1. The purpose of this paper is to set out a possible implementation plan, in current economic and political circumstances, for annual land value taxation (LVT) in Britain. It attempts to deal with all major transitional issues so as to minimise both political and administrative difficulties. It does not claim to present the only – or even the best – such plan. It is a basis for further discussion and does not aim to be prescriptive or to match a particular political party narrative.[i]
2. The paper omits discussion of the possible impacts on the banking system and how it might be dealt with, which is the subject of a separate paper by the Coalition for Economic Justice (CEJ). It also leaves out discussion of the “local or national tax” question, which is analysed separately. The CEJ has chosen a “national option” for LVT. However this does not mean that local taxes remain unreformed, only that to restrict LVT to local government enormously reduces the potential impact of LVT on the nation’s economy. Nor does choosing the national option mean that the Plan simply goes in one step to a wholesale reform of a raft of taxes (national and perhaps local) in one budget – although that would be possible.
3. It must be emphasised that LVT is not seen by its advocates as an additional tax but as a means of obtaining progressive reduction of many other taxes: a major tax shift. Naturally the choice of which taxes to abate, to what extent and how to do so is very much open to debate and ultimately the decision of the Government of the day. These choices greatly influence any eventual Plan.
4. The paper sets out a six-stage Plan, in approximate chronological order. At each stage in the Plan, the choices available are to some extent explained. However for the sake of brevity, the Plan only pursues one particular set of choices. Sometimes options exist to deal with some of the issues that arise (these issues are summarised at the end): the relative merits of each option are briefly discussed and a preferred option selected.
5. There are six suggested stages in this provisional LVT Implementation Plan. These are:-
1) Enact a National Land Rent Bill: establishment of a Vision for LVT.
2) Reform of Business Rates.
3) Replace Council Tax with a Housing Services Charge.
4) Introduction of national (UK-wide) LVT.
5) Moving to Annual Revaluations.
6) Phasing out remaining property taxes.
STAGE 1: The National Land Rent Bill: establishment of a Vision for LVT
6. Although it would be possible for a Government that had a clear commitment to full national LVT in its manifesto to include all necessary measures in a single Budget, CEJ believes this is an unlikely scenario. Hence, in order to secure popular support for a longer-term Plan, it is important for Government to establish at the outset its Vision of the purpose in shifting the burden of taxation onto Land and away from Labour and Capital. The nature of ‘economic rent’ as an ‘unearned increment’ which, unless diverted by way of tax, automatically accrues to those who happen to possess ‘land rights’, would be asserted in a preamble to the Bill that set out the moral case for reform.
7. Passage through a reforming Parliament of what has been called a National Land Rent Bill[ii] would thus establish the ethical and economic argument for LVT, without venturing into detail on implementation. In passing such a Bill, Parliament would be launching a series of detailed debates on all aspects of implementation. In particular, the dominance of the “local option” in debate needs to be countered: a strategy that sets out to reform local taxes alone is very likely never to proceed beyond that, thus sacrificing the chance to raise ten to fifteen times as much revenue from land values.
8. Most important, without such a commitment by Government, policy researchers and planners would continue to find it almost impossible to access official data-sets and other resources. Currently researchers are obstructed by lack of any in-principle official support for using land values as a basis for taxation, yet sceptics repeatedly cite lack of research as a reason to withhold such support!
9. Key elements of a National Land Rent Bill would be:
10. The less detail that appears in the Bill, the fewer the arguments that can be raised against it, also the greater the flexibility that is given to policy planners. However if a particular element helps secure cross-party support, necessary to achieve passage of the Bill (e.g. raising the basic income tax threshold by replacing it with LVT) then this can be specified as an early aim.
11. A further benefit of having this in-principle decision is that it will, of itself, start to affect the land (and financial) markets, as owners of land begin to realise that speculation in land will cease to be profitable. It gives market players an opportunity to invest in ways that are economically more healthy for the country, before they incur tax liabilities for not doing so.
STAGE 2: Shift National Non-Domestic Rates (NNDR) onto a Site Value Basis
12. Because businesses do not have a vote, the easiest first step electorally for a Government is to reform Business Rates (as opposed to Council Tax). ‘Site Value Rating’ (SVR) – the term used in Britain for LVT assigned to local government - could be introduced within a single Parliament. The Bill for it could be tabled within weeks of the passage of the National Land Rent Bill.
