[00:00] Arun Advani: For people who don't know, we haven't revalued council taxes since 1991. That is just embarrassing and ridiculous. Nobody thinks it makes sense. I mean, the most amazing part of it is that when we build new houses, we have to try and guess what we think this house would have been worth in 1991, which is just a stupid question.

[00:17] Ravi Gurumurthy: Hi, I'm Ravi Gurumurthy, CEO of Nesta, and we've got Arun Advani here in the latest in our podcast series, What Can Andy Burnham Do. We're going to be focusing today on the big question of taxation: Should we raise capital gains tax? Should we change the way we tax property? What can we do to tackle all the cliff edges that exist that create weird distortions? We're also going to be talking about what it means for devolution—what would a tax system that embraced devolution look like? Arun, welcome. Thank you for joining us.

[00:48] Arun Advani: Oh, thanks for having me.

[00:52] Ravi Gurumurthy: So we're going to talk all things tax and what Andy Burnham should do. Let's start with one of the most often suggested taxes, which is a wealth tax. The Greens in particular have suggested it. I don't think you think that there is a viable wealth tax that a government could do that's either desirable or feasible, but just outline what are the typical options that people talk about and why are they not possible?

[01:09] Arun Advani: So when you talk about a wealth tax, a wealth tax is saying: let's add up the value of all of the stuff that somebody owns—their house (if they own it jointly with a spouse, half the value of the house), take off the mortgage, the value of their pension, value of any money they have in a bank, if they have any stocks and shares, add up all of those things and then draw a line somewhere and say if their wealth is above that line, then on the bit above that line, let's tax that. So, for example, a policy that's been floating around for a little while is some people calling for a 2% tax above £10 million. They're saying if you have £11 million, you're paying 2% on just that top million.

If you look around the world at what wealth taxes of this sort look like, they've tended to be a relatively low threshold. So a threshold in UK terms is like £100,000, £500,000—those sorts of thresholds. That covers a very large share of the population. The impact of the choices that other countries have made when they have these low-threshold taxes is that because there's so many people who are covered by the tax, you have to keep the valuation of all these assets simple, right? You know roughly the value of your house; you don't know exactly the value. You know roughly the value you could get information on the value of your pension, but these things eventually add up to being complex.

Particularly what's complex is the value of a business. So if you're a plumber and you're running a small business, or you're a hairdresser running a small business, you have to work out: what's that business actually worth if you sold it to somebody else? Often not very much, but you still have to work it out. And so to solve the complexity that comes in here, countries have taken shortcuts in how they value things. So they say, okay, we'll value the business of a plumber at just what's the value of the tools that he or she has and the van and those things. For a plumber, that's basically right—in a world where they sell their business, if they're the one person running the business, once they sell it, all it's worth to somebody else is just the stuff.

The problem is that when you then use that same method for valuing the business someone very wealthy has, that's nowhere near the value of the business. Because it's not the computers and the tables and the chairs that make a big business valuable—it's often the know-how, the client lists, the special IP that they've built up. That's really where the value is. Taking these shortcuts means that wealth taxes aren't very effective when they're designed in that way for attacking the very wealthiest.

The alternative that at the Wealth Tax Commission—which I ran with Andy Summers and Emma Chamberlain—we talked about was: you could instead, if you wanted, have a wealth tax starting at a much higher threshold, say £10 million. Then you have a relatively small share of the population who you're covering, and you could then afford to do very high-quality market valuation. It's not easy; it takes some work, but you can do it. No one should pretend it's going to be simple. No one should pretend that you're not going to have some avoidance behaviors. We estimated that maybe 17% of the expected revenue you'd get if nobody changed their behavior would be lost due to changes in behavior for a 1% tax. So that's the way to think about it: you can definitely deliver it, but it will have some costs.

It will also take time. It takes a few years to build a new tax like this. We don't currently know who owns what shares, what properties, what businesses in an effective way, so you couldn't implement it immediately.

[04:04] Ravi Gurumurthy: Just because the actual valuation process would take time?

[04:06] Arun Advani: Valuation takes time. You literally just need computer systems to do it and forms to do it. Those things seem really boring and process-oriented, but they are a barrier. Particularly at the moment, in these discussions going around about whether Andy Burnham should create a wealth tax now: he's not going to have a wealth tax that gives him any money this side of an election. So if he really wants to try and do the work for it, he could, but he shouldn't expect any revenue from it this side of an election.

That's just a trade-off: do you want to burn political capital on that when there are other taxes on wealth that we have, like inheritance tax and capital gains tax, that could be improved? Particularly capital gains tax—there's lots of scope for improvement there.

[04:48] Ravi Gurumurthy: And in terms of the amount you'd actually raise if you went for this tax on assets above £10 million, what kind of estimate do you come up with?

[04:55] Arun Advani: So the Wealth Tax Commission, when we were running back in 2020, we said a 1% tax on assets above £10 million would be taxing at the time about 22,000 people, but it would raise about £10 billion. So you're not going to pay for the whole NHS budget on that.

[05:09] Ravi Gurumurthy: Is that an annual tax?

[05:10] Arun Advani: That's recurring at £10 billion a year. You should think about that as like 1.5p on standard rates of income tax. Now wealth has increased a little bit, so we think maybe it's about £12 billion. The ONS version of those data declined in quality, so it's a bit harder to know now exactly what the number is, but we're talking £10 to £12 billion after accounting for behavioral responses every year.

[05:35] Ravi Gurumurthy: And those behavioral responses—how did you get a handle on how big or small those are? That must be pretty difficult to estimate.

[05:40] Arun Advani: Yeah, so we couldn't estimate it directly for the UK because we don't have a tax like this in the UK. What you can do is look at all the things that people can do and then look at the scale of those responses in other countries where these things have happened.

It's worth saying that the estimate of losing about 17% of the revenue was even after you do a lot of fixes. In other countries, often the loss is even larger than that because they've made bad choices. One bad choice is to not value some asset classes. So in some countries, they say main homes are just exempt from the wealth tax. The problem with that is, of course, people spend more money on their main homes than other assets, right? Or you say, let's take a shortcut and value private businesses at a discount—work out the value of the private business and tax half that value. Well, suddenly my private business turns out to own a lot of my stuff because having it inside the business gives me a 50% discount. So I put all of my stuff inside the business instead of owning it personally.

In Switzerland, the tax has been designed in a way that's very local by Canton, which is actually quite small, and so people move Cantons because just moving across the border can reduce your tax. When those changes that you need to make in your life are quite small relative to the tax impact, you see quite a lot of impacts there.

In the best-case scenario where you don't make those bad choices, there are things you still can't stop people doing. You still won't stop the fact that if I have £12 million and you put a tax on me above £10 million, I'm suddenly going to give some more money to my kids. I was going to give it to them one day; I give it to them a bit sooner, and suddenly we're all below £10 million. You can't do much about that; that's a totally legitimate thing. I genuinely don't have that money if I've really given it to my adult children. But at that point, clearly you're not getting revenue from me, and so we have to account for those sorts of behaviors.

[07:25] Ravi Gurumurthy: Okay, let's talk then about something else that has been dubbed a wealth tax—I think by Wes Streeting—which is about CGT (Capital Gains Tax). You mentioned the idea of indexing the gains, equalizing the rate between capital gains tax and income tax, and then to avoid the behavioral responses you mentioned, your proposal is to have a death tax and an exit tax so you can't basically leave the country or hoard your assets until you die. Now, you basically think that's a good idea.

I want to start by unpicking the effects for different people and what this does. There are three different segments, I think. One is that you've got people who get taxed unfairly because their asset has only grown with inflation, yet they're getting taxed on a nominal increase—and your proposal would fix that; you'd pay nothing in that situation.

Then you've got the problem of the solicitor who becomes a partner in a firm, whose income is being taxed more as a capital gain because people can shift things around, and therefore they're being undertaxed—they're not being taxed at the right rate when it's basically income.

[08:46] Arun Advani: Just to give a practical example of what you see with that kind of behavior: the classic example is in management consultancy or IT consultancy. You work for a big company for a few years, get your skills, then go and set up a sole trading incorporated company as a consultant. You get paid into the business, keep the money mostly in the business, pay yourself a small salary, pay some dividends every year because they're taxed at a lower rate than salary would be, and then anything you don't need immediately, you keep stored up inside the business. At some point, you can liquidate the business and get that money out as a capital gain. That's basically work you were doing that should have been taxed at income tax rates, but you're getting a much lower capital gains tax rate now.

