The Gardener's Tithe

On income tax, its false promise of fairness, and why the only honest alternative is to tax the earth itself

Consider a person who plants vegetables. She has prepared the soil, purchased the seeds, tended the shoots through spring, and pulled the weeds by hand on the long evenings of July. In September she harvests. Before she can eat, the state arrives and takes a portion — not of anything she found lying about, not of any windfall, not of any value that accrued to her through the passage of time or the exertion of others, but of the direct material product of her own effort and her own time. This is income tax. It is, at its moral foundation, a levy on the act of living productively — a tithe extracted not from surplus or luck or inheritance but from the most basic expression of human agency: the decision to work, and to produce something where before there was nothing.

The objection will be raised immediately that the state provides services in return — roads, courts, hospitals, schools, the defence of the borders within which the garden exists. This is true. The exchange is real. But it does not answer the moral question, which is not whether the state should be funded, but whether the particular mechanism chosen to fund it is the fairest one available. The vegetable grower pays her tithe. The person who owns the field next to her garden and does nothing with it but wait for its value to rise as the town expands around it — as the roads the state builds with her tax money improve its access, as the schools funded with her tax money make the neighbourhood more desirable, as the community her neighbours build with their collective labour increases the attractiveness of every piece of ground within it — pays almost nothing in proportion to what she receives. The gardener is taxed on what she makes. The idle landowner is taxed, barely, on what he holds. This asymmetry is not incidental to the income tax system. It is structural to it. And it has been structural to it since William Pitt the Younger invented the thing in 1799 to pay for a war.

The history of income tax in Britain is, in the first instance, a history of embarrassment. It was introduced as a temporary measure — the language is exact: temporary, exceptional, necessitated by the emergency of the French Revolutionary Wars — at a rate of two shillings in the pound on incomes above sixty pounds per year.¹ Pitt understood that what he was doing was politically novel and morally contested: the idea that the state had a claim on a person's earnings, rather than merely on their property or their transactions, was not an established principle but an improvisation under pressure. The tax raised approximately six million pounds in its first year, which was enough to make it indispensable and enough to make it politically toxic, and these two facts existed in permanent tension for the following century.

Napoleon was defeated at Waterloo in 1815. The income tax was abolished the following year, in 1816, after a parliamentary vote that reflected genuine public revulsion at the idea of its continuation in peacetime. The revulsion was so thorough that Parliament voted to destroy the records — the tax returns, the assessments, the documentation of what had been collected and from whom — rather than allow them to remain available for future governments who might be tempted to revive the machinery. The records were burned. The tax was gone. And the principle that the state should not levy a charge on the act of earning was restored, temporarily and imperfectly, until Sir Robert Peel, facing a budget deficit and the political consequences of abolishing the Corn Laws, reintroduced it in 1842.²

Peel's reintroduction was also temporary, in the sense that he said it was temporary and that nobody believed him. The tax was levied at sevenpence in the pound on incomes above one hundred and fifty pounds — a rate of less than three percent, applying to a small fraction of the population, designed explicitly as a measure that would be phased out as the fiscal situation improved. It was never phased out. It has been with us, continuously, for a hundred and eighty-three years, growing in rate, in scope, in complexity, and in the proportion of the population it touches, until it is now the single largest source of revenue for the British government, raising approximately two hundred and fifty billion pounds per year, and affecting every adult in employment from the moment their earnings cross the personal allowance threshold — a threshold that, in the fiscal years from 2021 to 2028, has been frozen in cash terms while inflation has eaten its real value, pulling millions of people into the basic rate and hundreds of thousands into the higher rate who were never intended to be there.³

The fiscal history of the rate is its own kind of argument. In the First World War, the standard rate rose to six shillings in the pound — thirty percent — and a supertax on the highest incomes pushed the combined marginal rate above fifty percent for the first time. By 1941, the standard rate was ten shillings in the pound — fifty percent — and the top marginal rate had reached ninety-seven and a half percent, a figure that must be read twice to be believed, representing the state's claim to all but two and a half pence in every pound earned above a threshold which, at the time, the country's most successful entertainers, business owners, and professional people crossed without great difficulty.⁴ The post-war settlement reduced these rates from their wartime peaks, but the top rate remained above eighty percent through the 1960s and into the late 1970s, when Denis Healey's top rate of eighty-three percent on earned income — and ninety-eight percent on investment income, which was a separate and even more spectacular confiscation — produced the well-documented phenomenon of the "brain drain," the emigration of productive people to lower-tax jurisdictions, and the equally well-documented phenomenon of elaborate tax avoidance schemes that the top rates made economically rational for anyone capable of affording the lawyers to design them. Margaret Thatcher reduced the top rate to sixty percent in 1979 and to forty percent in 1988, where it remained until Alistair Darling introduced a fifty percent rate on incomes above one hundred and fifty thousand pounds in 2010, in the final months of the Brown government, at which point the rate had been through more than a century of improvisation without any of its architects ever having explained, coherently and with evidence, what level was correct and why.