13. Although NNDR (also known as Business Rates) are already fully assigned to local government, the tax is generally seen as a national one and is known as the Uniform Business Rate because it is set at a uniform national level for each of England, Wales and Scotland. Whilst councils could be given more control over the reformed rates at the same time as the valuation base is reformed, starting to implement LVT this way allows the whole process to be seen from the outset as a national measure: applying across the whole of Britain, if not UK-wide.
14. Revaluation for business rates already involves a crude form of the system for ‘computer aided mass assessment’ (CAMA) of property values that LVT requires, and regular (5-yearly) revaluations, hence it is also technically easier to make the changes to tax systems than it would be for Council Tax. Leaving Business Rates unreformed while introducing LVT alongside it would add significantly to the complexity and volume of work that tax administrators would need to do initially. However a ‘split-rate’ system (similar to that of some cities in Pennsylvania) could be used: a progressively higher rate set for the site-value element of business rates, with the building value element phased out and no new valuation for buildings and other ‘improvements’ (see below).
15. Since the rate is to be paid by the owners of all land used for business, SVR would trigger a requirement for registration of all remaining unregistered title to such land[iv]. This requires no primary legislation and can commence by tabling an Order once ‘stage one’ is complete: if there is to be collection from owners of land rent as revenue, then it is necessary to establish where to send bills. Completing the Register would take about one year.
16. Land used – not merely designated - for residential purposes would at this stage be exempt (also possibly land used for agriculture or forestry[v]) but all vacant sites, empty properties, second homes and ‘amenity’ land in private ownership and not part of a residential curtilage would be covered. If no title is claimed to such land after (say) six months, then it should automatically revert to the local authority, for disposal by auction or community use. Thus SVR extends the rate base considerably, allowing most tax bills to be reduced.
17. Much of this land (i.e. vacant sites etc.) is difficult to assess – more so than assessing rental value of residential or agricultural land sites – largely because it has been sterilised from open market activity by various factors[vi]. Therefore the option of self-assessment would be offered to owners, to ease the initial work of the official Valuation Office Agency (VOA), with the proviso that the owner’s valuation would form the basis of purchase price by the local authority, for onward sale by auction if appropriate. This would ensure self-assessments were reasonably well based on market conditions.
18. Site-value based bills could be raised within at most 2-3 years of enactment of the Bill, if the next routine five-yearly revaluation for business rates (due to be carried out in 2012/13 for 2015) were to be suspended, allowing officials to instead concentrate on the new system. Publication of provisional site-value rating lists in late 2014 would show, before the next general election, that there are many benefits to businesses and to communities.
19. This reform could be a significant factor in economic revival, since many sites would be brought forward for redevelopment well before the impact of the much higher tax bills on those under-used sites was felt by their owners. If combined with Tax Increment Financing (TIF) using (initially) existing NDR, as already planned by the Coalition, it would finance a major urban infrastructure investment programme at no cost in terms of general taxation.
20. The shift from NDR to SVR could be phased over 2-3 years, with the 2010 NNDR valuations used for a progressively smaller proportion of revenue, as site-value assessments are conducted by VOA to replace self-assessed bills. This would probably smooth the transitional impacts on small business in a more equitable way than current NDR transitional arrangements. In general, business occupiers who do not own long leases or freehold in their premises will pay less under SVR than under NDR.
21. Bills could be sent to occupiers (those who now get the bill for NDR) until and unless the site owner is known. Occupiers would be obliged by law to tell the tax authorities who they pay rent to, or to pay the full SVR bill. They would be entitled to deduct the SVR payable from their rent, provided they inform the authorities who gets the rent.
22. If the property industry knew that LVT was to be nation-wide, it is probable that private sector funding of the necessary land information systems would be forthcoming. The UK property market is currently relatively poorly served by property information, as a number of studies have noted. All other EU member states already have complete registers of land ownership and value. It is the absence of any UK government commitment to property tax reform that to a large extent holds back the development of better land information systems. In all other countries, it is the property tax system that dictates the form of these systems. However Britain could achieve the world’s first holistically designed and privately funded National Land Information Service (NLIS), by having its design specified to meet a wide range of user needs: tax administrators, land use planners, investors, home-buyers and business occupiers. A detailed analysis of these needs should commence upon the tabling of the SVR legislation.
