I was in the Land Value Campaign when we supported an exercise of this kind in 2005. Against our better judgement, we agreed that the valuations should be based on selling prices and not on gross annual rental values. Long story sort - the results got the proposal shot down in flames.
LVT must be assessed on gross annual values. Existing property taxes payable must be added to the net annual value. Same as the business rates before 1988.
A 1% rate of tax corresponds to a rate of about 20% on gross annual value. Land Value Taxes based on selling prices are set up for failure.
Thanks, Henry — you're raising the long-running question of which base to use.
For this analysis, we used capital land values: we can build a capital land value for every household from the ONS National Balance Sheet land totals and the Wealth and Assets Survey property values, and calibrate to official regional and national figures.
We don't model the valuation or revaluation machinery in this analysis.
Capital values are misleading. CV is depressed by the capitalisation of taxes actually payable, and then stamp duty has to be added on, and they are affected by interest rates and unstable anyway as they depend on the mood of the market.Council Tax rates vary hugely across the country.
On top of that there is the effect of the business rates on commercial property - which is relatively higher than the Council Tax, which means that land prices are corresponding lower. Conversion, agricultural land prices are artificially inflated.
This becomes a serious issue because land in all classes of use must be taxed at the same % rate.
I recently published a paper looking at something similar, but based on property values rather than land values alone, proposing to replace both council tax and transaction taxes such as stamp duty. https://doi.org/10.31235/osf.io/jt9vy_v1 - happy for any thoughts.
Hey Jack! Read it — nice work, and thanks for sharing.
A few things stood out: tackling stamp duty alongside council tax is a useful broadening, since transaction taxes capture much of the friction criticism that recurring charges avoid.
The first-time-buyer point is a sharp observation too — because relief means FTBs pay little transaction tax today, they're the group an annual levy hits hardest, which isn't obvious until you work it through.
One thing worth being explicit about for readers is the 10-year framing: since stamp duty is per-transaction, the comparison leans on an assumed holding period, so who comes out ahead shifts with how often someone moves.
Honestly, it was mainly pragmatism. The UK already has infrastructure for capital value assessment through lender valuations, AVMs, and Land Registry data. The paper builds directly on that.
Separating land from building values at scale requires machinery that doesn't yet exist here. I'd agree a pure LVT on rental values is theoretically superior, but the proposal was designed to be implementable within existing institutional capacity.
A tax on pure land value shouldn't pass through to tenants: land supply is fixed, so rents are set by what tenants will pay, not the owner's tax bill — the burden stays with the landowner. It's a tax on structures that can be passed on, since building supply responds over time.
We flagged it anyway for two reasons: our land values come from property values via a regional land-share, so the base isn't a perfectly clean land/building split; and the static model just doesn't model second-round effects either way — it puts the first-round burden on current landowners.
LVT can be passed on to tenants when they are paying below market levels. A surprising number are. Some landlords are lazy and don't keep the rents up to date for various reasons eg not wanting to disturb an arrangement that works. LVT would tend to bring rents up or down to the going rates.
LVT needs to be part of a bigger package of tax cuts all round to put more money in people's pockets.
It is hard to see the political viability of a policy that leaves 30% of households worse off. The 20% of land rich households in the bottom decile will be used to hijack the narrative. It is also hard to see the viability of a proposal that has London land values used to subsidise local authority profligacy in low land value areas like the north.
You're reading the distribution right — that income-poor, land-rich minority (around a fifth of the bottom decile) is exactly the group we flag in the post, and it's why most LVT proposals (Mirrlees, Fairer Share) pair the tax with deferral or transitional protection.
A flat national rate does redistribute across regions, and replacing council tax raises the question of how local government gets funded. We note in the post that the local-government funding redesign any reform would need is outside this static model.
Yes. The change to an LVT system has to be part of a bigger package. There is a surprisingly big space for manouevre. 9 of the 25 different UK taxes yield 95% of the revenue. The other 16 could simply be ditched.
LVT can replace Business Rates, Council Tax, Capital Gains Tax, Inheritance Tax, Stamp Duty and charges for planning consents - at a stroke.
