Land Value Tax in Practice - Grounds for Cautious Optimism?

In responding to the report, I would like first to identify points on which I think Neidle’s analysis is important and correct, or at least worth considering; second, to identify aspects of the analysis which I think are problematic or could be strengthened; and finally, to make some (very tentative) suggestions for addressing some of the questions raised in the report. My aim is to provide some constructive feedback from someone who has some familiarity with certain aspects of the subject matter of the report, having reflected on the topic of LVT and its implementation in the UK for a number of years, in the hope that I might be able to contribute in some small way to the ongoing debate.

Points of (partial) agreement

While the report is generally sympathetic to the idea of LVT and presents a solid case in favour of introducing it as a replacement for one or both of the Council Tax and Stamp Duty Land Tax (SDLT), it also raises a number of difficult questions relating to the implementation of LVT in England. For some proponents of LVT, the analysis is likely to be a sobering read. Nevertheless, I think Neidle is right to emphasise that LVT enthusiasts need to be prepared to accept imperfect solutions and compromises. While the principled case for LVT may be compelling – perhaps even overwhelming – the overall case for reform is only as strong as the detailed solutions to the problems surrounding its practical implementation. Sobering though it may be, Neidle’s analysis is a timely contribution to the debate and should be welcomed even by those who find its arguments and conclusions less than fully convincing.

A more specific point of agreement concerns the raft of mechanisms proposed by Neidle to mitigate some of the seemingly unfair and politically contentious effects that would accompany a sudden transition from the current system of residential property taxation to a pure, unadulterated LVT. An immediate transition from one system to the other would lead to apparent double taxation for those who would already have paid a hefty SDLT bill, having suffered the misfortune of acquiring their property just before the introduction of the LVT and the abolition of SDLT. While one might (as Neidle himself acknowledges) question how unfair this really is, given the partial capitalisation of SDLT into the purchase price, Neidle is surely right to insist that for many people the double-taxation issue is likely intuitively to seem grossly unfair, and would certainly be politically highly problematic. His suggestion that recently paid SDLT be credited against tax increases that result from the switch to LVT should therefore be given serious consideration.

When it comes to the very substantial increases in tax burdens that would follow from a sudden transition to LVT, one of Neidle’s suggestions is simply that the tax be phased in over a 10-year period, with 10% more of council tax and stamp duty abolished every year and 10% more of LVT applying. It is hard to disagree with Neidle on this point, given some of the eye-wateringly large - and politically explosive - increases that property owners in high value areas would face following the introduction of LVT, which would far exceed the kinds of amounts normally considered by tax policy experts to be within the range of acceptable losses.

On the other hand, one must also acknowledge the trade-offs that a lengthy phase-in period will inevitably entail. The current inefficiencies and injustices that derive from the dysfunctional Council Tax and SDLT will take that much longer to remedy. Perhaps more worryingly, a 10-year transition period would make it that much easier for the opponents of LVT to reverse the policy. A sudden, radical change might be harder to overturn. However, a sudden, radical switch to LVT would be profoundly risky for all sorts of reasons, and would only be justifiable and feasible if the country found itself in such a desperate state (in the aftermath of a financial crisis, for example) that it would be even more risky to stick with the status quo. In the meantime, a gradual, but steady and, above all, patient, roll-out of LVT seems like the only reasonable option.

But Neidle worries that the lengthy phasing-in of LVT would not be enough to ‘soften the blow’ on people who would face hefty increases in their tax bills. Apart from the questionable fairness of imposing an enormous tax increase on people who have recently bought at great personal cost, the huge numbers that apply in prime London and the commuter belt can spark hesitation in the minds of people who would otherwise be sympathetic to LVT. For this reason, Neidle proposes that rather than having one LVT rate that applies across the country, the tax should be made partially regional, with different rates applying in different regions. This would reduce the amount of redistribution from London and the South East to the rest of the country, which would make the differences between LVT charges faced by people in different regions less extreme.