STAGE 3: Replace Council Tax with Housing Services Charge
23. Some would argue that LVT should be applied last of all to owner-occupied residential land – if at all. However the longer that a revaluation for Council Tax (CT) is delayed, the more likely it is that it will need replacing eventually by something radically different. All sensible politicians must realise this. No Party now supports a complete scrapping of domestic property taxation: even the Liberal Democrats, for whom replacing CT with Local Income Tax (LIT) was until recently a major policy pledge, in their 2010 Manifesto only called for local pilots of LIT[vii].
24. The Mirrlees Review recommends replacement of NNDR with SVR but argues for CT to be replaced by a tax on the capital value of domestic properties, which it calls the Housing Services Charge (HSC). The name has its attractions, since it avoids ‘tax’. However the argument is not persuasive and we suggest that the term be adopted for a local form of LVT on all residential properties, with three major differences to the IFS/Mirrlees proposals:
25. This offers several political, social and economic benefits, compared to CT. Firstly, it is simpler to calculate site rental values than capital values, which implies detailed inspection of all properties. Secondly, there are fewer bills to send out: privately rented flat-dwellers only pay directly if they share in the equity[viii]. This leads to a third benefit, attached to a personal or Homestead Allowance (HA): the only way in which the owner of several residential properties can offload the LVT bill (at least in part) is to sell properties to their occupants, either fully or in part[ix]. This strongly counters the argument that LVT runs contrary to the British culture of home ownership.
26. The situation with social housing is more complicated. However most social housing is at the lower end of taxable value. Owner-providers of such housing (local authorities themselves or registered social landlords (RSLs)), who would be liable for the tax payable on the land occupied by their housing stock, could either move towards sharing the equity with their tenants or apply for registration of charitable status for their social housing land. Unlike the United States, where the charitable status of the taxpayer entitles it to exemption from property taxes, we would urge that only the charitable purpose of the site be allowed to attract any exemption: this would help ensure that social housing, which is already deemed ‘charitable’ in purpose, remains as such.
27. If HSC (the domestic version of SVR) is seen as a charge payable to the local authorities (county, unitary, district, parish) for communal, location-related services such as waste collection, policing, environmental health, public transport - and housing - then it is more likely to be seen as ‘fair’. If a basic standard of housing is seen as a human right, then a tax-free allowance (HA) against the HSC can be seen as an acknowledgement that owner-occupiers are largely meeting their own basic needs. For Councils that provide social housing (or RSLs), provision of those basic needs could be seen as a tax-exempting charitable contribution towards what the State would otherwise have to provide: rent paid by tenants is, in effect, in lieu of HSC. Most providers and tenants in the social housing sector would be glad to consider shared ownership.
28. As for the level at which HSC and HAs should be set, these are political decisions. The net yield from HSC would need to be equal to that from CT. HA would be set nationally as a ratio to local HB/market rents: a political decision for central (or devolved) government. It would almost certainly be electorally necessary to set it at a level which, when combined with the likely HSC rate, ensures that most households pay less than they do under CT at the time of the change.
29. As with the change from NNDR to SVR, the change from CT to HSC could be phased over several years. However this might be seen as adding to confusion among voters, less likely than businesses to understand split bills. We suggest that the change occurs at a time chosen by each of the devolved governments. Whether or not to change might itself be left for these governments to decide.
30. Asset-rich, income-poor hardship cases (like the ‘poor widow’) can be overcome by simply allowing any taxpayer on low income to defer their HSC bill until the property is sold, re-mortgaged or the owner dies. At that point, the tax authorities would have first call on unpaid amounts, through a Land Charge, with interest payable. This happens in some other countries.
31. Some might argue that the introduction of any residential form of LVT should be delayed to a later stage or not attempted at all, which would imply either continuing to use Council Tax (CT) as an ever more unfair property related tax, or simply scrapping it. Any reform or replacement of Council Tax is very visible to voters and hence very risky electorally, yet its yield is only equivalent to about 1/20th of total tax revenue – so why not reform other, less visible more nationally significant taxes, runs the argument.