[09:25] Ravi Gurumurthy: And no substantial investment was going into that business, exactly. But then there's a third thing, which is let's say you set up your business, you're an entrepreneur, and you'd then be paying CGT at 45% or whatever—that would be quite a big shift. The concern I would have is: does that not deter entrepreneurial activity? For your first two groups, great, you've solved those two problems. But how do you think about how to avoid deterring entrepreneurship in your change?

[09:54] Arun Advani: A few different things to think about there. One is actually our best evidence on what genuine entrepreneurs know when they're setting up businesses is that they basically don't know much about what capital gains tax rates look like. There's some good qualitative work that was done by HMRC many years ago where they were looking at reliefs that they were giving entrepreneurs and trying to work out: did these entrepreneurs know anything about these reliefs? They didn't know any of the stuff existed; they had no idea what the rate was.

Many of these people are basically thinking, "I've got my brilliant idea, I really believe in it, I want to make it happen. Obviously I want to make a profit doing it, but I'm not really aware of what the tax status looks like when I exit the business." Right now, they want to build a business and make it successful, and for the next 10 years, they see themselves running a business and making money through the business. They're not thinking as much about the end state. Obviously, at the point that you've made a big business, you care and want that rate to be as low as possible, but it's not a big part of what people are thinking at the start. So that's the first thing.

[10:49] Ravi Gurumurthy: Although if it was a 45% CGT rate—higher than any other country—presumably it might deter, if people know it exists.

[10:57] Arun Advani: Only if you know it exists and how it's taxed, but yes, absolutely.

Second thing: to the extent you're putting capital at risk, you're going to be getting this indexation allowance—this investment allowance—that's going to reduce your tax rate. If there's basically no capital in the business and what you're saying is, "Well, I'm just working really, really hard," well, yes, lots of people are working really, really hard. That's what income tax rates are for. Now, we might say income tax rates are too high or too low, but ultimately that's a different debate about what the income tax rate should be. If you're working really hard to build your business, you're working just like lots of other people are working, and it should be taxed like it's work.

If you're making a profit on that business beyond just the returns you're getting on the capital, it's either because you're incredibly hardworking and skillful, or it's luck. If it's luck, there's not a really compelling case to give you a lower tax rate. Or it's that you have some market power or monopoly power—your business has cornered something that means you'll be able to make profits beyond what everyone else can make. Again, in that case, there's not really a reason why we want to be giving you a lower tax rate.

Third thing: if you really are still worried as a policymaker and you say, "I hear all that, but I'm still really worried about this," despite the lack of evidence... and I think the answer genuinely is a lack of evidence; we just don't know a lot about what this effect looks like in practice because these kind of people are few and far between in terms of setting up big businesses. If you're really concerned about that, what you should do is take a bad relief that we have—literally called BADR (Business Asset Disposal Relief)—which currently says on the first million pounds that you get of a private business you run, you get a special lower tax rate. That tax rate used to be 10%, has now gone up to 18%. You get an 18% rate on the first million pounds, and then you pay the standard rate on everything else.

Now, if you have a business worth £2 million, getting half of the value of that business at a low rate is quite nice. But if you're really building the next unicorn and it's going to be worth a billion pounds, do you really care that you get a low rate on the first million pounds? Not at all, right? It's not having any impact on the tax you're paying.

So what you do is take that bad relief and flip it on its head. Say actually, we won't give you any relief if you're building a small business. But if you build a business that's above some huge size—really valuable, contributing to overall employment—then we'll give you a relief. If you build a company worth more than £100 million and then sell some share of it, we'll give you a special relief that gives you a low rate on the whole value. Flipping a bad relief on its head gives you relief on an unlimited amount, but only for a very narrow slice of people who are really building these giant companies. That would be a way to protect specifically that group if you're worried about it.

[13:52] Ravi Gurumurthy: Okay, so in some ways, the cumulative impact of both those changes would be: stay invested, build something big.

[13:58] Arun Advani: Yeah. If you're building something big and you're staying here and building it really big, then you're going to get a relief. If you build something that's small, then great, you did it, but there's no big wider impact for the UK that we should be giving you a lower rate for.

[14:09] Ravi Gurumurthy: What about the other slight risk? When Nigel Lawson did this in 1988, it was in the context of a tax-cutting budget where top rates were coming down. If you did it now on top of all the other tax changes that happened over the last two years, is there a slight risk that from a vibe perspective, it just feels like taxing wealth creation even more?

[14:36] Arun Advani: Yeah, so the best time to do this would have been yesterday, paired with other things. Even this capital gains tax reform—by giving an investment allowance—is giving you a lower tax rate if you're putting actual capital at risk, because it's bringing down the cost of capital. That is encouraging investment.

More than half of current capital gains taxpayers are better off or no worse off under this reform. The people who are paying more are the ones where there's very little capital in the business, mostly the example of the consultants and that sort of person. That's where the revenue is coming from.

It also raises quite a lot of money. If you do this overall package of reform, you get something like £12 billion a year, and you could use some of that money towards other things that are pro-growth. That could be giving money to the British Business Bank to go and do investments, or increasing the size of existing reliefs. You can think about the best way to support growth with that, which might send the signal that what we've got now isn't a capital gains tax policy that's pro-growth—it's one that's pro-shifting income between different sources. What we want is one that actually compensates you for capital at risk.

By giving an investment allowance that protects that capital and says you can make a certain return on that that is at least what you would have got by just sticking it in a bank, and we're only going to tax you if you make more profit than that—that's actually good for growth. You could spend the money on other things that would also support growth and pair well with a pro-growth budget.

[16:23] Ravi Gurumurthy: And the £12 billion figure—is that a material amount that you could raise? I've heard Treasury internal estimates on this change being lower than that. Do you know why that might be? Are they assuming some behavioral response that you've not accounted for?

[16:40] Arun Advani: It's really about what package of reforms you have. The number that people throw around a lot is that you lose money on a reform that just increases the tax rate. I think that's probably right; I would not start from here and say let's just increase the tax rate. That is definitely bad for growth, bad vibes, and highly likely to cost money because people are going to respond in ways that don't support more revenue coming from capital gains tax.

But the point is a package reform that says: look, we're going to actually reduce the cost of capital for people and reduce the tax rate for more than half of current capital gains taxpayers. The ones who are going to pay more are the ones not putting capital at risk. That is a way you can actually raise money.

Costing just a tax increase plausibly gets you to lower tax revenue. Costing it with these other things actually reduces some of the ways in which people avoid the tax now. Currently, there's a very strong incentive for someone in their 70s or 80s who has owned assets for a long time: if I sell my shares to try and get something more exciting, I have to pay a big capital gains tax upfront. It's not worth it; I'm probably better off having the capital gains tax written off when I die.

Or if I've built a huge fintech firm worth many billions of pounds, and I want to build my next business here, if I sell my company while I'm here, I'm going to have to pay a 24% rate right now. I could just move to Dubai and there's no charge on settling up. I've used UK resources and all the benefits of the UK to build this, and the UK is saying, "We're going to tax you, except if you leave. If you promise to leave, we'll give you a 0% rate as long as you promise not to come back for six years." That's what we do right now. Obviously, if you say that, lots of people take that incentive seriously and choose to leave.

If you shut down those kinds of margins and say, "We're not trying to tax you for exiting—you're welcome to exit, but you will have to settle up the tax bill you owe on the gains you've built up while you were here." It's sensible that we allow you to defer the tax until the point where you realize and get the money from those gains, but if you're leaving the tax system, it makes sense to say you do have to settle up and pay your restaurant bill for having been here. Other countries do that—we're one of only two countries in the G7 that don't.

[19:08] Ravi Gurumurthy: Now that you've raised £12 billion from this change, what would be the most pro-growth way of spending some or all of that £12 billion on tax changes?

[19:22] Arun Advani: In some sense, it's not even obvious that tax changes are what you want to do, right? One answer is that there are lots of places where we know there's a return to spending, like the state of transport in this country or potholes—things people complain about. There are real benefits to spending on some of those things.

Within the space of tax, I think the CGT reform overall is a good reform even before you think about how you spend that money. The fifth limb of that reform is being more generous on losses. People who invest might make losses. Why can't we be more generous on losses now? Right now, we can't be more generous because we don't want you to build up losses and manipulate the system.