VAT also needs to be phased out. The true yield is little more than half the headline figure, due to deadweight losses (at least 5% of GDP), consequential welfare costs, churning, abstraction of revenue from other taxes, and administration and compliance costs. A phase-out over 2 or 3 years would be needed for the economy to adjust (down to 12%, then 6%, then scrap).
A big hike in income tax and NI thresholds (both for standard and higher rates) would create 100s of thousands of jobs, especially at the low end
LVT does not work as a local tax. Local authories need a pot of money based on population and need factors eg age structure and geography (highways etc). Many services should just be charged for directly. And LVT assessments must be on gross annual rental value. 0.7% of prices is around 15% of gross annual rental value.
The number of losers now drops well below 30% and the policy looks like a genuine programme for getting the economy out of the mess it has been since the end of World War One.
"Restore" needs to take this up; migration is to a large extent a response to pressure on labour costs due to taxation and housing costs.
Phasing out VAT is a remarkable view. I'd phase out corporations tax first, and even income tax before VAT. How are any of these less worse than VAT (and why do zero economists agree?)?? And where would the money come from to replace VAT?
EVERY economist worthy of the name should know that VAT breaks all the principles of a sound tax, as set out in the Canons of Taxation. These were formulated over 250 years ago. They are obvious if you think about them for a moment. Many economists have criticised VAT on various grounds. It is not true that zero economists agree, and those who support VAT show themselves up as bad economists. It is an exceptionally bad tax
* administration and enforcement costs
* compliance costs
* discriminates against small traders due to complexity. Hits migrants especially hard
* fraud - the VAT Gap in Romania is 35%!
* evasion
* makes crime more attractive
* deadweight losses - about 5% of economy disappears due to reduction in demand
* consequential welfare costs
* churn
* incidence - traders are forced to absorb some of the tax to maintain volumes of sales - if they can not they cease trading
* abstraction of revenue from other taxes
At best estimate, the true yield from VAT is about half the headline figure. It might be as low a third. Conceivably it is nothing at all, since, by scrapping VAT, yields from other taxes go up and welfare costs go down.
The incidence of PAYE Income Tax and NI falls on employers and forms part of labour costs; it just takes a while to feed through. If they cannot carry it they get rid of labour eg by using things like self-service checkouts, running buses without conductors and trains without guards, care home staff at the lowest possible levels. Eventually the entry-level jobs vanish, which explains the very high unemployment rates among under-25s and over 50s. It's another disaster happening. Put it with VAT and we have a double disaster. And to the extent that the incidence of VAT falls on the end customers it becomes a cost to the government. Huge amounts of taxpayers money are handed out for people to pay back in VAT. That nets off to zero. Do you see what I mean? Expert economists rarely refer to it.
Which of these bullet points do you disagree with?
* administration and enforcement costs
* compliance costs
Estimated 6% of yield
* discriminates against small traders due to complexity. Hits migrants especially hard
* fraud - the VAT Gap in Romania is 35%!
* evasion
(sources HMRC and EU Tax Gap)
* makes crime more attractive
Crime is tax free earnings
* deadweight losses - about 5% of economy disappears due to reduction in demand
Source IFS based on 2009 VAT reduction as economic stimulus
* consequential welfare costs
* churn
Public sector employees and welfare/pension recipients are taxpayer-funded. VAT forms part of the price index to which pay and benefits are linked.
* incidence - traders are forced to absorb some of the tax to maintain volumes of sales - if they can not they cease trading
A Swedish study found that VAT cuts did not result in corresponding price cuts - however, businesses became more profitable. A cut in VAT from 25% to 12% for restaurant meals created 11,000 additional jobs in the sector.
* abstraction of revenue from other taxes
All taxes are ultimately taken from the same revenue stream. Any surplus is rent. (Standard Ricardian theory)
I love the quality of the discussion in this thread.
On VAT I think it has two advantages in that it captures revenue from people who create wealth outside the domestic tax system (e.g. if I make my money as capital gains in Singapore at 0% but spend a month a year living it up in London, VAT captures revenue from me for the UK that otherwise wouldn’t be captured)
Those are not arguments to disagree with but statements. Some of them are non-issues. The Public sector one, the migrants one, and the Romania one seem to be non-sequiturs, and the 5% of the economy one sounds like you pulled it out of your arse. The others are an argument for exactly what I support and you apparently oppose: harmonisation of rates and removal of exemptions.