32. Scrapping CT without at the same time replacing it with another property tax on residential land would lead to an immediate one-off increase in house prices and would make it much harder ever to introduce any residential property tax again, as the Irish Government is finding. Leaving it in place would, in addition, mean the deferral of any benefits accruing to the property industry and policy-makers from a nation-wide land valuation: without a residential form of LVT, land valuing some 40% of the national total would almost certainly remain unvalued, as well as untaxed.
STAGE 4: Introduction of National (UK-wide) LVT
33. For reasons given in a separate CEJ paper, we firmly believe that there should be a national LVT, at a single rate, levied on all land irrespective of its use. The main purpose of this is to enable other dysfunctional national taxes to be abated. This should be signalled from the outset (Stage One).
34. Having established site-value-based taxes to replace CT and NNDR, the administrative cost of introducing a national LVT would be very small. The tax rate could therefore also be small to start with. It could either be collected as a ‘precept’ on those paying SVR & HSC, through local billing authorities in the same way that NNDR is collected and passed – in full – to central government now; or it could be re-introduced as Schedule A[x] within the income tax system and a similar additional component of Corporation Tax.
35. In the case of both corporate and personal taxpayers, following completion of the Land Registers, it would only be necessary for them to declare their current landed assets. The tax authorities could then relate each site owned to its registered valuation. Under the Swedish system, the annual tax forms are sent out pre-completed by the authorities: all the taxpayer needs to do is certify that there have been no changes in ownership or other status, then return the form.
36. We would therefore favour an early move towards incorporation of SVR & HSC, as well as national LVT, within the central government tax systems: a single, simple tax return for all taxable entities (personal and corporate), with local authorities merely setting their tax rates and central government passing the SVR/HSC money due to councils straight to them. This happens now with CT in two-tier areas with districts/boroughs and counties, also with districts and their parishes.
37. One major advantage, many would say, to this tax simplification through use of LVT as a national tax, is the scrapping of the annual CT bill dropping through householders’ letter-boxes. If local taxes are presented as merely a component of the main national tax bill (through PAYE and Corporation Tax) then they are less visible. There would still be full accountability for local elected officials who set local tax rates – possibly more so than now – because (as explained in our paper on ‘local v. national’) the disparities in tax yield potential between poor and rich areas would reduce as the level of national LVT rises. ‘Equalisation’ would be needed much less and councils would collect far more of the revenue they need: possibly 75% on average.
38. There need not be any tax-free allowance on the national LVT, nor should there be any differential tax rates. The lowest value areas should pay the same ‘poundage’ (rate per unit value) as the highest value land: it is the relative valuation that achieves equity.
39. The main difference between SVR/HSC and this National (UK-wide) LVT is that the rate for the former would be set by local councils, whereas the UK Parliament would set the rate of LVT. Any LVT levied by/for devolved national Parliaments/Assemblies would of course be set by them too. The share of total potential annual land rent available for each tier of Government to use as revenue would need to be set by Parliament as well.
40. At this stage, if LVT is to be incorporated in existing national business and personal tax systems, the idea of voluntary Land Value Covenants (LVCs) could be explored. This involves a taxpayer undertaking to roll up all (or a portion) of their future tax liabilities in exchange for transferring the land rent due in future years to the state. If the terms were made attractive enough, the cost of LVCs to both taxpayers and government could be considerably less than the cost of existing taxes, since the tax is of course fixed and immovable. Once the benefits of LVT were seen, this voluntary approach to elimination of dysfunctional taxes and economic reform might prove irresistible to the general body of taxpayers. We see LVCs being introduced (if at all) only alongside UK-wide LVT, because they could be applied to any or all national taxes. To have LVCs associated with Stages 2 and/or 3 of this Plan would be too complicated for the local councils which mainly administer and benefit from SVR/HSC.
41. It is also at this point that Scotland’s government might wish to adopt an element of LVT instead of that portion of income tax that it is allowed to vary in relation to the UK rate. The Scotland Bill currently going through Parliament at Westminster allows up to 10% of income tax to be varied. Until the UK income tax systems make technical provision for this, it would be extremely difficult to achieve in practice, whatever the legislation says.
