the taxpayer base, the wealthy taxpayer base, there's a much greater space for policy gap between lawmakers', policymakers' intention and what actually can be achieved than there is, for instance, with the baseline employed individuals. This isn't helped by the fact that policymakers don't always want to achieve what good tax policy, what good outcomes might dictate. So Helen mentioned the need to promote investment or at least not discriminate against investment. One of the problems with taxing the wealthy in the UK is that a lot of the— what is regarded as avoidance by taxpayers is created through the— is generated through the existence of reliefs: the entrepreneurs' relief, the relief for inheritance tax, relief from capital gains taxes, and investment reliefs. So you have an inherent tension between achieving any sort of horizontal or vertical equity and the desire of policymakers to promote various forms of activities. So the complexity of the tax— the rich people's taxpayers' affairs, the opportunities to do different things, to create new instruments, to create new forms of holding wealth, all create challenges for tax policy. In addition to this, as I think has been implied by what Helen and Owen have said, geography and time both play significantly greater factors in thinking about policy for the wealthy than they do for other people. The wealthy do have the ability to move geographically. They have the ability to phase the timing, definitely of capital gains receipts, but also of income receipts to a certain extent. And just responding a little bit to Owen's comments about capital gains, it's interesting to see that on a 10-year horizon, raising capital gains tax in the US could well raise more money. One of the problems with any tax policy, particularly with capital gains tax policy, is the expectation of taxpayers, of individual taxpayers, to future policy change. And when, as with capital gains tax, you have the ability to defer realization, if you are, if not confident, at least believe that policy may change in the future, you will defer realizing capital gains. And while if there were indeed a steady state of tax policy over 10 years, you might achieve higher levels of tax. The expectation, the uncertainty about future political action creates a huge difficulty in achieving levels of tax at any particular point in time, and that was one of the issues behind the evaluation of the changes in the income tax, higher rate of income tax, at the end of the last government 2009-10 was that not that raising the income tax rate would not raise more money from those people who stayed there, but because the higher earners do proportionally pay a higher rate of tax, you don't need to lose many of the higher-rate taxpayers to lose a significant proportion of the tax base. And so the geographical and mobility aspects can disproportionately affect the yield from tax changes. At the same time, the territorial limitations on the tax base mean that the choices over what taxes you can change and the impact of that and the ability to collect taxes from UK residents or domestic residents, if you're looking at this for other jurisdictions, is practically limited. And Helen briefly referred to the non-dom rules. Now, the non-dom rules may be controversial, but the reality is that every tax system has to have territorial limitations on the application of its income tax for new arrivals and indeed for permanent residents, just through practicalities. So the policy constraints imposed by geography, the national territorial nature of tax, mean that there are practical policy limitations on what you can do to raise tax from the wealthy. The other point about the wealthy, and Helen's analysis shows the extent to which there is a differential between employment and business income, at least in the UK, is that where you do look at some of the more passive forms of income, while there may be good equity cases for increasing the tax charge, there are both societal and practical reasons why 2 of the largest generators of wealth and income for the better off, homes and pensions, are extremely difficult to tax. Partly that's about the difficulty of taxing capital gains, difficulty of taxing capital gains on an accruals basis, but also on the intrinsically long-term and difficult nature of pension taxation. So although we're not talking about taxation of wealth or we're not talking about wealth taxes, we are for a lot of the top 1% talking about the returns from their wealth, and for a lot of the, as it were, the lower part of the top 1%, a lot of that wealth is represented by homes and pensions. So again, there is a significant practical/policy constraint on the extent to which more income can be taken from rents and from investment income, because not all pensions are in in privileged schemes. So the combination of the variety of the wealthy's activities, their choices over what they can do with their wealth, the temporal element of both expectation and the ability to manipulate income gains over time, territorial limits on the reach of a domestic tax system, and the practicality of taxing some of these larger areas of income, wealthy income sources, homes and pensions, impose much greater policy constraints on taxing the wealthy than some of the commentators and indeed some of the questions which I've seen on Slido might imply. So far as administrative constraints, I mean, the HMRC tax administration perspective is concerned, HMRC always inevitably get a bad press because all tax administrations do, If you look at the tax gaps publication, you'll see that the tax gap from the wealthy is relatively small. In fact, it's much smaller, for instance, than the general population of small businesses and self-assessment from business, where the tax gap is significantly— is very significant. Now, you know, it can be said, well, HMRC are just not finding this income. To a large extent, that is just not true. HMRC and the increasing growth of the international agreements on exchange of information means that HMRC does have a handle on, to a very high extent, a very large degree, on the income which the wealthy do have, which is within the scope of UK tax. And I think the qualifier here is that the policy practical constraints which I refer to mean that HMRC neither can nor are required to chase down some sources of offshore wealth, of which there is a legitimate debate about how and whether they should be taxed. So by and large, although HMRC are always chasing the wealthy in the sense that the wealthy do have more resources, they do do more, you know, they involve more, they involve more, indulge in more complex schemes, by and large HMRC are on top of it. If you did want to get more tax from the wealthy within the existing rules, it is simply a matter of more skills, more intensity, more resource into HMRC, and that ultimately is just a political choice as to whether you do want to put more resources into chasing the wealthy as opposed to more resources into changing the tax laws. So, you know, broad conclusions for this. Is that the system is not perfect, as Helen's identified. Tax administration, by and large, is doing a reasonable job within the rules. There are a number of areas where the wealthy are always, as it were, ahead of the tax administration, but the gap, as it were, between the two is not widening. And to the extent you want to achieve both some of the outcomes which Helen's paper implies and the number of the questions on Slido ask for, there are very significant constraints on the ability to extend policy further and raise more money. The vertical and horizontal equity are limited both by the practicalities of collecting tax and imposing policy, and not to forget that they are constrained by the horizontal equity is constrained by politicians' endless desire to promote business, promote investment, and those forms of reliefs are inevitably and understandably always exploited by the better off, not always to achieve what policymakers have set out to achieve. So I will act as a bit of a dampener on the following the discussion which we're going to have, because most of my answers to the suggestions are going to be, well, that's fine in theory, but it won't work in practice. And I'll stop there and hand over to Richard.