To explain the treatment of losses right now: you might invest in a few different companies. Some do well, some do badly. We let you offset the losses you make on the companies doing badly against gains you make. So if you bought two companies for £1 million each, one went to zero and one went to £2 million, you're flat. You made no profit, so we don't tax you because you can offset the loss against the gain.

Suppose you only invested in one company and it lost money. Now you've lost that £1 million. We could let you offset it against the income you're getting, because you are worse off. Why don't we do that now? Because suppose you made the £1 million loss and £1 million gain, and you didn't sell the company with the gain—you sold the company with the loss, declared it, took it off your income tax, paid much less income tax for a few years, and then left the country and sold the company with the gain. We couldn't do anything to deal with that right now. Or you hang on to the ones that do well until you die, so there's no capital gains tax, and all the losses you harvested were used to offset your current tax bill.

The choices we make now mean we end up being overly punitive on losses. Other countries, like the US, let you offset your losses against income tax. Why can they do that? Because you can't leave the country without settling up your capital gains tax bill.

[21:41] Ravi Gurumurthy: This CGT change is part of a wider principle about trying to make sure that you pay the same amount of money wherever you earn it. There's a similar principle you could apply to National Insurance contributions (NICs), couldn't you? Some people end up paying National Insurance and some people don't, or pay lower rates. Give us the examples, and then what could we actually do about it?

[22:03] Arun Advani: A couple of really striking examples: one that I always find most obvious is someone working full-time on the minimum wage is paying income tax and National Insurance contributions. They then go home and pay rent to a landlord, and the landlord pays income tax and no National Insurance. It's hard to rationalize why you would want landlords to have a lower effective rate on that level of income.

Another one that is very odd: if you're a solicitor or an accountant working in a partnership as a junior employee, you pay income tax and NICs on your income. One day you get promoted to partner—that's a big success, you get a pay increase, and the NICs disappear! You're only now taxed on income tax; there are no NICs on your partnership income. It's a very odd situation that exists for historical reasons when National Insurance contributions weren't as large, so it didn't make as much quantitative difference. Now it really matters, and it changes how companies structure themselves. More companies are structured as partnerships specifically to make use of the fact that they can make lots of people partners and not pay NICs.

[23:03] Ravi Gurumurthy: And are partnerships less productive forms of economic organization?

[23:05] Arun Advani: One problem with a partnership structure is you don't have limited liability in quite the same way. You can have limited liability partnerships, but lots of them are general partnerships without that. Limited liability is important because I can afford to take risks in a world where I'm not going to have someone come after my house when things go wrong.

The other thing is that partnerships find it harder to own assets in a corporate way—effectively, the partners themselves have to own the assets, and that becomes quite messy in how they invest. So it does actually make it harder for them to invest in practice.

Because HMRC can see this is a bit of an issue where people set up partnerships that are really companies using a partnership structure, there's a whole load of anti-avoidance provision. We have reams and reams of legislation trying to stop people setting up fake partnerships, creating complexity for everybody.

[23:59] Ravi Gurumurthy: How much would that raise then, if you extended NICs?

[24:01] Arun Advani: Extending NICs to partnerships raises a couple of billion pounds—again, not life-changing, but definitely useful.

Going back to the landlords example: if you were to have the equivalent of National Insurance contributions on landlords, you raise something around £3 to £5 billion from doing that change. Doing the two of them annually gives you roughly £5 billion.

[24:50] Ravi Gurumurthy: So we're up to £17 billion in your package if you do both changes. Let's talk about simplification. One of the things that people always complain about is cliff edges. If you started to spend a bit of money on reducing or tapering those cliff edges, how would you do it?

[25:08] Arun Advani: One of the most egregious cliff edges that people end up talking a lot about is the cliff edge at £100,000. Now, that's a pretty high level of income; most people are not anywhere near that. But as a principle, it's a bit odd.

What happens right now is if you have young children who are pre-school age (nine months to just under school age), you can get free hours of childcare—the government gives you 15 to 30 hours of free childcare. That depends on your income staying below £100,000. If you go just over £100,000, suddenly you lose all of that free childcare.

Free childcare, especially in London and the Southeast, is very expensive. But across the whole country, losing the entire value of your childcare for going a pound over £100,000 leaves you definitely worse off, and it's kind of crazy. It's increasingly a problem compared to 5 years ago because the expansion of free childcare hours means more of your kids might be covered at any point in time, so the loss for going a pound over £100,000 is much larger. Childcare has also become more expensive. With all of those things happening, that cliff edge at £100,000 has stayed at £100,000 since 2017, so more and more people are affected by it.

What could you do? In the simplest world, if all you want to do is make your life easy, you could just move where that cliff is. That would still be a cliff edge, but it would affect fewer people. Better would be to design a taper that would taper away the value of the benefit. So rather than losing all of my free hours for going a pound over, you lose a bit of the hours as your income rises.

It is a hard policy to administer, it's worth saying, because the value of a certain number of hours of childcare depends on where you are in the country and how many children you have. Someone living in the Northeast with one child under five is going to have a much lower benefit than somebody of the same age and income who lives in London and has two or three kids under five. So you do have to design a taper that isn't crazy and accounts for those different circumstances. We've got some work on this coming out in a few weeks, but you can design things that deal with this problem, and whichever solution you design will be less egregious than the current cliff edge.

[27:42] Ravi Gurumurthy: One of the challenges the government faces is that the manifesto stops it raising taxes on about three-quarters of tax revenue because of commitments on VAT, National Insurance, and income tax. One possibility, though, is reforming property taxes and getting rid of council tax and stamp duty, replacing it, for instance, with a proportionate property tax or land value tax. Can you talk us through what proposals you think would work, and particularly what is viable within this time period, because some of these things are incredibly complicated to administer and value houses or land?

[28:22] Arun Advani: There's a few different things to think about in any kind of reform in this space. First, for people who don't know, we haven't revalued council tax since 1991. That is just embarrassing and ridiculous. Nobody thinks it makes sense. I mean, the most amazing part of it is that when we build new houses, we have to try and guess what we think this house would have been worth in 1991, which is just a stupid question! Why don't we just look at what it's worth now?

Revaluing that stuff is totally sensible. Some houses have increased in value more than others, so it's completely logical. That kind of process typically takes a couple of years. There are already plans to do something of a revaluation because at the last budget, Rachel Reeves introduced this high-value council tax surcharge, which works a bit differently to council tax, but in the end is trying to value properties. You could just extend that process to all properties. Wales did their revaluation about 10 years ago and it took them a couple of years, so you could do something like that.

A different question is: do you want to move the tax base from the value of the property to just the value of the land underneath the property? Just administratively, that's a bit of a pain. We see lots of sales of houses, so you can probably guess the value of your house based on what neighbors sold for recently. What's the value of the land under my house? I have no idea what the value of that bit is separately. You can estimate these things, but it feels less attached to what people know.

I think there's a different important thing we need to think about in the context of reforms to council tax. Reforming council tax in the sense that you reband people's houses is straightforward. Then there's a question of whether you want to move to what is overall more like a proportional tax on property.

At the moment, what we do is we have taxes that vary by your local area. Houses in lower-wealth areas of the country—roughly the North and West—have higher effective tax rates than areas in the Southeast. The reason for that is because if you're in Westminster, you need to pay for social care and all these other things, and if you're in Burnley, you need to pay for those same things. But in Westminster, houses are worth a lot, so you don't need to have a very high rate on those houses to raise the money you need. Westminster also gets loads of parking tax revenue!

Whereas in Burnley, you don't have that benefit, so you have to have a higher effective rate. If you moved to a purely proportional system, you would raise much more money in the Southeast and much less around the rest of the country. Then you have the question: how do you keep councils afloat in the rest of the country when you've massively reduced their tax base?

There's an easy answer to that: we already have the Barnett formula that covers transfers across the four nations. You could just build a "super Barnett formula" that deals with all 300-ish local authorities and makes transfers between them. But the problem for the current government is that at the same time, it's talking a lot about devolution. Meaningful devolution requires local authorities to have some control over what they're doing and what their tax base looks like. When you tell them to plan for the future, but they know that after every election a new government might change what all the transfers are, it's very hard for councils to be sure what their budget looks like over the next 10 years. That's why you want councils to have some control over their own tax base.