While I would hesitate to identify the idea of a partially regionalised LVT as a point of agreement, it is certainly an idea I would not want to reject outright. My hunch would be that the lengthy phase-in period would suffice to soften the blow of hefty LVT charges, particularly if a way can be found to reduce the required LVT rate down from the 1.46% rate2 needed to replace Council Tax and residential SDLT (about which I have more to say below). Nevertheless, a partially regional LVT is an idea about which I am inclined to be open-minded.

An easier and more obvious point of agreement is Neidle’s proposed solution to the classic problem case for the land value tax – the elderly person living in a large house without much income. The proposal to allow deferral of the excess of LVT over the previous CT charge will be familiar to anyone with more than a passing interest in LVT, and while questions concerning the specific design of such a policy are interesting and important, I won’t dwell on this issue any further than to register my agreement.

A final point of agreement: Neidle is surely right to suggest that although introducing LVT as a replacement only for SDLT would substantially reduce the tax increases faced by people in high value areas, doing so would probably not be viable. This is partly because the administrative costs of implementing LVT would not justify replacing the £11.5 billion of revenue currently generated from SDLT. It is also because unlike the abolition of Council Tax, from which more than 60% of people would benefit in immediate cash terms, nobody would benefit in this way from the abolition of SDLT. Lots of people’s bills would go up; none would go down. Replacing SDLT with LVT might be just as politically challenging as replacing Council Tax with LVT, albeit for different reasons.

Points of (partial) disagreement

Having identified the main points on which I would be inclined on the whole to agree with Neidle’s analysis and recommendations, I now turn to some aspects of his analysis which I feel could perhaps be strengthened. I begin, perhaps a little unfairly, with a quibble about Neidle’s answer to the question ‘Why a land value tax?’. On the face of it, the answer to this question seems extremely simple: The UK’s existing property taxes are ‘broken, unfair, and anti-growth’, as well as giving Britain comfortably the largest property tax as a percentage of GDP of any OECD country. Switching to LVT would be more equitable and efficient. Moreover,

‘taxing land is better for growth than taxing any other income or asset because normally when you tax something, you get less of it, but when you tax land, you do not. In fact, you incentivise more efficient use of the land’.

Why, exactly, are Britain’s existing property taxes broken, unfair, and anti-growth? Council tax is unfair because it is still based on property values from 1991, its top band is fixed at three times the bottom band, and its effective rate falls sharply as value rises (so that a very high-value Westminster home can pay roughly the same cash bill as an ordinary flat and less than a modest home in a high-tax northern authority). Business rates are inefficient and anti-growth because the tax falls partly on buildings and improvements (so that business investment can raise rateable value, leading to a higher tax burden), revaluations lag, relief thresholds produce cliffs, and empty-property relief has resulted in avoidance. Stamp duty is particularly inefficient because it reduces household mobility, results in inefficient use of land, and constrains economic growth.

This overview of the shortcomings of existing UK property taxes is perfectly fine as far as it goes. My quibble is that it fails to articulate what is arguably one of the most economically damaging consequences of the way the Council Tax is structured, which is simply that it leaves the bulk of the land value in the private sphere, capturing only a tiny fraction for public benefit. By over-taxing land in low value areas and under-taxing land in high value areas, the Council Tax simultaneously behaves like a wealth tax (falling almost entirely on the value of buildings) in the lowest value areas and a tax subsidy in the highest value areas (because the value of housing wealth is radically under-taxed relative to the value of other assets). The point is that this is not simply unfair; it is also highly distortionary because the favourable tax treatment of residential property reduces the user cost of capital for housing relative to other assets, which biases portfolios and credit toward housing and away from business capital.