The standard political defence of income tax is that it is progressive: those with higher incomes pay a higher rate, and this redistribution is a feature rather than a bug. The defence is largely false, and the ways in which it is false are not minor technical qualifications but structural distortions that run through the entire system from bottom to top.

Begin at the bottom. The personal allowance — currently twelve thousand five hundred and seventy pounds — removes the lowest earners from income tax entirely, which is genuinely progressive so far as it goes. But income tax does not stand alone. National Insurance contributions run alongside it and are not progressive in the same way: employees pay twelve percent on earnings between the Primary Threshold and the Upper Earnings Limit, and then two percent on everything above that limit.⁵ The person earning forty thousand pounds pays twelve percent National Insurance on most of their employed income. The person earning two hundred thousand pounds pays twelve percent on the lower band and two percent on everything above it, producing a combined National Insurance rate that falls as a proportion of total income as income rises. This is not progressive. It is the opposite of progressive. It is regressive from the point at which the Upper Earnings Limit is crossed, and the Upper Earnings Limit has been frozen — like the personal allowance, like most of the thresholds in the system — while earnings have continued, in nominal terms, to grow.

Move upward to the middle of the distribution, where a different and more operationally bizarre distortion operates. The personal allowance is tapered away for incomes above one hundred thousand pounds: for every two pounds earned above one hundred thousand, one pound of the personal allowance is removed, until the allowance has been entirely withdrawn at approximately one hundred and twenty-five thousand pounds. The effect of this taper is to create an effective marginal income tax rate of sixty percent on income between one hundred thousand and one hundred and twenty-five thousand pounds — higher than the forty-five percent top rate that applies on incomes above one hundred and twenty-five thousand, higher than any rate the Conservatives applied between 1988 and 2010, and almost certainly the result of nobody having designed it at all but of two policy decisions — the introduction of a higher personal allowance and the introduction of the taper — that happened to interact in a way that nobody at the Treasury thought through before the combination was legislated.⁶ The sixty percent band is not announced as policy. It is not defended as policy. It simply exists, as a consequence of the accumulated interaction of individual decisions made by people who were thinking about one thing at a time.

Move further down the income distribution, to the households receiving Universal Credit, and the effective marginal rate problem becomes not sixty percent but something closer to eighty. Universal Credit is withdrawn at sixty-three pence for every pound of net earnings above the work allowance. On top of this, the claimant pays income tax at twenty percent and National Insurance at twelve percent, meaning that their combined effective marginal rate — the proportion of each additional pound earned that does not remain in their pocket — is approximately seventy-five percent on earned income in the relevant range, and can be higher in households with particular combinations of benefits and circumstances.⁷ The person on the lowest rungs of the income distribution faces a marginal tax rate that would have been politically unacceptable if applied to the highest earners in any decade of the twentieth century. This is the progressive income tax system in its actual, operational form: at its steepest not at the top, where the wealthy have the advice and the structures to manage their exposure, but in the middle and the lower-middle, where the combination of income tax, National Insurance, and benefit withdrawal produces effective rates that make additional work economically irrational for a significant fraction of the people the system is supposed to support. As the essay The Slow Confiscation establishes, the household does not exist in financial isolation from these structures; it is shaped by them, pressured by them, and in many cases broken by the perverse incentives they collectively produce.

And then there is the most fundamental asymmetry of all, the one that sits above every technical distortion and which cannot be resolved by adjusting thresholds or rates: the systematic preference of the tax system for capital over labour. Income from employment is taxed at twenty, forty, or forty-five percent, plus National Insurance. Income from capital gains is taxed at eighteen or twenty-four percent on most assets, and at ten or eighteen percent on business assets qualifying for Business Asset Disposal Relief. Dividend income is taxed at eight and three-quarter, thirty-three and three-quarter, or thirty-nine and a half percent, but without any National Insurance charge. The person who earns their income by working — by producing vegetables, in the metaphor this essay has already established — pays a higher rate than the person who earns the same income from assets they hold, simply by holding them.⁸ The person who earned their money through work and has already paid income tax on it, then invested it and earned a return, will pay a lower rate on that return than they paid on the original income. The incentive of the system, considered as a whole, is to convert labour income into capital income by any available means — through incorporation, through salary sacrifice, through the use of trusts and family investment companies and the thousand other structures that the accountancy profession exists to construct and maintain. The conversion is economically rational and legally available. It is not available to the vegetable gardener who does not have surplus income to invest, cannot afford the accountant who would design the structure, and has no option but to receive her income in the taxable form in which labour income necessarily arrives.