Thanks very much. I've got a few slides to share. Let me get them up there. I hope everybody can see and hear me. So what I'm just summing up a little bit on this remarkable set of contributions. In fact, the debate makes me feel— go back over many years of debates on tax at IFS, right back to Al Murley's review and the dampener coming from the tax practitioners and the excitement of the tax reformers and the theorists. It's always a great and exciting kind of interaction here, and we've got— we're right at the heart of it in this fantastic set of presentations. What I wanted to do here is point out a little bit about the review, and then just a few points on what we've heard and how we might think about it. It's a remarkable set of contributions, all live. They all went up today. And I wanted to add that there's more just in this area than we've heard today. We've got a great paper in addition to the chapter and Owen's wonderful commentary. We've got a great paper by Aaron Varney and Andy Summers on measurement and taxing top incomes and wealth, looking a little bit at wealth taxation, its interaction with income taxation too. And we've also got an article from the great wealth tax reformers, Emmanuel Saez and Gabriel Zucman, based at Berkeley, who are proposing a modern progressive wealth tax and telling us how to think about it. And of course, we've heard Ed's view. And this is a very— it's the way the review is happening here is we're mixing kind of theory, evidence, and practitioners. But there's lots more going on in the review. I just want to remind you that it's incredibly related to this, just the philosophy and ethics of what's wrong with inequality. Why do we care about the top? Perhaps we care about their power and influence as much as the income directly that they hold. Of course, it's related to discussions on labour markets and inequality, which we already had one hugely important set of work. The role of firms— we saw the discussion of business organization, business ownership, and of course superstar firms and where the highest streams of profits are coming from. The role of globalization and trade, of immigration— I'll come back to that a little bit as it's so important at the top of the distribution— of geography, social mobility, where did the top income earners come from over the generations, and of course on supporting the poor. What we've got is a fantastic picture of top incomes and top income taxation, and in a way the figures speak the story. I picked on 3 that largely come out of the chapter, but this kind of picture that Helen gave, which shows the differences in effective tax rates Once you bring in national insurance contributions, both the employer and the employee contributions across sources of income, we saw how wage earners are treated differently from the self-employed. This is really the heart of part of the debate about getting the labour market to organize effectively. And then we see how wage earners and owner-managers are treated differently. That's a that's a huge impact on shifting between types of organization and how the business— how businesses and workers interact. And I'll come back to this. Of course, what Helen showed, and the work on capital gains particularly show, is that given this discussion of averages across type, if you dig deeper, there's even more heterogeneity. And I thought This is a startling figure really, which shows the top statutory tax rates across different forms of income here. Employment income is of course right at the top there, and as we go down through business incomes, we get right down to— once we get into entrepreneurs' relief and capital gains, we get down to very low levels of effective tax rates, which every one of these presenters has discussed, and it's really key in this debate. And of course, as Owen and Helen showed very clearly, this is absolutely important at the top because right at the top, business owners, active business income is very important, and active business income is taxed differently to wage earning income. Even though to many and to a large extent they may be doing something very similar. And if human capital is the main investment that's going on here, then why would you want to treat these differently? And that's really at the heart of much of this debate on reform. So what I did here, just a couple of slides of, you know, the picture— what does it look like? Um, there's incredible concentration of income at the top. as Helen said, the top 1% receiving 50% of fiscal income more than the bottom 55%, up from 6% in 1980. That was right at the bottom of the U-shaped inequality picture, that top tax share, top shares of top incomes that Helen gave. And it's actually fallen back, or at least stayed pretty stable over the last decade or so. or so, but it's still high. And to many, as you're going to see, and in many discussions asking people what really concerns them, that high level of concentration of income— and we'll see also wealth— is of great concern to many people, and it's really much of the focus here. I should add that the other big concern we all worry about is— one of the other big concerns we worry about is productivity, and that really captures some of this trade-off, especially the last one that Edward discussed on incentivizing entrepreneurial behavior, innovation, investment, and how do we balance these 2. And I think this is really at the heart of this trade-off right at the top. So we see business owners are important at the top, and average rates of tax rates are very different across different types of incomes in the top. What happens there is there's huge incentives for shifting, and these are largely inefficient changes in organizational form and shifting from labor income into capital gains and other forms of business income, meaning that the impact of changing these income taxes, statutory income tax rates at the top, are to some extent avoided or moved out of or shifted from quite easily, giving relatively high elasticities. And as we discussed, that means that the top tax rates that give you the most revenue, if you want, are lower than they would otherwise be. If we look at the top, we see