We have reason to think that the favourable tax treatment of owner-occupied housing is a key contributing factor to the UK’s longstanding productivity problem, which places significant constraints on the economy’s capacity to generate sustained real wage growth, maintain fiscal resilience, and finance high‑quality public services. Replacing the Council Tax with LVT would be a way of partially reversing the tax bias in favour of residential property, thereby reducing the extent to which the allocation of financial investment is distorted by the tax system, so that tax-induced over-investment in real estate is less likely to crowd-out more productive business investment.

This is not really intended as a criticism of Neidle’s presentation of the case for LVT, and it would be unfair if it were, given that Neidle states explicitly that the Tax Policy Associates report is about the details surrounding the implementation of LVT, not about the case for LVT. Nevertheless, a comprehensive articulation of the benefits gained by switching from Council Tax to LVT would make navigating the very considerable technical and political difficulties that will accompany any serious attempt to implement such a reform seem more likely to be worth the time, effort, and risk. The practical prize to be won from replacing the Council Tax with LVT is much more economically significant than merely replacing an outdated relic with something far more equitable.

Why Revenue Neutrality?

A slightly sharper criticism of Neidle’s analysis would be that having made the point that taxing income from land makes more economic sense than taxing income from other sources, he then undermines this point somewhat (at least to my mind) by stating his strong preference for revenue-neutrality over the longer term: Once LVT has been implemented,

‘policymakers would have to choose between keeping neutrality and reducing the rate, or keeping the rate and raising revenue. It seems to me imperative that this choice is made up front and locked into the way the land value tax works (and my strong preference is neutrality)’.

The fact that the UK’s trio of bad property taxes already generate more revenue as a percentage of GDP than the property taxes of any other OECD nation might seem to mitigate against the idea of using LVT to raise additional revenue over and above the revenue currently generated by existing property taxes. One might very well argue that aiming for revenue-neutrality with the existing property tax regime is ambitious enough.

However, the amount of revenue raised by UK residential property taxes (Council Tax and the residential component of SDLT) is nothing like as much as the total revenue raised from property taxes, which include Business Rates and commercial SDLT. Indeed, US residential property taxation raises significantly more revenue as a percentage of GDP than the Council Tax and residential SDLT3. One might plausibly argue that that UK’s fiscal (and general economic) situation is sufficiently desperate to justify using a residential LVT to raise additional revenues, which could then be used to fund reductions in rates of more economically damaging forms of tax, such as income tax or NICs. It is not only ‘vaguely left-wing politicians’ (as Neidle puts it) who might be inclined to aim for better than revenue neutrality.

This is still a very mild criticism. The most important thing, of course, is that LVT replaces the existing property taxes. This, as I have already emphasised, would go a long way to reversing the tax bias in favour of residential property, as well as removing the distortions generated by SDLT. This is obviously the prize on which the supporters of LVT should focus their attention, and in the pursuit of which they should focus their energies.

….Or is it? Should the focus necessarily be on replacing the outdated and unfair Council Tax, as well as the economically damaging SDLT? Until very recently, my view had aligned with the recommendations made in numerous reports on the UK property tax system published since the IFS-sponsored Mirrlees Review of 2011 – that business rates should be replaced by a commercial LVT and that the Council Tax and SDLT should be replaced by either a residential LVT or some sort of proportional property tax. Even if I preferred to think of radical reform of the property tax system as more of a first step in the right direction, rather than the final destination, it seemed completely obvious that one wouldn’t introduce LVT without first getting rid of the existing nonsense.

