Stamp duty is a terrible tax. The Tories want to abolish it for most first time buyers. But the evidence shows that cutting stamp duty increases house prices, and that previous attempts to provide relief for first time buyers were ineffective.
Council tax is also terrible tax – with Buckingham Palace paying less council tax than a semi in Blackpool.
We can solve both problems together, and tax land in a way that encourages housebuilding and economic growth. But that requires smart thinking and brave politics.
The problem with stamp duty
Stamp duty land tax (SDLT)1Apologies to all tax professionals, but I’m going to call SDLT “stamp duty” throughout this article. is a deeply hated tax.
It reduces transactions2The elasticities found in HMRC research are incredible; a 1% change in the effective tax rate results in almost a 12% change in the number of commercial transactions and a 5-7% change in the number of residential transactions. (Strictly semi-elasticities because they are by reference to absolute % changes in the tax rate, not percentage changes in the % tax rate)., distorts the housing market, and often stops people moving when they want to. Stamp duty makes it harder to borrow from a bank (because the stamp duty is “lost value”). All of this means it reduces labour mobility, results in inefficient use of land, and plausibly holds back economic growth.
And the rates are now so high that the top rates raise very little; HMRC figures suggest that increasing the top rate any further would actually result in less tax revenue.
It also makes people miserable.
Stamp duty only exists because, 300 years ago, requiring official documents to be stamped was one of the only ways governments of the time could collect tax. We have much more efficient ways to tax today – but stamp duty remains. Until four years ago HMRC even still used the Victorian stamping machine in the picture at the top of the page.3Hello tax professionals. Yes, I know stamp duty and SDLT parted ways in 2003… but the point about the antiquated nature of stamp taxes remains valid. And I like the picture.
The problem is that, like many bad taxes, politicians have become addicted to it. SDLT now raises £12bn each year – an amount that’s hard to ignore.
And there’s an even worse problem: abolition would inflate property prices.
The problem with abolishing stamp duty
The link between stamp duty and prices is clear when we look at the impact on house prices of the stamp duty “holidays” in 2021:
The spikes in June and September coincide with the ends of the “holidays”. A rush of people to take advantage of the discounted stamp duty.
Of course the “holidays” were temporary – but the chart suggests that there was a permanent upwards adjustment in house prices (probably due to the “stickiness” of house prices).
Previous stamp duty holidays had less dramatic effects. There’s good evidence that the 2008/9 stamp duty holiday did lead to lower net prices, but 40% of the benefit still went to sellers, not buyers. I’d speculate that the difference is explained by the much lower stamp duty rates at the time.4There’s some published research on the 2021 holiday, but it’s qualitative as it was completed too soon to catch the September heart attack. I’m not aware of anything more recent, which is a shame – 2021 was a brilliant double natural experiment.
A detailed Australian study looked at longer-term changes than the recent UK “holidays” – it found that all the incidence of stamp duty changes fell on sellers (and therefore prices). This is what we’d expect economically in a market that’s constrained by supply of houses.5i.e. because tax incidence theory says that where supply is inelastic and demand is elastic, the seller bears the incidence.
These effects mean that stamp duty cuts aimed at first time buyers may end up not actually helping first time buyers. An HMRC working paper found that the 2011 stamp duty relief for first time buyers had no measurable effect on the numbers of first time buyers.
The problem with council tax
Stamp duty isn’t our only broken property tax. Council tax is hopeless – working off 1991 valuations, and with a distributional curve that looks upside down.
We can see the problem immediately from the Westminster council tax bands:
The bands cap out at £320k – equivalent to about a £2m property today. So there are two bedroom apartments paying the same council tax as Buckingham Palace.6Meaning the Royal Residence at Buckingham Palace – most of the rest of the complex isn’t a dwelling, and pays business rates not council tax. I haven’t seen any data on the value of the Royal Residence, but safe to assume it is very high indeed.
And the top Band H rate – restricted by law to twice the Band D rate, is pathetically small compared to the value of many Westminster properties.
The problem is then exacerbated by the fact that poorer areas tend to have higher council taxes. Here’s Blackpool:
So Buckingham Palace pays less council tax than a semi in Blackpool.