that it's very concentrated in financial services in the UK. The rather dominant thing is Southeast London and financial services, with 44% of the top 0.1% working in financial services. Quite remarkable. We can think about agglomeration, but can agglomeration be too excessive? One fact, startling fact really, that Arun Advani and Andy Summers have made a focus on is that the vast majority of the 90%, they find, of observed rise in the top 1% share over the past 20 years— and there was a rise over that— has accrued to migrants, that is those not born in the UK, and mostly in financial services. So it gives you a different perspective of who's there in the UK at the top 1% and who are the most likely to move around. And of course, much of their income, as I say before, and much of the income at the top is likely to accrue to non-doms. We see as much as 8%, 12% of the top 1% unrecorded, undeclared in tax havens, but it's very hard to capture that given the rules that we have on non-doms, which I won't comment on too much here. And of course, as we also discussed, the transfers of wealth, inheritance, and gifts That's a big amount, and at the moment that's relatively lightly taxed, around 6% of GDP. And really, that's these things, you know, beginning to focus on where the big policy reforms could happen, all of course taken in the light of tax practitioners' concerns about constraint on policy. But let me think about that a little bit. I call it a balance of policies, and we know that people worry about the top 1%, so it's on the policy agenda. Capital gains— there's a great debate about whether and how we should include them in the measurement of top 1% income, but there's no question that capital gains taxation is a key issue for reform, however we want to think about cap counting it, just because of the way it's taxed at a differential rate and how business owners can exploit that or use that. What we've seen all the contributors here, and this harks back to the way we thought about reforms in the Mirrlees Review, is a move toward alignment of at least the marginal rates across labor income and capital income, including capital gains. That's a key policy recommendation, and I think we all have to— it's a behoove on ourselves to really think about how we could make this happen or get towards that. There's clearly, as I put here, base and death both matter. We need base reform, and we could go into great detail about why base reform, but that's really key. It's key in this debate about not unduly disincentivizing entrepreneurial and innovation behavior, but it's also key in making sure that there's no simple escape routes for tax revenue if we do want to increase taxes on the top. And clearly one important aspect of that is the removal of capital gains tax. uplift at death, which I notice is now becoming a hotly debated topic of reform both in the US and here. One thing we've learned as empirical researchers is that evidence really matters here. Owen gave such a brilliant discussion of that, and the discussion about— with Troop on exactly how we think about that is absolutely key. But behavioral elasticities really matter. The mechanical effect is usually not necessarily a very good guide, and we're often led to think there would be incredible escape from an increase in tax. But of course, if we change the base and we think about a slightly longer horizon, then we've seen how we can change our mind about that, and we need to think about that more carefully. But just to end, it's not just about tax. These concerns about the top of the distribution cannot all be addressed through tax alone. That's why it's a balance of policies. Indeed, that's why the Deaton Review is covering so many different aspects of what drives inequalities and top income inequality, and understanding the drivers of top incomes matter for policy. Helen made a comment on that. You know, what is happening at the top? Is it mainly property, innovation, business income? Is it driven by market power and restrictive practices? All those things matter, and tax policy has to go hand in hand with that type of thinking. And if you think about entrepreneurial activity and innovation, it's true the top 1% reflect innovation rents, and we don't want to unduly tax innovation. But innovation and agglomeration rents can be too high, and the way we might want to think about that is a set of complementary policies. And it is true now that in competition authorities, the thinking about competition also thinks about market power, of course, but the resulting inequality that may flow from that. So competition policy to promote entry and to reduce incumbent rents is part of thinking about how to address concerns about top income inequality. And R&D tax credits, tax allowances are perhaps a much better way of incentivizing entrepreneurial behavior, as are supporting things that complement innovation— education and skills. And perhaps the point that I made so clearly and is now coming out rather startlingly across many economies, both in North America and in Europe, Scandinavia for example, is how the innovators have come from the wealthiest backgrounds. The children from the top 1% might many, many more times likely to become entrepreneurs and innovators. That must be a completely hopeless outcome for productivity as well as inequality. And if there's something that this review can do is to try and bring all these ideas together and policies together. And then finally, a wealth tax, a tax on wealth transfers we just described. We thought we've discussed it here. We've started thinking about the role of inheritance and what have you. But I should turn— I won't spend any time on this, But if you look at the commentaries and articles by Aaron O'Banion and Andy Summers and Emmanuel Saez and Gabriel Zucman, you'll see that they are very strongly proposing a one-off tax on high wealth and an annual wealth tax on the very top millionaires. It's not what we're here to discuss today, but I know that it's something that everyone wants to put into this discussion about the taxation of top incomes. So let me hand back to Paul at this point to open up the discussion. Thank you.