Property is classically the tax base that won't move anywhere, so it's the thing you want to give them control over. But if you give them control over their tax base, because that wealth is worth less in the North, they are going to always need a higher rate. That's just a tension—there's no right answer that economics tells you. It's a trade-off: do you really believe deeply in devolution and want councils to have control (in which case you'll have higher rates where wealth is lower), or do you want a system that's purely proportional (in which case you make transfers)?

At the point you say you want a tax that's purely proportional based on property values, there's a different question: why are we taxing property wealth specifically rather than other sources of wealth? For the middle classes, property wealth is the wealth they have alongside their pensions. For the wealthiest people, their wealth is actually not mostly in property or pensions—it's in private businesses. Is there a reason you want to tax property wealth more than those other things? There are some answers: property wealth is special, you can see it, it's the local tax base, or maybe you get a consumption benefit from living in your house without paying VAT. But depending on where you land, you end up in different places on how you want the reform to look.

The upshot for a current government is there's a tension between devolution and what a purely proportional system looks like. At the same time, it's going to take a couple of years to do any version of this if you're doing a revaluation, so you shouldn't think of the impacts being this side of an election. Politically, the other difficulty with a purely proportional system is that voters in the Southeast are going to pay more tax, creating a whole area of the country where people are less willing to vote for your party.

[00:0Lev3:26] Ravi Gurumurthy: It feels like if you're going to do that, you do it in the first two years of an administration, not in the last year when people will be really aware of it just before an election.

I want to come on to the devolution question, because fiscal devolution is something Andy Burnham has talked about. On this point about fiscal transfers, isn't it always going to be the case that we have fiscal transfers in our system given our level of inequality, adding a source of unpredictability for councils?

[35:00] Arun Advani: It really depends on what share of the council budget is the unpredictable bit relying on government formulas versus the local tax base. In a world where you can get quite a long way on your local tax base and transfers are a more marginal thing, that's very different from a world where most funding comes from transfers.

The other thing we could do from a devolution perspective is actually undo something we've devolved. Part of the point of devolution is that local councils are responsive to their constituents and can do things people locally want. But then we impose on them really large obligations—like social care—that are government-mandated, where they have no control over delivery. Local government might be better at knowing priorities between parks and libraries or which roads to resurface, but the funding for social care doesn't need to come from them. Moving social care into an NHS-type system and taking it away from councils would free up their budgets for things they are actually better at delivering.

[36:38] Ravi Gurumurthy: I think this is really important because we need to decide what local government is for. If you think about four things local government does: safer/cleaner/greener local environment, economic development, transport, and skills. Safer, cleaner, greener is how you win and lose local elections, so there's an incentive to pay for that. Local transport is a very local good. But on children's and adult social services, you're never going to get an electoral penalty in the same way, so the natural incentive is to underfund it. There's a good argument for it being nationally funded, even if locally run.

If you went down that road of saying the areas for maximum discretion are safer/cleaner/greener and economic development, what would be the tax base that allows them to make those choices?

[38:00] Arun Advani: If property tax stays devolved, giving them more flexibility would help. Currently, it's "fake devolved" because we constrain the ratio between Band A, Band B, Band C, etc., so you can't choose how progressive or non-progressive you are locally. You also can't change rates quickly because of caps. If local government is accountable to local people, let them make those choices. If they increase rates too fast, people can vote them out.

I would give them much more flexibility in how they tax local properties. If we reduce the number of things they need to fund, their current revenue base is sufficient. Part of the reason councils are going bust now is that so much of the budget is spent on social care rather than local priorities or skills where a combined authority level has a better sense of what local industry needs.

[39:28] Ravi Gurumurthy: As we record this, it's being mooted that income tax might start with earmarking a certain amount to a given area. That strikes me as a potential stepping stone, but I struggle to see how it's useful. If existing grants are just relabeled as your portion of income tax, and income tax goes up, great; but if it goes down, are we really going to cut grants? We'll end up doing equalization again to deal with the Burnley vs. Westminster challenge.

[40:15] Arun Advani: I'm not very pro-devolution on something like income tax because it creates strong incentives for people to pick areas to live in based on lower tax rates while commuting elsewhere. You see that in other countries with devolved income tax—you get a lot of migration, underfunding the city centers while the commuter belts get rich income tax revenue. That's not what we're trying to achieve, so income tax is not the base I would choose to devolve.

My colleague Tim Leunig argues we should devolve more "regressive" or specific taxes like VAT, cigarettes, or alcohol. What do you think about that? In practice, people already travel across the channel for cheaper alcohol and cigarettes; if it's just driving to the next local authority, you'd get a lot of car trips to the nearest town for a tax break. You can't enforce something like that easily.

[41:21] Ravi Gurumurthy: What about business rates?

[41:28] Arun Advani: Government hasn't decided what it wants to do with business rates. While talking about devolving them, they're making rules about which businesses we like or don't like. There's a reasonable case for saying vape shops or gambling shops have negative externalities. Some economists argue taxes on land end up falling on the landlord, but in practice, if a landlord can rent to a newsagent or a vape shop, a higher tax on the vape shop creates a wedge that can discourage certain shops. You could allow local authorities to make those choices.

[42:59] Ravi Gurumurthy: Do other countries give local governments general freedom to establish any tax they want over and above national taxes?

[43:14] Arun Advani: There's an information and administrative problem. Even if you give local governments power to tax income, you don't want them administering a separate income tax system. You need central government to collect information. In the US, local and state governments add surcharges or local rates on top of income tax.

Starting from where the UK is as a centralized system, I'd start with property. Two other taxes I would devolve are tourist/hotel taxes and cordon/congestion charges for transport in and out of cities. Dense cities need to fund bus networks. Being able to say, "We're going to charge a fee to drive in, like the London congestion charge, but plow all revenue back into the bus network," is a sustainable model.

[44:46] Ravi Gurumurthy: To wind up, last couple of questions. If you could look abroad and steal anything from someone else's tax system, what would you import?

[44:57] Arun Advani: If you design capital gains tax right, it can be much better for growth and raise revenue. The most complex part is dealing with people leaving the country—settling up. The closest good example is Canada. Canada's version of capital gains tax does a fairly effective job of saying: if you leave for a bit and come back, you stay in the tax system; if you leave for a long time, you don't have to pay all the money on day one, but over a period we work out how to get the money owed. Because Canada shares a legal tradition with the UK, a lot of that could be imported fairly straightforwardly.

The other thing I would import is in inheritance tax. Currently, we tax inheritances based on the estate of the person who died, regardless of who receives it. If I have 10 children and you have one, the tax paid is the same despite my children receiving much less each. Ireland has a Capital Acquisitions Tax, which instead taxes the recipient based on how much they receive over their lifetime. That's a better design. It also avoids our problem where people rush to give away assets 7 to 10 years before dying, because under a recipient tax, you are taxed whenever you receive the money.

[47:14] Ravi Gurumurthy: Final question: What would you do if you were Andy Burnham in your first 100 days to signal the kind of tax reform you want?

[47:27] Arun Advani: There are many reforms you could think of for growth: sort out the VAT system, equalize how we tax income from different sources (work, partnership, self-employed), or reform property taxes. I would pick one of those areas and do comprehensive reform. If it's property taxes, take stamp duty, business rates, and council tax together into one coherent fix. If it's income from work, focus on all the ways we tax work.

Pick one naughty problem rather than salami-slicing a little bit everywhere. That allows you to tell a story at a budget that you are fixing an underlying problem, rather than just grabbing money because headroom went down.

[48:53] Ravi Gurumurthy: I'd also like "fiscal forward guidance," where a government telegraphs what it will cut or raise if fiscal headroom changes, or provides a roadmap for multi-year tax changes.

[49:28] Arun Advani: Absolutely. When doing reforms—like putting full NICs on partners—doing it all at once might be too big a shock. You can telegraph a multi-year roadmap: year one you do this much, year two this much, and by year three it's equalized. In 2010, the coalition government published a corporate tax roadmap showing what would change over time. Doing that gives certainty and plots a clear path to a better system.

[50:32] Ravi Gurumurthy: Arun Advani, thank you very much for joining us.

[50:36] Arun Advani: Thanks very much. Cheers.

[50:41] Ravi Gurumurthy: If you enjoyed this episode, please do like, share, and subscribe wherever you get your podcasts. Nesta is a research and innovation foundation based in the UK. We design, test, and scale solutions to the big challenges of our time. We're funded by a charity endowment and are politically neutral. For more, visit Nesta.