That’s why, if we plot property values vs council tax, we see a tax that hits lower-value properties the most:
In a sane world, this curve would either be reasonably straight (with council tax a consistent % of the value of the property), or it would curve upwards (i.e. a progressive tax with the % increasing as the value increases). This curve is the wrong way up.
The solution
The solution is to fix council tax and stamp duty at the same time.
Abolish stamp duty altogether, and change council tax to make it fairer… calibrating that change so that end of stamp duty doesn’t just send house prices soaring.7i.e. because economically we can expect the present value of future council tax payments to be priced into house prices, and if we increase council tax slightly at the low end and significantly at the high end, we should be able to undo the price effects of abolishing stamp duty. This is not an original proposal – it was one of the recommendations of the Mirrlees Review in 2010. Paul Johnson of the Institute of Fiscal Studies has also written about it.
But we can go further. The really courageous answer is to scrap council tax, business rates and stamp duty – that’s about £80bn altogether – and replace them all with “land value tax” (LVT)8A quick health warning: many of the people and websites promoting land value tax are eccentric. I once had a lovely discussion with someone from a land value tax campaign. After a while I asked what kind of rate he expected – 1% or 2% perhaps? His answer was 100%. Land value tax’s supporters remain one of the biggest obstacles to its adoption. They often suggest income tax/NICs, VAT and corporation tax could all be replaced with LVT – a look at the numbers suggests this is wildly implausible.. LVT is an annual tax on the unimproved9i.e. as if there was nothing built on it. value of land, residential and commercial – probably the rate would be somewhere between 0.5% and 1% of current market values10Meaning a higher % of the unimproved value; but it’s the % of market value that people will care about when the tax is introduced.. This excellent article by Martin Wolf makes the case better than I ever could.
There are two amazing things about LVT.
The first is that it has support from economists and think tanks right across the political spectrum. How many other ideas are backed by the Institute of Economic Affairs, the Adam Smith Institute, the Institute for Fiscal Studies, the New Economics Foundation, the Resolution Foundation, the Fabian Society, the Centre for Economic Policy Research, and the chief economics correspondent at the FT?
The second is that everyone says it’s politically impossible.
I wonder how true that is.
So let’s definitely not do land value tax. Let’s instead abolish stamp duty and fund it by adding some bands to council tax, so it more closely tracks valuations. Most people will pay a bit more tax, but not much more – and it’s worth it to get rid of the hated stamp duty. Whilst we’re at it, let’s update valuations more regularly, so it’s fairer. And why not make it apply to the unimproved value of land, so people aren’t punished for improving their property?
Everyone agrees business rates need reform – so let’s make similar changes to business rates.
What we end up with won’t be called “land value tax”, and won’t exactly be a land value tax. But it’s getting awfully close.
The price
That’s the price of abolishing stamp duty: some of us have to pay a bit more council tax (or, in my fantasy world, land value tax)11That would be quite unfair on someone who has just paid a large SDLT bill to buy an expensive property – they get punished under the new rules and the old. It would make sense to give some form of relief for recent SDLT… for example allowing SDLT to be written off over ten years worth of neo-council tax/LVT. So for example someone who paid SDLT nine years ago would get 1/10th of that credited against the new tax for one year. Someone who paid SDLT yesterday would get 1/10th of that credited for each of the next ten years. But this is one of many ways it could work.. That’s worth doing for a saner housing market that doesn’t hold back growth. And a land value tax should encourage house-building and actually boost growth.
But if all we do is abolish stamp duty, most or all of the tax saved by buyers will be eaten up in higher property values. It becomes a £12bn government handout to sellers.
There’s no free lunch. But there is an opportunity for a big pro-growth tax reform. It might even be popular.
Photo of original stamping machine is Crown copyright, and reused here under the Open Government Licence
Many thanks K for assistance with the economic aspects of this article.
- 1Apologies to all tax professionals, but I’m going to call SDLT “stamp duty” throughout this article.