Well, thank you very much indeed, Richard, and also to Edward Owen and Helen for a fantastic set of presentations and a heck of a lot of big and difficult issues to deal with. Which are very well reflected actually in the questions that we've got on Slido, which are some big and difficult questions. So I'm glad I'm the one relaying them rather than trying to answer them. So I am, as ever, going to at least start with those that will be voted to the top. I will then pick some others, but the The first one is from Rhys Harper of the Scottish Exchequer. He's asking, saying the Treasury rejected a proposal made by Office of Tax Simplification to align capital gains tax rates with income tax. Did Treasury give a satisfactory justification for that decision? I'm going to go to Helen and then to Edward on that one.

Right, so I haven't got the HMRC response in front of me, so I don't want to misquote quote them. And of course, ultimately, it's politicians that matter here as opposed to the Treasury, because they're the ones that get to decide what to do. But there's 2 things worth saying. I think one is, if you're just talking about adjusting capital gains tax rates, I think there is a real and genuine concern that if you just put up rates, you are going to discourage some forms of investment in a way that's undesirable. And that's not a made-up concern, it's a real concern. But I think the constraint there shouldn't be seen as a hard one, because we can do things to tax base that alleviate that problem. So one of the comments that's come up in the Q&A, which I think is completely right, is that at the moment we basically tax inflationary gains. Now I've been saying that's a problem for a long time. People recently said to me, don't worry, inflation's low. That's not true anymore, clearly. And even when it is low, it compounds pretty quickly over time. But we could take inflationary gains out of the tax base. In fact, I'd go a little bit further and basically rather than index for inflation, index for interest rates. just to say that you can make some returns on your capital before we start taxing you effectively. So big picture, if you just tax rates up, I think there are concerns, but we can do broader things to the tax base that make those concerns, alleviate them to a large degree. So I think there's no justification for saying we just can't do this. I think the point that Edward raised is really important about people have to believe that it's going to be a permanent tax change, because if not, they're just going to hold on to their gains and not do anything. That is hard, and you might think if a Conservative government did it, it'd be more likely to stick, but of course, actually remember that it was a Labour government that introduced entrepreneurs relief, the preferential rate for business owners, so it's not just a Conservative preference for cutting capital gains. My personal view, and maybe this is just a naive hope of an academic type view, but is that the more you fix the broader system and do something more systematic that reforms it, makes it more stable, and doesn't just tax inflation, the more likely you are for that regime to stick and not be fiddled with constantly. But I do think thinking about how the regime's— how permanent it will be matters for how well it will work.

Shall I come in, Paul?

Yes, please.

I mean, the simple answer is that the Treasury didn't give a satisfactory answer, but of course there are some quite good reasons why you shouldn't do this. It's also worth noting that Nigel Lawson, of course, did do this in 1988, under a Conservative government, and it didn't stick. And in a sense, what Helen has said explains why, because the tax system is to a large extent one size fits all, and capital gains come in a lot of different sizes. And there are capital gains, you know, and carried interest for private equity participants are always cited as one, which clearly should be taxed as income. But also, there are, at the same time, there are a lot of gains which should not be taxed at full income tax rates because they represent profits which have already borne tax to a greater or lesser extent, and therefore a full income tax rate would represent double taxation. There are good policy reasons not for doing it, and experience has shown that it doesn't stick, but there are also the practical reasons which we've talked about, which is that if you have excessively high capital gains tax rates, people will not realize the gains. We haven't talked about an accruals method of taxation, we might want to come back to that, and as Helen has said and as experience has shown, people don't believe that it's going to stick, and so there's no point in doing it because you won't raise the tax because no one will realize their gains unless they are unfortunate enough that they have to realize their gains for whatever reason.

