The Immortal Exception — Statecraft

Until 1992 the Dutch tax authorities levied no tax on beer but on wort, the unfermented intermediate product of the brewing process. Small breweries lost relatively more of that semi-finished product in production than large ones, and to compensate for this they were given a lower rate. The rationale was clear and the compensation justified: at the wort stage, the disadvantage of scale weighed more heavily on the small brewer. That disadvantage was attached to the wort stage. In 1992, under a European directive, the levy switched from wort to the finished product, and with that the ground beneath the lower rate fell away: the difference in production loss belongs to the unfermented intermediate and no longer counts in a levy on the beer. The disadvantage disappeared with the shift of the point of levy. The lower rate remained.

As early as 2008 an evaluation by the Ministry of Finance established that the disadvantage the scheme was meant to offset had disappeared, and the state secretary announced its abolition. That proposal never came. The Rutte III cabinet (2017–2022) tried again and wanted the advantage to lapse from 2023. In 2021 that abolition was voted down, after a petition with thousands of signatures, a letter of alarm and, according to NRC, a lobby by the brewers, who send every member of parliament and every member of the cabinet a package of beer once per cabinet term. The amendment that took the abolition off the table was submitted by Eelco Heinen, then a member of the House of Representatives for the conservative-liberal VVD, and co-signed by the Christian-democratic CDA, the ChristenUnie, D66 and the farmers’ party BBB. That same Eelco Heinen is now Minister of Finance, and thereby ultimately responsible for the very system the cabinet wants to simplify, one of whose smallest ineffective components he personally kept standing. A new evaluation, from December 2025, called the purpose of the scheme neither clear nor unambiguous. On Prinsjesdag 2026, Budget Day, Heinen as minister nevertheless proposed abolishing the scheme from 2028. Whether it survives this attempt too, parliament decides this autumn.

The small brewers’ scheme — the kleinebrouwersregeling — involves roughly two million euros a year. On the scale of the national budget, that is negligible. And precisely for that reason the scheme is so illuminating. If an exception of two million, whose own ministry has known since 2008 that its ground has gone, survives the combined pressure of a negative evaluation, a cabinet decision and a clean-up ambition, then something is at work that is bigger than beer. Then the question is why for twenty-five years — since the last major overhaul of 2001, which gave the Dutch income tax its current structure of three schedular “boxes” — it has proved impossible to change the tax system in any essential way.

Five causes, or one pattern

In her analysis in the Dutch daily NRC, Eline Nugteren names five causes why thorough tax reform keeps running aground. The special interests that organise themselves, the absence of money for compensation, the tendency to make the system more complicated instead, the fear of new execution debacles at an exhausted Tax and Customs Administration — the Belastingdienst, the national tax authority — and the absence of a well-matched duo at the Ministry of Finance, the minister and state secretary who must lead a reform together. Each of the five is correct. But they are not five causes. They are one pattern that shows itself in five places, and the pattern is familiar from every other file in which the Dutch state runs aground in its own structure.

The brewers’ scheme is the purest form of what elsewhere in this corpus is called the congealed outcome: a decision once taken for a reason remains standing after the reason has disappeared, and then begins to function as a given that nobody questions any more. The scheme compensated a disadvantage that has not affected the levy since 1992. What remains is a rate without a ground, which maintains itself because abolishing it costs political capital and leaving it standing costs nothing. The name bears the weight the case no longer can. It is still called a small brewers’ scheme, as if a small brewer’s interest were being protected, while what is in fact being protected is a historical residue whose original logic has dried up. Box 3 — the Dutch capital-income regime that taxes an assumed rather than actual return on savings and investments — has been the same phenomenon on a large scale: a notional yield that presupposed a reality that no longer existed, and that remained standing until the courts demolished it, the Supreme Court of the Netherlands striking the regime down in its Christmas judgment of December 2021 and setting off a recovery operation that runs into the billions. The difference between the brewers’ scheme and box 3 is only the scale, and the question whether a judge comes along.

The second manifestation was formulated sharply by Klaas Knot, the former president of De Nederlandsche Bank, the Dutch central bank, who points out in the article that special interests organise en masse while the general interest is felt more weakly, and that those who would profit from simplification barely feel that benefit and therefore organise poorly. That is the logic of collective action, and it is the political variant of a pattern that recurs everywhere in public execution: the measurable and organised crowds out the essential and unorganised. Every tax exception, once it exists, acquires a concentrated circle of beneficiaries who defend it. The benefits are sharply assigned; the costs are smeared out across everyone. The brewer feels his two cents per glass; the taxpayer does not feel his share of the two million. Simplification has no constituency, because its benefits are diffuse and its losers concrete. The agricultural exemption, which keeps increases in the value of farmland out of the tax and costs well over a billion a year, and the business succession scheme, which largely exempts the transfer of family firms from inheritance and gift tax, are the same structure writ large: ineffective according to the ministry, untouchable thanks to the lobby. The mortgage-interest deduction — which the civil-service report of 2023 doubts ever increased home ownership, and which drives up house prices and mortgage debt — still stands because one party has made it a political totem and abolition would mean losing face. The pattern is the same every time: an instrument that has lost its function but found its defender.