STAGE 5: Moving to Annual Revaluations
42. Land values are constantly changing, both in absolute terms and relative terms. Failure to regularly revalue for tax purposes guarantees that any property tax levied annually will become unfair. The longer the period between revaluations, the harder it becomes politically to insist on a revaluation. Eventually a property tax that is based on out-of-date valuations becomes untenable.
43. Modern property tax systems use sophisticated computer-aided mass assessments. By continually monitoring market transactions, which a properly functioning property market needs to do anyway, a government can both serve the market and satisfy this basic requirement of property taxation. Research indicates that the overall cost of managing ad hoc appeals against out-of-date valuations and of conducting periodic wholesale revaluation (e.g. for NNDR) is little different – indeed probably greater, taking account of costs to taxpayers – than the cost of ‘rolling revaluation’, whereby every year all or part of property tax registers are updated. Most countries with modern property tax systems revalue at 3-5 yearly intervals. Some academics and practitioners in the UK already favour annual revaluations.
44. We strongly suggest that legislation for LVT should specify that the land registers be designed as multi-purpose and open (not necessarily free at point of use), so that there is every incentive for property market players to contribute to their maintenance on a regular basis. The tax registers should be updated at not less than three-yearly intervals, starting from the time when NNDR is scrapped. Because commercial site values can fluctuate very greatly over space and time[xi] and because they attract the highest tax bills, we think that all land above a certain taxable value (say £50,000 per ha) should be revalued annually. However for most sites an annual indexing between full revaluations would be quite adequate.
STAGE 6: Phasing out Remaining Property Taxes
45. Although many national taxes are, in effect, collecting land rent to a degree, the most notable property taxes that would need to be phased out as LVT is introduced are Stamp Duty Land Tax (SDLT), Inheritance Tax and Capital Gains Tax (at least on UK-based landed assets). All are based on capital values, not annual rental values.
46. Because landed property transactions will initially incorporate a large component of ‘pre-LVT’ unearned capital gain, the phasing out of these taxes should be gradual, although it could begin at any time since legislation already allows for their rates and incidence to be varied. The initial national land valuation could establish a base value against which subsequent capital value gains can be (un)taxed, whereas gains made prior to the changeover date but not realised until sale would remain taxable. The aim should be to avoid ‘double taxation’ of that gain.
47. SDLT should be scrapped at the earliest possible time, at least on sales of principle residences, because it is a barrier to labour mobility and hence to a healthy economy. Assuming the first national tax which Governments would wish to merely abate is basic rate income tax (so that nobody whose full-time equivalent earnings are less than whatever national minimum wage is at the time pays any income tax) the next most appropriate tax to eliminate is probably SDLT. However these are largely political decisions.
48. What are known as “Section 106” payments by developers to planning authorities, as well as the proposed Community Infrastructure Levy (CIL) ought to also be scrapped upon introduction of full SVR (Stage 3). Again this will need to be done gradually, because they are not structured as taxes but embedded in legal agreements that form part of the granting of planning permissions.
49. The Swedish practice of allowing developers to defer payment of higher post-permission assessed property taxes (which arise as soon as permission is given but before developments are built) would act as a further incentive to developers to implement their permissions. The deferment ‘clock’ should start when the permission is given and run for a set period, calculated to accord with construction phasing agreed with the planning authority at that time.
Summary of Transitional Issues
50. In the Plan outlined above, certain issues have been touched upon - with most of them being ‘transitional’ (i.e. they are either inherently short-term in nature - related to the transition from current taxes towards LVT - or they are the result of understandable fears that any major change in policy causes). We now list these and summarise how they can be dealt with, in theory, by this implementation plan:-
1) Establishing early confidence in the efficacy of LVT. The National Land Rent Bill and having a coherent long-term (but flexible) Plan formally enacted from the outset helps achieve early popular understanding and support. By first reforming business rates, LVT maximises funding opportunities for the infrastructure that will benefit residential [voting] home-owners, whose Council Tax bills will mainly fall when LVT later comes in on other land. It seeks to demonstrate economic benefits that benefit the public by tackling the most valuable land first. It does this without “frightening the [electoral] horses” too much!
2) Data sets. The cost of the necessary datasets for property tax reform is often cited as a major obstacle to progress. The planned introduction of LVT would act as a spur to modernisation of the UK’s land information management systems, aiding many national policy objectives including climate change mitigation. Property industry/stakeholder confidence in the data systems will be established early on: these systems are likely to attract private sector funding, greatly reducing the cost of tax administration. The cost of an initial nation-wide land valuation is spread over several years by allowing the use of initial self-assessments.