Okay, I'm going to I'm going to give the panelists a warning that Jeremy Moore's question you see here, what would be the least bad way to take an additional 1% of GDP from the top 1%? I'm going to come to that at the end, so you've all got a moment to think about how you would if you wanted to get a lot of extra money from the richest. There's a question here from Kevin Langford which actually Owen really addressed to some extent, but I think it is worth I think everyone reflecting on this actually, which is the HMRC view that increasing the top rate doesn't really raise very much money. I mean, to what extent do we believe that, and to what extent is that just very uncertain, and what would most change that? Why don't you go first, Owen, since you sort of addressed this most directly, at least in the US context. Then I will actually come to all 3 of the other panelists as well on this question. I think it's actually rather an important one.

And I think it is a great question. In terms of the top rate, it depends on which tax base you're talking about. So, you know, given the chart that Helen showed on labor income, you know, we typically think that distortions are growing with the square of the tax rate. And so at higher rates, the distortions can be somewhat material. I haven't looked at the latest estimates of kind of the Laffer rates, and they are sensitive to what these elasticities are on labor income. And so, you know, something closer to 40%, 50% is getting, you know, maybe in the range that's covered by confidence intervals on labor income. For business income, it seems that the, you know, the rates are lower, especially on capital gains. And as I pointed out, you know, I totally am sympathetic to Edward's comment that there are a lot of ways to retime and kind of wait for politicians to change and defer gains. However, my view is that the data is kind of informative about how big that is. And so, you know, the mechanical effect in our estimates do indeed cut in half, but not, you know, by a factor of 5 or 7. And so I think the best way to answer these things is to look at what's actually happened. To the extent we can. And, and so the short answer is there are different answers for different types of income, and it really depends on where the current rate is. So for capital gains, the rate's lower, and so I think it's less compelling that you'll get less revenue, even though capital gains are more responsive than labor income. So it's a bit of a local answer.

Richard, do you want to go next on this one?

Yeah, yeah, sure. It's always been difficult because when we were looking at this some time ago, you know, it's very hard to get evidence if there's no variation. But of course, we don't want variation, as I think you all tried to say. But variation is very good because you can see how people respond. And we had a lot of variation in top tax rates, but it was a long time ago. And when 10 years ago we were trying to figure out the elasticity of taxable income, it was really— it turned out to be using the reforms way back really in the 1980s. And that's controversial for a number of reasons. It can go either way. We've had base reforms since which make the elasticities harder to shift, but also we've had— we're a much more joined-up international economy, and we've seen the importance— I tried to stress the importance of migrants in the distribution and non-doms. And so you can see that that might also make the elasticity or the behavioral effects bigger, and that really matters. You know, at the top, there are really 2 things that matter, Paul, and it's just the amount of income up there and how responsive on average people are up there. But it's really hard to get a good number there. When we did it, we came up with a number that made a little bit of a rise in the top rate get some additional revenue, not a huge amount. That was with the current tax base as it stood. But in fact, the confidence of that in a statistical sense was rather weak, and you could fit in a number of Since then, we've had variation in the UK that's got us a little bit closer to home on some of these issues, and I think it shows— and I think Helen made this point— that with the current tax base, responsiveness is quite relatively high. Remember the kind of responsiveness that Owen was given in capital gains that was used in the reform analysis was very, very responsive, which is why there was little revenue. What he's suggesting is something rather in the middle, an elasticity that looks just like what I would have expected, and it gets a lot more revenue but not the full mechanical amount. And these things matter hugely, thank goodness, otherwise there'd be, you know, there wouldn't be room for us empirical researchers. But I think that that is part of the the point here that you really have to think of base reform, and there is plenty of base reform to do. It's always going to be hard, but the empirics matter. And I think that's— I agree in a way, this debate that we're having, including that Edward's put forward here, you know, it's not black and white, this. And you— it— we look at behavior and we see what we can find, and then we think about about setting the rates. And I think there is room for base reform in a way that would allow us, if we so wished, to increase rates and align them rather like Helen's ambition, I'm going to say now, is.

Edward, Helen, did you have anything to add? Don't feel you have to if you don't.

We can't hear you, Edward. You can go first, I'll chip in.

Very briefly, a factor which we haven't talked about before, which is the sort of moral and compliance attitude at any time, you know, particularly for the wealthy who do have choices, the perception of how sensible it is to do this does have a moral compliance element, and I do think over the last 10 years we have seen as it were, a slightly reduced willingness to be seen to be doing outrageous things, to make space. And actually, I think if we put taxes up now and if the case was made well, I think we would probably raise more than we might have done in the heyday of the sort of gangbusting noughties.