But having reflected for a couple of weeks on the Tax Policy Associates report, I’m suddenly not so sure. The question of what to do about business rates is, for various reasons, a separate topic from the question of how to deal with the taxes on residential property, and can be set aside for now (as it is in the TPA report). SDLT is extremely damaging economically and is arguably the worst UK tax of all, as many economists and tax policy experts (including Neidle) have pointed out, so it simply has to go. This leaves the absurd, outdated relic that is the Council tax. Perhaps somewhat counterintuitively, it may be that as ridiculous as the Council Tax undoubtedly is, replacing it with LVT would be a mistake, given the huge tax increases for people in prime London this would entail, and the political backlash – not to mention risk of economic instability – that would inevitably follow. I agree that replacing only residential Stamp Duty would be a mistake, for the reasons highlighted in the TPA report: Everybody’s tax would go up (some very substantially), and nobody’s tax would go down – an absolute political nightmare. But what if LVT revenues were to replace part of the revenues currently generated by Council Tax, rather than the whole lot? Alternatively, what if LVT revenues were to replace revenues currently generated by some other economically harmful tax, like employer or employee NICs?

What I want to suggest is that there are many more options available to us than the two options considered in the TPA report. Let’s look in more detail at some of the possibilities. Since all options involve getting rid of residential SDLT, it is worth bearing in mind that this can be done immediately or gradually over a period of time (as modelled in the TPA report). Doing away with the tax at one fell swoop would make sense on purely economic grounds. Political considerations might mitigate in favour of phasing-out the tax over a period of 5-10 years.

Option 1: Replace residential SDLT and the portion of Council Tax related to the funding of social care

This option, which is similar4 to a proposal made in a recent article by Louis Haigh (who is now a member of Andy Burnham’s new cabinet), might be one of the more politically appealing policy options, since it would probably yield modest tax reductions for a large number of households in low value areas of the country (the new LVT + reduced CT bills would probably be a bit lower than current CT bills). The new tax bills would be somewhat higher in low value areas and lower in high value areas than if Council Tax were entirely replaced. (The policy could be tweaked by using LVT revenues to replace only the revenue currently generated by the Adult Social Care Precept, which would require a considerably lower rate). To make the proposed reform even more politically palatable, the new tax could be called a ‘Social Care Levy’, rather than a land value tax.

Option 2: Replace residential SDLT and reduce employer NICs

This option would involve using LVT revenues either to zero‑rate employer NICs for social care, or to reduce the rate of employer NICs in sectors with structurally low productivity and high labour intensity, such as hospitality, social care, and retail. At a time when labour‑intensive sectors are struggling with rising costs, chronic recruitment shortages, and structurally weak productivity growth, a plausible economic case could be made for a policy which might also attract a degree of support from the public.

Option 3: Replace residential SDLT, reduce CT, and reduce employer NICs

This option would seek to combine some of the benefits of the previous two options. If LVT revenues are set aside to replace the revenue currently generated by the Adult Social Care Precept, then sufficient revenues could be generated to fund modest reductions in employer NICs (or zero-rate employer NICs for social care) without requiring a problematically high rate of LVT.

Numerous additional options could of course be devised, including, for example, using LVT revenues to reduce income tax rates, raise income tax thresholds, reduce the rate of VAT, and so on. Some options would be more economically beneficial; others would confer more political advantages. The main point here is that it may be a mistake to focus exclusively on using LVT to replace both residential SDLT and the Council Tax. Other approaches might be both economically and politically more advantageous.

Land Value Tax or (Proportional) Property Tax?

My final comment on the TPA report concerns the question of the type of tax that should or should not be introduced. Neidle’s view is that LVT is strongly preferable to a Proportional Property Tax, although the latter is a plausible alternative that would be a significant improvement on Council Tax and SDLT. The main problem with the PPT is that

‘taxing the whole property also taxes extensions, rebuilding and renovation: improve the home and the tax base rises. So the PPT creates an incentive to not improve property. Or the flip side: it doesn’t create an incentive to improve property. We can put a number on this effect: a 0.48% annual tax is equivalent to imposing an approximately 16% up-front tax on the value created by home improvements.’