- 2The elasticities found in HMRC research are incredible; a 1% change in the effective tax rate results in almost a 12% change in the number of commercial transactions and a 5-7% change in the number of residential transactions. (Strictly semi-elasticities because they are by reference to absolute % changes in the tax rate, not percentage changes in the % tax rate).
- 3Hello tax professionals. Yes, I know stamp duty and SDLT parted ways in 2003… but the point about the antiquated nature of stamp taxes remains valid. And I like the picture.
- 4There’s some published research on the 2021 holiday, but it’s qualitative as it was completed too soon to catch the September heart attack. I’m not aware of anything more recent, which is a shame – 2021 was a brilliant double natural experiment.
- 5i.e. because tax incidence theory says that where supply is inelastic and demand is elastic, the seller bears the incidence.
- 6Meaning the Royal Residence at Buckingham Palace – most of the rest of the complex isn’t a dwelling, and pays business rates not council tax. I haven’t seen any data on the value of the Royal Residence, but safe to assume it is very high indeed.
- 7i.e. because economically we can expect the present value of future council tax payments to be priced into house prices, and if we increase council tax slightly at the low end and significantly at the high end, we should be able to undo the price effects of abolishing stamp duty.
- 8A quick health warning: many of the people and websites promoting land value tax are eccentric. I once had a lovely discussion with someone from a land value tax campaign. After a while I asked what kind of rate he expected – 1% or 2% perhaps? His answer was 100%. Land value tax’s supporters remain one of the biggest obstacles to its adoption. They often suggest income tax/NICs, VAT and corporation tax could all be replaced with LVT – a look at the numbers suggests this is wildly implausible.
- 9i.e. as if there was nothing built on it.
- 10Meaning a higher % of the unimproved value; but it’s the % of market value that people will care about when the tax is introduced.
- 11That would be quite unfair on someone who has just paid a large SDLT bill to buy an expensive property – they get punished under the new rules and the old. It would make sense to give some form of relief for recent SDLT… for example allowing SDLT to be written off over ten years worth of neo-council tax/LVT. So for example someone who paid SDLT nine years ago would get 1/10th of that credited against the new tax for one year. Someone who paid SDLT yesterday would get 1/10th of that credited for each of the next ten years. But this is one of many ways it could work.
49 responses to “Stamp duty is a terrible tax. We should abolish it – but there’s a price.”
Taxes should be a choice. The recent interest rate hikes affected people whether they could afford the hike or not. Far better to put additional sales tax on luxury items so that people could choose to shop or not and not punish people for being homeowners. There needs to be a glass ceiling for sales tax too, but it is much easier to work these days with modern accounting software.
good luck defining “luxury items”.
i’ve just picked up your article. In Scotland I have been promoting a property tax known as AGFRR , Annual Ground Floor and Roof Rent. It would replace all devolved taxes and set a zero rate for income tax on earned income. If Scotland was independent it would replace all taxes.
All public and private land and property is charged a rate per square metre according to certain land types on which each rate is set.
It raises do much that under devolution an average 3 bedroom semi would attract AGFRR of around £4000 but within the funds raised there is enough to fund a Universal Citizens Income of £200 per week for every adult and child.
The Scottish Government would not seek any funds from the Block Grant or Barnett Consequentials.
AGFRR can’t be avoided.
No valuations are involved. As a property lawyer for nearly 50 years I have witnessed how valuations are manipulated. It’s not the bankers who control the U.K. economy; it’s the charted surveyors!
During any transition, people change attitudes and priorities. The efficiency of the proposed tax (in addition to its fairness and ease of assessment and collection) will encourage landowners to invest and occupiers to change dwelling when their circumstances change. This can work with a CGT based on transaction price minus the assessed value of unimproved land, which has already been taxed.
I’ve thought on the best workable proxy for a LVT to replace council tax.
We live in an information age, property websites estimate for free individual property values based on local/recent transactions.
Rental values are also estimated freely. Estate agents could be contracted to perform valuations, this might give them more stable income streams than typical sales percentages.
In terms of estimating unimproved land values, this can be approximated by deducting rebuild costs. And again these valuations are already routinely perform by insurance companies using established methods (BCIS).
Taxable amount= % of property/rental value minus rebuild cost.