The third manifestation touches politics itself, and it is the most sobering. Jasper H. van Dijk of the Institute for Public Economics (Instituut voor Publieke Economie) observes that a tax measure can be adjusted quickly and is therefore suited to showing that something is happening, while the real solutions — the power grid, education, the roads and bridges — are slow and expensive. That is the countability trap at the level of government. The tax code is filled up with symbolic measures because they are visible and immediate, while the structural investment remains invisible until it pays off years later. The innovation box and the WBSO — respectively a reduced corporate-tax rate on profits attributed to innovation and a payroll-tax credit for research and development, together good for some five billion euros a year in tax reductions — produce no demonstrable growth in innovation, according to the Netherlands Court of Audit, and of the innovation box, according to calculations by the financial daily Het Financieele Dagblad, more than ninety per cent of the benefit ends up with three companies: ASML, MSD and Booking. An instrument erected to stimulate innovation broadly turns out in practice to be a discount for a handful of giants. The measurable gesture — we are doing something about innovation — crowds out the question of whether the instrument achieves its stated goal, and the beneficiary who remains then defends the scheme against correction.

The figure nobody would invent

Anyone inclined to dismiss base broadening — closing exemptions and deductions so that more of what is earned is actually taxed — as an ideological wish of those who want to strip the state bare must face a few numbers, and they are civil-service numbers. In 2023 the Ministry of Finance counted 116 tax schemes — the Dutch umbrella term for deductions, exemptions, reduced rates and other special provisions — together worth some one hundred and fifty billion euros a year. Of the 73 that had been evaluated, eleven proved effective and efficient; twenty-one, together worth thirty billion, were assessed negatively, and for forty-one of them the ministry could not establish whether they work. An updated overview followed in June 2026: of 124 evaluated schemes, 55 score negatively on at least one criterion, because they miss their goal, because the goal could be reached more cheaply, or because they are too complex for the tax authority or the taxpayer. Together those 55 represent some ninety billion euros. It is the tax authority’s own bookkeeping. The metres-high stack of studies, advisory reports and negative evaluations the article speaks of is the system’s documented knowledge of its own dysfunction.

And that is the point where the five manifestations converge into one diagnosis. The system knows what is wrong. It has known it in detail, in its own reports, for years. And it cannot convert that knowledge into a correction of its own behaviour. That is the classic condition of an organisation that cannot use its own knowledge to steer itself, and the tax system is merely the most visible place where it shows. The scheme that outlives its own rationale is the symptom from which the disease can be read: a system that accretes because every exception acquires an interest, and that does not shrink because clearing away an exception always strikes a concrete loser and never produces a concrete winner.

Cleaning up or designing

Here the question splits, and the split determines everything. One can try to clear away the existing follies, scheme by scheme — which is precisely what has failed for twenty-five years, because every individual scheme has its individual defender. Or one can put the question one level deeper and ask which architecture causes exceptions to keep forming. That is the difference between the mop and the tap. A system built up out of deductions, rates per category and exceptions to the exception offers every special interest a hook to fasten onto. Every bracket, every deduction, every exemption is a point of leverage for the lobby and a place where the congealed outcome can lodge itself. A system without those hooks cannot degenerate in this way — because there is nothing to rivet an exception to without everyone seeing that the principle is being abandoned.

That is an architectural observation, and it has been worked out elsewhere as a thought experiment in the form of a source-neutral system that takes steering and exception out of taxation and brings the levy back to one visible rate on visible events. The merit of such a design lies in the property that it removes the condition under which new mess arises. What this piece is about is not that design, which stands on its own and justifies itself. What this piece is about is the observation that precedes it: that the problem of the Dutch tax system is not the individual folly, but the mechanism that produces and protects follies, and that simplification will therefore never succeed as long as it is conceived as a cleaning operation instead of a redesign.

But with that, the hardest question has not yet been asked, only displaced. For suppose the design exists, and suppose it is good. There remains the fact that consensus on the diagnosis has existed for twenty-five years, that virtually every party puts simplification in its manifesto, that it is written into the coalition agreement of the current cabinet, and that nothing happens nonetheless. A system that agrees on what must change and still does not change it has an execution problem, and that is a question of public administration of the first order. The beer package arrives again this autumn, now with a proposal for abolition on the table. Why the reform everyone agrees on does not come is the subject of the second part.

This is Part I of a diptych. Part II, “Consensus Without a Deed”, moves the question from the tax system to government itself: why a system that agrees on a reform cannot carry it out.

Accountability and sources

Occasion and principal source: Eline Nugteren, “Waarom grondige belastinghervormingen al een kwarteeuw een dode letter zijn en de schatkist tientallen miljarden misloopt” (“Why thorough tax reforms have been a dead letter for a quarter of a century and the treasury misses out on tens of billions”), NRC, 7 August 2026. Figures drawn from the article cited and from the sources it cites, including the Netherlands Court of Audit and Het Financieele Dagblad. Primary sources: Ministry of Finance, civil-service report Aanpak fiscale regelingen (30 June 2023); letter to parliament from the State Secretary for Finance on negatively assessed tax schemes (23 June 2026); the evaluations of the small-brewery scheme (Parliamentary Papers II 2007/08, 31 200 IXB, no. 18, and December 2025); amendment Heinen et al., Parliamentary Papers II 35 918, no. 8 (2021); SEO, evaluation of the agricultural exemption (2024). Facts checked against primary sources as of 26 September 2026. All facts are paraphrased.

Jacob Huibers is an interim manager with more than twenty years of experience in the Dutch public sector. He has worked as cluster manager, cluster director and kwartiermaker — brought in to stand up new organisations — at municipalities ranging from fifty thousand to over two hundred thousand inhabitants, and at inter-municipal collaborative bodies across the social and physical domains. Statecraft is his platform for strategic reflection on public-sector execution, pillar IV of House of Viridian.

Responses and counter-arguments via Statecraft.nl.