I'll just say quickly that just on the HMRC estimate in particular, I think it's credit where credit's due. I think they're using the best estimates out there, including some of the ones that we've— Richard mentioned that we've created at IFS. It's just that because it's a group that's hard to study, there's a lot of uncertainty about that. But I think most economists think that if you did some of the basic reforms you've talked about, made it harder to switch into dividends or capital gains, because you just reduce some shifting opportunities, you'd have a higher rate be possible. Of course, you still wouldn't know some things about migration effects, but you know, you can have higher rates. I think also echoing what Edward said earlier, it's worth remembering that when we say that the top is responsive, we don't mean that every individual responds a lot to every tax change. it is the case that because the money, the sums involved, even a few people making changes to their affairs have some pretty big revenue effects. So if somebody moves as a result of the tax, you don't just lose the marginal pound, you lose all the pounds you would have raised. So I think even a small group being very responsive has a big effect because they're a small group to start with.

There's a couple of questions here, one about— I know they're different things, but there's one about avoidance and one about evasion, and I think we can't go through this We can't go through this panel without dealing with those issues. So there's a question here from Michael Lipton, somebody saying rising tax take means incentive to evasion have risen. Essentially, how much evasion do we think there is? What evidence do we have on that? And then a couple of questions down, there's a question about avoidance. The avoidance estimates are striking. In their book, Saez and Zucman wrote that the people dodge taxes as a policy choice and recommend a federal agency to combat it? Could that actually have a big effect on avoidance? So 2 rather big questions there. I'm going to start with Edward and then go to Owen on that one.

This is something where we get so hung up on the terminology. There is a much bigger area for the wealthy where the amount of tax actually raised is less than people, and that includes policymakers, think should be raised. Whether the tax that is not raised should be categorized as behavioral avoidance, clearly not evasion, because I want to leave evasion because that is illegal, that is breaking existing law. But how you categorize it really becomes less important what is the best way to actually get a closer match between what policymakers and the public would like to see as the outcome and what is actually achieved. And as I said, because of the much wider range of choices open to the wealthy, because of the extent to which they can do a lot more discretionary things with their cash, there is just a much wider range of activity here, and there always will be. I think the question is not what is the measurement, what is the actual answer, but where should we be devoting our efforts now to raise more tax in particular areas. I do feel this is a debate which gets far too emotional, and I do think there are some ridiculous numbers banded around, and it really doesn't matter how you characterize them. The real question is, can we actually get some money from them, not would we Like to?

Well, I do have an answer to that. I mean, can we in a reasonable way? Then you've also evaded the question about evasion.

Well, let me let me deal with the question about evasion because because evasion is simply saying the law law is clear and these people are breaking the law. By and large, the wealthy, at least in the UK, and there's a long strong local element to this, do not break the law. They do comply with the law simply because they have got. many economic interests, they're so embedded into the national system, into the laws, the regulations, the society, that actually breaking the law and doing something illegal for which, you know, there is both criminal and potentially, you know, jail sanctions for, that they don't do it. What they do is they push the boundaries of what the law allows as far as they possibly can with the aid of good advisors. So I I do believe the HMRC tax gap estimates of evasion by the wealthy. That's not to say that I think that it's right that these people should be allowed to maintain trusts in the British Virgin Islands or wherever it is. I'm just saying it's not illegal for them to do it. It's not evasion. If we can, and we are progressively finding ways where we can legitimately, lawfully, and effectively tax them, then we should be doing so. Where are the areas where we can actually do more in raising tax? It is, I think, a large extent— first of all, I don't want to come back to non-doms, but I think the non-dom rules have got further to be rolled back. Clearly, our Chancellor may have some conflicts of personal interest if he seeks to do that, but I don't want to go there. I do think some of the areas which Helen has talked about, which is the treatment of business reliefs, and I do think there is scope for exploring, at least in some areas, a greater degree of accruals basis of taxation, but only in some limited, really quite top-end areas of activity. Owen?

Yes, I think that generally the point by Emmanuel and Gabriel is quite right, that avoidance is a policy choice. So I think the base reforms that Richard was talking about, things like step-up at death or the ability to relabel income of different forms into lower rates, is really quite important for how easy it is to evade. In the US version, there's also how well-funded the IRS is. So the Internal Revenue Service is horribly underfunded, and the ability to track down and follow the money for quite complicated entities is very much limited by these resource constraints. And so I don't have as good of a sense on what the current status is in the UK. I'd imagine it's more reasonable than what we're doing. But that's very much a policy choice. We should do much better on that. On evasion, I think that the general trend of governments coordinating with each other and things like FATCA in the US or on the business side, thinking about the BEPS process is generally helping us do a bit better because there really is a lot of patterns that just don't line up with what you'd expect to see, things that are an order of magnitude larger than GDP in terms of activity that are hard to rationalize. And so I think some coordination is good. In terms of the magnitude of evasion, there are estimates from linking the Panama Papers to administrative data, or there's some nice work by Daniel Reck and Gabriel Zuckman and Max Risch and John Guyton, you know, doing that in the US. In general, some of these things are a little hard to get right. There's a lot of decision choices, but it's pretty clear that, you know, putting some resources behind doing a better thing is, is a quite reasonable thing, both for fairness and for revenue, especially if you're starting from a low base where you're not putting that much money behind it.