[00:00] Arun Advani: For people who don't know, we haven't revalued council taxes since 1991. That is just embarrassing and ridiculous. Nobody thinks it makes sense. I mean, the most amazing part of it is that when we build new houses, we have to try and guess what we think this house would have been worth in 1991, which is just a stupid question.

[00:17] Ravi Gurumurthy: Hi, I'm Ravi Gurumurthy, CEO of Nesta, and we've got Arun Advani here in the latest in our podcast series, What Can Andy Burnham Do. We're going to be focusing today on the big question of taxation: Should we raise capital gains tax? Should we change the way we tax property? What can we do to tackle all the cliff edges that exist that create weird distortions? We're also going to be talking about what it means for devolution—what would a tax system that embraced devolution look like? Arun, welcome. Thank you for joining us.

[00:48] Arun Advani: Oh, thanks for having me.

[00:52] Ravi Gurumurthy: So we're going to talk all things tax and what Andy Burnham should do. Let's start with one of the most often suggested taxes, which is a wealth tax. The Greens in particular have suggested it. I don't think you think that there is a viable wealth tax that a government could do that's either desirable or feasible, but just outline what are the typical options that people talk about and why are they not possible?

[01:09] Arun Advani: So when you talk about a wealth tax, a wealth tax is saying: let's add up the value of all of the stuff that somebody owns—their house (if they own it jointly with a spouse, half the value of the house), take off the mortgage, the value of their pension, value of any money they have in a bank, if they have any stocks and shares, add up all of those things and then draw a line somewhere and say if their wealth is above that line, then on the bit above that line, let's tax that. So, for example, a policy that's been floating around for a little while is some people calling for a 2% tax above £10 million. They're saying if you have £11 million, you're paying 2% on just that top million.

If you look around the world at what wealth taxes of this sort look like, they've tended to be a relatively low threshold. So a threshold in UK terms is like £100,000, £500,000—those sorts of thresholds. That covers a very large share of the population. The impact of the choices that other countries have made when they have these low-threshold taxes is that because there's so many people who are covered by the tax, you have to keep the valuation of all these assets simple, right? You know roughly the value of your house; you don't know exactly the value. You know roughly the value you could get information on the value of your pension, but these things eventually add up to being complex.

Particularly what's complex is the value of a business. So if you're a plumber and you're running a small business, or you're a hairdresser running a small business, you have to work out: what's that business actually worth if you sold it to somebody else? Often not very much, but you still have to work it out. And so to solve the complexity that comes in here, countries have taken shortcuts in how they value things. So they say, okay, we'll value the business of a plumber at just what's the value of the tools that he or she has and the van and those things. For a plumber, that's basically right—in a world where they sell their business, if they're the one person running the business, once they sell it, all it's worth to somebody else is just the stuff.

The problem is that when you then use that same method for valuing the business someone very wealthy has, that's nowhere near the value of the business. Because it's not the computers and the tables and the chairs that make a big business valuable—it's often the know-how, the client lists, the special IP that they've built up. That's really where the value is. Taking these shortcuts means that wealth taxes aren't very effective when they're designed in that way for attacking the very wealthiest.

The alternative that at the Wealth Tax Commission—which I ran with Andy Summers and Emma Chamberlain—we talked about was: you could instead, if you wanted, have a wealth tax starting at a much higher threshold, say £10 million. Then you have a relatively small share of the population who you're covering, and you could then afford to do very high-quality market valuation. It's not easy; it takes some work, but you can do it. No one should pretend it's going to be simple. No one should pretend that you're not going to have some avoidance behaviors. We estimated that maybe 17% of the expected revenue you'd get if nobody changed their behavior would be lost due to changes in behavior for a 1% tax. So that's the way to think about it: you can definitely deliver it, but it will have some costs.

It will also take time. It takes a few years to build a new tax like this. We don't currently know who owns what shares, what properties, what businesses in an effective way, so you couldn't implement it immediately.

[04:04] Ravi Gurumurthy: Just because the actual valuation process would take time?

[04:06] Arun Advani: Valuation takes time. You literally just need computer systems to do it and forms to do it. Those things seem really boring and process-oriented, but they are a barrier. Particularly at the moment, in these discussions going around about whether Andy Burnham should create a wealth tax now: he's not going to have a wealth tax that gives him any money this side of an election. So if he really wants to try and do the work for it, he could, but he shouldn't expect any revenue from it this side of an election.

That's just a trade-off: do you want to burn political capital on that when there are other taxes on wealth that we have, like inheritance tax and capital gains tax, that could be improved? Particularly capital gains tax—there's lots of scope for improvement there.

[04:48] Ravi Gurumurthy: And in terms of the amount you'd actually raise if you went for this tax on assets above £10 million, what kind of estimate do you come up with?

[04:55] Arun Advani: So the Wealth Tax Commission, when we were running back in 2020, we said a 1% tax on assets above £10 million would be taxing at the time about 22,000 people, but it would raise about £10 billion. So you're not going to pay for the whole NHS budget on that.

[05:09] Ravi Gurumurthy: Is that an annual tax?

[05:10] Arun Advani: That's recurring at £10 billion a year. You should think about that as like 1.5p on standard rates of income tax. Now wealth has increased a little bit, so we think maybe it's about £12 billion. The ONS version of those data declined in quality, so it's a bit harder to know now exactly what the number is, but we're talking £10 to £12 billion after accounting for behavioral responses every year.

[05:35] Ravi Gurumurthy: And those behavioral responses—how did you get a handle on how big or small those are? That must be pretty difficult to estimate.

[05:40] Arun Advani: Yeah, so we couldn't estimate it directly for the UK because we don't have a tax like this in the UK. What you can do is look at all the things that people can do and then look at the scale of those responses in other countries where these things have happened.

It's worth saying that the estimate of losing about 17% of the revenue was even after you do a lot of fixes. In other countries, often the loss is even larger than that because they've made bad choices. One bad choice is to not value some asset classes. So in some countries, they say main homes are just exempt from the wealth tax. The problem with that is, of course, people spend more money on their main homes than other assets, right? Or you say, let's take a shortcut and value private businesses at a discount—work out the value of the private business and tax half that value. Well, suddenly my private business turns out to own a lot of my stuff because having it inside the business gives me a 50% discount. So I put all of my stuff inside the business instead of owning it personally.

In Switzerland, the tax has been designed in a way that's very local by Canton, which is actually quite small, and so people move Cantons because just moving across the border can reduce your tax. When those changes that you need to make in your life are quite small relative to the tax impact, you see quite a lot of impacts there.

In the best-case scenario where you don't make those bad choices, there are things you still can't stop people doing. You still won't stop the fact that if I have £12 million and you put a tax on me above £10 million, I'm suddenly going to give some more money to my kids. I was going to give it to them one day; I give it to them a bit sooner, and suddenly we're all below £10 million. You can't do much about that; that's a totally legitimate thing. I genuinely don't have that money if I've really given it to my adult children. But at that point, clearly you're not getting revenue from me, and so we have to account for those sorts of behaviors.

[07:25] Ravi Gurumurthy: Okay, let's talk then about something else that has been dubbed a wealth tax—I think by Wes Streeting—which is about CGT (Capital Gains Tax). You mentioned the idea of indexing the gains, equalizing the rate between capital gains tax and income tax, and then to avoid the behavioral responses you mentioned, your proposal is to have a death tax and an exit tax so you can't basically leave the country or hoard your assets until you die. Now, you basically think that's a good idea.

I want to start by unpicking the effects for different people and what this does. There are three different segments, I think. One is that you've got people who get taxed unfairly because their asset has only grown with inflation, yet they're getting taxed on a nominal increase—and your proposal would fix that; you'd pay nothing in that situation.

Then you've got the problem of the solicitor who becomes a partner in a firm, whose income is being taxed more as a capital gain because people can shift things around, and therefore they're being undertaxed—they're not being taxed at the right rate when it's basically income.

[08:46] Arun Advani: Just to give a practical example of what you see with that kind of behavior: the classic example is in management consultancy or IT consultancy. You work for a big company for a few years, get your skills, then go and set up a sole trading incorporated company as a consultant. You get paid into the business, keep the money mostly in the business, pay yourself a small salary, pay some dividends every year because they're taxed at a lower rate than salary would be, and then anything you don't need immediately, you keep stored up inside the business. At some point, you can liquidate the business and get that money out as a capital gain. That's basically work you were doing that should have been taxed at income tax rates, but you're getting a much lower capital gains tax rate now.