If you want to mention any of the vaguely relevant ones like compliance and administration costs, please do refer to whether this is more or less per dollar raised than other taxes. Or don't bother: the answer is less, even much less.
If you have a proposal for a cheaper tax that raises even a quarter of what VAT raises, please do forward it HMT, they would be most interested.
Do you have an idea how the student exemption would influence this? As universities rely on overseas students to pay their bills, student numbers are rising rapidly, especially of overseas students (home students are, on paper, paying council tax at their parents' home).
Take Exeter, population just under 140k, HESA says 33k students many of whom will not be paying council tax if they live in accomodation, including private, that's only occupied by students.
We don't model the student exemption explicitly. Our council tax figure comes straight from the survey data, and our imputation doesn't capture the Class N student exemption as a specific rule.
The more robust point is on the reform side: once council tax is gone, the exemption is moot, and the LVT falls on whoever owns the land, not the student.
So we wouldn't capture a city-specific student effect like your Exeter example; that'd need the exemption modelled explicitly, which is a fair extension.
I'm confused by chart 1 and especially chart 5. If this is revenue neutral, shouldn't the bars on these charts add to zero? Chart 5 seems to show -660 revenue on average.
Also, shouldn't the sum of the bars in 1 and 5 equal each other? Why would arranging the data differently get a different net result?
I was in the Land Value Campaign when we supported an exercise of this kind in 2005. Against our better judgement, we agreed that the valuations should be based on selling prices and not on gross annual rental values. Long story sort - the results got the proposal shot down in flames.
LVT must be assessed on gross annual values. Existing property taxes payable must be added to the net annual value. Same as the business rates before 1988.
A 1% rate of tax corresponds to a rate of about 20% on gross annual value. Land Value Taxes based on selling prices are set up for failure.
Thanks, Henry — you're raising the long-running question of which base to use.
For this analysis, we used capital land values: we can build a capital land value for every household from the ONS National Balance Sheet land totals and the Wealth and Assets Survey property values, and calibrate to official regional and national figures.
We don't model the valuation or revaluation machinery in this analysis.
Capital values are misleading. CV is depressed by the capitalisation of taxes actually payable, and then stamp duty has to be added on, and they are affected by interest rates and unstable anyway as they depend on the mood of the market.Council Tax rates vary hugely across the country.
On top of that there is the effect of the business rates on commercial property - which is relatively higher than the Council Tax, which means that land prices are corresponding lower. Conversion, agricultural land prices are artificially inflated.
This becomes a serious issue because land in all classes of use must be taxed at the same % rate.
I recently published a paper looking at something similar, but based on property values rather than land values alone, proposing to replace both council tax and transaction taxes such as stamp duty. https://doi.org/10.31235/osf.io/jt9vy_v1 - happy for any thoughts.
Hey Jack! Read it — nice work, and thanks for sharing.
A few things stood out: tackling stamp duty alongside council tax is a useful broadening, since transaction taxes capture much of the friction criticism that recurring charges avoid.
The first-time-buyer point is a sharp observation too — because relief means FTBs pay little transaction tax today, they're the group an annual levy hits hardest, which isn't obvious until you work it through.
One thing worth being explicit about for readers is the 10-year framing: since stamp duty is per-transaction, the comparison leans on an assumed holding period, so who comes out ahead shifts with how often someone moves.
Good to see more modelling in this space.
Why have a tax on property values rather than the ad valorem tax on land, assessed on gross annual rental values?
Honestly, it was mainly pragmatism. The UK already has infrastructure for capital value assessment through lender valuations, AVMs, and Land Registry data. The paper builds directly on that.
Separating land from building values at scale requires machinery that doesn't yet exist here. I'd agree a pure LVT on rental values is theoretically superior, but the proposal was designed to be implementable within existing institutional capacity.
LVT assessment on annual rental values is easier than selling prices. VOA already value land and buildings separately.
Very interesting stuff, thank you for sharing! I do want to ask about this part:
> Nor does it capture rent pass-through to tenants;
Are you imagining that the tax on land gets passed through to tenants? Or that the tax on buildings *can* be?