Governments have understood for a long time that the income tax system is a tangle. Their response has been to attempt individual repairs, each of which creates new distortions, which require further repairs, which create further distortions, in a process that has been running continuously since 1842 and which has produced a tax code of extraordinary complexity, running to tens of thousands of pages, comprehensible in its entirety to approximately nobody, and navigable only by those wealthy enough to retain the specialists who have made it their life's work to understand the passages the Treasury has not yet blocked.

The fifty pence top rate, introduced in the final budget of the Brown government in 2010 and applying to income above one hundred and fifty thousand pounds, is a representative example of this dynamic. It was introduced to raise revenue and to signal, in the political circumstances of an approaching election and a financial crisis, that the burden of fiscal adjustment would be shared across the income distribution. HMRC's own analysis, published when the rate was subsequently reduced to forty-five percent by George Osborne in 2013, found that the fifty pence rate had raised substantially less revenue than projected, because a significant portion of the income that would have fallen within its scope had been accelerated — realised before the rate took effect — or deferred, restructured, or shifted offshore by the people best placed to do any of those things: the very people the rate was designed to tax.⁹ The rate existed for three years and raised, at most, a fraction of what its introduction was claimed to justify. Its reduction was resisted on the grounds that this would be a tax cut for the rich, which was true as a description and irrelevant as an objection, since the rate it replaced had not succeeded in taxing the rich to any significant degree. The episode illustrates, with particular clarity, the structural problem with income tax as applied to the wealthy: income is a legal category, not a physical substance, and the wealthy have access to a permanent staff of people whose professional purpose is to ensure that the substance accumulates in the category that attracts the lowest rate.

The High Income Child Benefit Charge — introduced in 2013, modified repeatedly since, and still generating confusion and underpayment — is a different kind of illustration, showing the perverse effects of attaching benefit tapering to income rather than to household income. A household with two earners each receiving forty-nine thousand pounds per year — a combined household income of ninety-eight thousand pounds — retains its full child benefit entitlement. A household with one earner receiving eighty thousand pounds per year — a lower combined income — loses a portion of its child benefit, and at ninety thousand pounds loses it entirely. The mechanism is not a coincidence of competing policies but a deliberate design choice, driven by the administrative difficulty of assessing household rather than individual income. The effect is to impose a very high effective marginal rate on the higher earner in a single-income household — one of whom is, statistically, likely to be the person who reduced their hours or left employment to care for children — at precisely the income levels where the decision to return to full-time work or to take on additional hours is being weighed.¹⁰ The incentive runs directly against the outcome the design claims to seek. This is not a design failure in the exceptional sense. It is the normal condition of a system that has been patched, adjusted, and extended for nearly two centuries without any moment at which the whole was reconsidered from its foundations. As the The Ceremony of Not Knowing describes in a different context, the system's complexity has become a form of institutional self-protection: too intricate to reform without pain, too embedded to abandon, tended by professions whose existence depends on its continuation.

There is an alternative. It has been available, intellectually, since 1879, when Henry George published Progress and Poverty, the most widely read economics text of the nineteenth century, which sold millions of copies across the English-speaking world and prompted more political movements, more reform campaigns, and more serious economic engagement than any work of economic theory before Keynes.¹¹ George's central argument was simple, and it has not been refuted: the value of land is not created by the person who owns it. It is created by the community that surrounds it — by the roads the state builds, the schools the taxes fund, the workers whose labour makes the neighbourhood productive, the businesses whose presence makes the location desirable, the population growth that makes any given square metre more contested and therefore more expensive. The landowner, who has done none of these things and whose contribution to the land's value is in most cases precisely nothing, captures this value through the simple mechanism of title. The worker, who has created value through her own effort, is taxed on the value she creates. The landowner, who has merely occupied the space in which others have been busy, is taxed at the modest rate of council tax — a figure that bears essentially no relationship to the actual value of the land, is assessed on an estimated property value from 1991, and has not been comprehensively revalued in England in the thirty-four years since.¹²

The land value tax that George proposed — and that economists from across the political spectrum, from Milton Friedman to Joseph Stiglitz, have subsequently endorsed — is a levy on the unimproved value of land: not on the buildings constructed on it, not on the income produced from it, not on the transactions made with it, but on the site value alone. Its theoretical properties are remarkable. Because the supply of land is fixed — no amount of taxation produces less land, as it might produce less labour or less capital — a land value tax creates no deadweight loss, the economic term for the productive activity that other taxes deter. A person taxed on their income has a reason to earn less; a person taxed on their land does not have a reason to own less land, because the tax is owed regardless of what the land is doing. The tax therefore falls without distortion on the holder, and, crucially, incentivises use: a landowner sitting on an undeveloped site in the centre of a city, banking on its increasing value while contributing nothing to the community that generates that value, would face the same annual tax bill as a landowner who developed the site and put it to productive use. The idle hold becomes economically costly. The speculative vacancy becomes financially unattractive. This matters in Britain in 2025 not as a theoretical consideration but as an urgent practical one: land value speculation, the holding of sites in expectation of planning permission and rising values without any requirement to develop them, is among the principal structural causes of the housing shortage. The tax system as currently designed actively rewards this behaviour. A land value tax would actively penalise it.