3) Hardship. Through a combination of Homestead Allowance (for principle residence of all households) and deferment of tax bills, hardship for almost all owner-occupiers is avoided. For tenants in social housing, charitable status applied to this land gains exemption for housing providers and hence helps keep rents low for tenants. However, there should be no other exemptions since the low land values associated with uses such as recreation and conservation under our ‘Plan-led’ system (which would be retained) will ensure that hardship does not arise: low values mean low tax bills.
4) Negative Equity. If Homestead Allowance is calculated such that only owner-occupiers of the most valuable residential sites pay more HSC than CT, the situation will be no worse than now. LVT should not, of itself, cause capital values (hence mortgage debts) to increase, for reasons given in our Banking paper. Few first-time buyers move straight into large homes on valuable sites. Most first-time buyers are earning, so their income tax thresholds will rise – assuming national LVT is used to achieve this. Therefore the situation ought to improve for almost every family now in negative equity.
5) Phasing out existing property taxes. The entire Plan is based around gradually phasing out CT, NNDR, SDLT and Section 106 contributions by developers. All work on CT, NNDR and SDLT valuations would cease early in the transition, largely offsetting costs of LVT. If introduced as a nation-wide tax on all land, the property industry is likely to pay for the necessary LVT systems. The overall cost of administering property taxes would fall.
(c) Copyright, Coalition for Economic Justice, February 2011.
[i] The Liberal Democrat Party reaffirmed support in 2007 for LVT ‘for the longer term’, also specifically for business rates to be converted to site value rating within a single Parliament. The Green Party also included LVT in its 2010 manifesto, as did the Co-operative Party, which sponsored some 40 candidates – most of whom were successful – including the former Prime Minister Gordon Brown. LVT is a long-standing part of Liberal, more recently Labour, tradition. Supporters include several sitting Conservative councillors: the Bow Group within the Conservatives recommended LVT to the Party’s Tax Reform Commission in 2006.
[ii] J D White MP tabled a Private Members Bill in 1936 with this title and purpose.
[iii] The arguments for the use of annual rental values, as opposed to capital values, are set out in a separate paper by CEJ: “Land Value Taxation - Valuation Principles”.
[iv] Currently some 50% of land (by area) of England & Wales is unregistered. However this probably represents less than 10% by value and 20% of the total eventual number of titles.
[v] We do not believe that agricultural and forestry land should be explicitly exempted from local property taxes. Local authorities have to provide services, at high cost, to owners and occupiers of properties made remote by the existence of such land. However because it is very much lower in value than urban land, the revenue yield from all rural ‘green’ sites (some 80% of the country by area) is only about 5% of the total. It is not worth the political or administrative effort of including currently untaxed rural properties within the LVT system initially, so we suggest that sites of under (say) £3,000/ha rental value (irrespective of size) be exempted.
[vi] All valuations for tax purposes depend upon the existence of property market evidence. Where ownership or potential use is hard to determine, sites rarely enter the market.
[vii] Former Lib Dem Treasury Spokesman and author of the Party’s 2007 Tax Policy Paper, Dr Vincent Cable MP, has never favoured scrapping Council Tax or LIT. His Party now accepts there should be LVT on all land in the long term. Their support for both LIT and LVT might be resolved by incorporating something similar to the former “Schedule A” (notional income from owner-occupation) within income tax. Sweden still has such a system.
[viii] As with commercial property, occupiers in possession of a lease would be liable to pay a proportion of the total LVT bill, since they share the benefit accruing to ownership. Connellan, in “Land Value Taxation in Britain: Experience and Opportunities” (2004) sets out a simple formula for assigning liabilities to freeholder, leaseholder and – potentially – tenants with lifetime secure tenure.
[ix] The minimum share in equity for a tenant/part-owner to be entitled to HA is a political decision.
[x] Schedule A of income tax was, until about 1960, based on the notional rental income deriving from owner-occupation of the taxpayer’s residence.
[xi] Some property investment managers revalue their portfolios every month. It is not the buildings that change in value so rapidly but the sites they stand on.