[09:25] Ravi Gurumurthy: And no substantial investment was going into that business, exactly. But then there's a third thing, which is let's say you set up your business, you're an entrepreneur, and you'd then be paying CGT at 45% or whatever—that would be quite a big shift. The concern I would have is: does that not deter entrepreneurial activity? For your first two groups, great, you've solved those two problems. But how do you think about how to avoid deterring entrepreneurship in your change?

[09:54] Arun Advani: A few different things to think about there. One is actually our best evidence on what genuine entrepreneurs know when they're setting up businesses is that they basically don't know much about what capital gains tax rates look like. There's some good qualitative work that was done by HMRC many years ago where they were looking at reliefs that they were giving entrepreneurs and trying to work out: did these entrepreneurs know anything about these reliefs? They didn't know any of the stuff existed; they had no idea what the rate was.

Many of these people are basically thinking, "I've got my brilliant idea, I really believe in it, I want to make it happen. Obviously I want to make a profit doing it, but I'm not really aware of what the tax status looks like when I exit the business." Right now, they want to build a business and make it successful, and for the next 10 years, they see themselves running a business and making money through the business. They're not thinking as much about the end state. Obviously, at the point that you've made a big business, you care and want that rate to be as low as possible, but it's not a big part of what people are thinking at the start. So that's the first thing.

[10:49] Ravi Gurumurthy: Although if it was a 45% CGT rate—higher than any other country—presumably it might deter, if people know it exists.

[10:57] Arun Advani: Only if you know it exists and how it's taxed, but yes, absolutely.

Second thing: to the extent you're putting capital at risk, you're going to be getting this indexation allowance—this investment allowance—that's going to reduce your tax rate. If there's basically no capital in the business and what you're saying is, "Well, I'm just working really, really hard," well, yes, lots of people are working really, really hard. That's what income tax rates are for. Now, we might say income tax rates are too high or too low, but ultimately that's a different debate about what the income tax rate should be. If you're working really hard to build your business, you're working just like lots of other people are working, and it should be taxed like it's work.

If you're making a profit on that business beyond just the returns you're getting on the capital, it's either because you're incredibly hardworking and skillful, or it's luck. If it's luck, there's not a really compelling case to give you a lower tax rate. Or it's that you have some market power or monopoly power—your business has cornered something that means you'll be able to make profits beyond what everyone else can make. Again, in that case, there's not really a reason why we want to be giving you a lower tax rate.

Third thing: if you really are still worried as a policymaker and you say, "I hear all that, but I'm still really worried about this," despite the lack of evidence... and I think the answer genuinely is a lack of evidence; we just don't know a lot about what this effect looks like in practice because these kind of people are few and far between in terms of setting up big businesses. If you're really concerned about that, what you should do is take a bad relief that we have—literally called BADR (Business Asset Disposal Relief)—which currently says on the first million pounds that you get of a private business you run, you get a special lower tax rate. That tax rate used to be 10%, has now gone up to 18%. You get an 18% rate on the first million pounds, and then you pay the standard rate on everything else.

Now, if you have a business worth £2 million, getting half of the value of that business at a low rate is quite nice. But if you're really building the next unicorn and it's going to be worth a billion pounds, do you really care that you get a low rate on the first million pounds? Not at all, right? It's not having any impact on the tax you're paying.

So what you do is take that bad relief and flip it on its head. Say actually, we won't give you any relief if you're building a small business. But if you build a business that's above some huge size—really valuable, contributing to overall employment—then we'll give you a relief. If you build a company worth more than £100 million and then sell some share of it, we'll give you a special relief that gives you a low rate on the whole value. Flipping a bad relief on its head gives you relief on an unlimited amount, but only for a very narrow slice of people who are really building these giant companies. That would be a way to protect specifically that group if you're worried about it.

[13:52] Ravi Gurumurthy: Okay, so in some ways, the cumulative impact of both those changes would be: stay invested, build something big.

[13:58] Arun Advani: Yeah. If you're building something big and you're staying here and building it really big, then you're going to get a relief. If you build something that's small, then great, you did it, but there's no big wider impact for the UK that we should be giving you a lower rate for.

[14:09] Ravi Gurumurthy: What about the other slight risk? When Nigel Lawson did this in 1988, it was in the context of a tax-cutting budget where top rates were coming down. If you did it now on top of all the other tax changes that happened over the last two years, is there a slight risk that from a vibe perspective, it just feels like taxing wealth creation even more?

[14:36] Arun Advani: Yeah, so the best time to do this would have been yesterday, paired with other things. Even this capital gains tax reform—by giving an investment allowance—is giving you a lower tax rate if you're putting actual capital at risk, because it's bringing down the cost of capital. That is encouraging investment.

More than half of current capital gains taxpayers are better off or no worse off under this reform. The people who are paying more are the ones where there's very little capital in the business, mostly the example of the consultants and that sort of person. That's where the revenue is coming from.

It also raises quite a lot of money. If you do this overall package of reform, you get something like £12 billion a year, and you could use some of that money towards other things that are pro-growth. That could be giving money to the British Business Bank to go and do investments, or increasing the size of existing reliefs. You can think about the best way to support growth with that, which might send the signal that what we've got now isn't a capital gains tax policy that's pro-growth—it's one that's pro-shifting income between different sources. What we want is one that actually compensates you for capital at risk.

By giving an investment allowance that protects that capital and says you can make a certain return on that that is at least what you would have got by just sticking it in a bank, and we're only going to tax you if you make more profit than that—that's actually good for growth. You could spend the money on other things that would also support growth and pair well with a pro-growth budget.

[16:23] Ravi Gurumurthy: And the £12 billion figure—is that a material amount that you could raise? I've heard Treasury internal estimates on this change being lower than that. Do you know why that might be? Are they assuming some behavioral response that you've not accounted for?

[16:40] Arun Advani: It's really about what package of reforms you have. The number that people throw around a lot is that you lose money on a reform that just increases the tax rate. I think that's probably right; I would not start from here and say let's just increase the tax rate. That is definitely bad for growth, bad vibes, and highly likely to cost money because people are going to respond in ways that don't support more revenue coming from capital gains tax.

But the point is a package reform that says: look, we're going to actually reduce the cost of capital for people and reduce the tax rate for more than half of current capital gains taxpayers. The ones who are going to pay more are the ones not putting capital at risk. That is a way you can actually raise money.

Costing just a tax increase plausibly gets you to lower tax revenue. Costing it with these other things actually reduces some of the ways in which people avoid the tax now. Currently, there's a very strong incentive for someone in their 70s or 80s who has owned assets for a long time: if I sell my shares to try and get something more exciting, I have to pay a big capital gains tax upfront. It's not worth it; I'm probably better off having the capital gains tax written off when I die.

Or if I've built a huge fintech firm worth many billions of pounds, and I want to build my next business here, if I sell my company while I'm here, I'm going to have to pay a 24% rate right now. I could just move to Dubai and there's no charge on settling up. I've used UK resources and all the benefits of the UK to build this, and the UK is saying, "We're going to tax you, except if you leave. If you promise to leave, we'll give you a 0% rate as long as you promise not to come back for six years." That's what we do right now. Obviously, if you say that, lots of people take that incentive seriously and choose to leave.

If you shut down those kinds of margins and say, "We're not trying to tax you for exiting—you're welcome to exit, but you will have to settle up the tax bill you owe on the gains you've built up while you were here." It's sensible that we allow you to defer the tax until the point where you realize and get the money from those gains, but if you're leaving the tax system, it makes sense to say you do have to settle up and pay your restaurant bill for having been here. Other countries do that—we're one of only two countries in the G7 that don't.

[19:08] Ravi Gurumurthy: Now that you've raised £12 billion from this change, what would be the most pro-growth way of spending some or all of that £12 billion on tax changes?

[19:22] Arun Advani: In some sense, it's not even obvious that tax changes are what you want to do, right? One answer is that there are lots of places where we know there's a return to spending, like the state of transport in this country or potholes—things people complain about. There are real benefits to spending on some of those things.

Within the space of tax, I think the CGT reform overall is a good reform even before you think about how you spend that money. The fifth limb of that reform is being more generous on losses. People who invest might make losses. Why can't we be more generous on losses now? Right now, we can't be more generous because we don't want you to build up losses and manipulate the system.