Great, thank you. There's a couple of questions from Hillary Cooper here. I'll take one of them very quickly, which is, would it be a quick win to put something equivalent to NI on high-income pensioners? And the answer is economically yes, politically probably not for the party that tried to do it. Economically, because most occupational pension income has never been subject to any National Insurance either, so they've got— it doesn't doesn't attract National Insurance at any point. There is a strong economic case for putting some additional levy on occupational pensions in payment, but I think the chances of that happening, given the propensity of pensioners to vote, and particularly to vote for this government, I think the chances of happening are close to zero. There's another question here also about the effect of the zero rate on capital gains at death. I don't know if anyone I don't know, Helen, if we've got any idea of what— or Edward— what might we raise? How might people behave if one were just to get rid of that relief at death?

So the OTS has a number that they said removing it they thought was about £1.6 billion, but for all the reasons we've been discussing, including the reasons that Owen set out, there's just a ton of uncertainty about that. It really would depend on how people would respond if they believed it was permanent and all that kind of stuff. I think what's worth bearing in mind is that the reason to do it— well, lots of reasons to do it— one reason to do it is to stop it being a backstop to other problems. So if you put a main rate of capital gains, you would get more capital gains at death, and that would therefore be a more valuable route for people to use. So what you don't want to do is have main rates go up and then push people to wait until they die. So in some ways, it's not just— it's doesn't matter just so much how much we're spending on it right now, it's that all the time it's there, it prevents us from doing other things to capital gains because it's an outlet for gains to go to. So yes, I'm uncertain about the number, but it's really about that backstop role it plays, I think.

Does anyone want to add anything to that?

Actually, yeah, one thing. So in the Biden administration's most recent budget proposals, if you are a tax nerd like myself and look at the Green Book, They have an estimate of how much step-up basis at death would raise. I don't remember the specific number off the top of my head. I remember it's on page 31, but I don't remember the number. The difficulty is that there's very little evidence on what happens if you make this policy reform. And our understanding of capital gains behavior is quite limited. And we did some back-of-the-envelope work in the paper with Larry Summers and Natasha and Eric Zwick, that like half of gains aren't taxed. So I don't know if it's the same in the UK, but it's a pretty large amount of resources. And economically, if you think there's a lock-in effect, especially towards death, that people are waiting for, there's some efficiency gains as well, potentially, from this policy.

Yeah, absolutely. I think the efficiency gains are potentially quite big. Do we have any idea, Helen, what fraction of capital gains actually end up not being taxed because they're kept until death? Do we have any idea at all?

I don't think we know. There are some numbers in the paper as well, but we know, for example, that most capital gains aren't taxed because they're in housing. So the big untaxed capital gain here is primary residences. I don't think we actually— I don't know, I have to double-check, but I don't think we have a great number on the death.

That does bring to mind the famous change in inheritance tax in Queensland when inheritance tax went from positive to zero. people managed to sort of hold on for a few extra days to avoid the inheritance tax. I wonder if you went in the other direction and I wonder if you'd have the same— you'd have the opposite effect if you took capital gains tax from zero to positive and people actually would leave this earth a few days earlier than they might otherwise have done. Let's— one last question before I come to Jeremy Moore's question. I don't know whether you wrote this, Helen, but it's certainly one for you. How do you change the tax base so as not to discourage investment while equalizing tax rates?