A tax on pure land value shouldn't pass through to tenants: land supply is fixed, so rents are set by what tenants will pay, not the owner's tax bill — the burden stays with the landowner. It's a tax on structures that can be passed on, since building supply responds over time.
We flagged it anyway for two reasons: our land values come from property values via a regional land-share, so the base isn't a perfectly clean land/building split; and the static model just doesn't model second-round effects either way — it puts the first-round burden on current landowners.
LVT can be passed on to tenants when they are paying below market levels. A surprising number are. Some landlords are lazy and don't keep the rents up to date for various reasons eg not wanting to disturb an arrangement that works. LVT would tend to bring rents up or down to the going rates.
LVT needs to be part of a bigger package of tax cuts all round to put more money in people's pockets.
It is hard to see the political viability of a policy that leaves 30% of households worse off. The 20% of land rich households in the bottom decile will be used to hijack the narrative. It is also hard to see the viability of a proposal that has London land values used to subsidise local authority profligacy in low land value areas like the north.
You're reading the distribution right — that income-poor, land-rich minority (around a fifth of the bottom decile) is exactly the group we flag in the post, and it's why most LVT proposals (Mirrlees, Fairer Share) pair the tax with deferral or transitional protection.
A flat national rate does redistribute across regions, and replacing council tax raises the question of how local government gets funded. We note in the post that the local-government funding redesign any reform would need is outside this static model.
Yes. The change to an LVT system has to be part of a bigger package. There is a surprisingly big space for manouevre. 9 of the 25 different UK taxes yield 95% of the revenue. The other 16 could simply be ditched.
LVT can replace Business Rates, Council Tax, Capital Gains Tax, Inheritance Tax, Stamp Duty and charges for planning consents - at a stroke.
VAT also needs to be phased out. The true yield is little more than half the headline figure, due to deadweight losses (at least 5% of GDP), consequential welfare costs, churning, abstraction of revenue from other taxes, and administration and compliance costs. A phase-out over 2 or 3 years would be needed for the economy to adjust (down to 12%, then 6%, then scrap).
A big hike in income tax and NI thresholds (both for standard and higher rates) would create 100s of thousands of jobs, especially at the low end
LVT does not work as a local tax. Local authories need a pot of money based on population and need factors eg age structure and geography (highways etc). Many services should just be charged for directly. And LVT assessments must be on gross annual rental value. 0.7% of prices is around 15% of gross annual rental value.
The number of losers now drops well below 30% and the policy looks like a genuine programme for getting the economy out of the mess it has been since the end of World War One.
"Restore" needs to take this up; migration is to a large extent a response to pressure on labour costs due to taxation and housing costs.
Phasing out VAT is a remarkable view. I'd phase out corporations tax first, and even income tax before VAT. How are any of these less worse than VAT (and why do zero economists agree?)?? And where would the money come from to replace VAT?
EVERY economist worthy of the name should know that VAT breaks all the principles of a sound tax, as set out in the Canons of Taxation. These were formulated over 250 years ago. They are obvious if you think about them for a moment. Many economists have criticised VAT on various grounds. It is not true that zero economists agree, and those who support VAT show themselves up as bad economists. It is an exceptionally bad tax
* administration and enforcement costs
* compliance costs
* discriminates against small traders due to complexity. Hits migrants especially hard
* fraud - the VAT Gap in Romania is 35%!
* evasion
* makes crime more attractive
* deadweight losses - about 5% of economy disappears due to reduction in demand
* consequential welfare costs
* churn
* incidence - traders are forced to absorb some of the tax to maintain volumes of sales - if they can not they cease trading
* abstraction of revenue from other taxes
At best estimate, the true yield from VAT is about half the headline figure. It might be as low a third. Conceivably it is nothing at all, since, by scrapping VAT, yields from other taxes go up and welfare costs go down.
VAT is a hidden disaster.
Wtf?
Can you elaborate please?
Sure, if you can find me one reputable economics paper that agrees even a little bit with your position on VAT.
CORRECTION LVT does not work as a LOCAL tax. Places like Sunderland and Whitehaven don't have much land value.