Obviously in terms of local funding there would need to be central redistribution, since most revenue would be collected from wealthy areas.
What about a CGT on gains when property is rezoned/ granted significant change of use Planning permission? Money could go entirely to local councils (making local development more popular with residents)
Please may you demonstrate how you get to the 1991 top band of £320k+ being equivalent to £2M in 2024?
Try https://www.nationwide.co.uk/house-price-index/
For the practical reason that managing transition is difficult, we need evolutionary reform of property taxes but consistent with
a long term direction of travel. Five suggestions are:
The SDLT threshold to rise annually in line with Inflation.
All property with residential planning permission be subject to Council tax (most holiday lets don’t pay Council Tax or rates as they claim Small Business Rate Relief as their RV is below £15k).
On Council tax, move incrementally from 1991 values.
Charge any residential property with a market value of more than, say, £2m a supplementary annual capital value tax of say 1% per annum (with similar operation and exemptions to AETD).
Extend AETD to cover most Trusts.
Even if SDLT was abolished for main residences, I would keep it for holiday lets and second homes so as not to imperil the supply of homes for families.
Why is it a good tax for holiday lets and second homes but not in general? It’s a transaction tax, it just reduces liquidity. If anything, businesses are more able to absorb large capital outlays than families and keeping it just complicates the tax system even further.
If you want to disincentivise holiday lets and second homes, why not just increase the rate of property tax on those?
Largely just a restatement of some of the points made, but the most significant issue with SDLT is one of cash flows and leverage. A saved £2.5k on SDLT means that an extra £25k can be added to the purchase price (assuming a 90% LTV). This is a factor for those trading up but is felt more keenly by first-time buyers (an extra £25k is a big chunk towards a £300k 2-bed flat).
A neat and elegant solution touched on here but not explored is to keep the tax (maybe even broaden it) but defer the payment for people who don’t already own a property*. This way the tax does not interact with the amounts being borrowed and the subsequent purchase budget. Deferral is a messy I accept, but there is no reason why it cannot be linked to the mortgage somehow – when the mortgage is renewed for the first time and some equity has been built up (due to repayments or property inflation) the tax becomes payable. There would still be inflationary pressure put on the housing market but only by first time buyers – those trading up or buying second homes would still have the cash flow pinch.
A subtle change but a big impact.
*delinking from the property and linking to the individual would be an excellent change. First time buyer discounts didn’t apply if buying with someone who already owned a property (or two).
Given it’s not mentioned in the body or yet in the comments perhaps I’m wrong but wouldn’t these proposed changes need to trigger a comprehensive rethinking of local authority funding? Because SDLT is a central tax but council tax and a proportion of business rates are retained locally?
Maybe that’s a happy dividend because there’s presumably as much consensus from think tanks and economists about a renewed approach to the local government settlement. What am I missing that means I’m the first person to raise this angle?
You are completely right, and it would have to be thought about very carefully. In my mind you set a minimum LVT rate which goes to central government and then permit local authorities to go upwards for their own funding. But neither I nor the team has local government expertise, so I wouldn’t want to take a position on this.
I don’t think abolition and replacement all-at-once would work politically. I think the LVT would have to be introduced gradually (say over two parliaments) and the other taxes reduced.
This would stop people who had paid SD recently getting annoyed. It would also reduce the temptation to give old people in big houses an unfair/complicated exemption because they can’t afford to pay the new tax.
Dan mentions this in footnote 11, you could have a transition period where people get tax refunds for recently paid stamp duty.
I think you’d need to give the old people some sort of exemption or you’d get crucified politically, people are (understandably) very emotive about their homes. It doesn’t have to be phased in though, you could defer some of it until death or sale, although you’d have to be very careful not to accidentally create a disincentive to move
I’ve been an advocate of ‘Georgism’ for some time, so it’s great to see the central idea being given some thought
I’ve always wondered if the capital gains exemption on property should be abolished and instead owners should pay CGT on gains above wage inflation during the period of ownership. It might gum up the property market too, particularly as there is no CGT when someone dies (I’d abolish that too), but it may temper, slightly, the rampant house price inflation we’ve had due to supply side issues.