To explain the treatment of losses right now: you might invest in a few different companies. Some do well, some do badly. We let you offset the losses you make on the companies doing badly against gains you make. So if you bought two companies for £1 million each, one went to zero and one went to £2 million, you're flat. You made no profit, so we don't tax you because you can offset the loss against the gain.

Suppose you only invested in one company and it lost money. Now you've lost that £1 million. We could let you offset it against the income you're getting, because you are worse off. Why don't we do that now? Because suppose you made the £1 million loss and £1 million gain, and you didn't sell the company with the gain—you sold the company with the loss, declared it, took it off your income tax, paid much less income tax for a few years, and then left the country and sold the company with the gain. We couldn't do anything to deal with that right now. Or you hang on to the ones that do well until you die, so there's no capital gains tax, and all the losses you harvested were used to offset your current tax bill.

The choices we make now mean we end up being overly punitive on losses. Other countries, like the US, let you offset your losses against income tax. Why can they do that? Because you can't leave the country without settling up your capital gains tax bill.

[21:41] Ravi Gurumurthy: This CGT change is part of a wider principle about trying to make sure that you pay the same amount of money wherever you earn it. There's a similar principle you could apply to National Insurance contributions (NICs), couldn't you? Some people end up paying National Insurance and some people don't, or pay lower rates. Give us the examples, and then what could we actually do about it?

[22:03] Arun Advani: A couple of really striking examples: one that I always find most obvious is someone working full-time on the minimum wage is paying income tax and National Insurance contributions. They then go home and pay rent to a landlord, and the landlord pays income tax and no National Insurance. It's hard to rationalize why you would want landlords to have a lower effective rate on that level of income.

Another one that is very odd: if you're a solicitor or an accountant working in a partnership as a junior employee, you pay income tax and NICs on your income. One day you get promoted to partner—that's a big success, you get a pay increase, and the NICs disappear! You're only now taxed on income tax; there are no NICs on your partnership income. It's a very odd situation that exists for historical reasons when National Insurance contributions weren't as large, so it didn't make as much quantitative difference. Now it really matters, and it changes how companies structure themselves. More companies are structured as partnerships specifically to make use of the fact that they can make lots of people partners and not pay NICs.

[23:03] Ravi Gurumurthy: And are partnerships less productive forms of economic organization?

[23:05] Arun Advani: One problem with a partnership structure is you don't have limited liability in quite the same way. You can have limited liability partnerships, but lots of them are general partnerships without that. Limited liability is important because I can afford to take risks in a world where I'm not going to have someone come after my house when things go wrong.

The other thing is that partnerships find it harder to own assets in a corporate way—effectively, the partners themselves have to own the assets, and that becomes quite messy in how they invest. So it does actually make it harder for them to invest in practice.

Because HMRC can see this is a bit of an issue where people set up partnerships that are really companies using a partnership structure, there's a whole load of anti-avoidance provision. We have reams and reams of legislation trying to stop people setting up fake partnerships, creating complexity for everybody.

[23:59] Ravi Gurumurthy: How much would that raise then, if you extended NICs?

[24:01] Arun Advani: Extending NICs to partnerships raises a couple of billion pounds—again, not life-changing, but definitely useful.

Going back to the landlords example: if you were to have the equivalent of National Insurance contributions on landlords, you raise something around £3 to £5 billion from doing that change. Doing the two of them annually gives you roughly £5 billion.

[24:50] Ravi Gurumurthy: So we're up to £17 billion in your package if you do both changes. Let's talk about simplification. One of the things that people always complain about is cliff edges. If you started to spend a bit of money on reducing or tapering those cliff edges, how would you do it?

[25:08] Arun Advani: One of the most egregious cliff edges that people end up talking a lot about is the cliff edge at £100,000. Now, that's a pretty high level of income; most people are not anywhere near that. But as a principle, it's a bit odd.

What happens right now is if you have young children who are pre-school age (nine months to just under school age), you can get free hours of childcare—the government gives you 15 to 30 hours of free childcare. That depends on your income staying below £100,000. If you go just over £100,000, suddenly you lose all of that free childcare.

Free childcare, especially in London and the Southeast, is very expensive. But across the whole country, losing the entire value of your childcare for going a pound over £100,000 leaves you definitely worse off, and it's kind of crazy. It's increasingly a problem compared to 5 years ago because the expansion of free childcare hours means more of your kids might be covered at any point in time, so the loss for going a pound over £100,000 is much larger. Childcare has also become more expensive. With all of those things happening, that cliff edge at £100,000 has stayed at £100,000 since 2017, so more and more people are affected by it.

What could you do? In the simplest world, if all you want to do is make your life easy, you could just move where that cliff is. That would still be a cliff edge, but it would affect fewer people. Better would be to design a taper that would taper away the value of the benefit. So rather than losing all of my free hours for going a pound over, you lose a bit of the hours as your income rises.

It is a hard policy to administer, it's worth saying, because the value of a certain number of hours of childcare depends on where you are in the country and how many children you have. Someone living in the Northeast with one child under five is going to have a much lower benefit than somebody of the same age and income who lives in London and has two or three kids under five. So you do have to design a taper that isn't crazy and accounts for those different circumstances. We've got some work on this coming out in a few weeks, but you can design things that deal with this problem, and whichever solution you design will be less egregious than the current cliff edge.

[27:42] Ravi Gurumurthy: One of the challenges the government faces is that the manifesto stops it raising taxes on about three-quarters of tax revenue because of commitments on VAT, National Insurance, and income tax. One possibility, though, is reforming property taxes and getting rid of council tax and stamp duty, replacing it, for instance, with a proportionate property tax or land value tax. Can you talk us through what proposals you think would work, and particularly what is viable within this time period, because some of these things are incredibly complicated to administer and value houses or land?

[28:22] Arun Advani: There's a few different things to think about in any kind of reform in this space. First, for people who don't know, we haven't revalued council tax since 1991. That is just embarrassing and ridiculous. Nobody thinks it makes sense. I mean, the most amazing part of it is that when we build new houses, we have to try and guess what we think this house would have been worth in 1991, which is just a stupid question! Why don't we just look at what it's worth now?

Revaluing that stuff is totally sensible. Some houses have increased in value more than others, so it's completely logical. That kind of process typically takes a couple of years. There are already plans to do something of a revaluation because at the last budget, Rachel Reeves introduced this high-value council tax surcharge, which works a bit differently to council tax, but in the end is trying to value properties. You could just extend that process to all properties. Wales did their revaluation about 10 years ago and it took them a couple of years, so you could do something like that.

A different question is: do you want to move the tax base from the value of the property to just the value of the land underneath the property? Just administratively, that's a bit of a pain. We see lots of sales of houses, so you can probably guess the value of your house based on what neighbors sold for recently. What's the value of the land under my house? I have no idea what the value of that bit is separately. You can estimate these things, but it feels less attached to what people know.

I think there's a different important thing we need to think about in the context of reforms to council tax. Reforming council tax in the sense that you reband people's houses is straightforward. Then there's a question of whether you want to move to what is overall more like a proportional tax on property.

At the moment, what we do is we have taxes that vary by your local area. Houses in lower-wealth areas of the country—roughly the North and West—have higher effective tax rates than areas in the Southeast. The reason for that is because if you're in Westminster, you need to pay for social care and all these other things, and if you're in Burnley, you need to pay for those same things. But in Westminster, houses are worth a lot, so you don't need to have a very high rate on those houses to raise the money you need. Westminster also gets loads of parking tax revenue!

Whereas in Burnley, you don't have that benefit, so you have to have a higher effective rate. If you moved to a purely proportional system, you would raise much more money in the Southeast and much less around the rest of the country. Then you have the question: how do you keep councils afloat in the rest of the country when you've massively reduced their tax base?

There's an easy answer to that: we already have the Barnett formula that covers transfers across the four nations. You could just build a "super Barnett formula" that deals with all 300-ish local authorities and makes transfers between them. But the problem for the current government is that at the same time, it's talking a lot about devolution. Meaningful devolution requires local authorities to have some control over what they're doing and what their tax base looks like. When you tell them to plan for the future, but they know that after every election a new government might change what all the transfers are, it's very hard for councils to be sure what their budget looks like over the next 10 years. That's why you want councils to have some control over their own tax base.