Land cannot be moved offshore. It cannot be restructured, incorporated, or placed in a trust in a jurisdiction with a lower rate. The complex avoidance industry that makes the income tax system so porous at the upper end — the accountancy firms, the tax counsel, the family investment companies, the salary-into-dividend structures — has nothing to offer the land value tax. The site is there. Its value can be assessed. The assessment can be updated regularly. The bill can be presented to whoever holds the title. The entire apparatus of complexity through which The Ledger That Forgot Its Name describes the financial system's capacity to make obligations disappear simply does not apply. The land stays. The tax follows it.

The political history of land value taxation in Britain is, for this reason, also a history of the people who prevented it. Lloyd George's People's Budget of 1909 — the budget that triggered the constitutional crisis of 1909 to 1911, the Parliament Acts, and the permanent curtailment of the House of Lords' power to block financial legislation — included, alongside its provisions for old age pensions and national insurance, a tax on the increment in land values and a duty on undeveloped land.¹³ The House of Lords rejected the budget, an act without precedent in two hundred and fifty years, and the constitutional crisis that followed was resolved in the Crown's favour only after two general elections, a threat to create sufficient Liberal peers to pass the legislation, and the Lords' eventual capitulation. The land tax provisions were subsequently repealed in 1920, before they had raised meaningful revenue, by a Parliament containing rather more landowners than it contained vegetable growers. The pattern has repeated. A land value tax was introduced in the Finance Act 1931, abolished in 1934. It was proposed by the Liberal Democrats in various manifesto iterations; by the Scottish Green Party; by a significant body of academic and think-tank opinion cutting across conventional left-right lines. It has never been implemented in England in any serious form, and the reason is not technical — there is no technical barrier to assessment; countries from Estonia to Singapore to various municipalities in Pennsylvania operate variants of the system with considerable success — but political: the people most exposed to a land value tax are, disproportionately, the people with the greatest political influence, the most available lawyers, and the longest institutional memory for the mechanisms by which radical land reform has been defeated before.¹⁴

The argument is sometimes made that income tax, whatever its flaws, is at least familiar — that the population understands it, has arranged its affairs around it, and would suffer disruption from any reform significant enough to matter. The argument is the argument of every established system against every improvement: the cost of change is visible and attributable; the cost of the status quo is distributed, invisible, and borne by people without the vocabulary or the political organisation to quantify what they are losing. As The Agreement That Wasn't explores in a different context, the social contract implied by a system and the social contract actually delivered by it are frequently not the same document, and the gap between them tends to be largest for the people least equipped to litigate the discrepancy.

The income tax system tells the public that it is progressive — that it takes more, in proportion, from those who have more. This is partially true in the middle of the distribution and false at both ends: false at the bottom, where the combination of benefit withdrawal and National Insurance produces marginal rates that would be politically unacceptable if described plainly; false at the top, where the availability of structuring options produces effective rates far below the headline figures; and structurally false throughout, in the fundamental sense that it taxes the creation of value rather than the unearned capture of it. The landowner who has watched the site value of his holdings increase by the efforts of the community around him — whose land is worth more because the state built a railway station nearby, because the local authority improved the schools, because the workers whose income tax funded all of this have created a neighbourhood worth living in — contributes, through the current system, a fragment of what justice would require. He is not a villain in the conventional sense. He is the beneficiary of a design, and the design is old enough and complex enough and politically protected enough that most people who might object to it have ceased to notice that it could have been any other way.

The vegetable grower is still being taxed. The field next door is still appreciating. The tithe on the gardener funds, in part, the infrastructure that makes the field more valuable, which accrues entirely to the person who owns it and does nothing with it. This is the moral structure of British taxation in 2025. It was the moral structure of British taxation in 1842. The governments in between have adjusted the rate, altered the thresholds, introduced reliefs and withdrawn them, created traps they did not intend and defended distortions they could not justify, and through all of it have left the fundamental architecture untouched — the architecture that taxes the act of producing and leaves the fact of holding largely alone. The argument for land value tax is not, at its foundation, a technical argument about deadweight loss and incentive structures, though the technical arguments are overwhelming. It is a moral argument about who made what and who owes what to whom. And that argument has been available, clearly stated, for a hundred and forty-six years. The failure to act on it is not a failure of knowledge. As with the The High Premise on which the institutions of democratic governance rest, the gap between what is claimed and what is delivered is not the product of ignorance but of interest — and the interests that benefit from the gap are the same interests that have been most reliably present whenever the gap was about to close.