I’m no economist so have no idea if this idea is a dud, or even what sort of effective rate of CGT would be appropriate – a flat or progressive rate would seem logical.
Great idea. How often and when would the value be rebased? There needs to be a mechanism to reassess on a regular basis. I suggest whenever the land/property is sold or rented. The LVT could also address the view that VAT on private education is unfair as some move closer to good schools and won’t pay VAT on the inflated house prices. They would pay via a good LVT.
How would the tax on land under blocks of flats be assessed and who would pay?
Land values vary considerably across the country. Could there be a mechanism to rebalance, perhaps called ‘levelling up”.
Have thad thoughts along similar lines for a long time – it would need a well thought out transition system to avoid freezing the property market, as nobody will want to pay stamp duty shortly before it’s abolished. Perhaps giving a form of credit for stamp duty paid in previous 5-10 years in some form: you paid £50k in stamp duty in 2019, until 2029 you can offset £5k against the new LVT / council tax up to 50% of that bill etc.
I agree – there’s now something on this in the footnotes
Two comments:
1) I’m not sure how undesirable some of your arguments against stamp duty are- making the market stickier (less transactions) and cooling the property market might be seen as benefits.
2) The thing that seems odd to me about stamp duty is that is paid buy the buyers, rather than the vendors. So it’s a cost on top of purchase (like vat) rather than a revenue based tax (like say capital gains tax, which is another area for serious reform).
As it is, stamp duty falls hardest on the poorest (those buying their first homes, or upscaling, typically with young families). If it fell to the seller, it would hit those downsizing (which might be a bad thing, as we probably want to encourage downsizing), or those selling inherited properties, or developers/landlords, which feels more progressive to me.
it doesn’t matter who legally pays it – the economic incidence will likely be unchanged, as prices adjust. Problem with the seller paying is that it’s much harder to enforce (at the moment the buyer’s land registration is tied to paying stamp duty)
Interesting article….but you didt expand on the statement ‘it would be politically impossible’.
Why is that the view when you outlined various organisations that support the abolition of Stamp Duty with a varient of Land Tax?
I thought he wrote that ironically.
There was a development land tax back in the 1970’s that was abolished by the Tories early 80’s but I don’t recall its effect on value though I’m sure there is some data somewhere!
The point on business rates is that it works against adding plant in a factory or where house as that then increases rateable value. An early win might be to stop taxing notional value but just determine a price per m2 or m3. Then no valuers needed so the tax office can use those staff elsewhere or save some cash!
The most obvious reaction to this would be from asset rich cash poor retirees. Seeing as abolishing the triple lock appears to be beyond the pale I think we can safely say that this ain’t going to happen.
Unfortunately this ship has sailed. There are retirees all over London who bought in pre gentrification areas who are now sitting on £1m plus houses and will struggle to pay a 0.5%-1% land tax. I call it the 1980’s Hackney teacher syndrome.
LVT would give local voters yet another reason to prevent development in their area as a local plan that allows nothing will reduce their LVT bill.
I’m not sure that makes sense. The reduction in property value will be more important to many people than the c1% of that they’ll save in reduced LVT. But in any cases realistic LVT has to be accompanied by planning reform.
Discussion of removing or introducing any tax needs a thorough evaluation of the practicalities of collection and the facility with which it can or cannot be avoided. Economists are not always very good at looking at this aspect (in theory tax credits are a great idea; the reality in the UK anyway proved very different – a whole topic in itself). It is relatively hard to avoid paying stamp duty land tax (and indeed stamp duty as it existed befor) and that in part is surely why it remains – easy money as low collection costs. Ditto council tax (and why the bands are so rarely reassessed). LVT sounds great in theory but may prove quite costly to administer and generate large numbers of appeals, as someone has already commented.
I have frequently suggested exactly this to friends in France. Do not do LVT, just consolidate and slightly modify the existing taxes before slowly increasing them to find abolition of other taxes. I think it might make sense to stop at (not) LVT, VAT and probably CIT and maybe PIT. But I’d be open to scrapping the last two as well!