Another problem is that although property is frequently valued, most properties are not valued very often, and accurate property valuation ‘has to be done by reference to the actual features of the inside and outside of the property in question: floor area, extensions, age, condition, fittings and refurbishment’. This means that property valuation cannot be done accurately from the desktop, unlike land valuation, which ‘is driven more by plot, permitted use, location and surrounding amenities, which can be determined on an automated basis from public and private data sources’. This creates the following paradox:

‘Land valuation is rarely undertaken in practice, but in principle land across England can be valued from the desktop. Property valuation is commonplace in practice, but cannot accurately be undertaken from the desktop’.

This leads Neidle to think that the valuation problem is likely to be just as challenging in the case of the PPT as it is in the case of LVT, and may even turn out to be less of an obstacle for LVT.

The question that has arisen in my mind – and it is still very much a question at this stage – is whether the two forms of tax could be combined in some way that resolves the valuation paradox. In posing and elaborating on this question, I am taking a leaf out of of Tim Leunig’s Substack by by writing up an idea that may not be good, or even remotely sensible. After all, sometimes even a bad idea can be a springboard for others to think more clearly about an issue5....

To flesh this out, let me restate the valuation paradox:

Land valuation is accurate in theory but difficult to verify in practice, because while it can be done from the desktop (based on location, planning status, amenities, transport, neighbourhood characteristics, and so on) it cannot be checked against a real market transaction (because urban land rarely sells bare).

Property valuation is verifiable in practice but inaccurate in theory, because while it can be checked against real sales (“the house next door sold for £X”), it cannot be done accurately from the desktop (because interior quality, condition, refurbishment, fittings, and construction variation are invisible and subjective).

Might this dilemma be addressed by way of a hybrid tax comprising a market-based land valuation and a formula-based building assessment? Under this approach, only the land would be valued. Buildings would not be individually appraised at all. Instead, the tax base would include a statutory building charge determined by a simple published formula based on objective characteristics such as floor area, age and building type, deliberately ignoring differences in quality, fittings and other subjective attributes. The building component would therefore not represent a market valuation, but a policy-defined element of the tax system that is transparent, predictable and inexpensive to administer.

The principal advantage of this approach is that it confines valuation to the one component that genuinely requires it. By fixing the building component according to a published formula, all remaining differences between observed sale prices and assessed taxable values are concentrated in a single residual. Although that residual still reflects both land valuation errors and unmeasured differences in building quality, it is easier to analyse statistically than a system in which both land and building values are uncertain. If residuals exhibit consistent spatial patterns, they may indicate that the land value model requires adjustment; if they appear largely random, they are more plausibly attributed to unmeasured building characteristics.

The hybrid approach would retain many of the economic and administrative advantages of a land value tax while replacing the most subjective aspect of property valuation—the estimation of individual building values—with a simple, objective schedule. Whether this would improve the accuracy of land valuation is ultimately an empirical question, but it has the potential to make the valuation process more transparent and less dependent on subjective judgement than either a conventional property tax or a pure land value tax.

A further advantage of a land value tax with a statutory building charge is that it would moderate the geographical distribution of the tax burden. As the TPA model demonstrates, under a pure LVT, properties in high-value locations face huge tax increases, while otherwise similar properties in low-value areas pay comparatively little. Adding a uniform statutory building charge introduces a modest element that depends on the physical presence of a building rather than solely on land value. For a given level of revenue, this results in slightly higher tax liabilities in lower-value areas and slightly lower liabilities in higher-value areas than under a pure land value tax. Although the land component remains the dominant determinant of liability, this moderation may improve perceptions of fairness and political acceptability without reintroducing the subjective valuation of individual buildings.

I have no idea whether this idea is worth exploring, or indeed whether something like it already has been thoroughly analysed in the valuation literature. However, the serious point here is that I think proponents of LVT should remain flexible on the question of the precise form of tax that is to be introduced. Neidle’s hunch that LVT is likely to be easier to apply across England than a proportional property tax may well be a good one. But if some form of property tax turns out to be easier to implement or politically more feasible than a pure LVT, then so be it.

This completes my response to the TPA report. Feel free to add your own comments and suggestions.