Property is classically the tax base that won't move anywhere, so it's the thing you want to give them control over. But if you give them control over their tax base, because that wealth is worth less in the North, they are going to always need a higher rate. That's just a tension—there's no right answer that economics tells you. It's a trade-off: do you really believe deeply in devolution and want councils to have control (in which case you'll have higher rates where wealth is lower), or do you want a system that's purely proportional (in which case you make transfers)?

At the point you say you want a tax that's purely proportional based on property values, there's a different question: why are we taxing property wealth specifically rather than other sources of wealth? For the middle classes, property wealth is the wealth they have alongside their pensions. For the wealthiest people, their wealth is actually not mostly in property or pensions—it's in private businesses. Is there a reason you want to tax property wealth more than those other things? There are some answers: property wealth is special, you can see it, it's the local tax base, or maybe you get a consumption benefit from living in your house without paying VAT. But depending on where you land, you end up in different places on how you want the reform to look.

The upshot for a current government is there's a tension between devolution and what a purely proportional system looks like. At the same time, it's going to take a couple of years to do any version of this if you're doing a revaluation, so you shouldn't think of the impacts being this side of an election. Politically, the other difficulty with a purely proportional system is that voters in the Southeast are going to pay more tax, creating a whole area of the country where people are less willing to vote for your party.

[00:0Lev3:26] Ravi Gurumurthy: It feels like if you're going to do that, you do it in the first two years of an administration, not in the last year when people will be really aware of it just before an election.

I want to come on to the devolution question, because fiscal devolution is something Andy Burnham has talked about. On this point about fiscal transfers, isn't it always going to be the case that we have fiscal transfers in our system given our level of inequality, adding a source of unpredictability for councils?

[35:00] Arun Advani: It really depends on what share of the council budget is the unpredictable bit relying on government formulas versus the local tax base. In a world where you can get quite a long way on your local tax base and transfers are a more marginal thing, that's very different from a world where most funding comes from transfers.

The other thing we could do from a devolution perspective is actually undo something we've devolved. Part of the point of devolution is that local councils are responsive to their constituents and can do things people locally want. But then we impose on them really large obligations—like social care—that are government-mandated, where they have no control over delivery. Local government might be better at knowing priorities between parks and libraries or which roads to resurface, but the funding for social care doesn't need to come from them. Moving social care into an NHS-type system and taking it away from councils would free up their budgets for things they are actually better at delivering.

[36:38] Ravi Gurumurthy: I think this is really important because we need to decide what local government is for. If you think about four things local government does: safer/cleaner/greener local environment, economic development, transport, and skills. Safer, cleaner, greener is how you win and lose local elections, so there's an incentive to pay for that. Local transport is a very local good. But on children's and adult social services, you're never going to get an electoral penalty in the same way, so the natural incentive is to underfund it. There's a good argument for it being nationally funded, even if locally run.

If you went down that road of saying the areas for maximum discretion are safer/cleaner/greener and economic development, what would be the tax base that allows them to make those choices?

[38:00] Arun Advani: If property tax stays devolved, giving them more flexibility would help. Currently, it's "fake devolved" because we constrain the ratio between Band A, Band B, Band C, etc., so you can't choose how progressive or non-progressive you are locally. You also can't change rates quickly because of caps. If local government is accountable to local people, let them make those choices. If they increase rates too fast, people can vote them out.

I would give them much more flexibility in how they tax local properties. If we reduce the number of things they need to fund, their current revenue base is sufficient. Part of the reason councils are going bust now is that so much of the budget is spent on social care rather than local priorities or skills where a combined authority level has a better sense of what local industry needs.

[39:28] Ravi Gurumurthy: As we record this, it's being mooted that income tax might start with earmarking a certain amount to a given area. That strikes me as a potential stepping stone, but I struggle to see how it's useful. If existing grants are just relabeled as your portion of income tax, and income tax goes up, great; but if it goes down, are we really going to cut grants? We'll end up doing equalization again to deal with the Burnley vs. Westminster challenge.

[40:15] Arun Advani: I'm not very pro-devolution on something like income tax because it creates strong incentives for people to pick areas to live in based on lower tax rates while commuting elsewhere. You see that in other countries with devolved income tax—you get a lot of migration, underfunding the city centers while the commuter belts get rich income tax revenue. That's not what we're trying to achieve, so income tax is not the base I would choose to devolve.

My colleague Tim Leunig argues we should devolve more "regressive" or specific taxes like VAT, cigarettes, or alcohol. What do you think about that? In practice, people already travel across the channel for cheaper alcohol and cigarettes; if it's just driving to the next local authority, you'd get a lot of car trips to the nearest town for a tax break. You can't enforce something like that easily.

[41:21] Ravi Gurumurthy: What about business rates?

[41:28] Arun Advani: Government hasn't decided what it wants to do with business rates. While talking about devolving them, they're making rules about which businesses we like or don't like. There's a reasonable case for saying vape shops or gambling shops have negative externalities. Some economists argue taxes on land end up falling on the landlord, but in practice, if a landlord can rent to a newsagent or a vape shop, a higher tax on the vape shop creates a wedge that can discourage certain shops. You could allow local authorities to make those choices.

[42:59] Ravi Gurumurthy: Do other countries give local governments general freedom to establish any tax they want over and above national taxes?

[43:14] Arun Advani: There's an information and administrative problem. Even if you give local governments power to tax income, you don't want them administering a separate income tax system. You need central government to collect information. In the US, local and state governments add surcharges or local rates on top of income tax.

Starting from where the UK is as a centralized system, I'd start with property. Two other taxes I would devolve are tourist/hotel taxes and cordon/congestion charges for transport in and out of cities. Dense cities need to fund bus networks. Being able to say, "We're going to charge a fee to drive in, like the London congestion charge, but plow all revenue back into the bus network," is a sustainable model.

[44:46] Ravi Gurumurthy: To wind up, last couple of questions. If you could look abroad and steal anything from someone else's tax system, what would you import?

[44:57] Arun Advani: If you design capital gains tax right, it can be much better for growth and raise revenue. The most complex part is dealing with people leaving the country—settling up. The closest good example is Canada. Canada's version of capital gains tax does a fairly effective job of saying: if you leave for a bit and come back, you stay in the tax system; if you leave for a long time, you don't have to pay all the money on day one, but over a period we work out how to get the money owed. Because Canada shares a legal tradition with the UK, a lot of that could be imported fairly straightforwardly.

The other thing I would import is in inheritance tax. Currently, we tax inheritances based on the estate of the person who died, regardless of who receives it. If I have 10 children and you have one, the tax paid is the same despite my children receiving much less each. Ireland has a Capital Acquisitions Tax, which instead taxes the recipient based on how much they receive over their lifetime. That's a better design. It also avoids our problem where people rush to give away assets 7 to 10 years before dying, because under a recipient tax, you are taxed whenever you receive the money.

[47:14] Ravi Gurumurthy: Final question: What would you do if you were Andy Burnham in your first 100 days to signal the kind of tax reform you want?

[47:27] Arun Advani: There are many reforms you could think of for growth: sort out the VAT system, equalize how we tax income from different sources (work, partnership, self-employed), or reform property taxes. I would pick one of those areas and do comprehensive reform. If it's property taxes, take stamp duty, business rates, and council tax together into one coherent fix. If it's income from work, focus on all the ways we tax work.

Pick one naughty problem rather than salami-slicing a little bit everywhere. That allows you to tell a story at a budget that you are fixing an underlying problem, rather than just grabbing money because headroom went down.

[48:53] Ravi Gurumurthy: I'd also like "fiscal forward guidance," where a government telegraphs what it will cut or raise if fiscal headroom changes, or provides a roadmap for multi-year tax changes.

[49:28] Arun Advani: Absolutely. When doing reforms—like putting full NICs on partners—doing it all at once might be too big a shock. You can telegraph a multi-year roadmap: year one you do this much, year two this much, and by year three it's equalized. In 2010, the coalition government published a corporate tax roadmap showing what would change over time. Doing that gives certainty and plots a clear path to a better system.

[50:32] Ravi Gurumurthy: Arun Advani, thank you very much for joining us.

[50:36] Arun Advani: Thanks very much. Cheers.

[50:41] Ravi Gurumurthy: If you enjoyed this episode, please do like, share, and subscribe wherever you get your podcasts. Nesta is a research and innovation foundation based in the UK. We design, test, and scale solutions to the big challenges of our time. We're funded by a charity endowment and are politically neutral. For more, visit Nesta.