Dan, successive governments and chancellors have messed about with our tax system for years and have demonstrated absolutely no inclination to think creatively in a way that might actually help the economy going forward. I don’t see any chance of seeing a radical change in the status quo.
My hobby horse is online retailing even though I make considerable use of its convenience. But it was always predictable that it would have a devastating impact on the high street, concentrate power with the major supermarkets and result in the rise of ‘white van man’.
I can see no possible reason why high street shops have to trade at such a disadvantage compared with online. The projections seem to show that the online trend will continue inexorably upwards and we face the prospects of more shop closures and yet further charity shop opening; how depressing.
Might a simple solution be a higher VAT rate for online business?
Would this not be unfair on renters who never plan on entering the property market but are still required to pay council tax?
Pretty much 100% transferred to owners, but if you preferred one could amend council tax to apply to the owner, who would then raise the rents.
How do you calculate the unimproved value of land? Won’t it lead to endless appeals? How do you separate out (for instance) the intangible effect of having a well-kept garden or clean tenement close from the value of the land? Can you persuade the electorate that the wealthy residents of the Georgian tenements of Edinburgh’s relatively dense New Town should (most likely) pay less tax than the residents of its suburban former council houses? I’m not opposed to LVT, but there do seem to be practical and political difficulties to address.
This is an objection to all taxes except eg poll taxes.
Since former council villas in several Edinburgh suburbs currently sell for less than the reinstatement cost, it might be argued their unimproved land value is negative. I don’t think that’s the case for plots in the New Town.
I suspect the answer is that the unimproved land value (i.e. just the land) is positive. In some cases but the market value might be zero given demolition or decontamination costs… I’m not sure how LVT would deal with that.
Assuming that the LVT will require frequent, detailed valuation exercises if it is to remain accurate and fair, won’t those present something of a practical hurdle to implementation given the (presumably) significant resources needed to carry them out?
The move away from stamp duty to LVT will encourage and push residential transactions . Such transactions have a very positive impact on related businesses (e.g house furniture, maintence , new kitchens etc ) altjough arguable why here this is real quality growth the activity will be welcomed .
The current stagnation in commercial property will welcome the move . Politically the middle income and retired settled pensioners will be the problem .
I agree. I can see perverse outcomes. Why would a good number of people on low or moderate incomes (or initially high incomes who then fall on hard times, have children, have caring responsibilities, or long-term illness) want to keep a large garden, when its very existence is a direct attack on their standard of living? Some will take the hit, year after year, but many others will feel obliged to build and build again all sorts of gim-crack structures to cover and “improve” the garden. “Build a very large shed, concrete your lawns, house an illegal lodger, and cut your LVT!”
The LVT will incentivise development, but what kind of development? Who defines what “improvement” is or isn’t? As estate agents know, property “value” is frequently intangible and in the eye of the beholder. And how often would LVT be revalued? The whole thing seems to me like an administrative nightmare. As outlined, it seems to fail the “easy to assess” test of a good tax, because it appears to involve micromanaging and valuing every square inch of the country. This will be great repeat work for an army of estate agents and land valuers, but if Governments haven’t been able to stomach a council tax revaluation since 1991, how likely are they to take on something so detailed and open to debate as LVT?
I think you’re misunderstanding. LVT is on the unimproved value of land. The tax won’t change whether you build a mansion or a hovel. Nobody values improvements. The existence of the tax *incentivises* you to improve the land, because just sitting on unimproved land costs you money.
Valuing land should be largely a desktop exercise.
I think the footnotes for footnotes are broken – footnote 2 is empty for example.
Politically impossible is an interesting indictment. If true, that must be because no major political party is as interested in fairer taxes as it is in rewarding its voting base – in a short term and measurable way.
Where would agricultural land sit in a new LVT?
Great article, as always. I enjoyed Martin’s Column when it came out, so am glad you also appreciate it!
Have you ever looked at introducing CGT on residential property?
To make this fair to those who have recently bought, should those who have recently paid SDLT (eg within the last 10 years) get a credit of some/all of that SDLT against any increase in Council Tax/LVT?
that is a great point that often comes up – I’ll add a footnote