NIC increases left about 70% of households worse off
The incidence of PAYE Income Tax and NI falls on employers and forms part of labour costs; it just takes a while to feed through. If they cannot carry it they get rid of labour eg by using things like self-service checkouts, running buses without conductors and trains without guards, care home staff at the lowest possible levels. Eventually the entry-level jobs vanish, which explains the very high unemployment rates among under-25s and over 50s. It's another disaster happening. Put it with VAT and we have a double disaster. And to the extent that the incidence of VAT falls on the end customers it becomes a cost to the government. Huge amounts of taxpayers money are handed out for people to pay back in VAT. That nets off to zero. Do you see what I mean? Expert economists rarely refer to it.
I dont think anyone is disagreeing about payroll tax, I said as much in another reply, save that there is no delay to feeding through.
OTOH I don’t think anyone understands your position on VAT, perhaps not even yourself.
Which of these bullet points do you disagree with?
* administration and enforcement costs
* compliance costs
Estimated 6% of yield
* discriminates against small traders due to complexity. Hits migrants especially hard
* fraud - the VAT Gap in Romania is 35%!
* evasion
(sources HMRC and EU Tax Gap)
* makes crime more attractive
Crime is tax free earnings
* deadweight losses - about 5% of economy disappears due to reduction in demand
Source IFS based on 2009 VAT reduction as economic stimulus
* consequential welfare costs
* churn
Public sector employees and welfare/pension recipients are taxpayer-funded. VAT forms part of the price index to which pay and benefits are linked.
* incidence - traders are forced to absorb some of the tax to maintain volumes of sales - if they can not they cease trading
A Swedish study found that VAT cuts did not result in corresponding price cuts - however, businesses became more profitable. A cut in VAT from 25% to 12% for restaurant meals created 11,000 additional jobs in the sector.
* abstraction of revenue from other taxes
All taxes are ultimately taken from the same revenue stream. Any surplus is rent. (Standard Ricardian theory)
I love the quality of the discussion in this thread.
On VAT I think it has two advantages in that it captures revenue from people who create wealth outside the domestic tax system (e.g. if I make my money as capital gains in Singapore at 0% but spend a month a year living it up in London, VAT captures revenue from me for the UK that otherwise wouldn’t be captured)
If your income arises in Singapore why do you owe anything to the government of a country halfway round the world?
Those are not arguments to disagree with but statements. Some of them are non-issues. The Public sector one, the migrants one, and the Romania one seem to be non-sequiturs, and the 5% of the economy one sounds like you pulled it out of your arse. The others are an argument for exactly what I support and you apparently oppose: harmonisation of rates and removal of exemptions.
If you want to mention any of the vaguely relevant ones like compliance and administration costs, please do refer to whether this is more or less per dollar raised than other taxes. Or don't bother: the answer is less, even much less.
If you have a proposal for a cheaper tax that raises even a quarter of what VAT raises, please do forward it HMT, they would be most interested.
Do you know what a "non-sequitur" is?
Do you have an idea how the student exemption would influence this? As universities rely on overseas students to pay their bills, student numbers are rising rapidly, especially of overseas students (home students are, on paper, paying council tax at their parents' home).
Take Exeter, population just under 140k, HESA says 33k students many of whom will not be paying council tax if they live in accomodation, including private, that's only occupied by students.
We don't model the student exemption explicitly. Our council tax figure comes straight from the survey data, and our imputation doesn't capture the Class N student exemption as a specific rule.
The more robust point is on the reform side: once council tax is gone, the exemption is moot, and the LVT falls on whoever owns the land, not the student.
So we wouldn't capture a city-specific student effect like your Exeter example; that'd need the exemption modelled explicitly, which is a fair extension.
FYI, the link is to the US policy engine site, not the UK one: https://www.policyengine.org/uk
Thanks for letting us know!
Putting a disproportionate burden on Northern Ireland, the poorest and most politically sensitive of the four nations, is an obvious non-starter.
I'm confused by chart 1 and especially chart 5. If this is revenue neutral, shouldn't the bars on these charts add to zero? Chart 5 seems to show -660 revenue on average.
Also, shouldn't the sum of the bars in 1 and 5 equal each other? Why would arranging the data differently get a different net result?