I'm not a mole behind it, I didn't actually write this. So without starting another half an hour lecture, I think one way to think of this is that if you take the return to investment, you can sort of think of it in 2 parts. So part of it is just what an economist would call the normal rate of return. You can imagine if you put your money in a bank account, you're partly getting a return just reflecting the fact you've got to wait for your money, you're just you can talk about doing normal savings. And when you're putting money in an asset in a business, part of the return you're getting is just a delay for waiting, or it's a compensation for inflation. And if you tax that part of the gain, then you're discouraging people from making those investments, whereas another part of the returns, or the returns above the normal rate of return, are the kind of profits you're making. So the sort of broad conceptual idea here is don't tax the normal rate of return, let people have the sort of normal investment returns, and do tax everything above that. There's different ways you can do that without getting into lots of details, but what you really want to do is make sure that the full cost of investment is deductible. So that matters how you treat capital allowances, making sure that people can deduct the full cost of their finance, so not just debt costs, but the equivalent opportunity cost of equity finance. It also matters about risk-taking. How you treat losses matters quite a lot here. So at the moment, the government basically takes an asymmetric stake in investment. It says, if you get a big profit, we'll tax you. If we get a loss, we'll offset them, but we usually offset profits at lower rates than we tax gains. And at the moment, if you put up the profits tax rate and you don't change losses, you're making that more asymmetric. But rather than just say, well, take somebody who's starting a business, let's give them low rates in the future if they're successful, instead you can basically say, well, let's give them more insurance upfront. and say if you make a loss, we'll allow you to offset those losses more flexibly. So basically you want to change the upfront treatment of investment so that we're not penalizing people who are investing and then increase the rates. But there's much more about that in the chapter and other places written that people can follow up.

Well, let's move on to the final—

Sorry, can I just briefly on that, and Helen and I will take it offline, but the reality is that most investment is made by companies, not by individuals, and most individuals make investment because they think they're going to make a million or a billion, not because they're worried about the tax treatment. So I don't think there's much we can do.

Well, just— but we'll take it offline. But just on that, I mean, if they're not worried about the tax treatment, then great, put their tax rates up, it's not affecting them. So it's— I think we're agreeing on the fact that the lower tax rates here aren't providing—

aren't making people I think that was the implication of what you were saying, Edward.

Yes, it was.

Yeah, great. Well, there we are. So that might be part of the answer to this last question. Just very briefly, if you did have to take an additional 1% of national income from the top 1%, how would you do it? I guess go Richard, Edward, Owen, Helen, but please answer briefly.

Brief, but they're all pie in the sky. So resources to collect the money, for sure. The big one is just coordination. I still feel we lack coordination across jurisdictions, and the reason it's pie in the sky is that, you know, the world is going in a different way. But in a practical sense, I will go back to the thing that we discussed right the way through this, and that is about NICs and capital gains tax. And I think there's room here, there really is room for raising more revenue there. You have to do it in a clever way, but I think, you know, the brains amongst us must be able to get that to work, and it's really important. There's a political dimension, but I think, you know, we've got to make the case.

Edward, you've run HMRC. If you've been also the politician in charge and were telling your colleagues to get some more money out of the rich, what would you have told them to do?

In terms of compliance activity, I think just a bit more, this is not a compliance issue, this is a policy issue. We've sort of answered the question, which is do away with all of these reliefs for investment, for enterprise, for agricultural property, all of this nonsense. The reality is CGT is always worth playing with, but it's never going to produce the money. It has to be on what, you know, we traditionally think of income. You should tax codgers like me on our rental income, our dividend income, our pension income, you know, more heavily. And as Helen's figures show, you get, you know, quite a bit more money.

Bring it on.

Excellent. Well, I think Edward has volunteering himself as the subject of this, of this policy. Owen.

The US version of it— Eric Zwick and I had this proposal called Party Like It's 1997, which basically rolled back the tax code to 1997 in the US, which had higher dividend taxes and estate tax with more teeth and lower exemptions. And so I think that direction— higher capital gains— that direction is reasonable. I think a balanced approach In the UK version, I think step-up at death and inheritance taxes are really our silver spoon taxes, you could think of them, I think are quite important. And I agree with Edward and Richard on capital gains taxes and other base broadeners.

I love it. Party like it's 1997.

Helen. So clearly, 1% of national income is a lot of money, especially if you're going to try and get that just from half a million people. no one tax rate you're going to change that's going to get you that. You have to do a package of things. So again, I think it has to be a combination of changing capital gains tax rates, dividend tax rates, rental incomes, a bit more compliance money, plus things outside of income tax. So to mention one UK one, council tax, horribly regressive, horribly out of date, big breaks at the top. We could get more from that group in a broader sense than just thinking about income. So I think we do have to look at the tax system as a whole. And if you want big money, you've got to do a bunch of things and be bold.

Well, that's probably quite a good place to leave it. Do a bunch of things and be bold. It's not a bad motto, actually, for many things. Thank you everyone for listening. Thank you to Richard, to Edward, to Owen, to Helen, particularly to Edward for getting up extremely early and Owen for getting up quite early in order to to do this. Thank you all for listening, and as I said, this is one of many outputs from the IFS Deaton Review of Inequalities. We have several more over the next couple of months, particularly focusing on things like education and the family, which— and early years issues, which are very different from this, but also very important in understanding inequality and understanding what you might want to do about inequality. So look out for those things coming up, and thank you very much for once again for listening.